Multilevel Inverter Based Active Power Filter for Harmonic Elimination

Abstract

In this paper, support vector machine (SVM) and fuzzy based feedback harmonic elimination technique is proposed. The proposed technique is to overcome the drawbacks of control rule selection of fuzzy logic. Here, the feedback error voltage and change of error voltage of the load is applied as the input of fuzzy logic. The Fuzzy Logic Controller (FLC) builds the logical rules; it depends on the input error voltage and change of error voltage. From the fuzzy rules, the basic SVM takes a set of input data and predicts optimum switching angle and frequency, for each given input. Then, the selected switching angle and frequency is applied to multilevel inverter and the harmonics of the system is eliminated. The proposed method is implemented in MATLAB/simulink working platform and the harmonic elimination performance is demonstrated by the comparison of Neuro Fuzzy Controller (NFC) and Adaptive Neuro Fuzzy Interference System (ANFIS).

DOI: http://dx.doi.org/10.11591/ijpeds.v3i3.4314