A hybrid DDAO-RBFNN strategy for fault tolerant operation in fifteen-level cascaded H-bridge (15L-CHB) inverter with solar photovoltaic (SPV) system

Abstract

This paper proposes a hybrid approach for fault-tolerant operation in fifteen-level cascaded H-Bridge (15LCHB) inverter with solar photovoltaic system (SPV). Fault-tolerant operation performs two tasks: detection (classification of normal or abnormal) and identification (identification of faults). The proposed intelligent controller is the combination of Differential Annealing Dynamic Optimization (DDAO) and Radial Basis Function Neural Network (RBFNN) named DDAO-RBFNN approach. The major intention of the DDAO-RBFNN approach is separated into fault detection with lessening of complete photovoltaic solar energy conversion system. The proposed system takes into account the failure of the entire autonomous photovoltaic solar conversion process and manages the output voltage (OV) owing to the conditions of partial shading (PS). The collected data set (normal, abnormal) is stored at workstation. RBFNN is utilized to compute the CMI switching pattern that changes the PWM carrier signal. DDAO get the voltage values from RBFNN and evaluates them with the data set, if there is a rate of change of voltage. Any deviation find in the rate of voltage change, it makes a decision based on standard or fault conditions. The proposed system is activated on MATLAB/Simulink working platform, then the efficiency is compared with existing approaches.