Performance Enhancement of Solar Photovoltaic-Maximum Power Point Tracking Using Hybrid Adaptive Neuro-Fuzzy Inference System–Honey Badger Algorithm

Abstract

The work evaluates the effectiveness of three maximum power point tracking (MPPT) techniques: pulse width modulation (PWM)-based, adaptive neuro-fuzzy inference system (ANFIS)-based, and a proposed hybrid ANFIS–honey badger algorithm (HBA) model that combines ANFIS with the HBA. Experiments and simulations were conducted to assess the performances of these techniques in terms of output current, output voltage, simulation output power, experimental output power, and efficiency. The experimental data are collected under a solar irradiance of 1000 W/m2 and a 25 °C temperature. The outcomes demonstrate the efficacy of the hybrid model-based approach MPPT technique outperforms both the PWM-based and ANFIS-based techniques, achieving an output voltage of 100 V, output current of 5 A, simulation output power of 500 W, experimental output power of 413.21 W, and an efficiency of 98.74%. The hybridization of ANFIS with the HBA demonstrates superior performance by combining adaptive learning and evolutionary optimization techniques. These findings highlight the potential of the proposed ANFIS–HBA-based MPPT technique in enhancing power extraction efficiency and output performance in solar photovoltaic (PV) modules. The outcomes of this research provide valuable insights for developing and optimizing MPPT techniques in solar PV systems and aid in the increased use of energy from renewable sources.