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

Table 2

Comparison of prediction accuracy for NARMA.

Method	Optimized parameters		MSE	RMSE	Accuracy improvement (%)
	SVM	0.1				1	1.0155e − 05	3.187e − 03	—
DE-SVM	1.8581	13.3556	1.0340e − 05	3.216e − 03	−1.82
PSO-SVM	0.0719	7.9756	1.0141e − 05	3.184e − 03	0.13
PIO-SVM	0.6372	12.5727	1.0140e − 05	3.184e − 03	0.14
OPIO-SVM	0.0772	13.4447	1.0075e − 05	3.174e − 03	0.79
DFPIO-SVM	0.0130	10.0047	9.9202e − 06	3.150e − 03	2.31