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 13 Quartile percent of the prediction error for Dataset 1 (minimum Std. deviation and IQR are bolded)

Model	Quartile percentile for Dataset 1: Kutubdia					Quartile percentile for Dataset 1: Cox’s Bazar
Model	Std	25%	50%	IQR	75%	Std	25%	50%	IQR	75%
MLR	0.6682	− 0.3211	− 0.0657	0.6059	0.2848	1.0808	− 0.6213	− 0.1224	0.9999	0.3786
Lasso	0.6794	− 0.3396	− 0.1186	0.6483	0.3088	1.0883	− 0.6093	− 0.1852	0.9948	0.3855
Ridge	0.6682	− 0.3211	− 0.0657	0.6059	0.2848	1.0808	− 0.6219	− 0.1228	1.0009	0.3790
Elastic Net	0.6745	− 0.3303	− 0.0960	0.6256	0.2953	1.0858	− 0.6117	− 0.1659	0.9865	0.3748
KNN	0.6683	− 0.2858	− 0.0286	0.5716	0.2858	1.0619	− 0.4573	0.0000	0.7718	0.3145
DT	0.6668	− 0.2919	− 0.0797	0.6025	0.3107	1.0531	− 0.3992	− 0.0798	0.6845	0.2853
RF	0.6445	− 0.2855	− 0.0602	0.5458	0.2603	1.0090	− 0.4448	− 0.0886	0.7188	0.2740
GBR	0.6381	− 0.2884	− 0.0584	0.5488	0.2605	1.0012	− 0.4446	− 0.0872	0.7179	0.2732
AdaBoost	0.6975	− 0.3864	− 0.0675	0.6025	0.2161	0.9716	− 0.7077	− 0.2351	0.9650	0.2573
XGBoost	0.6348	− 0.2915	− 0.0573	0.5569	0.2654	1.0001	− 0.4415	− 0.0765	0.7054	0.2640
LightGBM	0.6380	− 0.2859	− 0.0617	0.5409	0.2549	0.9972	− 0.4294	− 0.0798	0.6943	0.2649
CatBoost	0.6278	− 0.2902	− 0.0549	0.5408	0.2506	0.9953	− 0.4294	− 0.0803	0.6962	0.2668
LSTM	0.6485	− 0.3372	− 0.0902	0.5819	0.2447	1.0083	− 0.4855	− 0.1176	0.7614	0.2759
GRU	0.6520	− 0.3057	− 0.0773	0.5862	0.2804	1.0177	− 0.3890	− 0.1052	0.7974	0.4083