
Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)
Published: April 26, 2025
The conventional DC-DC architecture is often used in photovoltaic systems, with control based on incremental conductance-based maximum power point tracking (MPPT-IC) algorithms. Despite its simplicity and ability to ensure voltage balance across the output capacitors, this suffers several drawbacks. These drawbacks cause undesirable problems such as high ripples, overshoot, limited dynamic response. Therefore, paper proposes a three-level quadratic boost converter suitable solution replace inverters while combined an advanced MPPT method. new approach NARX neural network (NARAX-NN) This proposed strategy designed for accuracy, robustness, fast response compared MPPT-IC strategy. In work, MPPT-NARX-NN of different strategies (MPPT-IC, type 1 fuzzy logic (MPPT-T1FL), 2 (MPPT-T2FL)). comparison performed using MATLAB under operating conditions. Simulation results indicate that significantly improves operational performance other strategies, increasing reliability deployment systems. numerical show rise time by 96.43, 97.34, 94.50% MPPT-IC, MPPT-T1FL, MPPT-T2FL, respectively. Also, settling improved 50, 66.66, 6.66% Furthermore, increases average efficiency (%) 3.86 1.12% highlight effectiveness extracting energy, flexibility, improving system reliability, making solar PV systems more efficient promising indispensable solution.
Language: Английский