OECD ülkeleri arasında kurulu rüzgâr ve güneş enerjisi kapasiteleri için makine öğrenimi temelli tahmin modeli: Bir LSTM yaklaşımı
Loading...

Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
İklim krizinin ana nedeni olarak gösterilen fosil yakıt tüketimi, özellikle sanayileşme ve endüstriyel faaliyetlerin bir sonucu olarak ortaya çıkmaktadır. Bu değişimlerin etkilerini hafifletmek adına, çevresel politika önlemleri ve teknolojik inovasyonlar, özellikle rüzgar ve güneş enerjisi gibi yenilenebilir enerji kaynaklarına odaklanmıştır. Bu araştırma, makine öğrenimi temelli bir model olan uzun kısa vadeli hafıza (LSTM) algoritmasını kullanarak, kurulu rüzgar ve güneş enerjisi kapasitelerini tahmin etmeyi amaçlamaktadır. Analizler, 24 OECD üyesi ülkeyi içeren geniş kapsamlı bir zaman serisi veri setini kullanarak, 2000 ile 2020 yılları arasındaki sosyal, ekonomik, geleneksel ve yenilenebilir enerji kaynakları ve çeşitli çevresel politika teşvik ve yasalarını kapsamaktadır. Elde edilen sonuçlar, kurulu rüzgar enerjisi kapasitesi modeli yüksek doğrulukta tahminleme performansı sergilerken, kurulu güneş enerjisi kapasitesi modelinin kurulu rüzgar enerjisi modeline kıyasla daha düşük doğrulukta tahminleme performansı gösterdiğini ortaya koymaktadır. Bu çalışma, öngörülen güneş ve rüzgar kapasiteleri ile enerji dönüşüm süreçlerine ve politika oluşturmaya dair dikkat çekmeyi ve değerli bilgiler sunmayı amaçlamaktadır. Anahtar Kelimeler: Makine öğrenimi tahminlemesi, LSTM yöntemi, kurulu güneş enerjisi kapasitesi, kurulu rüzgâr enerjisi kapasitesi, çevresel politika
The main cause attributed to the climate crisis is the consumption of fossil fuels, particularly arising from industrialization and industrial activities. To mitigate the impacts of these changes, environmental policy measures and technological innovations have focused prominently on renewable energy sources, especially wind and solar energy. This research aims to predict installed wind and solar energy capacities using a machine learning-based model, specifically the Long Short-Term Memory (LSTM) algorithm. The analyses cover a comprehensive time series dataset, including social, economic, conventional, and renewable energy attributes, as well as various environmental policy incentives and laws. The dataset spans the years from 2000 to 2020 and encompasses 24 OECD member countries. The results reveal that the installed wind energy capacity model exhibits high accuracy in prediction performance, whereas the installed solar energy capacity model demonstrates lower accuracy compared to the installed wind energy model. This study intends to draw attention to and provide valuable insights into the predicted solar and wind capacities, aiming to contribute to discussions on energy transformation processes and policymaking. Keywords: machine learning prediction, LSTM method, installed solar energy capacity, installed wind energy capacity, environmental policy
The main cause attributed to the climate crisis is the consumption of fossil fuels, particularly arising from industrialization and industrial activities. To mitigate the impacts of these changes, environmental policy measures and technological innovations have focused prominently on renewable energy sources, especially wind and solar energy. This research aims to predict installed wind and solar energy capacities using a machine learning-based model, specifically the Long Short-Term Memory (LSTM) algorithm. The analyses cover a comprehensive time series dataset, including social, economic, conventional, and renewable energy attributes, as well as various environmental policy incentives and laws. The dataset spans the years from 2000 to 2020 and encompasses 24 OECD member countries. The results reveal that the installed wind energy capacity model exhibits high accuracy in prediction performance, whereas the installed solar energy capacity model demonstrates lower accuracy compared to the installed wind energy model. This study intends to draw attention to and provide valuable insights into the predicted solar and wind capacities, aiming to contribute to discussions on energy transformation processes and policymaking. Keywords: machine learning prediction, LSTM method, installed solar energy capacity, installed wind energy capacity, environmental policy
Description
Keywords
İşletme, Business Administration
Turkish CoHE Thesis Center URL
Fields of Science
Citation
WoS Q
Scopus Q
Source
Volume
Issue
Start Page
End Page
117
