The Application of Wavelet Theory to Power Quality Diagnostics

Martin Collins, Prof. W. G. Hurley

This project describes the novel application of wavelet theory to power quality diagnostics, and how wavelet theory can be used in an expert system/neural network environment for the classification of power quality problems. Wavelet theory is explained and it is shown by example how the wavelet transform can be used as a method of feature extraction in a software system for analysing current and voltage waveforms. It is illustrated that wavelet analysis is superior to Fourier analysis in applications involving non-stationary waveforms, and especially in the case of high frequency transients or impulses, as the wavelet transform gives both good time and good frequency resolution simultaneously. An expert system for power quality diagnostics is presented. The expert system controls the entire classification system and is used to broadly classify power quality waveforms, e.g. voltage sags, non-sinusoidal waveforms, high frequency transients, etc. The expert system is implemented using the declarative artificial intelligence language "Prolog". Neural networks are used by the expert system to make fine distinctions between waveforms of a particular type. Wavelet analysis is used as the method of feature extraction on high frequency transients, before they are inputted to the neural networks. Test cases are described and it is explained how the system is implemented in each case.

Start date: 1st of September 1993. This project has been completed.

Project Publications

Conference
M. P. Collins, W. G. Hurley, E. Jones, “The Application of Wavelet Theory in an Expert System for Power Quality Diagnostics”, Proceedings of the 30th Universities Power Engineering Conference, UPEC’95, University of Greenwich, vol. 1, pp. 274-277, September 1995.
M. P. Collins, W. G. Hurley, E. Jones, “The Application of Wavelet Theory to Power Quality Diagnostics”, Proceedings of the 29th Universities Power Engineering Conference, UPEC’94, University College Galway, vol. 1, pp. 129-131, September 1994.

• Disclaimer
• Privacy
• Copyright
• Accessibility