A hybrid framework for short-term load forecasting based on optimized InMetra Boost and BiLSTM

IF 9 1区 工程技术 Q1 ENERGY & FUELS
Qinghe Zhao, Shengduo Wang, Yuqi Chen, Jinlong Liu, Yujia Sun, Tong Su, Ningning Li, Junlong Fang
{"title":"A hybrid framework for short-term load forecasting based on optimized InMetra Boost and BiLSTM","authors":"Qinghe Zhao,&nbsp;Shengduo Wang,&nbsp;Yuqi Chen,&nbsp;Jinlong Liu,&nbsp;Yujia Sun,&nbsp;Tong Su,&nbsp;Ningning Li,&nbsp;Junlong Fang","doi":"10.1016/j.energy.2025.136582","DOIUrl":null,"url":null,"abstract":"<div><div>Accurate short-term load forecasting is essential for maintaining stable and efficient power grid operations, especially with the increasing complexity introduced by renewable energy sources. This paper proposes a novel hybrid model for day-ahead Short-Term Load Forecasting, combining an improved Boosting algorithm, InMetra Boost, and a Bidirectional Long Short-Term Memory model. InMetra Boost introduces an asymmetric penalty mechanism, allowing for more precise handling of positive and negative forecast deviations. The model's hyperparameters are optimized using the Tree-structured Parzen Estimator, and the temporal dependencies in load data are further captured by a BiLSTM model, whose architecture is refined via the Cuckoo Search algorithm. The proposed TPE-IMB-CS-BiLSTM (Tree-structured Parzen Estimator optimized InMetra Boost within BiLSTM tuned by Cuckoo Search) framework was evaluated on real-world Estonian grid data, demonstrating superior performance compared to traditional models. Firstly, the TPE-IMB model achieves a significant improvement, reducing MAE from 34.86 MW in the original boosting model to 30.38 MW, and RMSE from 47.79 MW to 40.45 MW. Secondly, the hybrid TPE-IMB-CS-BiLSTM model further enhances accuracy, reducing MAE to 27.77 MW and RMSE to 36.55 MW, significantly outperforming the TPE-IMB and vanilla BiLSTM models. Lastly, compared to other state-of-the-art models, the proposed model achieves the best performance with a 24.66 % reduction in MAE and 26.74 % reduction in RMSE, demonstrating superior robustness in handling complex and extreme data conditions.</div></div>","PeriodicalId":11647,"journal":{"name":"Energy","volume":"328 ","pages":"Article 136582"},"PeriodicalIF":9.0000,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Energy","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0360544225022248","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 0

Abstract

Accurate short-term load forecasting is essential for maintaining stable and efficient power grid operations, especially with the increasing complexity introduced by renewable energy sources. This paper proposes a novel hybrid model for day-ahead Short-Term Load Forecasting, combining an improved Boosting algorithm, InMetra Boost, and a Bidirectional Long Short-Term Memory model. InMetra Boost introduces an asymmetric penalty mechanism, allowing for more precise handling of positive and negative forecast deviations. The model's hyperparameters are optimized using the Tree-structured Parzen Estimator, and the temporal dependencies in load data are further captured by a BiLSTM model, whose architecture is refined via the Cuckoo Search algorithm. The proposed TPE-IMB-CS-BiLSTM (Tree-structured Parzen Estimator optimized InMetra Boost within BiLSTM tuned by Cuckoo Search) framework was evaluated on real-world Estonian grid data, demonstrating superior performance compared to traditional models. Firstly, the TPE-IMB model achieves a significant improvement, reducing MAE from 34.86 MW in the original boosting model to 30.38 MW, and RMSE from 47.79 MW to 40.45 MW. Secondly, the hybrid TPE-IMB-CS-BiLSTM model further enhances accuracy, reducing MAE to 27.77 MW and RMSE to 36.55 MW, significantly outperforming the TPE-IMB and vanilla BiLSTM models. Lastly, compared to other state-of-the-art models, the proposed model achieves the best performance with a 24.66 % reduction in MAE and 26.74 % reduction in RMSE, demonstrating superior robustness in handling complex and extreme data conditions.
基于优化InMetra Boost和BiLSTM的短期负荷预测混合框架
准确的短期负荷预测对于维持电网的稳定和高效运行至关重要,特别是随着可再生能源的日益复杂。本文提出了一种新的日前短期负荷预测混合模型,该模型结合了改进的Boost算法InMetra Boost和双向长短期记忆模型。InMetra Boost引入了非对称惩罚机制,允许更精确地处理正向和负向预测偏差。利用树结构Parzen估计器对模型的超参数进行优化,利用BiLSTM模型进一步捕获负载数据中的时间依赖关系,并通过布谷鸟搜索算法对其结构进行改进。提出的TPE-IMB-CS-BiLSTM (Tree-structured Parzen Estimator optimized InMetra Boost within BiLSTM,通过Cuckoo Search进行调整)框架在爱沙尼亚的实际网格数据上进行了评估,与传统模型相比,显示出优越的性能。首先,TPE-IMB模型实现了显著的改进,MAE从原增压模型的34.86 MW降低到30.38 MW, RMSE从47.79 MW降低到40.45 MW。其次,混合TPE-IMB- cs -BiLSTM模型进一步提高了精度,MAE降至27.77 MW, RMSE降至36.55 MW,显著优于TPE-IMB和普通BiLSTM模型。最后,与其他最先进的模型相比,所提出的模型达到了最佳性能,MAE降低了24.66%,RMSE降低了26.74%,在处理复杂和极端数据条件方面表现出卓越的鲁棒性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Energy
Energy 工程技术-能源与燃料
CiteScore
15.30
自引率
14.40%
发文量
0
审稿时长
14.2 weeks
期刊介绍: Energy is a multidisciplinary, international journal that publishes research and analysis in the field of energy engineering. Our aim is to become a leading peer-reviewed platform and a trusted source of information for energy-related topics. The journal covers a range of areas including mechanical engineering, thermal sciences, and energy analysis. We are particularly interested in research on energy modelling, prediction, integrated energy systems, planning, and management. Additionally, we welcome papers on energy conservation, efficiency, biomass and bioenergy, renewable energy, electricity supply and demand, energy storage, buildings, and economic and policy issues. These topics should align with our broader multidisciplinary focus.
×
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:481959085
Book学术官方微信