According to the chapter, which AI model (ANN, SVM, or RF) performed best in both training and testing phases for solar PV forecasting, and what were its RMSE values for training and testing?

According to the chapter, ANN performed best in both the training and testing phases for solar PV forecasting, with RMSE values of 1.4568 for training and 1.9874 for testing.

The Results and Conclusion portions of the chapter state that ANN outperformed SVM and RF on the evaluation metrics, including RMSE. Table 5.2 gives ANN's training-phase RMSE as 1.4568 and its testing-phase RMSE as 1.9874. One caveat: the same table lists RF's training-phase RMSE as 1.3166, which is lower than ANN's value, so the printed table is not fully consistent with the chapter's claim that ANN had the best training-phase RMSE.