Photovoltaic Power Estimation Using DBN, LSTM, CNN & GRU

GUI-based tool for photovoltaic power estimation using DBN, LSTM, CNN, and GRU with performance metrics, plots, and comparison results.

Description:

This MATLAB script estimates photovoltaic (PV) power output using four AI models: Deep Belief Network (DBN), Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and Gated Recurrent Unit (GRU). The program loads PV system data from a CSV file, trains the chosen models, and evaluates their performance. Results include statistical metrics and multiple visualizations for easy comparison.

Features:

• Reads input data (irradiance, temperature, wind speed, etc.) from CSV file
• Implements DBN, LSTM, CNN, and GRU for PV power estimation
• Calculates key metrics: RMSE, MAE, R²
• Saves per-model results into separate Excel files
• Generates plots:
• Correlation matrix of input features
• Time-series (Actual vs Predicted)
• Error histograms
• Scatter plots with y=x reference line
• Produces a final comparison table and bar chart across all models