iGlobusReport 17tex1

Design Report - iGlobus
submitted to
Indian Oil Corporation Limited
by
Solinas Integrity Pvt. Ltd.

IITM Incubation Cell, Chennai - 600113
Tamil Nadu, India
Summary
Design and development of a robot for inline inspection and conditional assessment
of oil and gas pipelines. The robot enables visualization of interiors of the pipeline
and also detects leakages at any point along the pipeline. The robot, also referred to
as iGlobus, is a spherical flow-propelled robot with no mechanical actuators. The pas-
sive locomotion of iGlobus is influenced by the flow characteristics of the fluid being
transported in that pipeline.
iGlobus carries a various set of sensors such as odometry sensors for localisation
and acoustic sensors for leak detection. These electronics are all integrated into the
robot with a custom-designed circuit board, that interfaces the sensors to an on-board
processing unit. This processor is programmed to ensure appropriate data acquisition,
noise removal and such pre-processing of the data, which is then stored into a memory
card/chip.
The data from memory card is later retrieved for detailed analysis through the soft-
ware developed. The software further processes the data for feature recognition and lo-
calization, thus bringing out the observations made by the sensors to report conclusions
on the defects and internal conditions of the pipeline. An interactive User-Interface is
also being developed for better visualisation of the inspection results and automation of
data processing using AI algorithms.
This report provides the overview of the design and analysis for the development of
iGlobus - Mechanical, Electrical & Electronics, and Software modules.
This report provides the overview of the use case for such a robot along with the
detailed design description.
Design Report i Solinas Integrity Pvt. Ltd.

CONTENTS
Contents
SUMMARY i
LIST OF TABLES iii
LIST OF FIGURES v
1. Introduction 1
1.1 Problem Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Technical Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Design Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Robot Design 4
2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 Detailed Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2.1 Mechanical Design . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2.2 Electronics Layout . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2.3 Software Layout . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.3 Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3. Test-bed Design 39
3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.2 Detailed Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.2.1 Mechanical Design . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.2.2 Sub-Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.2.3 Electronic Units . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.2.4 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . 56
3.3 Overall Test-bed Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.3.1 Testing Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . 58
ANNEXURE I 60
ANNEXURE II 65
Design Report ii Solinas Integrity Pvt. Ltd.

LIST OF TABLES
List of Tables
1 Stress Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2 Net Volume of Tank . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3 Acoustic Impedance of Common Materials . . . . . . . . . . . . . . . 59
4 Properties of Aluminium alloy 7075-T6 . . . . . . . . . . . . . . . . . 59
.ues
. . . . . . . . . . . . . . . .
5 Raspberry Pi 3 General Specifications . . . . . . . . . . . . . . . . . . 60
6 IMU General Specifications . . . . . . . . . . . . . . . . . . . . . . . 60
7 ISPK30I General Specifications . . . . . . . . . . . . . . . . . . . . . 60
8 PiCam General Specifications . . . . . . . . . . . . . . . . . . . . . . 61
9 MAX6675 General Specifications . . . . . . . . . . . . . . . . . . . . . 61
10 RTC (DS3231) General Specifications . . . . . . . . . . . . . . . . . . 61
11 M 20 EX SV General Specifications . . . . . . . . . . . . . . . . . . . 61
12 M 20 EX SV Operating Conditions . . . . . . . . . . . . . . . . . . . . 61
13 Buoyancy Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . 62
14 Technical Specifications of Pressure Transmitter . . . . . . . . . . . . . 65
15 Technical Specifications of Ultrasonic Flowmeter . . . . . . . . . . . . 65
16 Technical Specifications of Digital Turbine Flowmeter . . . . . . . . . 66
17 Technical Specifications of Temperature Transmitter . . . . . . . . . . 66
18 Max. Bending Stress and Max. Deflection Calculation. . . . . . . . . . . . . . . 67
19 Test Bed Properties... . . . . . . . . . . . . . . . . . . . . . . . . . . 68
20 Bill Of Materials for Test-Bed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
21 Bill Of Material Of Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Design Report 3 Solinas Integrity Pvt. Ltd.

LIST OF FIGURES
List of Figures
1 First And Third Design Concept of iGlobus . . . . . . . . . . . . . . . 7
. . . .
2 Threading And Snap fit Method of iGlobus . . . . . . . . . . . . . . . 8
. . . . . . . . . .
3 Cross Sectional View of Equivalent Stress Analysis of Shell . . . . . . 10
in SolidWorks 2019 Simulation
4 Cross Sectional View Of Total Deformation Analysis of Shell in . . . . . . 11
SolidWorks 2019 Simulation
5 Cross Sectional View Of Impact Analysis Of Shell . . . . . . . . . . . 12
6 Polyurethane Foam Layer . . . . . . . . . . . . . . . . . . . . . . . . . 12
7 Transitions in Fluid Pressure . . . . . . . . . . . . . . . . . . . . . . . 14
8 Transitions in Fluid Temperature . . . . . . . . . . . . . . . . . . . . . 14
9 Transitions in Fluid Velocity . . . . . . . . . . . . . . . . . . . . . . . 14
10 Raspberry Pi 3 PinOut diagram . . . . . . . . . . . . . . . . . . . . . . 18
11 IMU VN-100 IMU/AHRS . . . . . . . . . . . . . . . . . . . . . . . . 19
12 Acoustic Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
13 Microphone Circuit Layout . . . . . . . . . . . . . . . . . . . . . . . . 21
14 Camera Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
15 Electronics Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
16 Temperature Sensor: MAX 6675 + K Thermocouple . . . . . . . . . . . 23
17 Real Time Clock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
18 Saft M 20 EX SV Primary Li − MnO2 cell . . . . . . . . . . . . . . . . 25
19 Power Flow Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
20 Data Collection Diagram . . . . . . . . . . . . . . . . . . . . . . . . . 29
21 Software Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
22 Test-Bed Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
23 Elbow Joint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
24 Unequal (Left) and Equal (Right) Tee Joints . . . . . . . . . . . . . . . 41
25 Ball Valve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
26 PIG Launcher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

LIST OF FIGURES
27 Eccentric Reducer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
28 Flange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
29 Gasket . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
30 Supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
31 Centrifugal Pump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
32 Guides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
33 Baumer CTX Series Pressure Transducer (CTX 323.B26.0) Pressure
Transmitter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
34 Pressure Gauge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
35 Ultrasonic Flowmeter . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
36 Digital Turbine Flowmeter . . . . . . . . . . . . . . . . . . . . . . . . 55
37 Temperature Transmitter . . . . . . . . . . . . . . . . . . . . . . . . . 55
38 Data Acquisition Setup . . . . . . . . . . . . . . . . . . . . . . . . . . 56
39 Test-Bed Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

INTRODUCTION
Introduction
Pipelines in the modern industry are inevitable part of infrastructure. Pipelines play a
major role in transporting crude and refined petroleum, oil, natural gas and biofuels. It is
common that they cross populated areas. In case the pipeline is transporting hazardous
material, leak detection is necessary to reduce product loss and damage to environment.
Even small leaks can grow serious if left undetected. Therefore accurate leak detection
and localisation methods are of exigent need.
1.1 Problem Scope
Pipeline leaks can be caused by various reasons such as accidents, corrosion, bad main-
tenance, natural disasters, weld failures, acts of terrorists, etc. As flammable fluids may
explode when exposed to flame or sparks, even a small leak in the pipelines could result
in significant damage to environment and economy, apart from injuries and fatalities.
This could possibly be prevented by regular and watchful supervision of pipeline con-
dition and timely shutdown of pumping of fluids. Hence leak detection and localisation
methods play a vital role in maintaining the integrity of pipelines.
Exposure to any extreme or harsh conditions could weaken the integrity of the
pipeline structure. In many instances, structure of pipelines may undergo corrosion
or damage causing leakage. In some instances, pipelines may also develop blockages,
gas pockets, pitting, sedimentation, etc. Hence, many industries that rely heavily on
pipelines, such as petroleum, natural gas, manufacturing companies, and the like, could
face huge losses. More importantly, such scenarios could lead to discharge of industrial
effluents into nearby water and land resources, that may be potentially dangerous to the
environment. Therefore, continuous maintenance of pipelines remains a major concern
and and large set of resources are allocated for this purpose.
1.2 Technical Review
Tools for monitoring the condition of pipelines are in great in great regular demand
and are expensive. For instance, pipeline maintenance services in India are provided
by external agencies, which are expensive. Currently O&G pipelines are installed with
pressure sensors at regular intervals to identify the leakages. These pressure sensors are

INTRODUCTION
connected through SCADA systems which transmits the information about the pressure
in the pipelines in real time. Any leakage on the pipeline will reduce the pressure inside
the pipes which will be detected by these sensors. This system could detect any leakages
which is greater than 2% of the throughout. Any leakage smaller than this is difficult to
identify using the existing system. This threshold is maintained to prevent false alarms
due to pressure fluctuations inside the pipeline. iGlobus is being developed to address
this problem and identify smaller leaks and pilferage which go undetected by the
existing system.
Typically, non-destructive evaluation (NDE ) sensors based on magnetic flux leak-
age (MF L) are incorporated in Pipeline Inspection Gauge (PIG), also known as intelli-
gent or smart PIGs. The intelligent PIGs mainly serve two purposes: Firstly, the PIGs
clean the internal surface of the pipeline and remove blockages; Secondly, the PIGs pro-
vide surface anomalies details, such as cracks, pitting, weld defects and abnormalities
based on the NDE techniques (MF L, Ultrasonic, etc.). Moreover, pipeline inspection is
typically performed only during shutdown period, and the investigations are done using
techniques such as fluorescent-based visual inspection, laser profiling, etc. In addition
to this, MF L technology used in PIGs is reliable only in ferromagnetic or magnetic
conduits for characterizing defects and corrosion. NDE sensors including ultrasonic
and eddy current for fault (cracks, pitting, etc.) detection in both ferromagnetic and
non-ferromagnetic materials can be expensive to implement.
Therefore, iGlobus is conceptualized to be more affordable and easily deployable
robotic solution for inline inspection of pipelines. Unlike PIGs, the iGlobus can be
used in those pipelines also that are not designed for pigging maintenance activity, i.e.
unpiggable pipelines also. iGlobus also brings in a simpler spherical design which
weighs and sizes much less than a tenth of a conventional PIG. By hosting multiple
NDE sensors within, iGlobus performs similar pipeline inspection at a more affordable
price and at lower incurring costs.
1.2 Design Requirements
Based on the problem statement the design requirements can be classified into two
broad categories. These are:
• Design requirements for functionality.

INTRODUCTION
• Design requirements for Safety.
Functionality: These are the design requirements for iGlobus functionality wise.
These consist of the working constraints and size limitations.
1. Minimum diameter of the pipeline to be inspected : 10 inch.
2. Distance to be inspected : 200 kms.
3. Fluid inside the pipeline : Petrol, Diesel, Clear Naphtha, ATF.
4. Orientation of pipelines : Horizontal and Vertical pipelines.
5. Fluid Pressure : 5 bar to 90 bar.
6. Fluid Velocity : 7 km/hr(∼ 2m/s)
7. Type of flow : Turbulent
Safety: These are the design requirements with respect to the safety regulations and
protocols to be followed.
1. Insulation of robot with the view of preventing explosions.
2. Electronics connections inside the robot should be spark free.
3. Emergency shut down of robot.
4. Leak proofing of the robot even in elevated pressures.
5. Explosion proofing inside the shell.

ROBOT DESIGN
Robot Design
The shape of the robot is designed to be spherical as discussed in the earlier sections.
The spherical shape allows it to roll freely inside the cylindrical pipeline. The flow
of the fluid inside the pipeline provides the necessary thrust to roll the robot along its
direction. The structure of the robot is kept neutrally buoyant, so that it reaches its
terminal velocity i.e. the velocity of the material flowing as soon as soon as possible
after its deployment into pipelines.
2.1 Overview
This device includes a pair of outer shell which is spherical in shape. This spherical shape
makes the robot to roll inside the pipes or move in a flow without any obstruction in the
movement of the robot. The material of the outer shell is designed to withstand the static
pressure of 150bar and dynamic pressure of 90bar according to the high pressure inside
the pipeline. The material is selected based on its low acoustic impedance property.
Because the material should not affect acoustic leak signals which should propagate into
the shell to reach the sensor for detection. Port is provided for on/off switch, which is
covered by lid type closure. O-Rings are provided for sealing the closures. Inside the shell
various sensors are attached over the PCB board which senses the leak, orientation of the
robot and senses the temperature inside the robot. Battery is provided for long duration
operation which is minimum of 200km. The PCB board is attached inside the shell using
fasteners. Sensors inside the robot which kept near to the shell so that it can absorb the
leak signals easily.
Foam layer is provided at the outer surface of the shell. Foam is used to reduce the
contact between robot shell and pipe inner wall surface which avoid the chemical reaction
between these two metals. And also it avoids the spark produced by two metal contacts
and reduces the noise produced by the shell when it contact with the pipe wall. Foam
increases the overall surface area of the robot which helps the robot to get neutrally
buoyant. The foam thickness can be adjusted according to the pipe diameter and type of
fluid used inside the pipeline for making the robot neutrally buoyant.
The robot consists of multiple sensor units including acoustic sensors for leak and
pilferage detection, localization sensor for geo-tagging the defects. Finally, the
processing method includes data fusion by a processing unit to obtain a final data based

ROBOT DESIGN
on the raw acoustic data, visual data, and localization data indicative of defects in inner
surface of the pipeline.
The sensor data is stored in a memory unit of the device for further processing at a
later stage. The post-processing and data analysis method includes detecting and locating
defects such as - holes, leaks, cracks, pipe undulations, gas pockets, corrosion, blockages,
pilferage, pitting, or sedimentation.
2.2 Detailed Description
This section is subdivided into three broad categories,
1. Mechanical design which deals with various design concepts, material selection,
identifying optimum dimensions to with stand the pressure conditions inside the
pipeline, modularity, sealings etc.
2. The electronics layout consists of the details of the sensors, microcontroller and
other component details and specifications.
3. The Software layout comprises of data storage mechanisms, pre-processing, anal-

ysis, post processing and visualisation of the data
2.2.1 Mechanical Design
iGlobus is a robotic tool for inline inspection of pipelines for detection of leaks and
pilferage. It is propelled passively, in inflammable environment under high pressures.
The robot is illustrated in Figure 1. The robot is passively propelled inside a pipeline,
which transports a fluid. The pipeline can be a ferromagnetic or a non-ferromagnetic
pipeline and the fluid is can be any fluid such as oil, water or gas. The robot is propelled
based on the pressure differential in the pipeline, it will rotate using the flow of the fluid
in the pipeline. It collects information about the pipeline as data that is captured using
sensors for surveying the health of the pipelines. The robot primarily includes an outer
shell, Foam layer, sensors, Batteries and a processing unit.
Concept Design:
4-Design concepts had been made for iGlobus and the best concept among them has
been selected for product development.

