Deep neural networks model for accurate photovoltaic parameter estimation under variable weather conditions

Abstract

Estimating photovoltaic (PV) parameters is essential for accurate modeling and performance prediction of PV systems. This paper presents a deep neural network-based approach for determining the PV parameters via information from datasheets. The proposed technique is trained using thousands of data points generated from the PV module block in the MATLAB/Simulink library. The effectiveness of the model is evaluated using metrics such as Mean Absolute Percentage Error (MAPE), the coefficient of determination (R-squared), and Root Mean Square Error (RMSE). By utilizing the inherent pattern recognition and learning capabilities of neural networks, the model is able to estimate the PV parameters accurately. To evaluate the effectiveness of the proposed approach, the performance is subjected to different assessments including testing data, experimental data and commercial PV modules under standard test conditions (STC) as well as different weather conditions. The performance has been also compared with various recent algorithms reported in the literature. The results obtained from all assessments provide insights into the performance of the proposed approach. The findings demonstrate the effectiveness of the neural network-based method in estimating PV parameters, showcasing its potential as a viable alternative to traditional estimation techniques.