Experimental Validation of Photovoltaic Grid Connected NPC Multilevel Inverter Based on Neural Network MPPT

Abstract

Distributed energy resources (DER) became a suitable solution to meet part of energy demand and providing electricity in isolated sites. Photovoltaic (PV) energy generation is one of the solutions that presents several benefits for use as a distributed energy resource. This paper presents a PV grid connected system with a multilevel inverter. The later one offers several advantages by synthesizing the AC output terminal voltage from several levels of voltages, sinusoidal current waveforms with lower THD content, higher power quality and reduced stress on the switching devices. This provides higher efficiency in the power conversion stage and satisfies the grid’s connection criteria. On the other hand, a maximum power point tracker (MPPT) is very important in such us system to increase the system performance. In this paper, a control algorithm has been developed for the three-level single-phase inverter based on neural network maximum power point tracker to ensure a fast dynamic response. The proposed system is modeled and simulated under MATLAB/SIMULINK. As well as experimental results are presented to verify the feasibility of the proposed approach. The design and realization of the proposed system are presented and each part of the system will be discussed in detail.