Data Processing For Data Acquisition Systems: Analysis of Vibrations in Structures

DATA PROCESSING FOR DATA ACQUISITION
SYSTEMS: ANALYSIS OF VIBRATIONS IN

STRUCTURES
R. S. Gonçalves1, M. S. Vieira2, T. L. Ssntos2,J. G. S. de Jesus 2
1
Instituto Federal do Maranhão, Department of Mechanical and Materials, São Luís, Brazil
2
Instituto Federal do Maranhão, Department of Electrical and Electronics, São Luís, Brazil
e-mail: matheus.santos.vieira@gmail.com
Abstract: Currently, there is a lot of work and material about vibration analysis for
preventive maintenance in machines, but the application of this detection in other areas is
almost non existent, an example of field that has much to take advantage of this type of
analysis, is civil engineering and material analysis. The proposed software uses the hardware
system described to capture the data, and then uses them to detect the frequencies pertinent to
the process. This makes it relevant not only during the construction process, to avoid possible
failures caused by the use of heavy machinery, but also allows the monitoring of structures in
use, such as bridges, which, over time, may show cracks in their extension.
Keywords: Matlab, Processing, IHM, Vibrationss.
Introduction
With the cost reduction of microcontrolled platforms and the integration of their programming
and interfaces, these systems are being increasingly used for the measurement of physical
quantities. In some situations, there is a deficiency in processing capacity, however the size of
such systems is perfect for use in real-time data acquisition systems.
There is a notable concern from hardware system developers regarding data acquisition. On
the other hand, the creation of a high level interface that facilitates the understanding of the
user is left aside, thus paving the way for such development. This work have as objective the
creation of a software for capturing and processing the data of an acquisition system, whose
unique proposal is to analyze real-time vibrations in structures with an accelerometer and
perform the actions already described, as well as presenting the results at a high level for the
user.
The ADXL345 triaxial accelerometer and the chip ATMEGA 2560 form the hardware of the
acquisition system used in the present work. The vibrations printed on the three spatial axes of
the accelerometer were recorded, in matrix form, by a software wrote in Processing and the
mathematical analyzes were performed by a software wrote in MATLAB. This set
cooperatively executed the Fast Fourier Transform of the registered time signal (matrix data)
and returned to the user periodic graphical responses, facilitating the interpretation of the
results.
Experimental Procedure (or Computational Procedure)

The algorithm developed is done in two platforms, Matlab and Processing. Initially the data is
captured by the program written in the Processing platform, which reads them from the
acquisition plataform through the serial port (RS-232 model), emulated in a USB port, at a
transmission speed of 115200bps, 8 bits of data, one bit of start and stop and no parity bit, so
there is no verification of the data in this layer; immediately after this reading, the read value
is stored, byte by byte, in a text file, which in turn is closed after one minute of data writing
and another text file is started immediately afterwards. When the other text file is started, the
software written on the Matlab platform detects the initiation of a new write and starts to
process the file that was previously generated, applying a check on the data, indicating
possible treatable failures - rows without all columns, rows with extra columns or non-
numeric values, all possible and caused by errors in the serial port transmission. These errors
are then treated - the defective line is deleted and a copy of the previous line is made, except
in the case of the first line of data, in which case it is simply deleted - and then the Fast
Fourier Transform is applied, using the fourth data column to obtain the frequency values
correctly, since it contains the time between samples. With these data in hand, six graphs are
produced - three for the representation of the values in the time domain in relation to the
planes x, y, z of the accelerometer and three for the representation of the values in the
frequency domain also in relation to the planes x , y, z of the accelerometer - these are then
exported as figures to a separate folder and the corrected variables of the text files are then
exported in .mat to a folder reserved for them, finally the text file containing the original data
is discarded.
Results and Discussion

Preliminary results show the efficiency in the capture and analysis of the consistency of the
obtained data, generating a temporal and spectral graphic of the signal, as well as executing
the storing in regular intervals of one minute. In addition, the system identifies the
fundamental frequencies of the vibration signals obtained, classifying them according to the
standard SN 640312, thus generating alerts, when appropriate, and continuous reports of what
is captured.
In addition to the use in detection of structures, where the maximum frequency of analysis is
inferior to 300Hz, the algorithm was used for analysis of vibrations in machines, using a
smaller data collection interval, but arriving at a maximum analyzed frequency of
approximately 1kHz.
The development of this platform completely in Java has already begun, thus allowing the use
in any computer, depending only on the previous installation of the JVM, much more
accessible than the softwares used in the current development.
Conclusions
The integration of the computer platform with the hardware system presented
excellent performance and the high storage capacity of the computers allows the data to be
collected for long periods without significant losses, thus providing a light system, both for
the microcontrolled platform and the computer, which can perform the task in the background
without any hassles. The program is already being write in Java, where the same
functionalities will be offered, but with a data collection period of less than one minute.
Acknowledgments
The authors thank the Instituto Federal do Maranhão for the technological support.
References
[1] J. van der Geer, J.A.J. Hanraads, R.A. Lupton, The art of writing a scientific article, J. Sci.
Commun. 163 (2010) 51–59.
[2] W. Strunk Jr., E.B. White, The Elements of Style, fourth ed., Longman, New York, 2000.
[3] G.R. Mettam, L.B. Adams, How to prepare an electronic version of your article, in: B.S.
Jones, R.Z. Smith (Eds.), Introduction to the Electronic Age, E-Publishing Inc., New York,
2009, pp. 281–304.
Choose your preferred form of presentation:

( ) Oral
( X) Poster

Data Processing For Data Acquisition Systems: Analysis of Vibrations in Structures

Uploaded by

Copyright:

Available Formats

Data Processing For Data Acquisition Systems: Analysis of Vibrations in Structures

Uploaded by

Document Information

Original Description:

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

Data Processing For Data Acquisition Systems: Analysis of Vibrations in Structures

Uploaded by

Copyright:

Available Formats

DATA PROCESSING FOR DATA ACQUISITION

SYSTEMS: ANALYSIS OF VIBRATIONS IN

Keywords: Matlab, Processing, IHM, Vibrationss.

Experimental Procedure (or Computational Procedure)

Results and Discussion

Choose your preferred form of presentation:

You might also like