ROBOT DESIGN
In first concept Design, shells are connected by fasteners. Microphones are attached
near the wall of shell and other electronic components are mounted on the plates that is
attached with shell wall. Port are provided for the given requirements.
In the second concept design, shells are connected using threads. Microphones are
attached near to the shell wall using some supports. Other electronic components are
mounted on a plate and everything is supported which makes it as a single component.
In the third concept design, shells are connected using threads. Electronic components
are mounted on PCB board. Battery and PCB board mounted on the inner shell. Port are
provided for data acquisition and on/off switch.
Fourth concept is same as of third concept except snap fits were used instead of
threading.
Concept Selection:
Third and fourth concept is selected for iGlobus design because of the given factors
below.
1. Stress concentration will be low because it has no fasteners.
2. Mounting the microphone and other sensor is easy and reliable when compared with
other concepts.
3. The impact and minimum deformation on the outer shell won't affect the electronic
components located inside the shell.
4. Portway can be easily made.
5. Assembling and disassembling the shells and other components are easy.
Detailed Design:
The outer shell includes a pair of domes sealed together using Threading’s. The robot is
designed so as to withstand high pressure, harsh conditions, and operate in inflammable
environment. The outer shell is constructed out of Aluminium alloys 7075-T6 such
that it can withstand heavy load and has less acoustic impedance. Aluminium alloy
7075-T6 has good material properties such as it has high yield strength which can stand
high pressure and has low density which makes the robot light weight. It has low
acoustic impedance property so that the sensors which is inside the shell can absorb the
acoustic leak signals easily. The shell is designed to withstand the high pressure of
150bar in static condition for hydrostatic pressure testing and it can withstand the
impact inside the pipes.

ROBOT DESIGN
Figure 1: First and Third Concept Design of iGlobus
The outer shells are connected with each other using threading according to the third
concept. Inside the shell space for mounting the sensors are provided. Sensors are
mounted over the PCB board and processing unit. The PCB board is then attached with
the shell using fasteners. Batteries are provided for the robot to run long distance.
The PCB board has a number of electronic components including processing unit.
There are two PCB boards which is called as parent board and child board. Two boards
are connected with each other. The parent board is carrying the child board. The
electronic components mounted on the parent mounting surface includes plurality of
acoustic sensors, IMU, temperature sensor to inspect inner surface of the pipeline. The
microphone is used to receive acoustic data reflected from the inner surface of the
pipeline. The received data is stored in the memory card/chip for further processing. The
child PCB board carries a processor, power distribution unit and a memory storage unit.
Way for port is provided at the outer shell and lid is covering the port. Port has on/off
switch, recharge port for batteries, port for data acquisition and LED for indication of
working of the robot. Lid covers the port using thread. O-Rings are provided at the
closures which makes the sealing tight makes the robot water proof. Thread seal tapes
are provided at the threads which tighten the sealing and avoids the opening of robot due
to vibration.

ROBOT DESIGN
Figure 2: Threading and Snap fit Method of iGlobus
Shell Design :
Shell covers the payload of the robot and protects it from harsh pipeline conditions.
The shell also acts as a transporter for the payload. The spherical shape of the shell
ensures smooth traverse of the robot along the pipeline. The shell design is based on
numeric calculations and simulations performed. The selection of material is a critical
part of the design of the shells. Material selection is based on the following require-
ments:
• Material should withstand a pressure up to 90 bars.
• It should withstand the hydrostatic pressure testing of 150 bar in static

condition.
• Acoustic impedance of the material should be low.
• Lower density.
• Higher strength even at elevated temperatures.
• Non-reactive to acids, alkalis and especially the contents of the pipeline.

ROBOT DESIGN
• Non-inflammable.
• Should have Good Impact resistance.
Acoustic impedance of the shell material is an important characteristic for the func-
tionality of the robot. Lower acoustic impedance assists the transmission of acoustic
waves through the shell material to the acoustic sensors. The acoustic impedance of
some materials commonly used for manufacturing are given in Table 3 of Annexure I.
Based on the data given in Table 3 of Annexure I., Polyoxymethylene and Aluminium
alloy 7075 is suggested since they have low acoustic impedance. Polyoxymethylene
(POM) is also known as Polyformaldehyde or acetal. It is a highly crystalline ther-
moplastic known for its high flexural and tensile strength, stiffness, hardness and low
creep under stress. It also has low coefficient of friction, excellent chemical resistance
and outstanding fatigue strength, but only moderate heat stability and insufficient flame
resistance. Because of moderate heat stability and insufficient flame resistance, this
material cannot be used for making the shell of iGlobus.
Aluminium alloy 7075-T6 is a light metal, about the third of the density of steel,
copper and brass. It has very good corrosion resistance to common atmospheric and
marine atmospheres. Addition of alloy makes the material to stand high pressures. The
thermal conductivity of aluminium is around four times that of steel and its specific heat
twice that of steel. The heat is conducted away faster and a greater heat input is
necessary to bring the same mass of aluminium to a given temperature, compared
with steel. This helps to reduce hot spots where significant localised property loss
could occur, so extending the serviceability period. Providing an additional safety
feature in case of mishap. Hence, aluminium alloy 7075-T6 is a suitable material for
making the shell for iGlobus. The properties of Aluminium Alloy 7075-T6 are given in
Table 4 of ANNEXURE I.
The physical properties of aluminium, notably high thermal conductivity, specific

heat, and reflectance and its low emissivity, provide resistance to structures against
temperature rise comparable or superior to that provided by steels in the early stages
or in a non-engulfing fire. Even when temperatures do increase to intense incendiary
levels, aluminium does not burn in air nor will it support combustion. When tested in
accordance with ASTM or British standards, aluminium provides the highest ratings for

ROBOT DESIGN
resisting flame spread because it is not easily ignitable under atmospheric conditions and
does not support flame spread.
Thickness Calculation:
Pressure ∗ Out er d ia.

Allowable St ress =
4 ∗ Wall T hickness
Pressure ∗ Out er d ia.
Wall T hickness =
4 ∗ Allowable St ress
Simulation are done using Solidworks 2019 and the data is combined.
Maximum principal stress theory is the one that is used out of the five failure the-
ories. Considering it is used for ductile materials. Aluminium alloy 7075-T6, being a
ductile material, maximum principal stress is used to predict the failure of material at
uniaxial or biaxial load condition. According to maximum principal stress, if one of the
principle stress σ 1 (maximum principal stress), σ 2 (minimum principal stress) or σ 3
exceeds the yield stress, the material starts to yield. Using this maximum principal
stress theory and results of analysis in Solidworks 2019 the failure of the material at
various thickness can be calculated.
Figure 3: Cross Sectional View of Equivalent Stress Analysis of Shell in

Solidworks 2019 simulation
Static structural analysis of the aluminium alloy 7075-T6 is done in Solidworks 2019
on a globus design which connected by threading with various thickness to find the
thickness at which the material should not fail at given pressure of 150bar in static
condition. The robot is fixed at one side which is used as the boundary condition.
Meshing is done using Tetrahedron method and body sizing as 3mm. Boundary
ROBOT DESIGN
Figure 4: Cross sectional view of Total Deformation of Shell in SolidWorks

2019 simulations
conditions are applied as 150bar pressure on the outer surface of the sphere and one side
as fixed support. Equivalent stress and total deformation are calculated as results. The
yield stress of aluminium alloy 7075-T6 is 505MPa and the result from analysis should
not exceed 505MPa so the material starts to yield. A factor of safety of 1.2 is taken and
result is calculated to improve the reliability of design. Hence the maximum allowable
yield stress is 430.83MPa. 11 mm thickness of aluminium alloy shows the best result
according to the simulation which has the total equivalent stress of 427.8 MPa and has
low total deformation. The result of simulation shown in a cross-sectional view. The
equivalent stress analysis and total deformation analysis of the sphere in Solidworks 2019
Workbench is shown in Figure 3 and Figure 4, respectively.
When the robot moves inside the pipeline it hits the wall of the pipeline at high
pressure of 90bar in a velocity of 2m/s (which is the flow rate of fluid inside the
pipeline). This impact may cause damage to the robot shell. If the robot shell gets
damage at the impact, it will cause a serious damage in the pipeline. Hence impact test
has been done on the robot according to the condition of the pipeline.
During the impact test the robot is allowed to hit the wall in a velocity of 2m/s
surrounded with a pressure of 90bar. The simulation has been done with various
thickness of robot. 11mm thickness of shell gives the result of 389.6MPa equivalent
stress which is lower than the allowable stress. The equivalent stress analysis and total

ROBOT DESIGN
deformation analysis of the sphere in Solidworks 2019 Workbench is shown in Figure 3
and Figure 4, respectively.
Figure 5: Cross Sectional view of Impact Analysis of shell
Foam Layer:
The robot is cover with polyurethane foam layer. This layer will prevent sparks
generated due to the contact of robot with the pipeline surface. It acts like a cushion
between the robot and the pipeline’s internal surface. Also, it will increase the volume
of the robot, increasing the drag force on it due to the fluid flow, thereby increasing the
terminal velocity. The thickness of this polyurethane layer depends on the pipeline to
be inspected. The foam layer makes the robot neutrally buoyant because it increases the
surface volume of robot with low weight. Calculations for neutral buoyancy has been
made with different thickness of foam layer for different fluid types.
Figure 6: Polyurethane Foam Layer

ROBOT DESIGN
iGlobus being passively propelled in the pipeline reduces the designing and implementing
any actuators in the robot. However, since there is external forces acting on the robot, it is
important to calculate the effects caused due to those. The flow of the fluid inside tha
pipeline propels the robot. Hence, terminal velocity of the ball is one of the most important
term to be calculated. The fluid velocity and the density of the ball being constant parameters
to calculate the terminal velocity of the ball inside the pipeline. The terminal velocity will
help us determine the time taken by the robot to cover a certain distance (between to
insertion and retrieval points). Considering the fact that the distance between the insertion
and retrieval points is known to us, we can estimate the time taken by the robot to reach the
retrieval point from the insertion point. Using this data, the battery life or discharge rate
required for the power source can be estimated. It is necessary to add some factory of safety
value to the calculations although using a battery way more than the required specifications
might increase the weight or overall density of the robot. The terminal velocity was
calculated using the following
formula:
s
2 · fR
VT = VR −
fd · AC · ρ f
where, VT = Terminal Velocity
VR = Relative Velocity of fluid w. r. t. sphere
fR = Rolling Friction
fd = Drag Coefficient
AC = Cross sectional area of sphere
ρ f = Density of fluid
Fluid flow simulation and analysis is done in Solidworks 2019 flow simulation software.
This simulation and analysis is used to understand the flow of turbulence fluid inside a
pipeline with sphere inside it and also helps to understand the pressure and temperature
variation, velocity difference of fluid inside the pipeline. For that the pipeline of 0.273m
diameter is designed in Solidworks and assembled with spherical ball inside it. The
simulations for terminal velocity were performed in Solidworks. The results have been
displayed in the figures below:

ROBOT DESIGN
Figure 7: Transitions in Fluid Pressure
Figure 8: Transitions in Fluid Temperature
Figure 9: Transitions in Fluid Velocity
Design Iteration:
Stress concentration acts more on threads of the robot which is used to tightly close the robot.
Due to wearing while opening and closing the robot and stress concentration on the
threading, it get damage easily. Threading is not a reliable method for closing hence design
and simulation work on snap fit method is in progress (which is fourth design concept that
proposed)

ROBOT DESIGN
2.2.2 Electronics Layout
A schematic diagram of the placement electronic modules of the robotic device is illus-
trated in Figure 7. The electronic modules primarily include - a processing unit, sensor
modules and a power source. The sensor modules include acoustic sensors, an inertial
measurement unit, a temperature sensor. The acoustic sensor module includes a
microphone and an amplifier. The acoustic sensors are used to detect and record the audio
inside the pipeline to determine pressure gradient. The pressure gradient is determined by
detecting any difference in the acoustic signal, which may indicate the presence of an
opening, a hole or a leak encountered in the pipeline. The audio recordings are stored in the
memory card/chip for further processing. The post-processing may involve methods, such
as Fourier transformation, Hilbert Transformation, Kalman filter, or any other existing
techniques. A camera along with a LE D flash light is used for obtaining visual data of the
pipeline. The power supply to LE D flash light is controlled using a switch. The camera
will capture images of the pipeline and store it in a memory card/chip.
Further, iGlobus includes inertial measurement unit (IMU), which contains both an
accelerometer (3 axis) and a gyroscope (3 axis). The accelerometer is configured to
measure the distance travelled and the gyroscope gives the number of bends inside the
pipeline.
Microprocessor :
Figure 10: Raspberry Pi 3 PinOut diagram

ROBOT DESIGN
Raspberry Pi (RPi) is a low cost 85.60 mm x 56.5 mm x 17 mm sized computer that

can be later plugged to any display as monitor. It has the ability to interact with a
wide array of electronic modules. Raspberry Pi 3 (Model B) has been chosen as the
microprocessor for iGlobus. It has I/O capabilities coupled with a gigabyte of RAM
and quadcore 900 MH z ARMv7 CPU . It has 40 General Purpose Input Output (GPIO)
pins. It runs on DebianOS based on Linux. Specifications of the microprocessor are
given in Table 6 in ANNEXURE I.
Sensors :
A number of sensors and sensor packages are mounted on the robot for data collec-
tion. An Inertial Measurement Unit (IMU), acoustic sensors, temperature and pressure
sensors .Detailed description is given for all the sensors used.
Inertia Measurement Unit (IMU) contains an inbuilt accelerometer, gyroscope,
magnetometer and barometric pressure sensor. An accelerometer senses each revolu-
tion as a repeated pattern of acceleration. Alternately, a magnetometer will sense the
magnetic changes as its sensor approaches and retreats from pipeline during each rev-
olution. Anomalies found by any of the sensors (such as a leak found by the acoustic
sensor, or corrosion found by the magnetic sensors) can have their locations along the
pipeline determined by noting the data obtained from the accelerometer combined with
the data from gyroscope. Data is logged on the memory card/chip and typically the
analysis is done by an external CPU after the sensor unit has been removed from the
pipeline and thereby the data is downloaded into the computer from the SD card.
Figure 11: IMU VN-100 IMU/AHRS

ROBOT DESIGN
The IMU recommended to be used is VN-100 RUGGED. It is a miniature, high-

performance Inertial Measurement Unit (IMU) and also consists of an Attitude Head-
ing Reference System (AHRS). AHRS provides the pitch and roll angles relative to the
Earth’s gravity vector and heading angle relative to north. Incorporating MEMS sensor
technology, the VN-100 combines 3-axis accelerometers, 3-axis gyros, 3-axis magnetic
sensor, a barometric pressure sensor as well as a 32-bit processor. The VN-100 is con-
sidered both an IMU, in which it can output acceleration, angular rate, and magnetic
measurements along the X, Y, & Z axes of the sensor, as well as an AHRS in which
it can output filtered attitude estimates of the sensor with respect to a local coordinate
frame. General specifications of IMU are given in Table 7 in ANNEXURE I.
The VN-100 has a built-in microcontroller that runs a quaternion based Extended
Kalman Filter (E KF ), which provides estimates of both the attitude of the sensor as well
as the real-time gyro biases. Vect orNav uses a quaternion based attitude filter because
it is continuous over a full 360◦ range of motion such that there are no limitations on the
angles it can compute. However, the V N − 100 also has a built-in capability to output
yaw, pitch and roll angles, in which the sensor automatically converts from quaternions
to the desired attitude parameter. Outputs from the VN-100 include:
1. Attitude
∗ Yaw, Pitch, Roll
∗ Quaternions
∗ Direction Cosine Matrix
2. Angular Rates
∗ Bias Compensated
∗ Calibrated X ,Y and Z gyro measurements
3. Acceleration
∗ Calibrated X ,Y and Z measurements
4. Magnetic
∗ Calibrated X ,Y and Z measurements
5. Barometric Measurements

