Design and Analysis of LSTM-MPPT using PI Controller under Dynamic Loading Condition

The surge in renewable energy production in developing countries due to climate change has led to an increase in clean and green solar PV technology. However, the limited efficiency of solar PV modules has prompted researchers to focus on generating maximum power from them, particularly under partial shading conditions. The use of Maximum Power Point Tracker (MPPT) has become popular for this purpose. However, the performance of solar PV under partial shading conditions while maximizing power output is still an area that needs improvement. This paper proposes a Particle Swarm Optimization (PSO) algorithm based on the Perturb and Observe (P&O) MPPT technique, with an improved search space. The proposed algorithm has been optimized using Fuzzy Fokker Planck solution and tested under four different concentrations of partial shading. The proposed algorithm has been validated using both hardware setup and simulation studies. Simulation results obtained using MATLAB show that the proposed MPPT technique offers higher and better performance in terms of Maximum Power Point (MPP) convergence and improves the efficiency of the proposed system under partial shading conditions, making the system more robust.