Temperature Prediction of Photovoltaic Panels Based on Support Vector Machine with Pigeon-Inspired Optimization

Table 1

Comparison of prediction accuracy for the Mackey–Glass time series.

Method	Optimized parameters		MSE	RMSE	Accuracy improvement (%)
	SVM	0.1				1	4.1113e − 04	2.03e − 02	—
DE-SVM	17.7584	1.5187	1.2462e − 04	1.12e − 02	69.69
PSO-SVM	16.2988	0.6093	6.6701e − 05	8.18e − 03	83.78
PIO-SVM	18.6471	0.6140	6.6069e − 05	8.13e − 03	83.90
OPIO-SVM	16.8580	0.6129	6.6346e − 05	8.14e − 03	83.86
DFPIO-SVM	16.7831	0.6151	6.5455e − 05	8.09e − 03	84.08