A New Harmonic Signal Analysis Technique For Automated Source Detection And Identification System

Shamsudin, Nur Hazahsha (2017) A New Harmonic Signal Analysis Technique For Automated Source Detection And Identification System. Project Report. UTeM, Melaka, Malaysia. (Submitted)

Text A New Harmonic Signal Analysis Technique For Automated Source Detection And Identification System.pdf - Submitted Version Restricted to Registered users only Download (155kB)

Abstract

With the widespread use of the distributed harmonic sources, quantifying the individual harmonic impacts of the multiple harmonic producing loads is becoming increasingly important. Power harmonic can cause many problems such as equipment failure, data and economical losses. Thus, it is necessary to detect, identify and obtain the information of the source of the harmonic signal. A real time automated monitoring system is needed to reduce diagnostic time, rectify failures and identify the harmonic source. This research proposes a new technique of time-frequency distribution for real time harmonic source identification system. The harmonic signal analyzed is based on IEEE Std. 1159-2009 and consists of multi-frequency components and magnitude variations. Thus, time-frequency distributions (TFDs) are very appropriate to be used that represent the signals jointly in time and frequency representation (TFR). In this research, parameters of the harmonic signal are estimated from the TFR and then used to calculate characteristics of the signal. Based on the signal characteristics, the popular linear and bilinear TFDs will be compared to identify the best TFD in terms of accuracy, computation complexity and memory used of data processing by simulation using MATLAB. The linear TFD are such as spectrogram, Gabor transform, S-transform while the bilinear are Wigner Ville distribution and Smooth-Windowed Wigner Ville distribution. To verify the performance of the proposed technique, harmonics with various characteristics will be generated, tested and evaluated based on parameters given. In addition, the technique will also be tested using real signals from power line system. At the end of this research, a real time harmonic source identification system using the best TFD will be presented to detect, identify and provide information of the harmonic sources.

Actions (login required)

View Item