Study on harmonic analysis based on IIR orthogonal filter banks

Abstract

Nowadays, with the wide use of power electronic devices and nonlinear loads in industry, much more harmonic in the power system exists, resulting in the reduction of power quality. It is important to detect the harmonic accurately for suppressing it. The Fourier Transform method has some limitations for the non-stationary harmonic analysis. In this paper, the method based on filter banks to analyze harmonic is proposed, which can achieve multi-resolution analysis for the harmonic. Firstly, the IIR orthogonal filter banks with multi-channel is constructed, and the amplitude-frequency characteristics of IIR and FIR are compared. Then, the two channel tree filter banks is constructed by the IIR orthogonal filter banks, which possesses better frequency characteristic than FIR orthogonal filter banks. The multi-resolution analysis of the harmonic can be achieved by the Porous algorithm, which is improved with Mallat algorithm. Finally, the signal models are established and the simulations are done. By utilizing the filter banks to discompose harmonic and comparing with the Fourier Transform, the simulation results show that the filter banks can analyze the non-stationary harmonic effectively.