Power quality improvement using a 31-level multi-level inverter with bio-inspired optimization approach

In recent years, the power quality (PQ) improvements have been explored through various approaches. The employment of electronic devices with renewable energy sources has expanded the harmonics level of voltage and current. Due to harmonics, the PQ of a specific electrical system gets affected. At critical load conditions, the traditional PQ mitigation approaches fail to develop the performance of the system. Therefore, in this work, the Spider Monkey Optimization convolutional neural network (SM-CNN)-based 31-level multilevel inverter (MLI) is used. This method balances the reactive power demands and enhances real power in the grid-tied photovoltaic (PV) system. The maximum power point tracking (MPPT) algorithm depending on radial basis function neural networks (RBFNNs) is used to maximize PV power. For strengthening the voltage level of the PV and to generate higher DC voltage with a minimized switching loss, an integrated boost fly back converter (IBFC) is introduced. The presented technique is implemented in the MATLAB/Simulink platform to figure out the estimation of PQ issues. The suggested MLI lessens the total harmonic distortion (THD) value to 2.45% with an improved power factor.