Table 3 Performance of support vector regression across DR strategies.

From: Lightweight machine learning framework using temporal features for electric vehicle demand response forecasting on edge devices

Model

Strategy

MAE

MSE

MAPE

CV RMSE

R2

SVR

Peak

clipping

3.919342

29.980715

11.03

0.118

0.600357

SVR

Valley

filling

3.909713

28.58261

13.92

0.143

0.766542

SVR

Load

shifting

4.021035

30.174104

10.68

0.155

0.677378

SVR

Strategic

conservation

3.826682

27.840096

13.33

0.069

0.730547

SVR

Strategic

load growth

3.756972

27.49661

9.61

0.198

0.919789

SVR

Flexible

load shape

4.01228

31.303994

14.45

0.088

0.67134

Back to article page