Performance Analysis of an Optimized ANN Model to Predict the Stability of Smart Grid

Table 3

Accuracy for every fold using repeated random test-train split.

S.No.	Model	Accuracy Fold1	Accuracy Fold2	Accuracy Fold3	Accuracy Fold4	Accuracy Fold5	Accuracy Fold6	Accuracy Fold7	Accuracy Fold8	Accuracy Fold9	Accuracy Fold10
1	Logistic regression	0.81405	0.81527	0.8145	0.81838	0.8146	0.81561	0.8131	0.81116	0.81361	0.81272
2	KNN	0.813889	0.82283	0.81522	0.82078	0.81656	0.81267	0.81639	0.81561	0.81628	0.81789
3	Naïve Bayes	0.832556	0.8315	0.82444	0.83039	0.83056	0.83256	0.83239	0.82828	0.83311	0.83156
4	Decision tree	0.83305	0.83156	0.83183	0.837	0.83328	0.83244	0.83178	0.8315	0.8345	0.83139
5	SVM	0.927667	0.92706	0.92622	0.92944	0.92806	0.92628	0.92789	0.92767	0.92844	0.92911
6	Random forest	0.907056	0.90939	0.90572	0.90983	0.90928	0.909	0.90772	0.91183	0.90817	0.91056
7	XGBoost	0.92983	0.92894	0.92806	0.9275	0.93406	0.93206	0.92817	0.93133	0.93078	0.92794
8	Optimized ANN	0.9571	0.9705	0.9724	0.9771	0.9802	0.981	0.9781	0.9836	0.9821	0.98

Bold face represents the highest accuracy achieved for every model among different folds.