Power Prediction of Combined Cycle Power Plant (CCPP) Using Machine Learning Algorithm-Based Paradigm

Table 6

Comparison of results achieved by each of the algorithms on both the dataset splits. GBRT outperforms the rest of the algorithms by achieving the least RMSE and AE such that the number of trees is 100.

(a)

RMSE
Data splits	KNN	GBRT	NN	DL	LR
Training data 90%	3.323	2.641	4.168	3.616	4.263
Training data 80%	3.534	2.851	4.088	3.823	4.347

(b)

AE
	KNN	GBRT	NN	DL	LR
Training data 90%	2.374	1.931	3.345	2.826	3.402
Training data 80%	2.499	2.052	3.256	2.928	3.469