ROBOT DESIGN
Acoustic Sensor : Acoustic sensor is used to detect the presence of the leak and
size of the detected leak. It is small (0.4in x 0.3in) breakout board. ICS43434 is a digital
microphone that outputs I2S audio as a stream of 24-bit serial words that can be directly
read by any microcontroller with an I2S port. It has low noise and high sensitivity. The
I²Sinterface allows the ICS-43434 to connect directly to digital processors, such as DSPs
and microcontrollers, without the need for an audio codec in the system. The ICS-43434
has a high SNR of 65dBAand a wideband frequency response. The sensitivity tolerance
of the ICS-43434is ±1 dB, which enables high-performance microphone arrays without
the need for system calibration. It has a sample rate of 23-51.6 kHz.
Figure 12: Acoustic Sensor
Figure 13: Microphone Circuit Layout
The sensor can be used in any application with a noisy environment and requires
a small, narrowband frequency response. It can easily be mounted in small and tight
spaces.
Safety Measures :
• Both the microphones are connected on the non-conducting platform to avoid

short circuiting with the other sensors.
• When one microphones goes out of service another microphone is immediately

turned on to continuously detect the leak and record the data on the SD card,

ROBOT DESIGN
without any loss. Suitable programming is done on raspberry pi to ensure other
microphone turns on when one microphone stops working.
ELECTRONICS LAYOUT:
Figure 15: Electronics Layout
Temperature Sensor :
Figure 16: Temperature Sensor: MAX 6675 + K Thermocouple
A temperature sensor is installed in the device for temperature measurement of the

inner environment of the device. The MAX 6675 Module +K Type Thermocouple Sen-
sor makes use of the Maxim MAX 6675K-Thermocouple with a digital converter IC to

ROBOT DESIGN
provide a microcontroller-compatible digital serial interface (SPI compatible) for giv-
ing an accurate temperature compensated measurement of the supplied K-Type thermo-
couple sensor. It has a 12-bit resolution providing temperature readings from 0◦ C to
1024◦ C with a resolution of 0.25◦ C). Additionally, to detect the voltage and calculate
the temperature, dedicated thermocouple sensor module - MAX 6675 module is used.
This module, constantly monitors the output voltage from the thermocouple, calculates
the corresponding temperature and outputs this to a SPI interface. The SPI interface
can be used to get the temperature on Ras pberryPi.

ROBOT DESIGN
Temperature Sensor performs cold-junction compensation and digitizes the signal from
a type-K thermocouple. In order to avoid cold-junction related errors, no heat gen-
erating device or components are to be placed near MAX 6675. The specifications of
Temperature sensor are given in Table 11 in ANNEXURE I.
RTC :
RTC (DS3231): A real-time clock (RTC) is an IC that keeps an updated track of the
current time. This information can be read by a microprocessor, usually over a serial
interface to facilitate the software performing functions that are time dependent. RTCs
are designed for ultra-low power consumption as they usually continue running when
the main system is powered down. They usually interface to a microprocessor circuit
by an SPI or I2C serial bus, and may contain a number of other functions like backup
memory, a watchdog timer for supervising the microprocessor and countdown timers to
generate real time event. The general specifications of the RTC module used are given
in Table 10 in ANNEXURE I.
Figure 17: Real Time Clock
The SDA pin of real time clock is connected to the SDA pin of Raspberry pi and SCL
pin of real time clock is connected to the SCL pin of RaspberryPi.
Safety Measures : In the embodiment of iGlobus, we will be using two RTCs. Only
one RTC will work at a time. When one RTC gets heated up, then that RTC will go to
sleep mode until it cools again. In the meanwhile, the microprocessor is programmed
so as to activate the other RTC in order to continue the clock.
Power Supply : The power platform consists of a non-conducting surface to protect

the circuitry of sensor platform or to prevent the shorting of power source. Most conve-
niently, this is a long life battery. The power source provides stored power to perform

ROBOT DESIGN
continuous measurements by the sensors. A non-rechargeable Lithium - MnO2 cell is

preferred based on its performance, size, weight and appropriate rating, which delivers
suitable voltage to power the sensors. It can store enough energy to power the sensors
for the intended period of time. General Specifications of the battery to be used are
given in Table 12 in ANNEXURE I.
Figure 18: Saft M 20 EX SV Primary Li − MnO2 cell
Saft M 20 EX SV Primary Li − MnO2 cell is ideally suited for applications requiring

high energy and long operating life, with stable voltage under high discharge in -40◦ C
to +72◦ C environment. The cell is certified according to ATEX and IECEx for use in
potentially explosive atmospheres. Saft battery is being used for experimental purpose.
Later, high capacity battery will be used for 250 km. General Specifications of the
battery to be used are given in Table 13 in ANNEXURE I.
Features:
• High drain/high pulse capability.
• High voltage response, stable during most of the lifetime of the application even
after long dormant periods.
• High capacity and very high pulse capability.
• Low self-discharge with long operating life (<1% after 1 year at +20◦ C).
• Superior resistance to corrosion.
• Low magnetic signature.
• Integrated safety vent.
• Non-corrosive electrolyte.

ROBOT DESIGN
Safety Standards:
• Safety: UL 1642 (File MH 61234)
• IEC 60086-4
• ATEX/IECEx: IEC 60079-0, IEC 60079-11
• Quality: ISO 9001, Saft World Class Continuous program
• Environment: ISO 14001
Description: This power source is specifically chosen for iGlobus since it follows
all the norms of IEC 60079-0 and IEC 60079-11. Two batteries are to be connected in
series to increase the voltage range to be supplied to the Microcontroller. All the cells in
the battery are of the same electromechanical system, cell design and rated capacity and
made by the same manufacturer (IEC 60079-0:2007). All the batteries are arranged and
operated within the allowable limits defined by the cell and the battery manufacturer.
All cells are constructed or arranged so as to prevent leakage of electrolyte. When it
is necessary for the user to replace the cells of batteries contained within an enclosure,
the relevant parameters to allow correct replacement is legibly and durably marked on
or inside the enclosure, either the manufacturer’s name and the part number, or the
electrochemical system, nominal voltage and rated capacity is also included.
The cell or battery construction follows one of the following norm of IEC 6007911:
2006, that is, the battery is sealed Li − MnO2 battery does not require addition of elec-
trolyte during their entire life and have a sealed metallic enclosure. Li − MnO2 battery
ensures that the internal surfaces are not adversely affected by the action of an elec-
trolyte. Battery containers will be provided with sufficient ventilation openings. A seal
will be provided between each terminal pole and the lid of the cell to prevent leakage of
the electrolyte. The external surface temperature of the cell or battery will not exceed
the value specified by the cell or battery manufacturer (IEC60079-7:2006). If the bat-
tery has to be disconnected from its associated equipment in the hazardous area then it
is capable of being isolated safely.
Power Flow: The electronic components mounted in the embodiment include IMU,
temperature sensor, pressure sensor, processing unit, memory unit, battery, LED flash-
light, real time clock. The acoustic sensor includes microphone .The microphone is

ROBOT DESIGN
used to receive acoustic signal reflected from the inner surface of the pipeline. The
processing unit is configured to log the sensor data from the inner surface of the pipeline.
All the sensor data is stored in a memory unit. With this configuration for the data flow,
the sensors are all powered by microprocessor - Raspberry Pi, which is powered by the
battery mentioned above. The power flow diagram is shown in Fig: 24.
Figure 19: Power Flow Diagram
2.2.3 Software Layout
After collecting data mentioned in the previous sections we will be mostly working on
data analysis and noise filtration of data in the Software Section. It has four sub-parts
i.e.:
• Data Acquisition
• Data Processing
• Data Flow
• Data Representation

ROBOT DESIGN
Data Acquisition :
Here is the data acquisition method explained for visual and acoustic inspection of
pipelines containing a moving fluid. The method comprises of :
a. recording by acoustic sensors, acoustic data indicates defects in the pipeline;
b. capturing interior visuals by camera, visual data implies images of internal defects
of the pipeline;
c. Determining location by an inertial measurement unit sensors (IMU ), based on dis-
tance travelled in the pipeline and number of bends traversed in the pipeline.
From Acoustic Data, we will be mostly understanding source or the type of defect
and size of defect. From IMU Data, we will be obtaining the localization of the defect.
From Video Feed, we are getting visualisation of the whole of the pipe interior, digital
images of defects and estimation of corrosion. The sensor data is stored in a memory
unit of the device for processing after retrieval of iGlobus from the subject pipeline. The
post-processing and data analysis method includes detecting and locating defects such
as - holes, leaks, cracks, pipe undulations, gas pockets, corrosion, blockages, pilferage,
pitting, or sedimentation.
Data Processing Algorithms :

After retrieving iGlobus we gather our data collected from various sensors. Since
these are raw data, there are a series of data analysis involved to come for the final
result of the report, which we are covering here in this section. The data processing
mostly involves noise removal, quantification and localisation of the anomaly based on
acoustic, IMU and video input.
Acoustic Data (Noise Filtration) :

Band-Pass Filter : FFT Filtering is a process that involves mapping a time signal
from time-space to frequency-space in which frequency becomes an axis. By mapping
to this space, we can get a better picture for how much of which frequency is in the
original time signal and we can ultimately cut some of these frequencies out to remap
back into time-space. Such filter types include - low-pass: where lower frequencies are
allowed to pass and higher ones get cut-off; high-pass: where higher frequencies pass
and lower ones get cut off; and band-pass: which selects only a narrow range or ’band’
of frequencies to pass through.

ROBOT DESIGN
IMU
RTC
Microprocessor
Acoustic
Sensor
Camera
Memory
Card
Temperature
Sensor
Pressure
Sensor
Figure 20: Data Collection Diagram
The Low Pass Filter has a constant output voltage from DC (0 Hz), up to a specified
Cut-off frequency, ( f c ) point. This cut-off frequency point is 0.707 or −3dB(dB =
−20 log ∗VOU T /IN ) of the voltage gain allowed to pass.
The frequency range below this cut-off point ( f c ) is generally known as the Pass
Band as the input signal is allowed to pass through the filter. The frequency range
above this cut-off point fc is generally known as the Stop Band as the input signal is
blocked or stopped from passing through.
Passive High Pass Filter is the exact opposite to the low pass filter. This filter has
no output voltage from DC (0 Hz), up to a specified cut-off frequency ( f c ) point. This
lower cut-off frequency point is 70.7% or −3dB(dB = −20 log ∗VOU T /IN ) of the voltage
gain allowed to pass. The frequency range ’below’ this cut-off point ( f c ) is generally
known as the Stop Band while the frequency range ’above’ this cut-off point is generally
known as the Pass Band.
Band-pass filters are important, because we can model some aspects of articula-
tion and hearing in terms of the actions of band-pass filters. Unlike low-pass high-pass

ROBOT DESIGN
filters, which have a single cutoff frequency, band-pass filters have two cut-off frequen-
cies, one for the low end of the pass band and one for the high end of the pass band. A
band-pass filter is like a combination of a low-pass filter and a high-pass filter, where
the cutoff frequency of the low-pass filter is higher than the cutoff frequency of the
high-pass filter.
Kalman filter : Kalman filter works on the principle that it should perform the
technique of the Bayesian recursive least square error. The use of this technique makes
it possible to assess a signal transmitted by a distorted channel, while gaining some
noises at the same time.
Formulation of this filter in order that it can remove the noise and distortion is based
on the state-space. In this regard, x(m) vector is taken to show the desired vector and
y(m) vector shows the output noisy vector.
x(m) = Ax(m − 1) + e(m)
y(m) = H x(m) + n(m)
In order to apply these equations we have to first filter using the following algorithm.
The observed output, noisy and distorted, is taken as the input of the Kalman filter.
Having implemented the following algorithm, we will be able to achieve an appropriate
assessment of the desired original signal devoid of any noise or distortion.
IMU Data :
Kalman Filter : Since the IMU gives data with some noise, we propose to overcome
it with Kalman Filter. Estimating position with the help of state space equations: To
model the robot position, we wish to know its x, y and z coordinates and its orientation.
These three parameters can be combined into a vector called a state y variable vector.
The Kalman filter is a smarter way to integrate measurement data into an estimate by
recognising that measurements are noisy and that sometimes they should be ignored or
have only a small effect on the state estimate. It smooths out the effects of noise in
the state variable being estimated by incorporating more information from reliable data
than from unreliable data. The user can tell the Kalman filter how much noise there is
in the system and it calculates an estimate of the position taking the noise into account.

ROBOT DESIGN
The K f calculation steps are described below:
• 1. The prediction step :
X (k) = F (k).X (k − 1) + B(k − 1).U (k − 1)
P(k) = F (k).P(k − 1).F (k).T
• 2. Where F (k) is the state transition model which is applied to the previous state
x(k − 1); B(k − 1) is the control-input model which is applied to the control vector
u(k − 1).
y(k) = z(k) − H (k).X (k)
P(k) = F (k).P(k − 1).F (k).T + Q(k − 1)
where y(k) represents the innovation or measurement residual; H (k) is the obser-
vation model which maps the true state space into the observed space; P(k − 1)
represents the predicted (a priori) estimate covariance; Q(k − 1) is the covariance
of the process noise.
• 3. Update Step :
K(k) = P(k).H .T (k).S(k − 1)
X (k) = X (k) + K(k).Y (k)
P(k) = (1 − K(k)H (k))P(k)
AHRS (Attitude and Heading Reference System) : An IMU module ensures the
measurements of linear acceleration, angular velocity and the induction of earth mag-
netic field in three mutually orthogonal directions. The axes of all the sensors are par-
allel, sharing the same origin. Such solution of the sensors seating defines the IMU
body’s coordinate system called the body frame.
The IMU body frame with respect to an external coordinate system is determined.
The orientation values are the Euler angles called pitch, roll and yaw. Their precise de-
termination allows to eliminate the influence of gravity acceleration and Coriolis force
from subsequent navigation. Traditionally, the most commonly used method for Euler
angles determination is performing attitude computation on the basis of mean sensor

ROBOT DESIGN
readings made within few minutes during which the IMU module remains stationary.
This solution is sufficient however, it consumes a lot of time. Often, even a slight influ-
ence on the sensor system during this process can cause irregularities in the calculations.
As a consequence of these irregularities the incorrect determination of the initial IMU
orientation can be noticed. This will have a significant impact on the results of further
navigation. The use of AHRS algorithm for sensor orientation determination eliminates
these two factors.
In pipeline, flow makes the iGlobus to rotate abruptly on any of the axis. So it spins
on any of the axis of rotation X, Y, Z. Along with this data, in order to accurately identify
the position of iGlobus with the translational motion as well, we have introduced a
system called AHRS, which is able to change the reference system from the IMU frame
to the global reference system.
An AHRS integrates rate information from the gyroscopes to provide 3D orienta-
tion. Reference data – such as gravity (provided by the accelerometers) and the Earth’s
magnetic field (provided by the magnetometer) – is used to compensate for the drift cre-
ated by this integration. This makes the AHRS a much more cost-effective alternative
to high-grade inertial measurement units (IMUs).
Video Input :
Denoise : Gaussian Blurring, Median Blurring are techniques used in removing small
quantities of noise. In those techniques, we take a small neighbourhood around a pixel
and perform operations like gaussian weighted average, median of the values, etc., to
replace the central element. In short, noise removal at a pixel is made local to its neigh-
bourhood.
Noise is generally considered to be a random variable with zero mean. Consider a
noisy pixel,
P = P0 + n
where P0 is the true value of pixel and n is the noise in that pixel. We can take a large
number of same pixels from different images and computes their average. Ideally, we
should get,
P = P0
since mean of noise is zero.

