A novel technique for forecasting the optimal production of a wind generator installed at any geographical point located within a very large area

Journal of Electrical Systems and Information Technology

Table 1 CNN and LSTM parameters

CNN parameters	LSTM parameters
INPUT_WIDTH: 3 to 72 h
LABEL_WIDTH: 1 to 24 h
Filters: 10 to 300	Units: 10 to 1000
Units: 10 to 1000	Units: 10 to 1000
Learning_rates: \({\boldsymbol{ }10}^{-2},\boldsymbol{ }{10}^{-3},\boldsymbol{ }{10}^{-4}\)	Learning_rates: \({10}^{-2},\boldsymbol{ }{10}^{-3},\boldsymbol{ }{10}^{-4}\)