Fault Classification with Convolutional Neural Networks for Microgrid Systems

Table 9

Comparison of accuracies for the developed model with literature and conventional approaches.

References	Approach	Validation accuracy (%)	Testing accuracy (%)	Detection time
[10]	Decision tree			Not available
	k-nearest neighbor
	Support vector machine
	Naive Bayes
[11]	Decision tree	Not available
	Over current relay
	Differential relay
	Random forest
[13]	Cosine radial basis function network (RBFN)
	Gaussian Euclidian RBFN
	Manual fusion RBFN
	Dynamic fusion RBFN
Conventional approaches (SA mode)	Support vector machine
	k-nearest neighbor
	Decision trees
Conventional approaches (SA mode)	Support vector machine
	k-nearest neighbor
	Decision trees
Proposed approach (SA mode)	Convolutional neural networks
Proposed approach (GC mode)	Convolutional neural networks