A Novel Bat Algorithm-Optimized Quantum Neural Network-Based MPPT Techniques for Enhanced Photovoltaic in Grid Connected System Performance

Renewable energy source (RES) production has increased nowadays due to the depletion of fossil fuel reserves and the rise in greenhouse emissions. Incorporating RES into the grid may produce additional problems in power quality, system stability, harmonics and reliability owing to their sporadic nature. The RES is capable of operating in both ON and OFF electrical grid modes since ON-grid power systems are linked to the utility grid, while off-grid systems are not. However, because of the environmental variability, it can be challenging to obtain optimal voltage and extract maximum RES power. It is crucial to derive an optimal solution to obtain the maximum power from RES. Therefore, this paper proposes a bat algorithm for tracking (BAT) with a quantum neural network (QNN) for extracting maximum power from a PV panel. The Bat Optimization Algorithm (BOA) is used to select optimal parameters for QNN. The suggested method is designed via MATLAB/Simulink, and various parameters are considered, in which total harmonic distortion (THD) and efficiency analysis of the proposed method with an ant lion optimized cuckoo search (ALO-CS) algorithm are compared. Simulation and hardware implementation results demonstrate superior performance compared to the ALO-CS method. The proposed approach achieves a peak efficiency of 93.6% in simulation and 93.4% in hardware, with convergence times of 0.1[Formula: see text]s and 0.16[Formula: see text]s, respectively. Harmonic distortion is significantly reduced, with THD values of 0.43% in simulation and 0.7% in hardware using a current control technique. These results confirm that the QNN-BOA method provides faster tracking, higher efficiency and better power quality than conventional maximum power point tracking (MPPT) techniques, making it highly suitable for hybrid RES applications.