1st International Conference on Advances in Science, Engineering and Robotics Technology 2019 (ICASERT 2019)

IoT Based Smart Energy Management in

Residential Applications
Md. Rakibul Hasan Eklas Hossain Hossain Mansur Resalat Faruque
Department of Electrical and Department of Electrical Engineering Department of Electrical and
Electronic Engineering and Renewable Energy Electronic Engineering
Khulna University of Engineering & Oregon Institute of Technology Khulna University of Engineering &
Technology Klamath Falls, USA Technology
Khulna, Bangladesh eklas.hossain@oit.edu Khulna, Bangladesh
rakibul.eeekuet@gmail.com resalat_f@yahoo.com
Tipu Sultan
Department of Electrical and
Electronic Engineering
International Islamic University
Chittagong, Bangladesh
ssultanstipu@yahoo.com

Abstract—The lack of credibility of data acquired through system energy losses. Furthermore, the integration of the
traditional means from a conventional metering system has SEMS with the internet enables the consumers to observe their
reduced the efficacy of the energy management system. For consumption of electricity and take action according to their
overcoming the drawbacks of the conventional metering system, budget by remotely controlling the electrical equipment. It
this paper proposes a smart IoT based energy management also provides a feasible mean to the electric suppliers for
system (EMS), which not only ensures transparency and producing bills with minimal error.
credibility of data but also offers flexibility to the consumers by
offering the facility of remotely controlling their home With the advancement of technology, the accessibility of
appliances. Electricity consumption and its corresponding cost the internet is increasing day by day. Apart from being used
is calculated automatically and then the data is sent to a server for communication and entertainment purposes, the internet
in a periodic manner. For carrying out the task, NodeMCU has also provides a bunch of scopes in power system management,
been utilized, which has the combined functionality of both development, control, etc. Internet of things (IoT) is a cutting-
microcontroller and Wi-Fi modules. By means of its edge technology that interconnects the 'things' over the
microcontroller function, constant readings of energy meter are internet [1]. These ‘things’ incorporate everything we can
collected and saved in NodeMCU; then the data is transferred perceive of connecting-goods, buildings, appliances,
to a remote server on the internet via Wi-Fi. Consumers can machines, vehicles, plants, animals, and even human beings.
observe real-time data on the server to gain a clear idea of
It is estimated that the number of objects connected to the
electricity usage, therefore, they can reduce or increase the
internet will reach 50 billion by 2020 [2]. IoT is a bidirectional
consumption as per their budget. This proposed system also
provides the functionality of controlling the home appliances approach, where objects export data to a distant server and the
remotely over the Internet. Furthermore, electricity suppliers users can control the objects through the server by software
can produce bills from the server data automatically, making it programs and it is mainly done over Wi-Fi networks. The
a feasible solution. This paper also proposes and analyzes traditional electric power network has many limitations,
individual sub-metering of home appliances of high rating to however, by integrating with IoT it can be transformed into an
observe and control their energy consumption. A detailed cost intelligent power network [3]. The application of IoT can
analysis of the system has been performed for ensuring its cost- effectively improve the accuracy of real-time monitoring and
effectiveness. According to the analysis, the total cost associated control of smart grid [4, 5]. The informatization of power grid
with the system is BDT 1064, which is fairly economical system can be sped up by utilizing IoT and thus is beneficial
compared with existing systems. The add-ons of this proposed for effective management of the power grid infrastructure [6].
system can be also integrated with the existing prepaid meters It can also be used for precise and accurate estimation of
of the market, eradicating its major drawbacks by providing energy consumption [7]. Recent studies show that IoT can also
added functionality. be integrated with smart energy meters and can increase their
accuracy [8-10]. IoT based smart energy meter has been
Keywords— IoT, Smart meter, Energy management, Home proposed in [11-13]. However, it does not have the
automation, Residential tariff.
functionality of remotely controlling the home appliances. An
I. INTRODUCTION efficient billing system can also be developed using IoT [14,
15]. It also possesses great potentiality in its application in the
Bangladesh being a developing country, the energy control system of smart homes [16-18].
demand is increasing by leaps and bounds. In order to meet
with this rising energy demand, there has been a prolific This paper proposes a SEMS based on IoT, which ensures
increase in the implementation of alternative energy sources the credibility of smart meter data as well as provides the
for generating more electricity, however, this is proving to be facility for controlling home electrical appliances remotely.
insufficient due to system energy loss. Therefore, a smart The energy usage data of the consumer is stored in
energy management system (SEMS) is necessary to ensure NodeMCU, which is then transferred to a remote server on the
proper electrical power management, which will reduce the internet using Wi-Fi. NodeMCU is an open-source