ROBOT DESIGN
Dehaze :
The video feed of iGlobus in fluid will also entail haze due to lack of sufficient illu-
mination, atmospheric noise and scattered particles, as well as selective absorption of
different wavelengths of invisible spectrum causing colour imbalance and subsequently
poor contrast. Therefore, the video feed requires dehazing activity as the data pre-
processing for noise filtration.
Methods of Image Dehazing :
• Dark Channel Prior
• DehazeNet
• Non-Local Image Dehazing
In the atmospheric scattering model, we assume that the haze is independent of the
scene radiance and it attenuates the signal of the viewed scene, i.e., it introduces an
additional component to the image called ambient light or airlight. So the hazy image
is modelled as a convex combination of the actual scene radiance and the airlight. We
thus get an under determined system of three equations - one for each colour channel
- and four unknowns - transmission and three channel radiance intensities - per pixel.
Most of the approaches to solving this system require additional information such as
multiple images or assume some sort of local patch-based prior.
Ian Sun et al (2009) proposed an empirical algorithm for estimating transmission
in the Atmospheric Scattering model of Image Formation. They used the observation
that most local patches in haze-free outdoor images contain some pixels which have
very low intensities in at least one color channel. They also model the transmission as
inverse exponential of depth and so a high quality relative depth map is also obtained.
Haze Mode
The common hazy image formation model is :
I(x) = t (x) · J(x) + [1 − t (x) · A],
where x is the pixel coordinates, I is the observed hazy image, and J is the true
radiance of the scene point imaged at x. The airlight A is a single colour representing the

ROBOT DESIGN
airlight in image areas where t = 0. The scene transmission t (x) is distance-dependent :
t (x) = e − β d(x),
where β is the attenuation coefficient of the atmosphere and d(x) is the distance of the
scene at pixel x . Generally, β is wavelength-dependent and therefore t is different per
colour channel. This dependency has been assumed negligible in previous single-image
dehazing methods to reduce the number of unknowns and we follow this assumption.
The transmission t (x) acts as a mating coefficient between the scene J and the airlight
A. Thus, per-pixel x, E q.(1) has three observations I(x) and four unknowns: J(x) and t
(x), resulting in an under-determined estimation problem.
Bolun Kai et al (2016) proposed a Convolutional Neural Networks (CNN) based
deep architecture, whose layers are specially designed to embody the established as-
sumptions/priors in image dehazing called ’DehazeNet’ to estimate the transmission at
each pixel, a transmission map, in a given hazy image and then subtracting the atmo-
spheric light times transmission to recover underlying dehaze image. Here the layers
of Max out units are used for feature extraction, which can generate almost all haze-
relevant features. They also proposed a novel nonlinear activation function BReLU
- Bilateral Rectified LinearUnit to prevent overestimation of haze. Again the trans-
mission estimated here can be used to recover an approximate relative depth map by
modelling it as inverse exponential of the depth.
One of the best state-of-the-art image dehazing algorithms is one by Dana Berman
et al(2016). The ’Non-Local Image Dehazing’ algorithm differs from previous image
dehazing methods in that it doesn’t assume any local patch based priors. It uses the
observation that degradation depends on the distance of the point from the camera ex-
pressed in terms of transmission coefficients that control scene attenuation and amount
of haze in each pixel, ie,varying distances correspond to different transmission coeffi-
cients. They assume that colors of a haze-free image are well approximated by a few
hundred distinct colors that form tight clusters in RGB space. This is to reduce the
computation in frequency domain.
Therein, a key observation is that pixels in a given cluster are often non-local, spread
over the entire image plane and located at different distances from the camera, which is
very realistic as light converging on a point on the camera lens is coming from many di-

ROBOT DESIGN
rections sand angles and not just perpendicular, although the perpendicular component
may be the biggest contributor. Once we cluster the pixels according to their intensities,
each cluster in a clear image becomes a line in RGB space which they call haze-lines.
Using the haze-lines we get the distance maps (depth maps) and then use it to estimate
transmission map and subsequently get the haze-free image. This algorithm is linear in
the size of the image and requires no training
Data Flow :
This section deals with how the different raw data we collected are flowing from
storage to the database. The raw data collected in different formats are processed with
different algorithms. The three kinds of data formats are .csv, .wav and .mp4 for IMU,
which are for Acoustic, Video and IMU data respectively.
IMU data is first sent to Kalman filter for noise removal from the data from ac-
celerometer, gyrometer and magnetometer data. Acoustic data is sent to different noise
filters, Band Pass Filter for removing any kind of ambient noise from the environment
of particular upper and lower frequency and one Kalman filter for estimating the data in
magnitude format. Video Data is then processed with denoise filter and then dehazing
because of limited illumination inside pipeline, and denoised for better visual inspection
and further analysis.
All these data when collected will go to memory card, from which we will be able
to access data on our inbuilt GUI. Graphical User Interface is a visualisation tool for
the current state of the pipeline, with a dashboard depicting various information about
location of defects and the current state of pipeline. Spectrogram and Audiogram syn-
chronised with location from IMU and a Geographical Information System Maps for
each Inspection.
In Post-Processing, the raw data we have IMU for localisation in .csv format. Acous-
tic graphs and plots in the form of spectrogram which is a plot of frequency versus time
domain. Audiogram which is like a magnitude vs time domain plot. We also get a better
high definition video as the final output from the algorithms.

ROBOT DESIGN
Software Layout :
Figure 21: Software Layout
Data Representation : We have planned a layout for a Graphical user interface for
data representation purposes. Users can have access to any of the inspection video and
report instantly through data.
The GUI consists of four upload buttons: Original Video, Processed Video, Spectro-
gram, Audiogram and GIS Mapping.
Original Video will have the actual inspected video in display and we can capture any
frame at any point of time.
Processed Video is with better video quality denoised and dehazed video.
Spectrogram is the spectrum of the acoustic data after removing noise from the raw data
presented in the form of a plot, from where we can find the type of leak and pilferage
according to the frequency.
Audiogram is the magnitude v/s time domain plot of acoustic data and it can be used to
know the existence of leak, when it observes peak features.
GIS Mapping : After Collecting heading coordinates we get the latitude and longitude
on global positioning system. We then mark on the map giving these data that tells us
the pattern iGlobus travelled in the pipeline.
Finally, a report will be generated with the details of leak detected, its location and its
characteristics.

ROBOT DESIGN
Spectrogram
A spectrogram is a representation of how the frequency content of a signal changes
with time. Time is displayed along the x − axis, frequency along the y − axis, and the
amount of energy in the signal at any given time and frequency is displayed as a level
of grey. During regions of silence, and at frequency regions where there is little energy,
the spectrogram appears white; dark regions indicate areas of energy.
According to the hypothesis at constant pressure, the frequency of the acoustic emis-
sions having maximum amplitude depends on the diameter of the hole. The frequency
decreases with increase in the hole diameter. Thus making it easier to characterise the
size of each leak with help of acoustic and pressure sensors.
Audiogram
The acoustic signals are generated at the location of the leaks. When the robot moves
across the leak, the amplitude of the acoustic signal recorded by the robot increases near
the hole. The spike in the signals indicates that there is an anomaly at that location.
2.3 Application
The robot is used to inspect interiors of the pipeline. It needs to be inserted inside the
pipeline whose integrity has to be examined. Here we describe the complete operation
procedure for the robot and its application. The operations have been divided into three
broad categories:
• Pre-Insertion
• Insertion
• Retrieval
Pre-Insertion: Here, we describe the tasks required to be performed before inserting

the robot inside the pipeline for inspections:
1. The microprocessor is switched on along with all the sensors.
2. The shells are closed by tightening the screw-able fasteners, with the rubber seal-
ings in place.
3. Helium gas is pumped inside the robot, from the valve provided until the required
pressure is obtained inside.

ROBOT DESIGN
4. Additional adhesives are used if required while pumping Helium.
5. The foam/neoprene covering the outer shells is fixed using adhesives as per the
pipe diameter.
Insertion: Here, we describe the insertion procedure for the robot inside the pipeline
for inspection:
1. The fluid flow inside the pipeline is diverted or stopped as convenient.
2. The valve to be used for insertion is opened.
3. The robot is carefully placed inside without dropping it.
4. The valve is closed and the fluid flow is resumed.
Retrieval: These are the last set of tasks to be performed on the field, prior to data
post-processing:
1. Wait for the robot to reach the retrieval point.
2. Robot is diverted to an empty pipe section.
3. The valve is opened.
4. The robot is extracted from the valve and cleaned by letting into a solvent.
5. It is then carefully opened and the memory card/chip is taken for post-processing.

TEST BED DESIGN
Test-bed Design
Test-bed for iGlobus is designed based upon the safety standards and guidelines pro-
vided by ASME B31.4.2002. Overall length of the pipeline is 25x21meters. Different
diameters of pipes are connected in loop. valves are fitted depending upon the flow
conditions of the fluid inside the pipeline. The capacity of the tank is calculated based
upon the total volume of pipeline, valves and fittings. A PIG (Pipeline Inspection
Gauge) launcher and receiver is used to launch and receive iGlobus. Upriser and down-
comers are given in the test loop to test the vertical movement of robot. Pump is selected
on the basis of flow rate and maximum pressure inside the pipe. Supports span for
pipeline is made based upon the calculations. Electronic components such as a pressure
gauge, pressure transmitter, non-intrusive flow meters and ultrasonic non-intrusive
pingers are used. The maximum working pressure inside the pipe is kept as 15 bar.
The artificial leakages and corrosion in the pipeline are to be simulated.
3.1 Overview
The test-bed of the pipeline is designed to test iGlobus. To check the performance of
iGlobus it should pass through the pipeline with the conditions being the same as that of
an oil and gas pipeline. Water is used as the working fluid to ensure safety. The overall
pipeline length is decided to be 25x21meters based availability of space. Pipe with
different diameter such as 10-inch, 16-inch, 24 inch are connected in loop. Pipe with a
minimum diameter of 10 inches is selected keeping in mind the size of the robot. The
diameter of the robot is 6 inches so it can flow through the pipe along with a foam
coating. End type of the pipeline is flat or bevel to provide ease in welding. The outer
layer of the pipe which goes underground is coated with dual fusion bonded epoxy or
three-layer polyethylene to protect against corrosion. PIG launcher and receiver are
used to launch and receive iGlobus to provide ease of application of the robot in actual
pipelines without the need for any special equipment. Valves are selected on the basis
that they should not slow down, stop or prevent the inspection and operation by the
robot in any way. Elbow joints with a radius of 1.5D are used in the turnings of 90
degrees and 45 degrees. Uprisers and down-comers are provided to test the vertical
movement of robot inside the pipeline. Various diameter of the pipe are connected using
reducers. Centrifugal pump is selected based upon its mechanism,

TEST BED DESIGN
performance and efficiency. Required flow rate and pressure head can be achieved using
this type of pump. The overall volume of the fluid inside the pipeline including valves
and fittings are calculated. Capacity of the tank is selected based upon the volume of
liquid inside the largest pipeline and safety volume inside the tank. Concrete supports
are designed for pipe to be supported. The distance between two supports are based
upon calculations. Guides and anchors are used to retain the pipes on the supports.
3.2 Detailed Description
This section is subdivided into three broad categories.
1. Mechanical design which deals with design of testbed, valve selection, pump
selection, launcher and receiver for iGlobus, tank capacity estimation, design of
supports for pipeline, etc.
2. Subsystems used and their descriptions/specifications.
3. Electronic layout deals with selection of electrical components such as Pres-

sure gauge, pressure transmitter, non-intrusive flow meters and ultrasonic non-
intrusive pingers, etc.
3.2.1 Mechanical Design
iGlobus should be tested inside the pipeline to check its work efficiency and perfor-
mance. Hence, a small setup of pipeline with different diameter of 10inch, 16inch and
24inch is constructed as a loop to check the functioning of the robot. the pipe loops of
different diameter are connected using reducer. 3way ball valve is provided at the
junction of two different diameter pipe to change the flow direction. Water is used as
the working fluid instead of oil or gas. Tank is used to store the water for supply in the
pipeline. The capacity of the tank is calculated based on the total volume of fluid inside
the largest diameter pipeline including joints, fittings, valves, launcher and receiver
along with the safe volume of fluid inside the tank. Tank outlet or discharge is connected
with eccentric reducer using flanges. Reducer is connected with 6” pipeline and check
valve are used to allow or stop the flow of liquid from tank to pump. Eccentric reducer is
used to connect pipes of different diameters. This 6-inch pipe is connected to the pump
using eccentric reducer with the help of flanges. Tank and

