Photovoltaic Power Forecasting Using Support Vector Machine and Adaptive Learning Factor Ant Colony Optimization

Due to its flexible and clean nature, distributed Photovoltaic (PV) power is essential for solving energy and operational coordination challenges in integrating solar energy with other power sources. However, with rapid development complexities have emerged including in climate prediction, resulting in longer training times than those for the signal model. The Adaptive Leaning Factor Ant Colony Optimization (ALFACO) is used to optimize the ability to predict PV power values by linearly decreasing the weight and involved an adaptive learning factor to enhance the optimizing capability of the ACO. The proposed method, a combination of ALFACO and Support Vector Machine (SVM) can improve the robustness of the model, making it less likely to get stuck in local optima and making the SVM more capable of handling noisy or complex datasets. The proposed method, ALFACO-SVM achieves superior performance metrics including Root Mean Square Error (RMSE) of 0.0635, Coefficient of Determination <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R²</tex> of 0.8920, Mean Absolute Error (MAE) of 0.0091, Median Absolute Error (MedAE) of 0.5620, Mean Squared Error (MSE) of 0.0045 and Explain Variance Score (EVS) of 0.0095, compared to existing methods such as Long Short-Term Memory (LSTM), Artificial Neural Network (ANN), LSTM and Bias Compensation LSTM (BCLSTM).