978-1-7281-3445-1/19/$31.00 ©2019 IEEE

development kit composed of microcontroller and ESP8266 calculate the consumed energy and these data are extracted
Wi-Fi module, where microcontroller part does the necessary from the meter through a LED. Table 1 shows the
calculations and control and ESP8266 helps in specification of the watt meter used in this paper.
communication over Wi-Fi. The corresponding cost of the
consumed electricity is automatically updated, reducing the TABLE I. SPECIFICATION OF THE SINGLE PHASE STATIC WATT
errors that occur in the conventional energy meter system. HOUR METER
This proposed method also reduces the complexity of bill Supply System Range
payment and the consumer can check his/her energy usage at Voltage 0 to 240V (Phase to neutral)
any time. Furthermore, it gives consumers the ability to Current 5A to 30 A
remotely control any home appliance. Hardware Frequency 50Hz ± 5%
implementation of the system along with necessary software
Temperature −10°𝐶 𝑡𝑜 50°𝐶
configuration has been carried out in this research.
Additionally, a comparative cost analysis with the existing
energy meters available in the market has also been shown in In order to perform IoT operation, a microcontroller along
this paper. with Wi-Fi module is needed. The microcontroller sends the
The remaining part of the paper is organized as follows: in data to predefined internet server using Wi-Fi. NodeMCU is
section II, the present condition of energy metering system in such a development board that provides combining the
Bangladesh is represented. The tools and apparatus required functionality of having both I/O ports like microcontroller and
for the proposed system and its working principle are Wi-Fi module. This kit is based on the ESP8266 Wi-Fi
discussed in section III. The problems that occur during the module. It is programmed in an Arduino environment. Here,
operation of the proposed smart energy meter and the microcontroller does all the calculations necessary,
corresponding solutions are presented in section IV, which is whereas ESP8266 provides the functionality of
followed by a cost analysis in section V. Finally, in section VI, communication via Wi-Fi. The pin diagram of the NodeMCU
the key features along with future scope are described as a is shown in Fig. 1.
conclusion. Another required apparatus for the proposed system is the
optocoupler, which provides isolation between relatively high
II. PRESENT CONDITION OF ENERGY METERING SYSTEM IN voltage energy meter side and low voltage microcontroller
BANGLADESH circuit. The electrical signal from two isolated circuits is
A common practice in Bangladesh is that a person from transferred using an optocoupler. Data from power circuit is
utility services goes to individual consumers to collect the fed to optocoupler and then microcontroller extracts the data
energy meter data. The bill is prepared using those data and from the optocoupler. The model used in this work is
the bill paper, which contains all the information about the optocoupler – 4N25. Other components used for the proposed
usage of electricity in that month, is sent to individual IoT-based smart energy meter are a 220V – 9V transformer
consumers for payment. Finally, consumers pay their bill to step down the supply voltage, IC 7805 to regulate dc
according to the bill paper, which was sent to them by going voltage, diodes for rectification purpose, capacitors for
to a particular station or a bank. Therefore, this process is very smoothing the rectified dc, and some light emitting diodes
much hassling and requires a lot of time and steps, and since (LED) to emulate the home appliances.
there is direct involvement of human in collecting the meter
Fig. 2 shows the functional block diagram of the whole
data, the credibility of this system is compromised.
system. In the functional block diagram of a conventional
There are system losses occurring in the power network meter, only the ac power supply, energy meter, and loads
due to technical and non-technical aspects [19, 20]. Technical block exist. However, some additional blocks are needed to be
reasons include losses in power generation equipment, added here to realize the IoT and energy management features.
transmission lines, distribution lines, etc.; whereas non- The energy consumption data is taken from the energy meter
