Comparison and Optimization of Neural Networks and Network Ensembles for Gap Filling of Wind Energy Data

Table 10

Summary of the results for the calculation of the income of investment.

Target variable	Input variables	Network type	RMSE (kWh)	𝑟𝑆
𝑊SU1	𝑊(CO1--CO5)	Ensemble; (2 × MLP + 1 × RBF)	1010	0.97

SU 1 using an ensemble from the previously optimized individual networks.