An RBF Neural Network Combined with OLS Algorithm and Genetic Algorithm for Short-Term Wind Power Forecasting

Table 1

The evaluation of the accuracy of the three methods in wind power forecasting.

Season	Forecasting method	Maximum absolute percentage error	Mean absolute percentage error
	Proposed RBF neural network-based method	20.5026%	2.4676%
Winter day	Persistence method	21.4370%	2.7579%
	Back propagation neural network method	47.4301%	4.3943%
	Proposed RBF neural network-based method	57.4755%	15.4433%
Summer day	Persistence method	113.0435%	35.4214%
	Back propagation neural network method	108.1397%	25.2375%
	Proposed RBF neural network-based method	66.9832%	7.3247%
Spring day	Persistence method	116.4384%	8.4948%
	Back propagation neural network method	76.1072%	8.4283%
	Proposed RBF neural network-based method	121.7294%	29.0453%
Autumn day	Persistence method	186.6667%	39.5734%
	Back propagation neural network method	146.1742%	36.7328%