基于深度学习方法的多步太阳能发电预测

Zemouri Nahed, Bouzgou Hassen, A. Chouder, Douak Mohamed
{"title":"基于深度学习方法的多步太阳能发电预测","authors":"Zemouri Nahed, Bouzgou Hassen, A. Chouder, Douak Mohamed","doi":"10.1109/ICATEEE57445.2022.10093688","DOIUrl":null,"url":null,"abstract":"For the effective use and integration of solar energy into the power system, precise forecasting of photovoltaic power generation is required. Currently, forecasting methods can accurately predict PV power with different step horizons, but they fail to maintain a good resolution of accurate forecasting when we increase the number of steps ahead. This paper addresses the forecasting of PV performance of a system with multi-step ahead by considering 1, 2 and 3 steps ahead of 15-minute scenarios using Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN) which are kinds of Deep Learning method . The originality of this study is that the forecasted value of solar power is put against the real data issued from the PV system. All the obtained results are evaluated using different statistical indicators such as Mean Absolute Percentage Error (MAPE), Normalized Mean Square Error (NMSE), Coefficient of determination (R2) and Forecast skill (FS). The low values of the different statistical indicators justify the goodness of the proposed approaches.","PeriodicalId":150519,"journal":{"name":"2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Multi-Step Solar Power Forecasting using Deep Learning Methods\",\"authors\":\"Zemouri Nahed, Bouzgou Hassen, A. Chouder, Douak Mohamed\",\"doi\":\"10.1109/ICATEEE57445.2022.10093688\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"For the effective use and integration of solar energy into the power system, precise forecasting of photovoltaic power generation is required. Currently, forecasting methods can accurately predict PV power with different step horizons, but they fail to maintain a good resolution of accurate forecasting when we increase the number of steps ahead. This paper addresses the forecasting of PV performance of a system with multi-step ahead by considering 1, 2 and 3 steps ahead of 15-minute scenarios using Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN) which are kinds of Deep Learning method . The originality of this study is that the forecasted value of solar power is put against the real data issued from the PV system. All the obtained results are evaluated using different statistical indicators such as Mean Absolute Percentage Error (MAPE), Normalized Mean Square Error (NMSE), Coefficient of determination (R2) and Forecast skill (FS). The low values of the different statistical indicators justify the goodness of the proposed approaches.\",\"PeriodicalId\":150519,\"journal\":{\"name\":\"2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE)\",\"volume\":\"5 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-11-26\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICATEEE57445.2022.10093688\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICATEEE57445.2022.10093688","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

为了有效地利用和整合太阳能到电力系统中,需要对光伏发电进行精确的预测。目前的预测方法可以准确地预测不同步距的光伏发电功率,但当我们增加步距时,预测方法不能保持良好的准确预测分辨率。本文利用长短期记忆(LSTM)和卷积神经网络(CNN)这两种深度学习方法,通过考虑提前1、2和3步的15分钟场景,解决了多步预测系统PV性能的问题。本研究的创新之处在于将太阳能发电的预测值与光伏系统发布的实际数据进行对比。采用平均绝对百分比误差(MAPE)、归一化均方误差(NMSE)、决定系数(R2)和预测技能(FS)等不同的统计指标对所得结果进行评价。不同统计指标的低值证明了所提出方法的良好性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Multi-Step Solar Power Forecasting using Deep Learning Methods
For the effective use and integration of solar energy into the power system, precise forecasting of photovoltaic power generation is required. Currently, forecasting methods can accurately predict PV power with different step horizons, but they fail to maintain a good resolution of accurate forecasting when we increase the number of steps ahead. This paper addresses the forecasting of PV performance of a system with multi-step ahead by considering 1, 2 and 3 steps ahead of 15-minute scenarios using Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN) which are kinds of Deep Learning method . The originality of this study is that the forecasted value of solar power is put against the real data issued from the PV system. All the obtained results are evaluated using different statistical indicators such as Mean Absolute Percentage Error (MAPE), Normalized Mean Square Error (NMSE), Coefficient of determination (R2) and Forecast skill (FS). The low values of the different statistical indicators justify the goodness of the proposed approaches.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信