The integration of shunt bypass diodes in photovoltaic (P-V) module to mitigate hot spots frequently leads to the emergence of multiple in the PV array characteristics. Researchers consistently strive to develop, integrate, and refine innovative techniques inspired by various natural processes to achieve a global optimum that enhances the overall efficiency of PV systems. However, these techniques face challenges in adapting parameters to strike a delicate balance between exploration and exploitation, which is essential for circumventing local optima, reducing computation times, and refining precision to optimize energy capture. In this context, this paper introduces a groundbreaking new adaptive Maximum Power Point Tracking (MPPT) controller inspired by the social behavior and reproductive tactics observed in bonobos (BO). This innovative approach is underpinned by two key strategies: fission and fusion, with dynamic parameter adjustment in real-time. this enables for efficient exploration and exploitation of the search space, following the positive and negative phases of the BO. This method was compared with three methods: PSO, DE, and ICS, and evaluated through six simulation scenarios, ranging from 1 to 6 peaks, as well as three experimental scenarios: one uniform and the other two involving partial shading, using an Arduino board and a buck converter. According to the comparative analysis, the new BO algorithm outperforms the three other approaches in all performance evaluation parameters. It shows an average improvement in convergence time of more than 39.18 % and an average precision exceeding 99 %, with minimal oscillation in steady-state operation. This translates into an average MPE efficiency of over 96.66 %. Additionally, the experimental results confirm the findings from the simulations.
1 June 2025 in International Journal of Power Electronics and Drive Systems (IJPEDS)
Photovoltaic systems (PVS) exhibit variability in their maximum power point (MPP) output due to variations in irradiance and cell temperature. This can lead to reduced efficiency, as maximum power point tracking (MPPT) algorithms often have slow response times and limited ability to adapt to rapidly changing environmental conditions. New algorithms are therefore needed to capture more energy and improve the efficiency of these systems. In this context, this article presents a new method for temperature parametric (TP) and its implementation using a digital PI controller, a buck converter, and MATLAB-Simulink. This innovative approach relies on detecting the MPP by continuously measuring the cell temperature of the PV panel (𝑇𝑐𝑒𝑙𝑙) and solar irradiance (S). A 3D linear regression model connects these two parameters with the maximum current (𝐼𝑚𝑝𝑝), enabling real-time monitoring of the MPP. We have applied this new method on two different types of PV (POLY-40W and BPSX330J) under a range of environmental conditions, including stable and dynamic scenarios. The results of the simulation demonstrate the superiority of our approach compared to the hill climbing (HC) for perturbation steps of HC (1%) and HC (2%). Our method achieves faster convergence time 0.009 s and high MPPT efficiency at 98.18%, fewer steady-state oscillations, and better detection.
1 March 2025 in International Journal of Power Electronics and Drive Systems (IJPEDS)
This paper presents a comparative study of five maximum power point tracking (MPPT) control techniques in photovoltaic (PV) systems. The algorithms evaluated include classical methods, such as perturb and observe (P&O) and incremental conductance (IC), as well as intelligent approaches such as fuzzy logic (FL), artificial neural networks (ANNs), and adaptive neuro-fuzzy inference system (ANFIS). Intelligent methods provide faster response times and fewer oscillations around the maximum power point (MPP). The structure of the PV system includes a PV generator, load, and DC/DC boost converter driven by an MPPT controller. The performance of these techniques is analyzed under identical climatic conditions (same irradiation and temperature) in terms of efficiency, response time, response curve, accuracy in tracking the MPP, and others considered in this work. Simulations were performed using MATLAB-Simulink software, demonstrating that ANNs and ANFIS outperform traditional methods in dynamic environments, with FL being computationally intensive. P&O exhibited significant oscillations, while IC a showed slower tracking speed.