Measurement of power system harmonic based on adaptive Kaiser self-convolution window

Abstract

As the rapid development in power electronic devices, the harmonic pollution becomes one of principle power quality problems in power system. The fast Fourier transform (FFT) is widely used for analysing and measuring power system harmonics. However, the limitation of the FFT such as an aliasing effect, spectrum leakage picket-fence effect, would contribute to inaccuracy results. Furthermore, the real power system harmonic is actually a non-stationary signal while FFT is a tool for stable signal. This study focuses on a novel FFT based method for time-varying power system harmonic measurement. The harmonic signal is preprocessed by infinite impulse response filter bank and Teager-Kaiser energy operator for fast detection of instability onset time. Then an adaptive Kaiser self-convolution window-based interpolated FFT algorithm is used to estimate each harmonic component. The results of both simulation and practical implementation show that the proposed method is suitable to deal with time-varying harmonic and achieves a higher accuracy compared with the traditional FFT-based techniques.