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 12 The evaluation metrics for 14 models on both validation and test segment for Dataset 2 (best results are bolded)

Weather station	Model	Validation dataset			Test dataset
Weather station	Model	MSE	MAE	R²	MSE	MAE	R²
Kutubdia	MLR	0.5070	0.5316	0.7825	0.4912	0.5293	0.7849
	Lasso	0.5213	0.5372	0.7764	0.5007	0.5344	0.7807
	Ridge	0.5070	0.5316	0.7825	0.4912	0.5293	0.7849
	Elastic Net	0.5182	0.5379	0.7777	0.4988	0.5349	0.7815
	KNN	0.3972	0.4631	0.8296	0.3875	0.4604	0.8303
	DT	0.4665	0.5002	0.7999	0.4543	0.4980	0.8010
	RF	0.4208	0.4737	0.8194	0.4138	0.4756	0.8187
	GBR	0.3509	0.4242	0.8495	0.3447	0.4256	0.8490
	AdaBoost	0.5370	0.5475	0.7697	0.5159	0.5464	0.7741
	XGBoost	0.3427	0.4222	0.8530	0.3429	0.4249	0.8498
	LightGBM	0.3411	0.4221	0.8537	0.3447	0.4250	0.8490
	CatBoost	0.3309	0.4150	0.8580	0.3305	0.4164	0.8552
	LSTM	0.3832	0.4554	0.8356	0.3739	0.4549	0.8362
	GRU	0.3860	0.4589	0.8345	0.3709	0.4558	0.8375
Cox's Bazar	MLR	0.5679	0.5610	0.8032	0.5434	0.5538	0.8070
	Lasso	0.5833	0.5707	0.7979	0.5587	0.5623	0.8015
	Ridge	0.5679	0.5610	0.8032	0.5434	0.5538	0.8070
	Elastic Net	0.5803	0.5712	0.7989	0.5582	0.5633	0.80171
	KNN	0.4500	0.4891	0.8441	0.4401	0.4891	0.8436
	DT	0.5235	0.5337	0.8186	0.4973	0.5254	0.8233
	RF	0.4548	0.4988	0.8424	0.4451	0.4943	0.8419
	GBR	0.3913	0.4528	0.8644	0.3833	0.4504	0.8638
	AdaBoost	0.5908	0.5761	0.7953	0.5696	0.5702	0.7977
	XGBoost	0.3949	0.4526	0.8632	0.3843	0.4502	0.8634
	LightGBM	0.3904	0.4497	0.8647	0.3834	0.4470	0.8638
	CatBoost	0.3713	0.4398	0.8714	0.3744	0.4415	0.8670
	LSTM	0.4411	0.4901	0.8472	0.4342	0.4903	0.8457
	GRU	0.4676	0.5087	0.8380	0.4575	0.5089	0.8375