Journal of Electrical and Computer Engineering / 2019 / Article / Tab 4

Research Article

Electricity Theft Detection in Power Grids with Deep Learning and Random Forests

Table 4

Experimental parameters for comparative methods.

MethodsInput dataParameters

LRFeatures (1D)Penalty: L2
Inverse of regulation strength: 1.0
SVMFeatures (1D)Penalty parameter of the error term:
Parameter of kernel function (RBF): 0.0005
GBDTFeatures (1D)The number of estimators: 200
RFFeatures (1D)The number of trees in the forest: 100
CNNRaw data (2D)The max depth of each tree: 30
The same as the CNN component of the proposed approach