technical losses occur mainly due to customer level meter through the optocoupler. There is a LED in the energy meter
tampering. In order to reduce the losses occurring at the named ‘Cal’, which blinks a certain fixed times per Kilo Watt
customer level, the prepaid meter has been introduced in some
cities and the rest are on the way. Although the prepaid
metering system offers more advantages than the conventional
meter system, it also has some major drawbacks. The
customers need to purchase a card to refill the account but
purchasing the card is quite troublesome sometimes. The
consumers also need to predict the monthly bill and recharge
accordingly, which can create a situation of uncertainty.
Furthermore, prepaid energy meters do not provide the
consumer with the ability to control electrical appliances
remotely.
III. IOT BASED SMART ENERGY METER
IoT-based smart energy meter solves the issues of prepaid
energy metering by minimizing the complexities and
mitigates the non-technical losses by ensuring the credibility
of data. It also brings new important features, such as real-time
viewing of consumption data and remote controlling of home
appliances. A single phase static watt-hour meter is used to Fig. 1. Pin diagram of NodeMCU.
 reduce it, a smoothing capacitor is used in parallel. Finally,
this 9V DC is transformed into 5V DC by using the 7805
regulator IC that is fed to the NodeMCU. The hardware
implementation of the prototype is shown in Fig. 4.
Another important feature of this system is the control of
home appliances over the Internet. A user of this proposed
system can check the status of his home appliances and can
take any action as he wants; that means he can turn on or off
any home appliance remotely using the internet. In order to
simulate this, three LED’s are connected to NodeMCU and
can be controlled over the Internet. The LEDs are connected
to the general purpose input/output (GPIO) pins. Some
predefined commands have been configured in the Arduino
script to make the LED either on or off. When those
commands are transmitted to the NodeMCU from the website,
the LEDs will respond according to the specified commands
and thus, they can be controlled easily. In real life application,
this feature can be implemented by using a relay circuit. By
Fig. 2. Block diagram of the IoT based smart energy management.
controlling the relay, electrical appliances like fan, light etc.
Hour (kWh) for a particular energy meter. The meter used in can be turned on or off as required.
this project has such a functionality, where the ‘Cal’ LED When a user logs in to his account, he will see a webpage,
blinks 3200 times per kWh. This LED data is then fed to a which is shown in Fig. 5. This webpage has been developed
microcontroller through the optocoupler. The number of using the basic HTML and bootstrap library which are open-
blinking is observed by microcontroller programming and by source resources. Consumed unit of electricity and associated
using this data, the total electricity units consumed and cost are shown on the webpage. The consumer can remotely
associated cost are calculated.
The main feature of this system is to collect data over the
internet. Therefore, the consumed unit of electricity and
associated cost data need to be transmitted over the internet.
This is done by writing the required Arduino code in the
NodeMCU. In order to practically implement this system,
NodeMCU needs internet connectivity and to store data on the
internet, hence, a website domain is needed. However, in this
prototype, the website is hosted locally and Wi-Fi hotspot
connection is given to the NodeMCU to simulate the
prototype. The Wi-Fi service set identifier (SSID) and
password are written in the Arduino script. Therefore,
NodeMCU gets connected to local Wi-Fi whenever Wi-Fi is
available. The updated data are transmitted at a certain time
interval.
The power supply of NodeMCU is taken from the main
AC power supply. In order to use the 220V AC supply as 5V
DC supply, voltage step down transformer and a rectifier
circuit are used, which is shown in Fig. 3. In this circuitry, a Fig. 4. Hardware implementation of the prototype of IoT based smart
step-down transformer is used to convert 220V AC to 9V AC. energy management system.
This 9V AC is then transformed into 9V DC by a full bridge
rectifier circuit. The DC produced in the output of full bridge
rectifier has a significant amount of ripples and in order to

