An optimal design of single tuned filter in distribution systems

Abstract

This paper proposes an optimal design of single tuned shunt passive filter for the Underground Ramsis Metro Substation (URMS) using linear programming technique as a conventional technique and fuzzy linear programming (FLP) technique as a modern technique. The optimal filter parameters are obtained to overcome the harmonics generated in the voltage and current waveforms, due to the non-linear characteristics of the rectifiers, which distort the voltage and current waveforms at the point of common coupling (PCC). The harmonic levels are analyzed at different loading conditions using the fast Fourier transform (FFT) on MATLAB optimization program. The proposed optimal design of harmonic filter is based on minimizing the cost of the filter parameters and its losses. Both techniques are applied on the system under study. From the results, the THD levels of the line current and the bus voltage are reduced to be within the standard limits. A comparison between the optimal design filters is carried out using the LP and FLP techniques. However, the lowest cost of filter is obtained using the FLP technique.