TEST BED DESIGN
pump are fixed in concrete support to restrain against vibration and moments. 2m
distance is maintained between tank and pump. Pump is connected with 10 inches pipe
using flanges. The 10 inches pipe is made of carbon steel or mild steel, IS3589 as grade
with flat or bevel end type. The outer layer of the pipe is coated with dual fusion bonded
epoxy or three-layer polyethylene to protect against corrosion. An unequal tee joint is
fitted 3.7m away from the pump. 6 inches pipeline from unequal tee joint is connected
with kicker valve to PIG launcher. PIG launcher is used to launch iGlobus. This pipeline
is used to pressurize the PIG launcher by changing the flow direction of fluid through this
pipe. 3.3m away from the unequal tee joint, a bypass valve is connected by which the
water flows through the pipes. 0.5m away from the bypass valve an equal tee joint is
attached. This equal tee joint is connected with PIG launcher and main valve is attached
in between PIG launcher and equal tee joint using flanges. Main valve allows the robot
to pass into the pipeline. On the other side of equal tee joint, a 10-inch pipe of 10 m
length is connected. It is connected with 3way ball valve to change the flow direction
to riser and downcomer or either to different diameter pipe. After 4m away from the
3way ball valve downcomers are attached which is 45degree elbow joint that attached
with 3m pipe which goes underground. The bend radius of 45degree elbow joint is 1.5D
which is around 159mm. the underground pipe are attached with another 45degree
elbow joint which brings the pipeline to surface which is also called upriser. Again, 4m
pipe is connected which is attached with 3way ball valve is also used to control the flow
direction. 3way ball valve is connected with 16-inch pipe using reducer. It forms a loop
of 16inch diameter pipeline and 24-inch diameter pipeline. 90degree elbow joint are
connected with 24-inch diameter pipe with bend radius of 914mm which changes the
direction to upriser and downcomer. 10 m pipe is connected with 3way ball valve in 10-
inch pipe which is fitted with equal tee joint further which is connected with main valve
and pig receiver using flanges. 0.5 m away from tee joint, a bypass valve is connected.
3.5 m away from the bypass valve, unequal tee joint is connected. PIG receiver and
unequal tee joint is connected with 6 inches pipe which is also connected with kicker
valve. This allows the water to change the flow direction which makes the robot to get in
trap in the PIG receiver. iGlobus should move inside the pipeline without slowdown or
stop, hence ball valves are used in the areas where iGlobus will be passing through.
Fluid flowing inside the pipeline is then allowed to flow inside the tank, so the water
can be reused
TEST BED DESIGN
without any wastage. Artificial leakages and corrosion with various diameters are made
in the pipeline to check the leak detection performance of robot. Maximum operating
pressure inside the pipeline is maintained at 15 bar and pipe thickness is selected based
upon the maximum operating pressure.
Pi ∗ D
T hickness(t ) =
2∗S
where, Pi = Maximum operating pressure
D = Outer diameter of the pipeline
S = Design factor * minimum Yield Strength (Design factor for carbon steel = 0.72)
For 10” pipe Minimum Thickness Required (t) = 1.0342 mm
For 6” pipe Minimum Thickness Required (t) = 0.6374 mm
This formula gives the minimum thickness of the pipe that can work in a given
operating pressure. But the thickness should be selected within the standard values.
According to the standard we select sch.30 pipe which has the thickness of 7.8 mm for
a 10 inches diameter pipe, sch.20 pipe which has the thickness of 7.8 mm for a 16 inches
diameter pipe, sch.10 pipe which has the thickness of 6.3 mm for a 24 inches diameter
pipe and sch.10s/10 pipe which has the thickness 3.404 mm for a 6 inches diameter pipe,
The selected wall thickness shall also be checked to ensure that the diameter to
thickness (D/t ) ratio does not exceed 100 inches inorder to avoid damage to pipe during
handling and transportation.
D/t = 273.05/7.8
= 35 < 100

TEST BED DESIGN
Figure 22: Test-Bed Layout
Elbow joints are used to change the direction of flow between two pipes. Elbow
joints are available with an angle of 22.5◦ , 45◦ and 90◦ . If the pipes are of same diameter
then normal elbows are used otherwise reducer elbows are used. These elbow joints are
fixed with pipe either by butt welding or socket welding. Long radius elbows 1.5D is
selected for 10 inch, 24 inch and 6 inch.
Figure 23: Elbow Joint

Equal tee joints and unequal tee joints are used in these pipelines. These tee joints
are selected based on the thickness of the pipes with which they connect. In equal tee
joint all the sides are connected with 10 inches pipes hence tee joint is selected with 10
inches diameter for all the sides. Unequal tee joints have two different diameters. One
side it is 10 inches diameter and on the adjacent sides it is 6 inches diameter because of
different pipe sizes.

TEST BED DESIGN
Figure 24: Unequal (Left) and Equal (Right) Tee Joints
3.2.2 Sub-Systems
Here we discuss in detail about the various sub-systems used in the test-bed:
Valves Fluid passing inside the pipeline should pass through the valves. Valves are
used to allow the flow to pass and to control it. There are many types of valves are
used in pipeline. These valves are used in both 10 inches, 16 inches and 6inch pipes
so they are selected for those dimensions of pipe. In a test-bed, valves are selected
based on the operating pressure it can hold and it should allow the robot to pass through
it without any slowdown or stopping. Hence ball valves are selected as the valves to be
used on the path of the robot. The ball inside the valve has dimensions about the same
as pipeline dimension. So the robot will move inside the valve without any obstruction.
The ball valve’s ease of operation, repair, and versatility lend it to extensive industrial
use, supporting pressure up to 1,000 bar (100 MPa; 15,000 psi) and temperatures up to
752◦ F (400◦ C). Gate valves are used on the pipes where the robot does not pass. 3way
ball valve are used at the junction where the direction of flow is controlled to different
diameter of pipe.
Figure 25: Ball Valve

TEST BED DESIGN
PIG Launcher and Receiver PIG Launcher and Receiver system is used in oil and
gas pipelines to launch and receive the PIG for cleaning the pipeline. iGlobus is also
launched and received using PIG launcher and receiver in test-bed. This eliminated
the need of manufacturing and installing a separate launching mechanism for launching
the robot. The same procedures used in launching and receiving the PIG can be used
to launch and receive the robot. It has kicker line through which the fluid enters into
the launcher. It is tightly closed using closure door. Drain valve is used to drain the
launcher and pressure relief valve is used to remove the pressure inside the launcher
and receiver before the closure door is opened. Pressure gauge is used to measure the
pressure inside the launcher before it’s launch. Pressure inside the launcher and pipeline
should be equalised before launching the robot into the pipeline. Vent valves are used
to remove the air inside the launcher. During the time while the launcher is filling with
fluid, the vent valve are allowed to open so the air inside the launcher are removed. PIG
receiver also has the same components as PIG launcher.
Figure 26: PIG Launcher
Reducers Reducers are used to connect the pipes with different dimensions. Reducer
allows for a change in pipe size to meet hydraulic flow requirements of the system, or
to adapt to existing piping of a different size. The length of the reduction is usually
equal to the average of the larger and smaller pipe diameter. There are two main types
of reducers.
• Concentric reducers
• Eccentric reducers

TEST BED DESIGN
Concentric reducers are mostly used in vertical pipe. In horizontal liquid piping, eccen-
tric reducer are commonly used. This configuration prevents air pocket accumulation at
the upstream end of the reducer. Concentric reducer is recommended for vertical inlet
(suction) piping or horizontal installations where there is no potential for air vapour ac-
cumulation. When the source of supply is above the pump, then the eccentric reducers
must be placed with the flat side down. When the source of supply is below the pump,
then the eccentric reducers must be placed with the flat side up. In case of long hori-
zontal pipe runs, air pockets are avoided by installing the eccentric reducer with the flat
side up. On the other places in pipeline the bottom of reducer is kept flat to keep the
bottom of pipe in the same level.
Figure 27: Eccentric Reducer
Flange A flange is a method of connecting pipes, valves, pumps and other equipment
to form a piping system. It also provides easy access for cleaning, inspection or mod-
ification. Flanges are usually welded or screwed. Flanged joints are made by bolting
together two flanges with a gasket between them to provide a seal. Pipe flanges are
manufactured in all the different materials like stainless steel, cast iron, aluminium,
brass, bronze, plastic, etc. but the most used material is forged carbon steel and have
machined surfaces. There are many types of flanges are available. Welding neck flanges
are used in test-bed. Flanges have many classes based on the pressure it can withstand.
They are 150, 300, 400, 600, 900, 1500 and 2500.

TEST BED DESIGN
The maximum pressure for class 150 flange is 285 psi. Maximum pressure for class
300 flange is 740 psi. When ordering a flange, it should mention with NPS, class,
schedule and material. ASTM A105 is a forged carbon steel flange, while A182 is a
forged stainless-steel flange.
Figure 28: Flange
Gaskets Flange gaskets are used to create a static seal between two flanges faces, at
various operating conditions, with varied pressure and temperature ratings. Material for
gaskets can be divided into three types. They are:
Figure 29: Gasket
1. Non-metallic type
2. Semi-metallic type
3. Metallic type

TEST BED DESIGN
Spiral wound metallic gasket or metallic ring type gaskets are used in the proposed
test-bed because of its effective sealing. It can hold high pressure and temperatures and
hence it is used in refinery operations.
Supports A pipe support is a designed element that transfers the load from the pipe
to the supporting structures. The load includes the weight of the pipe, the content
that the pipe carries, all the pipe fittings attached to pipe, and the pipe covering such as
insulation. The four main functions of a pipe support are to anchor, guide, absorb shock,
and support a specified load. Pipe supports used in high or low temperature applications
may contain insulation materials. The overall design configuration of a pipe support
assembly is dependent on the loading and operating conditions. Maximizing the number
of supports required which in turn reduce the total cost of erecting these pipe supports.
Safety of design is checked using maximum deflection and maximum bending stress.
The maximum bending stress should not exceed the allowable stress. Calculations are
done with various lengths and diameters of pipe to evaluate the results of deflection and
bending stress in pipeline. There must be extra supports where the load concentration is
high such as valves, elbows etc. The calculations are as follows:
Figure 30: Supports

TEST BED DESIGN
Maximum Bending Stress,
(0.0624W L2 + 0.1248Wc L)D

Sb = Nm−2
I
Maximum Deflection,
(5W L4 + 8Wc L3 )
y=
384E I
Where, W = Uniformly distributed weight of pipeline in N/m
Wc = Concentrated weight on pipeline in N
L = Span length in m
D = Outside diameter of pipe in m
d = Inside diameter of pipe in m
E = Modulus of elasticity of pipe in N/m2
I = Moment of Inertia of pipe in m4
Total weight = weight of pipe (Wp ) + weight of fluid (W f )
Wp = π /4(D2 − d 2 )∗ density of pipe material
W f = π /4d 2 ∗ density of fluid in N/m
(∴ since the pipe joints is not considered to carry flanges, it will not carry any
concentrated loads)
Total weight = Wp + W f
Moment of Inertia of pipe I = π (D64−d
4 4)
According to the calculation and pipeline model, 26 concrete supports are used in
this test bed.
As the pipe temperature changes from the installation condition to the operating
condition, it expands or contracts. In the general term, both expansion and contraction
are called thermal expansion. When a pipe expands it has the potential of generating
enormous force and stress in the system. However, if the piping is flexible enough, the
expansion can be absorbed without creating undue force or stress.
Given = 32.804 foot long of Carbon Steel pipe,
Maximum operating temperature = 60◦ C(140F )
Minimum operating temperature = 15◦ C(60F )

TEST BED DESIGN
Temperature at the time of installation = 26◦ C(80F )

Calculations are as follows :
60◦ C(140F ) 1.064" per 100 ft of Carbon Steel pipe
15◦ C(60F ) 0.448" per 100 ft of Carbon Steel pipe
Difference = 0.616" per 100 ft of Carbon Steel pipe for temperature 15◦ C(60F ) to
60◦ C(140F )
Therefore, 32.804 feet of pipe = 0.3367" = 8.382 mm
Due to the temperature change and stress acting on the pipeline makes the pipe to
expand or contract. It gives a longitudinal or compressive stress over the pipe. The
longitudinal or compressive stress should not exceed the allowable stress of the pipe,
else it may cause damages. Hoop’s stress is calculated to find the overall stress acting
on the pipeline and this stress should below or equal to the allowable stress.
Formula for Hoop’s Stress (Sh ),
P(D − t )
Sh =
2t
Longitudinal Compressive Stress (Sl ), Psi(MPa)
Sl = E α (T2 − T1 ) − γ Sh
Where, E = Modulus of Elasticity,

T1 = Minimum Temperature
T2 = Maximum Temperature
γ = Stress factor
α = Coefficient of thermal expansion

TEST BED DESIGN
STRESS VALUES
Parameters Units 10-inch pipe 16-inch pipe 24-inch pipe

Yield strength of carbon
steel MPa 414 414 414
Allowable stress MPa 138 138 138
Diameter of pipe mm 273.05 406.4 609.6
Thickness of pipe mm 7.8 7.9 6.35
Internal pressure MPa 1.5 1.5 1.5
Minimum Temperature 0C 26 26 26
Maximum Temperature 0C 60 60 60
Modulus of elasticity MPa 2.00E5 2.00E5 2.00E5

Coefficient of Thermal
Expansion m/m0c 1.00E-5 1.00E-5 1.00E-5
Stress factor 0.3 0.3 0.3
Hoop's stress MPa 25.504807692 38.317307692 58.004807692

Longitudinal or
Compressive stress MPa 6.03E1 5.65E1 5.06E1
Table 1: Stress Values
From the above calculation we get to know that the design is safe because the lon-
gitudinal stress and hoop’s stress is below the allowable stress.
Tank Tank is used to store the water that passes through the pipeline. The storage
capacity of the tank is based on the volume of fluid that over all bigger diameter
pipeline carries and safety volume of fluid inside the tank. Volume of fluid inside the
pipeline includes pipes, valves, elbow joints, tee joints, reducers, PIG launcher and
receiver, etc. are added with the safety volume of tank and gives the overall storage
capacity of tank. Volume of fluid every component and its quantity are mentioned below:

TEST BED DESIGN
VOLUME OF TANK
Parameters Units For 1 piece (liters) For total (liters)
10-inch pipeline 40m 585.2 2340.8
6-inch pipeline 20m 222.2 444.4

16-inch pipeline 6m 1300 780
24-inch pipeline 35m 2920 10220
Elbow joint 10 inch 90 deg 4 14.6 58.4
Elbow joint 24 inch 90 deg 2 420 840
Elbow joint 24 inch 45 deg 4 210 840
Equal tee joint 2 30 60

Unequal tee joint 2 12 24
10 inch valves 7 27 189
6 inch valves 3 7 21
16 inch valves 2 89.6 179.2
Reducer 6x10 3 7.7 23.1

Reducer 10x16 2 110.17 220.34
Reducer 16x24 2 410 820

PIG launcher 1 100 100
PIG Receiver 1 100 100
Total liter of water 17455.84

Tank capacity with safety
volume 20000 liter
Table 2: Net Volume of Tank
From the above calculations and safety volume of tank, 20000 litres storage capacity
water tank is used to store the fluid. The pipeline connection is like a loop so liquid
inside the pipeline is allowed to flow back inside the tank. Hence the fluid is reused
without any wastage.