Fig. 3. Rectifier circuit to convert 220V AC to 5V DC. Fig. 5. Webpage showing the consumed electricity and their associated
cost along with the feature of controlling the appliances remotely.
control any home appliance using this webpage. For this The existing tariff rates in Bangladesh for different
instance, in order to control the green, yellow, and red LEDs categories according to DPDC [21], is shown in Fig. 6. The
over the Internet, two buttons have been configured as ON or variation of service charge s and demand charge according to
OFF for each LED and they can be turned on or off DPDC [20] is presented in Figure 7.
individually just by clicking on the corresponding buttons. A
comparative analysis of the proposed system with existing
works is tabulated in table II. The proposed system does not
require any additional microcontroller as the NodeMCU has
the dual functionality of microcontroller and a Wi-Fi module.
Furthermore, the proposed system also has the feature of
controlling home appliance remotely.

TABLE II. COMPARISON OF PROPOSED METHOD WITH EXISTING

WORKS

Authors Microcontroller Optocoupler Functionality of

controlling of
home appliance
remotely
Abate et STM32F205RB Not used Not available
al. [11]

Sahani et ATMega328 MOC3071 Not available

al. [12]
Fig. 6. Existing tariff rates in Bangladesh [21].
Singh et ATMega328 and Not used Not available
al. [13] PIC16F877P

Proposed Built-in 4N25 Available

microcontroller in
NodeMCU

The proposed system has been designed to determine the

electricity consumption and calculate the corresponding cost
for residential consumers only. Let us consider 𝑏 is number of
blinks of the ‘Cal’ LED in a month. As the ‘Cal’ LED blinks
3200 times per kWh, the total energy consumption (kWh) is
𝑏
𝑎= (1)
3200

If 𝑠 and 𝑑 are the service charge and demand charge per

month in Bangladeshi Taka (BDT), respectively, then the
associated cost of that month 𝑐 can be calculated using the
following equation
𝑐 = (𝑎 × 𝑘) + 𝑠 + 𝑑 (2)
Fig. 7. Variation of demand charges and service charges in Bangladesh
Here, 𝑘 is per unit price of energy in (BDT), whose value [21].
depends on 𝑎 . According to Dhaka Power Distribution
Company (DPDC), the value of k for different consumed unit IV. PROBLEMS AND SOLUTION
(a) is shown in table III [21]. NodeMCU needs power from the main supply to operate;
thus when power interruptions occur, it will be turned off.
TABLE III. TARIFF RATE FOR RESIDENTIAL CONSUMERS However, there is no possibility of data loss as the data is
stored periodically in the internal memory of the NodeMCU.
Consumed Unit (𝒂) Per Unit Rate of Energy (𝒌) Each time the NodeMCU starts, it extracts the most recently
updated data from the memory. This system needs an Internet
0-75 3.80 BDT connection to send data to the web server and there might be
the possibility of having no internet connection at a certain
76-200 5.14 BDT time. However, this problem is solved as NodeMCU sends the
updated data whenever it regains internet connectivity. Data
201-300 5.36 BDT
are always updated and stored in the internal memory of
301-400 5.63 BDT
NodeMCU and is sent to the server whenever the internet is
available. Therefore, there is no possibility of data loss. The
exchange of data over the public domain such as the Internet
401-600 8.70 BDT
is very much susceptible to unauthorized access [22].
Therefore, the system will be synchronized to a distant server
601-above 9.98 BDT
and it can be programmed so that any undesirable attempt will
inform the consumer as well as the electricity provider. In this implementation along with interfacing with the internet has
way, the risk of data theft can be reduced to a minimum. been presented in detail. For ensuring the cost effectiveness, a
detailed cost analysis has been conducted. The overall cost of
V. COST ANALYSIS the proposed IoT based smart energy management system is
There are various types of energy meters available in the BDT 1065 and add-ons cost is BDT 635. Hence, the proposed
market, their price ranging from BDT 2000 to 10000. system is cost effective and the add-ons can be also integrated
Different prepaid energy meters available in the market and with the existing prepaid meters, which can expunge its major
their corresponding prices are shown in table IV. drawbacks by providing added functionality. Therefore, the
presented design in this paper, possesses high potentiality in
TABLE IV. PRICE OF AVAILABLE PREPAID METERS IN MARKET its application in energy metering system of Bangladesh.

