The efficiency of photovoltaic (PV) systems often depends on the effectiveness of the cells or modules. With the advancements in PV panels, DC-DC converters, and loads/batteries, there is
an increasing focus on maximum power point tracking (MPPT) algorithms' operation under various irradiance conditions. These algorithms are continuously improved to ensure the PV
system's optimum performance and to address the non-linearity of PV characteristics under different conditions. Current MPPT algorithms, including conventional and soft computing
methods, are crucial for maximizing PV output but still face challenges such as tracking accuracy under complex conditions (e.g., rapidly and gradually changing weather), efficiency and stability
concerns (oscillations around the maximum power point and convergence time), and the tradeoffs between tracking speed and complexity. To overcome these shortcomings, a modified duty
cycle sweeping (MDCS) algorithm which is formulated based on a hill-climbing algorithm is proposed in this work. The goal is to alleviate the limitations of the conventional perturb and
observe (P&O) algorithm, specifically (1) the diverging direction of tracking during the dynamic change in irradiance and (2) failure to follow the global peak (GPMPP) when a rapid increase in
irradiance occurs. This failure occurs due to incorrect decisions taken and trapped at the local peak power (LP). MDCS algorithm is proposed to avoid the diverging direction by incorporating
the information of change in duty cycle in the tracking process, change in global peak (GPMPP), and the change in output power (PO). To verify the performance, the proposed method was
compared to the conventional P&O and particle swarm optimization (PSO) method under various environmental conditions using MATLAB/Simulink® software for simulation verification. A
DC-DC single-ended primary inductance converter (SEPIC) converter was constructed for experimental validation and the proposed MPPT algorithm is implemented into TMS320F28335
DSP controller board. Both simulation and experimental validations were performed under similar irradiance conditions. The results demonstrate that the proposed MDCS algorithm
effectively adapts to changing weather conditions by automatically directing the operating point of the system towards the optimal or nearest MPP position. The MDCS algorithm is regarded as
optimal in various aspects, with each verification measurement demonstrating outstanding tracking accuracy and efficiency exceeding 95% across the majority of tested scenarios.
