Research Article

Wind Power Prediction Based on Nonlinear Partial Least Square

Table 1

Analytical results for working condition I.

ModelRMSECVRMSEPLVsHidden nodesWidth

PLS256.6612265.58350.69030.66702--
BP212.8876248.99090.77440.6842-1-
RBF471.8434429.79470.77390.0259-1-
SVM328.6061240.52290.77500.7250--0.6
BPE-NPLS288.2057232.04490.62620.711214-
RBFE-NPLS206.2611225.59730.77240.732482-

The bold means the minimum error or the maximum correlation among models.