Product Title Market Price

REFERENCES
[1] S. N. Han, I. Khan, G. Myoung, N. Crespi, and R. H. Glitho, “Computer
Liberty Prepaid Energy Meter BDT 9880 Standards & Interfaces Service composition for IP smart object using
realtime Web protocols : Concept and research challenges,” Comput.
Techno Prepaid Energy Meter Class 1.0 BDT 5070 Stand. Interfaces, vol. 43, pp. 79–90, 2016.
[2] M. Swan, “Sensor Mania! The Internet of Things, Wearable
Elmeasure Prepaid Energy Meter BDT 5000 Computing, Objective Metrics, and the Quantified Self 2.0,” J. Sens.
Actuator Networks, vol. 1, pp. 217–253, 2012.
Meter by Yueqing Haixin Electronic Company BDT 2540 [3] G. Bedi, G. K. Venayagamoorthy, R. Singh, R. R Brooks, and K. C.
Wang, “Review of Internet of Things (IoT) in Electric Power and
Energy Systems,” IEEE Internet of Things Journal, vol. 5, pp. 847-870,
2018.
Cost of different equipment required for the proposed [4] L. Li, H. Xiaoguang, C. Ke and H. Ketai, "The applications of WiFi-
system and the overall cost is shown in Table V. In our based Wireless Sensor Network in Internet of Things and Smart
proposed design we used the Universal Energy Meter which Grid," 6th IEEE Conference on Industrial Electronics and
is a conventional energy meter. Additional equipment such as, Applications, pp. 789-793, 2011
NodeMCU, optocoupler, and voltage step down circuitry are [5] Miao Yun and Bu Yuxin, "Research on the architecture and key
technology of Internet of Things (IoT) applied on smart
used as add-ons so that this conventional energy meter can grid," International Conference on Advances in Energy Engineering,
function as IoT based smart energy meter. Furthermore, it has pp. 69-72, 2010.
the facility to provide the consumers the ability to control their [6] Q. Ou, Y. Zhen, X. Li, Y. Zhang, and L. Zeng, "Application of internet
home appliance remotely. The aforementioned add-ons are of things in smart grid power transmission," FTRA International
also compatible with the existing prepaid meters available in Conference on Mobile, Ubiquitous, and Intelligent Computing, pp. 96-
the market. Therefore, the consumers are able to pay their bills 100, 2012.
without having to predict their monthly consumption of [7] F. Englert, P. Lieser, A. Alhamoud, D. Boehnstedt and R. Steinmetz,
"Electricity-Metering in a Connected World: Virtual Sensors for
electricity with the added facility of remotely controlling their Estimating the Electricity Consumption of IoT Appliances," 3rd
home appliance. From data presented in table III and IV, it can International Conference on Future Internet of Things and Cloud, pp.
be said as the add-ons cost is very cheap, therefore, it is very 317-324, 2015.
much cost-effective integrate them with conventional or [8] D. Alahakoon, and X. Yu, “Smart Electricity Meter Data Intelligence
prepaid energy meters. for Future Energy Systems: A Survey,” IEEE Transactions on
Industrial Informatics, vol. 12, pp. 425-436, 2016.
TABLE V. COSTING OF DIFFERENT EQUIPMENTS OF THE IOT BASED [9] P. D. Talwar, and S. B. Kulkarni, “IoT Based Energy Meter Reading”
ENERGY MANGEMENT SYSTEM International Journal of Recent Trends in Engineering & Research
(IJRTER), vol. 2, pp. 586-591, 2016.
Market
Equipment [10] E. Hossain, J. Hossain and F. Un-Noor, "Utility Grid: Present
Price
Challenges and Their Potential Solutions," IEEE Access, vol. 6, pp.
Universal Energy Meter BDT 430 60294-60317, 2018.