TEST BED DESIGN
Pump Centrifugal pump is used to move the fluid inside the pipeline. The pump is
selected based on the maximum operating pressure and flow rate inside pipe. Pump
curve chart is used to identify the efficiency, flow rate, horse power, net positive suction
head required (NPSHr) etc. pressure and flow rate is controlled by controlling the rpm
of the pump. According to the performance of the pump impeller size can be changed.
Figure 31: Centrifugal Pump
Guides and Anchors Guides and anchors are providing axial movement of pipe while
restricting the angular and lateral movement of the pipe. It is supported on the concrete
support which is fixed. Number of guides and anchors are selected based on the number
of concrete supports used.
Figure 32: Guides

TEST BED DESIGN
3.2.3 Electronic Units
Pressure Transmitter Pressure transmitters are used to monitor the pressure at the in-
let and the outlet of the water pipelines. Pressure transmitters are used to monitor water
pressure at the various sections of pipelines. Pressure transducers convert pressure into
an analog voltage signal. The conversion of pressure into an electrical signal is achieved
by the physical deformation of strain gauges which are bonded into the diaphragm of the
pressure transducer and wired into a Wheatstone bridge configuration. Pressure applied
to the pressure transducer produces a deflection of the diaphragm which introduces
strain to the gauges. The strain will produce an electrical resistance change propor-
tional to the pressure. The output of the pressure transducer is 0-10 Volt. The pressure
transmitter recommended to be used is Baumer CTX Series Pressure Transducer (CTX
323.B26.0). Pressure transducers are located at the inlet and outlet of the 60-meter water
pipeline to continuously measure the fluid pressure inside the pipeline. The technical
specifications of the pressure transmitter are provided in Table 14 of ANNEXURE II.
Few features are discussed as follows:
• Excellent price/performance ratio.
• Compact and light weight.
• Stainless steel or brass construction.
Figure 33: Baumer CTX Series Pressure Transducer (CTX 323.B26.0) Pressure Trans-
mitter

TEST BED DESIGN
Pressure G a ug e Measurement of pressure is a key point for a network for an oper-

ating water pipeline. The pressure gauge in a test-bed is installed in easily accessible
places, so it is convenient to read and to maintain it in a proper working condition. For
maintaining our test bed, Bourdon tube pressure gauge is used, in which the primary
element is an elastic metal tube. As the pressure inside the tube increases, the oval tube
tends to become circular and this causes it to uncoil slightly. The pressure gauge is
placed near each valve to continuously monitor the pressure near the valves for safety
operation. The pressure gauge recommended to be used is : 4"Dial SS Diaphragm seal
pressure gauge (0 to 40 kgcm−2 ) 1/2"BSP bottom connection Model: 232.50.100 +
990.10 WIKA MAKE. Few key features of the pressure gauge given below:
• Excellent load-cycle stability and shock resistance.
• All stainless-steel construction.
• German Lloyd, Gosstandart and DVGW approval.
• Scale ranges up to 0 - 40 bar.
Figure 34: Pressure Gauge
Ultrasonic Flow Meter Ultrasonic flow meter is the non-intrusive flow meter, which
is designed in such a way that can detect in-situ flow rates from outside the pipework,
without any interruption to flow or process. Ultrasonic flow meter is selected so that it
does not restrain the flow of iGlobus inside the pipeline. Ultrasonic flow meter can mea-
sure almost all light homogeneous fluids including water. The flow meter recommended

TEST BED DESIGN
is BASE BCUFM2000 series Clamped on Ultrasonic Flow meter is a cost effective flow
measurement device for closed pipe liquid systems.
Figure 35: Ultrasonic Flowmeter
The Base BCUFM2000 series Wall-Mount Clamp-on ultrasonic flow meter provides
abundant capabilities for accurate liquid flow measurement from outside of a pipe. It
utilises technologies such as ultrasonic transmission/receiving, digital signal processing
and transit-time measurement. The proprietary signal quality tracking and self-adapting
techniques allow the system to optimally adapt to different pipe materials automatically.
The flow meters of the Base BCUFM2000 series is carefully designed so that their
operation is simple and easy. The unique clamp-on fixture design makes the installation
very simple and no special skills or tools are required. Due to the non-intrusive nature
of the clamp-on technology, there is no pressure drop, no moving parts, no leaks, no
risk of contamination, no risk of corrosion, no pressure dependency. Ultrasonic flow
meter will be located at 5 * pipe diameter at downstream side and 10 * pipe diameter
at upstream side within the pipeline. It is recommended to place a maximum of two
ultrasonic flow meter on the pipeline.
Type of Transducer
Type M: Medium Size Transducer (Magnetic)
For Pipe size 2"to 28" (DN 50 - DN 700 mm)
Turbine Flow Meter Turbine flow meter is an ideal device for flow rate measurement.
Turbine flow meter can endure high pressure and high and low temperature. It offers
a high turn-down with minimum uncertainty and excellent repeatability. Turbine flow
meter is placed before the iGlobus launching station to measure the volumetric total

TEST BED DESIGN
flow. The turbine flow meter selected to be used is : DN 250, LET 1000 Digital Turbine
Flowmeter for Water Application
Figure 36: Digital Turbine Flowmeter
Temperature Transmitter Temperature transmitters convert RTD sensor signal to 0 -

10V output signal. A temperature transmitter draws current from the remote DC power
supply in proportion to its sensor input. A thermocouple input transmitter will draw
4mA of current from a DC power supply when measuring the lowest temperature of the
process. As the temperature rises, the thermocouple transmitter will draw proportion-
ally more current, until it reaches 20mA. This 20mA corresponds to the thermocouple’s
highest sensed temperature. The transmitter’s internal signal circuitry (powered by the
portion of the 4-20mA) determines the temperature range that the output current signal
will represent. The temperature transmitter is inserted before the iGlobus launching
station through a hole of 6mm diameter.
Figure 37: Temperature Transmitter

TEST BED DESIGN
3.2.4 Data Acquisition
The sensors are mounted with a microcontroller called Sensor Nodes, which are con-
nected on a common Wi-Fi. All the data is logged in the server with the help of the
router. Which is then published in the GUI with the help of web application which can
then be accessed 24x7 by the user.
Figure 38: Data Acquisition Setup
The WSN is built of "nodes" – from a few to several hundreds or even thousands,
where each node is connected to one (or sometimes several) sensors. Each such sensor
network node has typically several parts: a radio transceiver with an internal antenna or
connection to an external antenna, a microcontroller, an electronic circuit for interfacing
with sensors and an energy source, usually a battery or an embedded form of energy
harvesting. A sensor node might vary in size from that of a shoebox down to the size of
a grain of dust, although functioning "motes" of genuine microscopic dimensions have
yet to be created. The cost of sensor nodes is similarly variable, ranging from a few to
hundreds of dollars, depending on the complexity of the individual sensor nodes. Size
and cost constraints on sensor nodes result in corresponding constraints on resources
such as energy, memory, computational speed and communications bandwidth. The
topology of the WSNs can vary from a simple star network to an advanced multi-hop

TEST BED DESIGN
wireless mesh network.
Procedure The Raspberry Pi is interfaced with sensors and reads data from it through
its GPIO pins. After this, the data is relayed to a web server. A brief outline of the
process:
Router with DHCP enabled. This is so that our Pi have a unique IP address to be
able to connect to it. Laptop and Pi are connected via a router, that now share a local
area network and can identify each other using their unique IP address.
To use SSH, we will need the IP address of the Pi. then we run IP Scanner command
to get the IP addresses of the devices on the network.
Steps for Raspberry Pi,
• Set up a Raspberry Pi for the first time
• Use SSH to remotely access the terminal window on another device
• Set up a webserver on an external computer to handle requests and store data in

MySQL database
• Connect a Sensor to the Raspberry Pi and record.
When the Raspberry Pi sends a request to the server, this file identifies and stores
the sent value in a variable. In case the value is above our set threshold, it sends a URL
request to another file on the same server, which sends a GCM push notification to all
registered users. The code is responsible for entering the values as well as the date and
time of the moment it was registered into the table in our database.
Setting up Server Bitnami Wamp stack loaded. While installing Bitnami WAMP
Stack, you will be prompted to create a MySQL “root” account by entering a password.
We configured Apache server to run .py files by editing httpd.conf file in the
apache install directory.
Login to PHPMyAdmin using the password with username root and create a new
database called log.
We can always display the data collected by sensor and distributed by the network
on the desktop application. Hence the whole process of transferring data was automated
from pipe sensor to graphical user interface.

TEST BED DESIGN
3.3 Overall Test-bed Layout
Figure 39: Test-Bed Layout
3.3.1 Testing Procedure
The overall length of test bed 25x21m. Fluid from the tank is allowed to flow inside
the pipe by opening the valve. The pressure at the suction side should be maintained
higher than the pressure inside the pump. These pressure differences make the pump to
work without any cavity or bubble formations. Pressure gauge is used at the suction
side as well as discharge side of the pump to measure the pressure. At suction side it
is used to measure the pressure that comes around the net positive suction required given
in pump curve. Flow control valve is provided near the pump which controls the flow
and pressure according to the requirements. Pressure gauge at the discharge is used to
measure the pressure exerted by the flow control valve. Flow meter is set near to the
pressure gauge to check the flow rate of liquid. Bypass valve is in open position while
kicker valve and mainline valve is in closed position. so that the liquid will flow through
the bypass valve in to the pipeline. iGlobus is launched using the PIG launcher. Vent
valve in launcher is opened to vent the launch trap to atmospheric pressure. When the
trap is completely vented, the closure door is opened and the robot is inserted inside the
trap, adjusting itself into the reduction previous to main valve with

TEST BED DESIGN
first cup of tool. The closure door is closed and secured. The air from the trap through
vent valve is purged by slowly opening kicker valve. When the purge is done, vent valve
is closed to allow pressure equalisation between trap and pipeline, then the kicker valve
is closed. The mainline valve is opened, then the kicker valve. iGlobus is now ready to
be launched. Bypass valve closed partially. This will increase the gas flow rate through
kicker valve and behind the robot. Bypass valve is closed slowly until the tool goes out
of the trap inserting itself into the pipe current indicated by the PIG-SIG passage
indicator. When the robot is launched from the trap and comes into the mainline, the
bypass valve is opened completely. The robot will flow into the pipeline receiving the
acoustic signals and visual data. Artificial leaks and corrosions in various dimensions
are made in the pipeline to check whether the robot is collecting the data or not. The
robot will move in vertical pipelines and moves underground using pipelines that buried
underground. Upriser and downcomers are used to make the test bed vertical under the
soil. The robot will cross all these setups and finally reach the receiver. The trap is
empty at atmospheric pressure. Bypass valve, vent valve and drain valve are open.
Mainline valve and return valve are closed. The closure door is closed and secured. The
trap is drained, drain valve is closed and return valve is opened slowly. Once it is drained,
the pressure is made to equalise in the trap closing vent valve with the kicker valve
opened. With return valve still opened, mainline valve is opened. The trap is now ready
to receive the robot. When the robot arrives, this will stop between mainline valve and
the entrance tee piece of the trap. Bypass valve is closed partially. This will force
iGlobus to introduce itself into the trap increasing the flow through return valve. After
iGlobus is in the trap, shown by the PIG-SIG passage indicator, the bypass valve is opened
and the mainline valve and return valve are closed. The vent valve and drain valve are
opened to vent the trap to atmospheric pressure. After the trap is vented and drained
with vent valve and drain valve opened, the closure door is opened and the robot is
removed from receiver. The closure door is closed and secured. The fluid will flow to
the tank and reused again in pipeline.

ANNEXURE I
ANNEXURE I
Materials Acoustic Impedance

(g/cm2 sec ∗ 105 )
Aluminium 17.10
Copper 41.61
Iron 45.43
Cast Iron 37.44
Steel 310 45.45
Titanium 27.32
Polyoxymethylene (or) Acetal 2.9
Table 3: Acoustic Impedance of Common Materials
Parameters Properties
Density 2770 kg/m3
Coefficient of Thermal Expansion 2.36 ∗ 10−5K −1
Young’s Modulus 7.1 ∗ 1010 Pa
Poisson’s Ratio 0.33
Bulk Modulus 6.9608 ∗ 1010 Pa
Shear Modulus 2.6692 ∗ 1010 Pa
Tensile Yield Strength 5.05 ∗ 108 Pa
Compressive Yield Strength 5.05 ∗ 108 Pa
Tensile Ultimate Strength 5.72 ∗ 108 Pa
Melting temperature 477◦ C
Thermal Conductivity 196w/m − k
Specific Heat Capacity 714.8J/(kg − k)
Table 4: Properties of Aluminium alloy 7075-T6

ANNEXURE I
Parameters Specifications
Microprocessor Broadcom BCM2837 64bit Quad Core Processor
Processor Operating Voltage 3.3V
Raw Voltage input 5V, 2A Power Source
Maximum current through each I/O pin 16mA
Maximum total current drawn from all I/O pins 54mA
Flash Memory (Operating System) 16 Gbytes SD memory card
Internal RAM 1 Gbytes DDR2
Clock Frequency 1.2GHz
GPU Dual Core Video Core IV r Multimedia
Co-Processor. Provides Open GLES 2.0,
hardware-accelerated Open VG,
and 1080p30 H.264 high- profile decode.
Capable of 1Gpixel/s, 1.5Gtexel/s or
24GFLOPs with texture filtering and
DMA infrastructure.
Ethernet 10/100 Ethernet
Wireless Connectivity BCM43143 (802.11 b/g/n)
Wireless LAN and Bluetooth 4.1)
Operating Temperature −40◦ C to +85◦ C
Table 5: Raspberry Pi 3 General Specifications
Size 36 mm x 33 mm x 9 mm
Weight 15 grams Single Power
Supply 4.5 to 5.5 V Communication Interface
Serial RS − 232 & T T L
Table 6: IMU General Specifications
Peak sensitivity, Ref V /(m/s) 82 dB Peak
sensitivity, Ref V /µ bar -39 dB Operating
frequency range 200 - 450 kH z Resonant
frequency, Ref V /(m/s) 350 kH z Directionality
± 1.5 dB
Gain 26 dB
Preamp DC power 5-7 V
Table 7: ISPK30I General Specifications

ANNEXURE I
Size 25mm ∗ 23mm ∗ 9mm
Sensor Resolution 3280 ∗ 2464 Pixels
Video mode 1080 p30, 720 p60, 640 ∗ 480 p90
Sensor Omni Vision OV 5647
Sensor Image Area 3.76 ∗ 2.74mm
Pixel Size 1.4µ m ∗ 1.4µ m
S/N ratio 36dB
Dynamic range 67dB at 8x gain
Sensitivity 680mV /lux − sec
Focal length 360mm ± 0.01
Table 8: PiCam General Specifications
Accuracy ±2 p pm from 0◦ C to + 40◦ C
Accuracy ±3.5 p pm from −40◦ C to + 85◦ C
Interface Fast (400kH z)I2C Interface
Operating voltage 3.3V
Digital Temp Sensor Output ±3◦ C Accuracy
Table 9: MAX6675 General Specifications
Operating voltage (V DC) 3 to 5.5 V
Temperature range (degree C) 0 to 1024
Cold junction temperature range -20 to +80
Resolution 0.25◦ C
Dimension 6 ∗ 5 ∗ 3cm
Table 10: RTC (DS3231) General Specifications
Electrical Characteristics of Cell

Ratings Nominal Capacity (at 150mA, +20◦ C, 2.0V cut-off)
12.4Ah Open circuit voltage (at +20◦ C)
3.2V Nominal Voltage (Under 1mA at +20◦ C)
3.0V Nominal Energy (at 150mA, +20◦ C, 2V cut off)
35W h
Pulse Capacity Up to 8.0A
Table 11: M 20 EX SV General Specifications
Operating Conditions Ranges

