Current Characteristics Estimation of Si PV Modules Based on Artificial Neural Network Modeling.

Area of Science:

• Renewable Energy
• Materials Science
• Electrical Engineering

Background:

• Outdoor evaluation of photovoltaic (PV) systems is complex due to environmental variables like temperature and irradiance.
• Accurate diagnosis of PV modules is crucial for maintaining optimal energy generation and system efficiency.

Purpose of the Study:

• To develop and validate an artificial neural network (ANN) model for predicting the performance of silicon (Si) PV modules.
• To evaluate PV module performance using statistical metrics and regression analysis, bypassing traditional analytical models.

Main Methods:

• An artificial neural network (ANN) was trained using temperature, irradiance, and voltage as inputs to predict current output.
• The model generates current-voltage (I-V) curves by repeating measurements, employing a data-driven, black-box approach.
• This method avoids the need for conventional PV module parameters like series and shunt resistance.

Main Results:

• The ANN model successfully predicted I-V curves for Si PV modules across a range of irradiance and temperature conditions.
• Performance evaluation using mean bias error, mean square error, and regression analysis demonstrated the model's validity.
• The data-driven approach proved effective in capturing PV module behavior without relying on physical device equations.

Conclusions:

• The proposed ANN model offers a robust and accurate method for evaluating PV module performance.
• This data-driven approach simplifies the diagnosis and prediction of PV systems, enhancing operational efficiency.
• The validated algorithm shows significant potential for real-world application in PV system monitoring and maintenance.