A New Support Vector Regression Model for Equipment Health Diagnosis with Small Sample Data Missing and Its Application

Table 7

Prediction effect of missing data of Pump A for three different prediction models.

	Single-variable prediction		Multiple-variable prediction		Dynamic weight combination prediction
	RMSE	MAPE	RMSE	MAPE	RMSE	MAPE
GA-SVR	0.0486	0.28	0.0423	0.22	0.0138	0.08
SVR	0.0737	0.40	0.0500	0.25	0.0303	0.16
BPNN	0.0920	0.52	0.0644	0.36	0.0547	0.28