Operating Temperature Range −40◦ C to +72◦ C
Storage Temperature Recommended +30◦ C max
Allowable −55◦ C to +90◦ C
Table 12: M 20 EX SV Operating Conditions

ANNEXURE I
Parameter Unit
s s
Outer
Diameter
with
neoprene m 0.16 0.17 0.18 0.19 0.2 0.21 0.22 0.23 0.24 0.25
Outer
Diameter m 0.1524 0.1524 0.1524 0.1524 0.1524 0.1524 0.1524 0.1524 0.1524 0.1524
Inner
Diameter m 0.128 0.128 0.128 0.128 0.128 0.128 0.128 0.128 0.128 0.128
Cross
section
area m2 0.080384 0.090746 0.101736 0.113354 0.1256 0.138474 0.151976 0.166106 0.180864 0.19625
volume of 0.002143 0.002571 0.003052 0.003589 0.004186 0.004846 0.005572 0.006367 0.007234 0.008177
ball m3 573 137 08 543 667 59 453 397 56 083
Volume of 0.000754 0.000754 0.000754 0.000754 0.000754 0.000754 0.000754 0.000754 0.000754 0.000754
material m3 884 884 884 884 884 884 884 884 884 884
Density of kg/
petrol m3 719 719 719 719 719 719 719 719 719 719
Density of kg/
diesel m3 832 832 832 832 832 832 832 832 832 832
Density of kg/
kerosene m3 810 810 810 810 810 810 810 810 810 810
Density of kg/
atf m3 840 840 840 840 840 840 840 840 840 840
Density of kg/
crude oil m3 870 870 870 870 870 870 870 870 870 870
Density of
clear kg/
naptha m3 1100 1100 1100 1100 1100 1100 1100 1100 1100 1100
Density of kg/
Water m3 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000
Pipe
diameter m 0.26525 0.26525 0.26525 0.26525 0.26525 0.26525 0.26525 0.26525 0.26525 0.26525
Velocity of
fluid (V) m/s 2 2 2 2 2 2 2 2 2 2
Relative
velocity of
fluid wrt
sphere m/s 2 2 2 2 2 2 2 2 2 2
Drag
coefficient 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
Density of kg/
sphere m3 2810 2810 2810 2810 2810 2810 2810 2810 2810 2810
payload
mass kg 1.1 1.1 1.1 1.1 1.1 1.1 1.1 1.1 1.1 1.1
Mass of 2.121224 2.121224 2.121224 2.121224 2.121224 2.121224 2.121224 2.121224 2.121224 2.121224
the sphere kg 512 512 512 512 512 512 512 512 512 512
Mass of
the
Neoprene
rubber kg 0.00467 0.01152 0.01923 0.02784 0.03741 0.04799 0.05962 0.07236 0.08626 0.10136
Payload
mass +
sphere
mass +
polyuretha 3.225894 3.232744 3.240454 3.249064 3.258634 3.269214 3.280844 3.293584 3.307484 3.322584
ne kg 512 512 512 512 512 512 512 512 512 512
Kinematic
viscosity m2/
(petrol) s 8.00E-7 8.00E-7 8.00E-7 8.00E-7 8.00E-7 8.00E-7 8.00E-7 8.00E-7 8.00E-7 8.00E-7
Kinematic
viscosity m2/
(diesel) s 5.00E-6 5.00E-6 5.00E-6 5.00E-6 5.00E-6 5.00E-6 5.00E-6 5.00E-6 5.00E-6 5.00E-6

ANNEXURE I
kinematic
viscosity m2/
(kerosene) s 2.71E-6 2.71E-6 2.71E-6 2.71E-6 2.71E-6 2.71E-6 2.71E-6 2.71E-6 2.71E-6 2.71E-6
kinematic
viscosity m2/
(atf) s 7.90E-6 7.90E-6 7.90E-6 7.90E-6 7.90E-6 7.90E-6 7.90E-6 7.90E-6 7.90E-6 7.90E-6
kinematic
viscosity m2/
(crude oil) s 2.90E-06 2.90E-06 2.90E-06 2.90E-06 2.90E-06 2.90E-06 2.90E-06 2.90E-06 2.90E-06 2.90E-06
kinematic
viscosity
(clear m2/
naptha) s 9.69E-07 9.69E-07 9.69E-07 9.69E-07 9.69E-07 9.69E-07 9.69E-07 9.69E-07 9.69E-07 9.69E-07
kinematic
viscosity m2/
(Water) s 3.35E-6 3.35E-6 3.35E-6 3.35E-6 3.35E-6 3.35E-6 3.35E-6 3.35E-6 3.35E-6 3.35E-6
Dynamic
viscosity m2/
(petrol) s 5.75E-4 5.75E-4 5.75E-4 5.75E-4 5.75E-4 5.75E-4 5.75E-4 5.75E-4 5.75E-4 5.75E-4
Dynamic
viscosity m2/
(Diesel) s 4.16E-3 4.16E-3 4.16E-3 4.16E-3 4.16E-3 4.16E-3 4.16E-3 4.16E-3 4.16E-3 4.16E-3
Dynamic
viscosity m2/
(kerosene) s 2.20E-3 2.20E-3 2.20E-3 2.20E-3 2.20E-3 2.20E-3 2.20E-3 2.20E-3 2.20E-3 2.20E-3
Dynamic
viscosity m2/
(atf) s 6.64E-3 6.64E-3 6.64E-3 6.64E-3 6.64E-3 6.64E-3 6.64E-3 6.64E-3 6.64E-3 6.64E-3
Dynamic
viscosity m2/
(crude oil) s 2.52E-03 2.52E-03 2.52E-03 2.52E-03 2.52E-03 2.52E-03 2.52E-03 2.52E-03 2.52E-03 2.52E-03
Dynamic
viscosity
(clear m2/
naptha) s 1.07E-03 1.07E-03 1.07E-03 1.07E-03 1.07E-03 1.07E-03 1.07E-03 1.07E-03 1.07E-03 1.07E-03
Dynamic
viscosity m2/
(Water) s 3.35E-3 3.35E-3 3.35E-3 3.35E-3 3.35E-3 3.35E-3 3.35E-3 3.35E-3 3.35E-3 3.35E-3
Reynolds
number
(petrol) 6.63E05 6.63E05 6.63E05 6.63E05 6.63E05 6.63E05 6.63E05 6.63E05 6.63E05 6.63E05
Reynolds
number
(Diesel) 1.06E05 1.06E05 1.06E05 1.06E05 1.06E05 1.06E05 1.06E05 1.06E05 1.06E05 1.06E05
Reynolds
number
(kerosene) 1.96E05 1.96E05 1.96E05 1.96E05 1.96E05 1.96E05 1.96E05 1.96E05 1.96E05 1.96E05
Reynolds
number
(atf) 6.72E04 6.72E04 6.72E04 6.72E04 6.72E04 6.72E04 6.72E04 6.72E04 6.72E04 6.72E04
Reynolds
number
(crude oil) 1.83E05 1.83E05 1.83E05 1.83E05 1.83E05 1.83E05 1.83E05 1.83E05 1.83E05 1.83E05
Reynolds
number
(clear
naptha) 4.13E06 4.13E06 4.13E06 4.13E06 4.13E06 4.13E06 4.13E06 4.13E06 4.13E06 4.13E06
Reynolds
number
(Water) 1.58E05 1.58E05 1.58E05 1.58E05 1.58E05 1.58E05 1.58E05 1.58E05 1.58E05 1.58E05
Relative
density kg/
(petrol) m3 2.09E03 2.09E03 2.09E03 2.09E03 2.09E03 2.09E03 2.09E03 2.09E03 2.09E03 2.09E03
Relative
density kg/
(diesel) m3 1.98E03 1.98E03 1.98E03 1.98E03 1.98E03 1.98E03 1.98E03 1.98E03 1.98E03 1.98E03
Relative
density kg/
(kerosene) m3 2.00E03 2.00E03 2.00E03 2.00E03 2.00E03 2.00E03 2.00E03 2.00E03 2.00E03 2.00E03

ANNEXURE I
Relative
density kg/
(atf) m3 1.97E03 1.97E03 1.97E03 1.97E03 1.97E03 1.97E03 1.97E03 1.97E03 1.97E03 1.97E03
Relative
density kg/
(crude oil) m3 1.94E03 1.94E03 1.94E03 1.94E03 1.94E03 1.94E03 1.94E03 1.94E03 1.94E03 1.94E03
Relative
density
(clear kg/
naptha) m3 1.71E03 1.71E03 1.71E03 1.71E03 1.71E03 1.71E03 1.71E03 1.71E03 1.71E03 1.71E03
Relative
density kg/
(Water) m3 1.81E03 1.81E03 1.81E03 1.81E03 1.81E03 1.81E03 1.81E03 1.81E03 1.81E03 1.81E03
Coefficient
of rolling
friction 1.00E-03 1.00E-03 1.00E-03 1.00E-03 1.00E-03 1.00E-03 1.00E-03 1.00E-03 1.00E-03 1.00E-03
Rolling
friction
force
(petrol) N 4.40E-02 5.27E-02 6.26E-02 7.36E-02 8.59E-02 9.94E-02 1.14E-01 1.31E-01 1.48E-01 1.68E-01
Rolling
friction
force
(diesel) N 4.16E-02 4.99E-02 5.92E-02 6.97E-02 8.12E-02 9.40E-02 1.08E-01 1.24E-01 1.40E-01 1.59E-01
Rolling
friction
force
(kerosene) N 4.21E-02 5.04E-02 5.99E-02 7.04E-02 8.21E-02 9.51E-02 1.09E-01 1.25E-01 1.42E-01 1.60E-01
Rolling
friction
force (atf) N 4.14E-02 4.97E-02 5.90E-02 6.94E-02 8.09E-02 9.37E-02 1.08E-01 1.23E-01 1.40E-01 1.58E-01
Rolling
friction
force
(crude oil) N 4.08E-02 4.89E-02 5.81E-02 6.83E-02 7.97E-02 9.22E-02 1.06E-01 1.21E-01 1.38E-01 1.56E-01
Rolling
friction
force (clear
naptha) N 3.60E-02 4.31E-02 5.12E-02 6.02E-02 7.02E-02 8.13E-02 9.35E-02 1.07E-01 1.21E-01 1.37E-01
Rolling
friction
force
(Water) N 3.81E-02 4.57E-02 5.42E-02 6.37E-02 7.43E-02 8.61E-02 9.89E-02 1.13E-01 1.28E-01 1.45E-01
Drag force
(petrol) N 1.16E01 1.30E01 1.46E01 1.63E01 1.81E01 1.99E01 2.19E01 2.39E01 2.60E01 2.82E01
Drag force
(diesel) N 1.34E01 1.51E01 1.69E01 1.89E01 2.09E01 2.30E01 2.53E01 2.76E01 3.01E01 3.27E01
Drag force
(kerosene) N 1.30E01 1.47E01 1.65E01 1.84E01 2.03E01 2.24E01 2.46E01 2.69E01 2.93E01 3.18E01
Drag force
(atf) N 1.35E01 1.52E01 1.71E01 1.90E01 2.11E01 2.33E01 2.55E01 2.79E01 3.04E01 3.30E01
Drag force
(crude oil) N 1.40E01 1.58E01 1.77E01 1.97E01 2.19E01 2.41E01 2.64E01 2.89E01 3.15E01 3.41E01
Drag force
(clear
naptha) N 1.77E01 2.00E01 2.24E01 2.49E01 2.76E01 3.05E01 3.34E01 3.65E01 3.98E01 4.32E01
Drag force
(Water) N 1.61E01 1.81E01 2.03E01 2.27E01 2.51E01 2.77E01 3.04E01 3.32E01 3.62E01 3.92E01
Terminal
velocity
(petrol) m/s 1.88E00 1.87E00 1.87E00 1.87E00 1.86E00 1.86E00 1.86E00 1.85E00 1.85E00 1.85E00
Terminal
velocity
(diesel) m/s 1.89E00 1.89E00 1.88E00 1.88E00 1.88E00 1.87E00 1.87E00 1.87E00 1.86E00 1.86E00
Terminal
velocity
(kerosene) m/s 1.89E00 1.88E00 1.88E00 1.88E00 1.87E00 1.87E00 1.87E00 1.86E00 1.86E00 1.86E00
Terminal
velocity
(atf) m/s 1.89E00 1.89E00 1.88E00 1.88E00 1.88E00 1.87E00 1.87E00 1.87E00 1.86E00 1.86E00

ANNEXURE I
Terminal
velocity
(crude oil) m/s 1.89E00 1.89E00 1.89E00 1.88E00 1.88E00 1.88E00 1.87E00 1.87E00 1.87E00 1.86E00
Terminal
velocity
(clear
naptha) m/s 1.91E00 1.91E00 1.90E00 1.90E00 1.90E00 1.90E00 1.89E00 1.89E00 1.89E00 1.89E00
Terminal
velocity
(Water) m/s 1.90E00 1.90E00 1.90E00 1.89E00 1.89E00 1.89E00 1.89E00 1.88E00 1.88E00 1.88E00
Downward
force due 31.64602 31.71322 31.78885 31.87332 31.96720 32.07099 32.18508 32.31006 32.44642 32.59455
to gravity N 5161 3661 8761 2861 4561 4361 4661 4061 3061 4061
Buoyancy
force 15.119458 18.135229 21.527510 25.318449 29.530192 34.184889 39.304686 44.911731 51.028173 57.676157
(petrol) N 714 653 551 052 8 44 617 975 158 812
Buoyancy
force 17.495674 20.985411 24.910832 29.297565 34.171238 39.557479 45.481918 51.970182 59.047899
(diesel) N 061 782 794 523 4 853 31 202 955 66.7407
Buoyancy
force 17.033048 20.430509 24.252132 28.522870 38.511488 44.279271 50.595970 57.486537 64.975921
(kerosene) N 064 067 888 281 33.267672 799 432 653 216 875
Buoyancy 17.663901 21.187194 25.150360 29.579272 39.937840 45.919244 52.469895 59.615668 67.382437
force (atf) N 696 588 032 884 34.499808 236 448 492 224 5
Buoyancy
force 18.294755 21.943880 26.048587 30.635675 41.364191 47.559217 54.343820 61.744799 69.788953
(crude oil) N 328 109 176 487 35.731944 673 464 331 232 125
Buoyancy
force (clear 23.131299 27.745135 32.934995 38.734762 52.299552 60.132343 68.710577 78.068136 88.238906
naptha) N 84 77 28 11 45.17832 69 92 43 96 25
Buoyancy
force 21.028454 25.222850 29.940904 35.213420 47.545047 54.665767 62.464161 70.971033 80.217187
(Water) N 4 7 8 1 41.0712 9 2 3 6 5
Drag
force+Buoy
ancy force
Drag
force+Buoy
ancy force
(diesel) 3.09E01 3.61E01 4.18E01 4.82E01 5.51E01 6.26E01 7.08E01 7.96E01 8.91E01 9.94E01
Drag
force+Buoy
ancy force
Drag
force+Buoy
ancy force
(atf) 3.12E01 3.64E01 4.22E01 4.86E01 5.56E01 6.32E01 7.15E01 8.04E01 9.00E01 1.00E02
Drag
force+Buoy
ancy force
Drag
force+Buoy
ancy force
(clear
Drag
force+Buoy
ancy force
Gravitation
al
force+Drag
Gravitation
al
force+Drag
(diesel) 4.50E01 4.68E01 4.87E01 5.07E01 5.29E01 5.51E01 5.75E01 6.00E01 6.25E01 6.53E01

