Research Article

An Experimental Investigation of FNN Model for Wind Speed Forecasting Using EEMD and CS

Table 3

The multi-step ahead forecasting errors of the four models in Jan.

StepsErrorsWTG 25WTG 26WTG 27WTG 28
Model 1aModel 2bModel 3cModel 4dModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4

One-step aheadMAE0.840.630.600.460.900.600.630.461.110.770.790.640.890.620.620.50
MAPE7.475.745.464.3110.666.886.975.089.867.197.345.938.476.036.124.77
RMSE1.150.850.800.641.140.800.840.622.141.091.220.911.230.830.830.68

Two-step aheadMAE1.080.710.730.570.960.690.790.531.360.971.300.641.170.680.750.54
MAPE9.796.676.805.2810.988.339.156.0312.179.3011.236.2811.406.847.405.13
RMSE1.490.950.980.741.240.881.050.693.251.332.970.851.600.920.980.74

Three-step aheadMAE1.080.770.790.631.260.760.750.641.180.981.070.811.030.760.800.62
MAPE10.737.277.415.9015.348.988.837.2611.6610.0010.548.0110.097.667.966.10
RMSE1.411.021.070.831.600.990.990.831.591.321.511.131.411.011.070.83

Four-step aheadMAE0.980.850.860.710.890.810.870.691.560.991.150.871.100.830.850.70
MAPE9.198.088.196.7710.449.7311.058.1214.3010.5511.538.9610.818.538.667.10
RMSE1.301.131.130.941.161.071.130.922.971.321.851.141.441.111.150.93

Five-step aheadMAE1.020.940.920.811.390.900.890.791.981.121.221.001.100.910.930.81
MAPE9.558.928.797.6317.6710.7511.299.3518.2811.8612.5310.3810.979.379.498.36
RMSE1.361.221.211.031.721.191.181.033.031.491.731.311.481.221.241.09

Six-step aheadMAE1.181.011.010.901.140.930.990.871.491.151.411.131.180.970.970.90
MAPE10.859.619.568.6413.2311.3412.2010.3514.5112.2314.0111.7112.959.9910.129.22
RMSE1.581.311.321.181.501.241.291.152.601.512.621.541.561.301.311.19

FNN model; bEEMD-FNN model; cCS-FNN model; dCS-EEMD-FNN.