Real-Time Harmonic Signal Detection And Sources Location Identification System For Power Distribution System

• Main Author: Mat Kassim, Nabilah
• Format: Thesis
• Language: English; English
• Published: 2017
• Subjects: T Technology (General); TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utem.edu.my/id/eprint/21455/1/Real%20-%20Time%20Harmonic%20Signal%20Detection%20And%20Sources%20Location%20Identification%20System%20For%20Power%20Distribution%20System.pdf; http://eprints.utem.edu.my/id/eprint/21455/2/Real%20-%20Time%20Harmonic%20Signal%20Detection%20And%20Sources%20Location%20Identification%20System%20For%20Power%20Distribution%20System.pdf; http://eprints.utem.edu.my/id/eprint/21455/; http://libraryopac.utem.edu.my/webopac20/Record/0000107866
• Institution: Universiti Teknikal Malaysia Melaka

One of the major issues of power quality is harmonic distortion,which refers to the distortion in the waveforms of the power system.This issue is even more prevalent nowadays,as new electrical power equipment is being constantly developed and introduced into the market,and these systems contribute towards the distortion of the electrical power supply.High levels of harmonic distortion may cause excessive heating of the components in the electrical power equipment as a result of current surge,which leads to equipment failure.This in turn,increases downtime in production facilities,and may lead to severe financial losses.Thus,it is essential to trace and identify harmonic sources in the power system so that precautionary steps can be taken to prevent the detrimental impact of harmonic distortion.Therefore,this research is focused on analysing harmonic signals and identifying the location of harmonic source by utilizing periodogram and spectrogram techniques.The harmonic signals are generated from the rectifier load to the test system using MATLAB/Simulink program.There are two forms of representation used for the harmonic signals generated in this study which are power spectrum,and time-frequency representation.The former and latter representation is obtained from the periodogram and spectrogram technique, respectively.The following signal parameters are also estimated,namely,the root mean square voltage,root mean square current,fundamental root mean square for voltage and current,total harmonic distortion,total non-harmonic distortion,and total waveform distortion for voltage and current.The harmonic signals are detected according to the procedure outlined in the IEEE Standard 519-2014,and the equations for harmonic source identification are formulated based on the characteristics of the harmonic source either at the downstream,upstream or both sides of the point of common coupling in the test system.The performance of the harmonic source identification techniques is verified using four test cases and the radial system network.A real-time monitoring system is developed upon completion of the verification process in order to detect the harmonic signals and identify the location of the harmonic source in an actual power system.A total of 25 signals with different characteristics are generated,tested and evaluated in order to determine the accuracy of the signal parameters of the monitoring system.The average mean absolute percentage error is determined to be 0.711 percentage. Based on the results,it can be concluded that the real-time harmonic signal detection and source identification system developed in this study is reliable and accurate to analyse harmonic signals,and identify the location of harmonic sources in power systems.