Wind speed prediction for site selection and reliable operation of wind power plants in coastal regions using machine learning algorithm variants

Sustainable Energy Research

Table 9 Creating and comparing 14 models using tenfold cross-validation and hyperparameter tuning with Hyperopt optimization for Dataset 1 (best results are bolded)

Weather station	Model	10-fold cross-validation			Hyperparameter tuning with Hyperopt
Weather station	Model	MSE	MAE	R²	MSE	MAE	R²
Kutubdia	MLR	0.4174	0.4325	0.5782	0.4174	0.4325	0.5782
	Lasso	0.9899	0.7127	-0.0003	0.4310	0.4477	0.5645
	Ridge	0.4174	0.4325	0.5782	0.4174	0.4325	0.5782
	Elastic Net	0.9256	0.6878	0.0648	0.4240	0.4397	0.5716
	KNN	0.4350	0.4291	0.5604	0.4096	0.4172	0.5863
	DT	0.7904	0.5412	0.1976	0.4229	0.4250	0.5727
	RF	0.4021	0.4147	0.5937	0.3919	0.4086	0.6039
	GBR	0.3855	0.4089	0.6105	0.3789	0.4030	0.6174
	AdaBoost	0.6720	0.5978	0.3225	0.4626	0.4598	0.5320
	XGBoost	0.3980	0.4073	0.5976	0.3809	0.4041	0.6152
	LightGBM	0.3798	0.4018	0.6163	0.3789	0.4020	0.6173
	CatBoost	0.3745	0.3984	0.6218	0.3744	0.3990	0.6218
	LSTM	0.3964	0.4173	0.5995	0.4350	0.4501	0.5604
	GRU	0.3984	0.4194	0.5973	0.4050	0.4229	0.5908
Cox's Bazar	MLR	1.1323	0.7412	0.4182	1.1323	0.7412	0.4182
	Lasso	1.9466	1.1068	-0.0002	1.1479	0.7460	0.4102
	Ridge	1.1323	0.7412	0.4182	1.1323	0.7413	0.4182
	Elastic Net	1.8116	1.0660	0.0693	1.1418	0.7431	0.4134
	KNN	1.1381	0.6638	0.4152	1.0675	0.6511	0.4516
	DT	1.9969	0.8122	-0.0265	1.0338	0.6485	0.4691
	RF	0.9779	0.6291	0.4976	1.0116	0.6406	0.4802
	GBR	0.9615	0.6286	0.5061	0.9546	0.6251	0.5095
	AdaBoost	1.0962	0.8532	0.3716	0.98144	0.7329	0.4375
	XGBoost	0.9982	0.6294	0.4872	0.9524	0.6209	0.5107
	LightGBM	0.9472	0.6192	0.5135	0.9468	0.6184	0.5137
	CatBoost	0.9462	0.6164	0.5140	0.9382	0.6162	0.5180
	LSTM	1.0051	0.6588	0.4835	0.9943	0.6464	0.4892
	GRU	1.0067	0.6569	0.4827	1.0042	0.6552	0.4839