A hybrid control topology for cascaded H-bridge multilevel inverter to improve the power quality of smart grid connected system: NBO-RERNN approach

Abstract

This manuscript proposes a hybrid control technique for the grid-connected photovoltaic (PV) generation system with a cascaded multilevel inverter (CMLI). The proposed hybrid approach integrates the Namib beetle optimization (NBO) and recalling-enhanced recurrent neural network (RERNN) approach, known as the NBO-RERNN technique. The cascaded multilevel inverter (MLI) is intended to increase the optimum control signal using the proposed controller. The cascaded multilevel inverter (MLI) is designed by using the minimum count of switches. The key purpose of the proposed approach is to improve power regulation or maximal solar energy conversion system (SECS) and to achieve good power quality (PQ) of the system. The proposed NBO technique generates the optimum control signal dataset offline. Based on the fulfilled dataset, the RERNN generates the best CMLI control signals inonlinemanner. The resultant control signals are used to regulate cascaded multilevel inverter insulated gate bi-polar switches (IGBT). For this purpose, the proposed NBO-RERNN control topology defines converter switching states by modeling generation system operating modes. The parameter variations of the system and external disturbances are mitigated, and the load demands are fulfilled optimally by using this control technique. The NBO-RERNN control topology is implemented in MATLAB, and its performance is compared with existing approaches. The simulation reveals that the proposed method has lower THD than the existing one.