A comprehensive country-based day-ahead wind power generation forecast model by coupling numerical weather prediction data and CatBoost with feature selection methods for Turkey

Cem Özen; Ali Deniz

doi:10.1177/0309524X221078536

A comprehensive country-based day-ahead wind power generation forecast model by coupling numerical weather prediction data and CatBoost with feature selection methods for Turkey

Cem Özen^*
, Ali Deniz

^*Bu çalışma için yazışmadan sorumlu yazar

Meteoroloji Müh.Bölümü

Istanbul Technical University

Araştırma sonucu: Dergiye katkı › Makale › bilirkişi

6 Atıf (Scopus)

Özet

A country-based day-ahead wind power generation forecast (WPGF) model with a grid selection algorithm and feature selection models was proposed in this study. Atmospheric variables extracted from 300, 500, 700 hPa pressure levels, and surface level of ERA5 reanalysis data with 2.5° spatial resolution were used to train/validate the categorical boosting (CatBoost) model. A special grid selection algorithm was proposed by considering Turkey’s spatial distribution of wind power plants. The day-ahead forecasts of ECMWF’s HRES (High-resolution) were used as the test subset, therefore, paving the way for the operational use of the model. The proposed model could be considered much as a specialized machine learning based downscaling method for country-based WPGF due to using numerical weather prediction model outputs as its input. Results showed that the proposed model that uses fewer features has outperformed the other models with a normalized root mean square error of 7.6% and coefficient of determination of 0.8989.

Orijinal dil	İngilizce
Sayfa (başlangıç-bitiş)	1359-1388
Sayfa sayısı	30
Dergi	Wind Engineering
Hacim	46
Basın numarası	5
DOI'lar	https://doi.org/10.1177/0309524X221078536
Yayın durumu	Yayınlandı - Eki 2022

Bibliyografik not

Publisher Copyright:
© The Author(s) 2022.

Finansman

We would like to thank Dr. Deniz Demirhan and ECMWF for providing ECMWF-HRES data which enable us to evaluate the operational process of the proposed model. The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study has resulted as a part of a project named “Building a Short-term Wind Power Generation Forecast Model by Coupling Numerical Weather Prediction Models and Machine Learning Algorithms” which is supported within the scope of PhD thesis projects of the Scientific Research Projects (BAP) Institution of the Istanbul Technical University with a grant number of MDK-2020-42646. The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study has resulted as a part of a project named “Building a Short-term Wind Power Generation Forecast Model by Coupling Numerical Weather Prediction Models and Machine Learning Algorithms” which is supported within the scope of PhD thesis projects of the Scientific Research Projects (BAP) Institution of the Istanbul Technical University with a grant number of MDK-2020-42646.

Finansörler	Finansör numarası
Istanbul Teknik Üniversitesi	MDK-2020-42646
European Centre for Medium-Range Weather Forecasts

BM SKH

Bu sonuç, aşağıdaki Sürdürülebilir Kalkınma Hedefine/Hedeflerine katkıda bulunur

SKH 7 Erişilebilir ve Temiz Enerji

Belgeye Erişim

10.1177/0309524X221078536

Diğer Dosyalar ve Linkler

Link to publication in Scopus

Alıntı Yap

@article{2dc8e4c361a44774b577833fca3720d9,

title = "A comprehensive country-based day-ahead wind power generation forecast model by coupling numerical weather prediction data and CatBoost with feature selection methods for Turkey",

abstract = "A country-based day-ahead wind power generation forecast (WPGF) model with a grid selection algorithm and feature selection models was proposed in this study. Atmospheric variables extracted from 300, 500, 700 hPa pressure levels, and surface level of ERA5 reanalysis data with 2.5° spatial resolution were used to train/validate the categorical boosting (CatBoost) model. A special grid selection algorithm was proposed by considering Turkey{\textquoteright}s spatial distribution of wind power plants. The day-ahead forecasts of ECMWF{\textquoteright}s HRES (High-resolution) were used as the test subset, therefore, paving the way for the operational use of the model. The proposed model could be considered much as a specialized machine learning based downscaling method for country-based WPGF due to using numerical weather prediction model outputs as its input. Results showed that the proposed model that uses fewer features has outperformed the other models with a normalized root mean square error of 7.6\% and coefficient of determination of 0.8989.",

keywords = "ECMWF HRES, ERA5, Wind power forecast, feature selection methods, machine learning, wind energy",

author = "Cem {\"O}zen and Ali Deniz",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2022.",

year = "2022",

month = oct,

doi = "10.1177/0309524X221078536",

language = "English",

volume = "46",

pages = "1359--1388",

journal = "Wind Engineering",

issn = "0309-524X",

publisher = "SAGE Publications Inc.",

number = "5",

}

TY - JOUR

T1 - A comprehensive country-based day-ahead wind power generation forecast model by coupling numerical weather prediction data and CatBoost with feature selection methods for Turkey

AU - Özen, Cem

AU - Deniz, Ali

N1 - Publisher Copyright: © The Author(s) 2022.

PY - 2022/10

Y1 - 2022/10

N2 - A country-based day-ahead wind power generation forecast (WPGF) model with a grid selection algorithm and feature selection models was proposed in this study. Atmospheric variables extracted from 300, 500, 700 hPa pressure levels, and surface level of ERA5 reanalysis data with 2.5° spatial resolution were used to train/validate the categorical boosting (CatBoost) model. A special grid selection algorithm was proposed by considering Turkey’s spatial distribution of wind power plants. The day-ahead forecasts of ECMWF’s HRES (High-resolution) were used as the test subset, therefore, paving the way for the operational use of the model. The proposed model could be considered much as a specialized machine learning based downscaling method for country-based WPGF due to using numerical weather prediction model outputs as its input. Results showed that the proposed model that uses fewer features has outperformed the other models with a normalized root mean square error of 7.6% and coefficient of determination of 0.8989.

AB - A country-based day-ahead wind power generation forecast (WPGF) model with a grid selection algorithm and feature selection models was proposed in this study. Atmospheric variables extracted from 300, 500, 700 hPa pressure levels, and surface level of ERA5 reanalysis data with 2.5° spatial resolution were used to train/validate the categorical boosting (CatBoost) model. A special grid selection algorithm was proposed by considering Turkey’s spatial distribution of wind power plants. The day-ahead forecasts of ECMWF’s HRES (High-resolution) were used as the test subset, therefore, paving the way for the operational use of the model. The proposed model could be considered much as a specialized machine learning based downscaling method for country-based WPGF due to using numerical weather prediction model outputs as its input. Results showed that the proposed model that uses fewer features has outperformed the other models with a normalized root mean square error of 7.6% and coefficient of determination of 0.8989.

KW - ECMWF HRES

KW - ERA5

KW - Wind power forecast

KW - feature selection methods

KW - machine learning

KW - wind energy

UR - https://www.scopus.com/pages/publications/85124594760

U2 - 10.1177/0309524X221078536

DO - 10.1177/0309524X221078536

M3 - Article

AN - SCOPUS:85124594760

SN - 0309-524X

VL - 46

SP - 1359

EP - 1388

JO - Wind Engineering

JF - Wind Engineering

IS - 5

ER -

A comprehensive country-based day-ahead wind power generation forecast model by coupling numerical weather prediction data and CatBoost with feature selection methods for Turkey

Özet

Bibliyografik not

Finansman

BM SKH

Belgeye Erişim

Diğer Dosyalar ve Linkler

Parmak izi

Alıntı Yap