A PSO-Enhanced Neural Network Framework for Robust Maximum Power Point Prediction in Solar PV Arrays
Nitin Machindra Pardeshi
, Prof. Ganesh G. Mhatre
Artificial neural network (ANN), Solar Photovoltaic (PV), Maximum power point tracking (MPPT), Particle Swarm Optimization (PSO), Perturb and observe
This paper introduces a hybrid Artificial Neural Network (ANN)–Particle Swarm Optimization (PSO) framework for accurate Maximum Power Point Tracking (MPPT) in photovoltaic (PV) systems. While conventional ANN-based controllers often struggle with weight initialization and architectural tuning under dynamic weather conditions, the proposed approach employs PSO to optimize both network topology and initial parameters, thereby improving convergence and prediction reliability. The hybrid design mitigates common issues of overfitting and slow learning, achieving a balanced trade-off between computational cost and tracking accuracy. The framework is validated in MATLAB/Simulink using real-world PV datasets recorded under diverse climatic conditions, including clear and partially cloudy skies. Comparative analysis against established MPPT strategies—Perturb and Observe (P&O), Fuzzy Logic Controllers (FLC), and standard ANN methods—demonstrates superior performance of the PSO-enhanced ANN, with faster convergence, improved stability, and higher robustness. Experimental results indicate average tracking efficiencies of 99.6% in stable irradiance and 99.3% under variable shading, highlighting its capability to maximize energy yield in grid-connected PV systems. The study concludes that integrating swarm intelligence with ANN provides a scalable and adaptable MPPT solution, with future scope in large-scale solar farms and inclusion of additional environmental predictors for further accuracy enhancement
"A PSO-Enhanced Neural Network Framework for Robust Maximum Power Point Prediction in Solar PV Arrays", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.10, Issue 9, page no.b88-b98, September-2025, Available :https://ijsdr.org/papers/IJSDR2509114.pdf
Volume 10
Issue 9,
September-2025
Pages : b88-b98
Paper Reg. ID: IJSDR_304947
Published Paper Id: IJSDR2509114
Downloads: 00050
Research Area: Science and Technology
Country: Raigad, Mumbai, India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave
