A rule-based expert system for harmonic load recognition

This paper presents a new method for identifying the different types of single phase nonlinear loads as sources of harmonic disturbances in a power system. The method combines the use of signal processing and artificial intelligence techniques. Fast Fourier transform and fractal analyses have been used to extract features of the harmonic signatures of the various nonlinear loads from the sampled input current waveforms. Intelligent and automatic harmonic load recognition process is achieved by using a rule-based expert system. The expert system has been verified using real measurements and the results show that the system give accurate identification of the single phase nonlinear loads such as personal computer, fluorescent lights, uninterruptible power supply and oscilloscope.