A 3-layered feedforward back-propagation ANN-based SVPWM control for neutral point clamped converter for PV grid integration

Abstract

Most of the power electronic converters based on the devices such as Silicon Controlled Rectifiers (SCRs) have been broadly utilized in home, business, and modern use in recent years. Despite their many benefits, these power electronic converters have major issues such as pulling harmonic current and the reactive part of the current from the supply, as well as having a highly nonlinear characteristic. The harmonics produced by the current supplied by these nonlinear elements cause voltage distortion at the common coupling point, which is causing problems for the functioning of number of sensitive instruments and other consumer appliances. Artificial Neural Networks (ANN) are a type of Artificial Intelligence (AI) approach that has been applied to improve the efficiency and regulation of the converter. In order to avoid the need for a Digital Signal Processors (DSP) by avoiding the online timing computations for various voltage space vectors in various regions and sectors and produce higher pulse resolution, an ANN-based space vector pulse width modulation (SVPWM) technique is proposed in this paper. The analysis of a 3-layered feedforward back propagation ANN algorithm based SVPWM control for NPC converter used to integrate PV source to grid has been evaluated and found to be better as compared to traditional techniques.