An efficient AOA-RERNN control approach for a non-isolated quasi-Z-source novel multilevel inverter based grid connected PV system

Abstract

In this paper, an effective control strategy based non-isolated Quasi-Z-Source (QZS) novel multilevel inverter topology (NIQZS-NMLI) is proposed for interfacing photovoltaic (PV) system. The proposed approach is the combination of Archimedes optimization algorithm (AOA) and Recalling-Enhanced Recurrent Neural Network (RERNN) called AOA-RERNN approach. Here, the modelling design of non-isolated QZS-NMLI topology is developed with new storage devices to distribute the maximum power from the photovoltaic power generating system. This novel multilevel inverter topology reduced the number of switches and the total harmonic distortion of the system, and also it is used to achieve the higher boost capability, lesser voltage stress across the active switching devices, and greater modulation index for the inverter. The objective function is determined depending on its controller parameters with constraints, like voltages, current, power and modulation. These parameters have been employed as the inputs of AOA-RERNN approach. This AOA-RERNN approach increases the voltage profile, power supply and decreases the power oscillations when sharing the power to the load. The maximum power distribution is guaranteed to the load by RERNN depending on the extraction of maximum power from the PV source. The proposed approach is executed in MATLAB/Simulink site; its performance is analyzed with existing approaches.