Comparative MPPT performance of particle swarm optimization, grey wolf optimizer, and crested porcupine optimizer under partial shading conditions

In this study, Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), and Crested Porcupine Optimizer (CPO) algorithms were comparatively evaluated for global maximum power point tracking in photovoltaic systems under standard test and partial shading conditions. The MATLAB/Simulink model consists of four series-connected 250 W PV modules, bypass diodes, a boost DC–DC converter, a PWM generator, and an MPPT controller that directly determines the duty cycle of the converter. The algorithms were tested under three operating scenarios: standard test conditions, mild partial shading, and severe partial shading. To ensure a fair comparison, identical PV system parameters, converter settings, duty-cycle limits, and a common true GMPP reference were used for all algorithms. The performance of each method was assessed using steady-state average power, common-reference-based MPPT efficiency, settling time, steady-state power oscillation, and energy yield. Under standard test conditions, PSO showed the most balanced performance in terms of tracking accuracy and low oscillation, achieving a steady-state power of 998.972 W. Under mild partial shading conditions, CPO provided the best combined performance, with a steady-state power of 485.000 W, a power oscillation of 0.000055%, and a settling time of 0.15273 s. Under severe partial shading conditions, GWO achieved the steady-state power closest to the common GMPP reference, with 363.798 W, whereas PSO provided the fastest response and the lowest steady-state oscillation. The results indicate that no single algorithm is superior under all operating conditions. Therefore, the most suitable MPPT method should be selected by jointly considering tracking accuracy, convergence speed, steady-state stability, and application-specific requirements.