ANNEXURE I
Gravitation
al
force+Drag
Gravitation
al
force+Drag
(atf) 4.52E01 4.70E01 4.89E01 5.09E01 5.31E01 5.53E01 5.77E01 6.02E01 6.28E01 6.56E01
Gravitation
al
force+Drag
Gravitation
al
force+Drag
(clear
Gravitation
al
force+Drag
Table 13: Buoyancy Calculation

ANNEXURE II
ANNEXURE II
Parameters Values
Pressure Range 0-25 bar
Output 0-10 Volt
Accuracy 0.5 % FS
Supply Voltage 10-30 VDC
Parameters IP 65
Electrical Connection 4- Pin Din Connector
Process Connection G1/4"BSP(M)
Operating Temperature 40 - 100◦ C
Material Process Connection SS 1.4404 AISI 316L
Housing Material SS 1.4301 AISI 304
Table 14: Technical Specifications of Pressure Transmitter
Parameters Values
Type Any Homogeneous and Sonically Conductive Liquid
Temperature Range -40◦ C to 160◦ C
Turbidity Less than 10000ppm (10g/I)
Type Carbon Steel, Stainless Steel, Cast Iron, Copper etc.,
Diameter DN15 to DN 6000
Minimum Straight Pipe Upstream Side = 10 X pipe diameter
Downstream Side = 5 X pipe diameter
Repeatability 0. 2 %
Accuracy ± 1%
Linearity 0.5%
Response Time 1-9999 second, user – configurable
Velocity 0 – 32 m/s
Measure Principle Transit Time
Analog Output 4-20mA
Input 5 Channel Analog 4-20mA
Display 2 x 20 English Letters LCD display
Power Supply DC8 to 36V
Power Less than 2w
Temp. Transmitter -30◦ C to 80◦ C
Temp. Transducer -40◦ C to 160◦ C
Table 15: Technical Specifications of Ultrasonic Flowmeter

ANNEXURE II
Parameters Values
Media Water
Line size DN250
Process Flow range 100 - 1000 m3 /hr
Process Pressure 8 - 20 bar
Process Temperature 26 - 90 degree C
Accuracy ±1%
Display Digital
Display Mount Integral
Display Content Current and total flow
MOC SS316
Rotor SS316
Power Supply 24V DC
Output 4-20 mA
End Connection Flange End
Table 16: Technical Specifications of Digital Turbine Flowmeter
Parameters Values
Element Pt 100 Ohms at 0◦ C, 3wire, Simplex
Standard DIN 43710 Class "A"
Insulation MGO Filled
Sheath SS 316, 6mm Dia
Leads Silver plated copper Wire Teflon Insulation
Transmitter Head Mount Transmitter Novas 4-20mA
Output 0 - 10 V DC
Head Pressure Die cast Aluminum,
weather proof to IP65 with 3/4" ET (F)
Cable gland entry 1 No. screwed Cap and Chain
Accessory SS 1/4" BSP (M) Adjustable Adaptor
Length below Head 300mm
Temperature 0 – 100◦ C
Thermowell Fabricated Thermowell
from SS316 5/8” (16± 0.5mm Dia.)
Process Connection adaptor welded on Top
Process Connection 1/2" NPT (M)
Total Immersion Length 240mm (Length below Process Connection)
Table 17: Technical Specifications of Temperature Transmitter

ANNEXURE II
MAXIMUM BENDING STRESS AND MAXIMUM DEFLECTION CALCULATION

Values (10inch Values (16inch
Parameters Units pipe) pipe) Values(24inch)
Outer diameter m 0.27305 0.4064 0.6096
Inner diameter m 0.25745 0.3908 0.594
Modulus of elasticity N/m2 2.00E11 2.00E11 2.00E11
Moment of Inertia m4 5.72E-5 1.94E-4 6.67E-4
Density of pipe material kg/m3 7.85E3 7.85E3 7.85E3
Density of water kg/m3 1.00E3 1.00E3 1.00E3
Weight of pipe kg-m 5.10E1 7.66E1 1.16E2
Weight of fluid kg-m 5.20E1 1.20E2 2.77E2
Total weight of pipe Kg-m 1.03E2 1.97E2 3.93E2
Length 1 m 5 5 5
Length 2 m 6 6 6
Length 3 m 7 7 7
Length 4 m 8 8 8
Length 5 m 9 9 9
Length 6 m 10 10 10
Length 7 m 11 11 11
Length 8 m 12 12 12
Maximum bending stress
L1 N/m2 7.53E6 6.30E6 5.49E6
L2 N/m2 1.08E7 9.07E6 7.90E6
L3 N/m2 1.48E7 1.23E7 1.08E7
L4 N/m2 1.93E7 1.61E7 1.40E7
L5 N/m2 2.44E7 2.04E7 1.78E7
L6 N/m2 3.01E7 2.52E7 2.20E7
L7 N/m2 3.64E7 3.05E7 2.66E7
L8 N/m2 4.34E7 3.63E7 3.16E7
Maximum deflection L1 m 7.19E-4 4.04E-4 2.35E-4
Table 18: Maximum Bending Stress and Maximum Deflection Calculation
Design Report 67 Solinas Integrity Pvt. Ltd

ANNEXURE II
Test Bed
Parameters Units Values
10-inch pipeline m 40
Outer diameter of 10-inch pipe m 0.27305
Inner diameter of 10-inch pipe m 0.25745
Thickness of 10-inch pipe m 7.80E-3
Outer diameter of 6-inch pipe m 0.16828
Inner diameter of 6-inch pipe m 0.16116
Thickness of 6-inch pipe m 7.11E-3
Length of Pipe m 10 to 12
Material Carbon Steel Or Mild Steel
Grade API 5L X-60
End Type Flat or Bevel
Bend Radius of Elbow Joint 1.5D
Cross Section Area of Pipe material m2 5.85E-2
Volume Inside 10-inch pipe for 10m m3 0.5203 585.2 liter
Volume inside 6-inch pipe for 10m m3 0.2222 222.2 liter
Liquid Type Water
Density of Water Kg/m3 1000
Velocity of Fluid m/s 2
Maximum Pressure bar 15
Kinematic viscosity of Water m2/s 6.59E-07
Dynamic viscosity of Water m2/s 6.54E-04
Reynolds Number 4.58E05
Flow Rate (Q) m3/s 0.117
volume of 10inch pipe for 60m m3 3.1218 3511.2 liter
volume of 6inch pipe for 20m m3 1.3332 444.4 liter
Equal tee joint (volume) m3 0.03001 30 liter
Unequal tee joint (volume) m3 0.01204 12 liter
Valves Ball valve and butterfly valve
Volume in 10-inch valve m3 0.0279 27 liter
Volume in 6-inch valve m3 0.00712 7 liter
Volume of 16-inch valve m3 0.0896 89.6 liter
Pig launcher (length) m 1.98 100 liter
Pig receiver (length) m 1.98 100 liter
Elbow joint 10 inch 90 deg m3 0.0146 14.6 liter
Elbow joint 24 inch 90 deg m3 0.42 420 liter
Elbow joint 24 inch 45 deg m3 0.21 210 liter
Reducer 6x10 m3 0.0077 7.7 liter
Reducer 10x16 m3 0.11017 110.17 liter
Reducer 16x24 m3 0.41 410 liter
Table 19: Test Bed Properties

ANNEXURE II
iGLOBUS TEST-LOOP BILL OF MATERIALS

NO. OF
NO. OF
10
PIPES
PART NAME DESCRIPTION GRADE SIZE QUANTITY UNIT METER
TO BE
LENGTH
COATED
PIPE
Carbon Steel IS3589/
Pipe seamless pipe/ Bevel SCH 30 10 INCH Diameter 63 meter 7 2
end/.

Pipe SCH 10 6 INCH Diameter 20 meter 2 -
seamless pipe/ Bevel end

Pipe seamless pipe/ Bevel SCH 20 16 INCH Diameter 29 meter 3 2
end/
Pipe SCH 20 24 INCH Diameter 26 meter 3 2
seamless pipe/ Bevel end
10 INCH Diameter,
Elbow joint SCH 30 Bend Radius 1.5D, 6 each
Bevel end
Bend Angle 90 DEG
6 INCH Diameter,
Bevel end
Bend Angle 90 DEG
24 INCH Diameter,
Bevel end
Bend Angle 90 DEG
10 INCH Diameter,
Elbow joint SCH 30 Bend Radius 1.5D, 4 each 2
Bevel end
Bend Angle 45 DEG
16 INCH Diameter,
Bevel end
Bend Angle 45 DEG
24 INCH Diameter,
Bevel end
Bend Angle 45 DEG
Equal tee Carbon Steel IS3589/ 10 INCH Diameter

SCH 30 2 each
joint Bevel end Connection
Unequal tee Carbon Steel IS3589/ 10 INCH AND 6 INCH

SCH 30 2 each
joint Bevel end Diameter Connection
ANNEXURE II
ASME B16.34, ASME
2-Way Ball B16.10, ASME B16.5/ API CLASS
10 INCH Diameter 2 each
Valve 598, API 607/ Carbon 150
Steel
ASME B16.34, ASME
B16.10, ASME B16.5/ API CLASS
Gate Valve 10 INCH Diameter 2 each
598, API 607/ Carbon 150
Steel.
ASME B16.34, ASME
B16.10, ASME B16.5/ API CLASS
Gate Valve 6 INCH Diameter 3 each
598, API 607/ Carbon 150
Steel
ASME B16.34, ASME
3-Way Ball B16.10, ASME B16.5/ API CLASS
Valve 598, API 607/ Carbon 150
Steel.
ASME B16.34, ASME
3-Way Ball B16.10, ASME B16.5/ API
CLASS150 16 INCH Diameter 2 each
Valve 598, API 607/ Carbon
Steel
ASME B16.34, ASME
Flow control B16.10, ASME B16.5/ API CLASS
Valve/ Gate 10 INCH DIAMETER 1 each
valve 598, API 607/ Carbon 150
Steel
Carbon Steel/ Eccentric
Reducer SCH 20 16X10 2 each
Reducer
reducer.
Reducer
Reducer SCH30 8X6 1 each
Reducer
Slip-on weld Carbon Steel/ ASME CLASS

flanges B16.5/ ASTM A105 150


Spiral Wound
Metallic
ASME B16.20/ Thickness For 10 INCH, 16 INCH
Gasket Or 18 each
3.2mm Flanges
Metallic Ring
Type Gasket
1-meter height from
Supports CONCRETE SUPPORT. NA 38 each
the ground surface
Maximum 20 bar
Pump CENTRIFUGAL PUMP NA 1 each
pressure
18000 or 20000 Liter

Tank 1 each
Storage
ANNEXURE II
Guides and Based on the pipe
23 each
anchors Diameter
ASME B31.4/ 6 INCH
Kicker line/ closure door/
12 INCH Diameter/ 2
PIG Launcher pressure relief valve/ 1 each
Meter Length
vent valve/ pressure
gauge connection
ASME B31.4/ 6 INCH
Kicker line/ closure door/
12 INCH Diameter/ 2
PIG Receiver pressure relief valve/ 1 each
Meter Length
vent valve/ pressure
gauge connection
Figure 20: Bill of Materials for Test-Bed
Figure 21: Bill of Materials of Tools

iGlobusReport 17tex1

Uploaded by

Copyright:

Available Formats

iGlobusReport 17tex1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

iGlobusReport 17tex1

Uploaded by

Copyright:

Available Formats

Design Report - iGlobus

Indian Oil Corporation Limited

Solinas Integrity Pvt. Ltd.

Design Report i Solinas Integrity Pvt. Ltd.

LIST OF TABLES iii

Design Report ii Solinas Integrity Pvt. Ltd.

Design Report 3 Solinas Integrity Pvt. Ltd.

Design Report 4 Solinas Integrity Pvt. Ltd.

Design Report 5 Solinas Integrity Pvt. Ltd.

1.1 Problem Scope

1.2 Technical Review

Design Report 1 Solinas Integrity Pvt. Ltd.

1.2 Design Requirements

• Design requirements for functionality.

Design Report 2 Solinas Integrity Pvt. Ltd.

• Design requirements for Safety.

1. Minimum diameter of the pipeline to be inspected : 10 inch.

2. Distance to be inspected : 200 kms.

3. Fluid inside the pipeline : Petrol, Diesel, Clear Naphtha, ATF.

4. Orientation of pipelines : Horizontal and Vertical pipelines.

5. Fluid Pressure : 5 bar to 90 bar.

6. Fluid Velocity : 7 km/hr(∼ 2m/s)

7. Type of flow : Turbulent

1. Insulation of robot with the view of preventing explosions.

2. Electronics connections inside the robot should be spark free.

3. Emergency shut down of robot.

4. Leak proofing of the robot even in elevated pressures.

5. Explosion proofing inside the shell.

Design Report 3 Solinas Integrity Pvt. Ltd.

Design Report 4 Solinas Integrity Pvt. Ltd.

2.2 Detailed Description

This section is subdivided into three broad categories,

3. The Software layout comprises of data storage mechanisms, pre-processing, anal-

2.2.1 Mechanical Design

Design Report 5 Solinas Integrity Pvt. Ltd.

Design Report 6 Solinas Integrity Pvt. Ltd.

Figure 1: First and Third Concept Design of iGlobus

Design Report 7 Solinas Integrity Pvt. Ltd.

Figure 2: Threading and Snap fit Method of iGlobus

• Material should withstand a pressure up to 90 bars.

• It should withstand the hydrostatic pressure testing of 150 bar in static

• Acoustic impedance of the material should be low.

• Higher strength even at elevated temperatures.

• Non-reactive to acids, alkalis and especially the contents of the pipeline.

Design Report 8 Solinas Integrity Pvt. Ltd.

• Should have Good Impact resistance.

The physical properties of aluminium, notably high thermal conductivity, specific

Design Report 9 Solinas Integrity Pvt. Ltd.

Pressure ∗ Out er d ia.

Figure 3: Cross Sectional View of Equivalent Stress Analysis of Shell in

Figure 4: Cross sectional view of Total Deformation of Shell in SolidWorks

Design Report 11 Solinas Integrity Pvt. Ltd.

Figure 5: Cross Sectional view of Impact Analysis of shell

Figure 6: Polyurethane Foam Layer

Design Report 12 Solinas Integrity Pvt. Ltd.

where, VT = Terminal Velocity

VR = Relative Velocity of fluid w. r. t. sphere

Design Report 13 Solinas Integrity Pvt. Ltd.

Figure 7: Transitions in Fluid Pressure