[11] F. Abate, M. Carratù, C. Liguori, M. Ferro, and V. Paciello, "Smart
NodeMCU V-3 Development Kit
BDT 500 meter for the IoT," IEEE International Instrumentation and
Add-ons Measurement Technology Conference (I2MTC), pp. 1-6, 2018.
4N25 Photo Transistor
BDT 15 [12] B. Sahani, T. Ravi, A. Tamboli, and R. Pisal, “IoT Based Smart Energy
Optocoupler
Meter,” International Research Journal of Engineering and
Voltage Step Down Circuitry BDT 120 Technology (IRJET), vol. 4, pp. 96-102, 2017.
Total BDT 1065 [13] A. Singh, and R. Gupta, “IoT based smart energy meter,” International
Journal of Advance Research and Development, vol. 3, pp. 328-332,
VI. CONCLUSION 2018.
[14] P. Mahajan, S. Tatia, and P. Jadhav, “Smart Meter: An IOT Based
In this paper, integration of IoT with energy management Smart Meter for Live Electricity Monitoring and Bill Payment,”
system has been demonstrated to make a more effective and International Journal of Electrical, Electronics and Computer Systems
reliable system compared with the conventional energy (IJEECS), vol. 5, pp. 49-51.
management system. Instead of collecting data from door to [15] N. N. Sasane, N. S. Sakat, K. N. Shital, V. R. Kaushik, and P. K. Pallav,
door, the key feature of the proposed system offers a more “IOT based energy meter billing and monitoring system - A case
study,” International Research Journal of Advanced Engineering and
accessible way of collecting data from a server through Science, vol. 2, pp. 64-68, 2017.
internet, which is automatically updated after a short time
[16] S. Chen, C. Lai, Y. Huang, and Y. L. Jeng, "Intelligent home-appliance
interval via Wi-Fi. The NodeMCU used in this system has the recognition over IoT cloud network," 9th International Wireless
dual functionality of a microcontroller and a Wi-Fi module, Communications and Mobile Computing Conference (IWCMC), pp.
which eliminates the need for an additional microcontroller. 639-643, 2013.
Furthermore, remote control of domestic appliances has also
been made possible through this system. The total hardware
[17] W. T. Cho, Ying-Xun Lai1, C. F. Lai, and Y. M. Huang, “Appliance-
Aware Activity Recognition Mechanism for IoT Energy Management
System,” The Computer Journal, vol. 56, pp. 1020-1033, 2013.
[18] M. Wang, G. Zhang, C. Zhang, J. Zhang and C. Li, "An IoT-based
appliance control system for smart homes," Fourth International
Conference on Intelligent Control and Information Processing
(ICICIP), pp. 744-747, 2013.
[19] F. I. Bappy, M. J. Islam, A. K. Podder, D. R. Dipta, H. M. R. Faruque,
and, E. Hossain, “Comparison of Different Hybrid Renewable Energy
Systems With Optimized PV Configuration to Realize the Effects of
Multiple Schemes,” 1st International Conference on Advances in
Science, Engineering and Robotics Technology (ICASERT), in press.
[20] E. Hossain, I. Khan, F. Un-Noor, S. S. Sikander, and M. S. H. Sunny,
"Application of Big Data and Machine Learning in Smart Grid, and
Associated Security Concerns: A Review," IEEE Access, vol. 7, pp.
13960-13988, 2019.
[21] Tariff Rate. [online] Available at:
https://dpdc.org.bd/article/view/52/Tariff%20Rates [Accessed Apr.
11, 2019].
[22] M. S. Rahman, M. S. Hossain, E. H. Rahat, D. R. Dipta, H. M. R.
Faruque, and F. K. Fattah, "Efficient Hardware Implementation of 256-
bit ECC Processor Over Prime Field," International Conference on
Electrical, Computer and Communication Engineering (ECCE), pp. 1-
6, 2019.