Photovoltaic (PV) panels exhibit a non-linear current-voltage characteristic with a Maximum Power Point (MPP) that varies due to environmental factors such as solar radiation and ambient temperature. In this study, an Artificial Neural Network (ANN)-based MPPT method, called the ANN-based Adaptive Reference Voltage (ARV) method, is proposed to determine the optimal operating point of the PV panel. The ANN-based ARV method is a voltage-controlled approach that can adapt to changing atmospheric conditions. The performance of the proposed method is evaluated using both a normal Proportional-Integral (PI) controller and an anti-windup PI controller. Comparative analysis is conducted with the widely used Perturb and Observe (P&O) and Incremental Conductance (INC) methods in the MATLAB/Simulink environment, considering three different atmospheric scenarios with varying radiation levels according to EN50530 standards. The proposed method demonstrates superior efficiency with overall results of 99.4%, 95.9%, and 96% in scenario 1, scenario 2, and scenario 3, respectively. Particularly, the proposed method exhibits notable superiority in rapidly changing atmospheric conditions.

Maximum Power Point Tracking (MPPT) algorithms are crucial for maximizing power extraction from photovoltaic (PV) systems. Traditional MPPT methods often exhibit suboptimal performance under partial shading conditions. Hence, advanced MPPT algorithms have been developed to enhance efficiency in such scenarios. The voltage scanning-based MPPT algorithm is notable for its superior performance under partial shading, characterized by high tracking speed and efficiency. This study introduces a novel enhancement to this method—namely, a voltage skipping algorithm designed to further improve tracking speed. Unlike conventional scanning techniques, this approach dynamically calculates skipping voltages during operation, eliminating the need for prior knowledge of PV panel characteristics. Performance evaluation was conducted using a MATLAB/Simulink model of a PV system comprising 4 series panels and a boost converter, subjected to simulations under 5 distinct partial shading scenarios. Comparative analyses with established optimization algorithms such as particle swarm optimization, cuckoo search algorithm, and grey wolf optimization highlight the proposed method's effectiveness in terms of tracking speed and maximum power output. The proposed algorithm has worked with high efficiency of 99.28 %, 99.61 %, 99.13 %, 99.16 % and 99.58 % in 5 different scenarios, respectively. By using the proposed method, a significant superiority has been achieved over other methods, with tracking speeds of 0.26s, 0.22s, 0.2s, 0.22s and 0.26s, respectively.

Maximum Power Point Tracking (MPPT) algorithms are used to obtain maximum power from PV (Photovoltaic) systems in variable atmospheric conditions. Partial shading conditions (PSCs) have an important effect causing the maximum power points of PV systems to change. In this study, a new hybrid MPPT algorithm was proposed by combining Voltage Scanning (VSC) and Cuckoo Search Algorithm (CSA) methods to obtain maximum power from a PV system running under PSC. To test the proposed method, a 252.6 kW grid-connected PV system with 609 panels, including 7 serial panels and 87 parallel strings, was created in the MATLAB/Simulink environment. This system was tested under 10 different PSCs by using VSC, Particle Swarm Optimization (PSO), CSA, and the proposed MPPT methods. The average tracking efficiency and speed of the proposed method under these conditions were obtained as 99.74% and 0.3s, respectively. Under PSCs, the proposed method achieved the highest tracking efficiency of 99.92%, while the highest tracking speed that it achieved was 0.22s. On the other hand, the superiority of the proposed hybrid MPPT algorithm was clearly demonstrated by comparing it to the VSC, PSO, and CSA methods.