An ensemble of artificial neural network models to forecast hourly energy demand

IF 2 3区工程技术 Q2 ENGINEERING, MULTIDISCIPLINARY Optimization and Engineering Pub Date : 2024-03-25 DOI:10.1007/s11081-024-09883-7

{"title":"An ensemble of artificial neural network models to forecast hourly energy demand","authors":"","doi":"10.1007/s11081-024-09883-7","DOIUrl":null,"url":null,"abstract":"<h3>Abstract</h3> <p>We propose an ensemble artificial neural network (EANN) methodology for predicting the day ahead energy demand of a district heating operator (DHO). Specifically, at the end of one day, we forecast the energy demand for each of the 24 h of the next day. Our methodology combines three artificial neural network (ANN) models, each capturing a different aspect of the predicted time series. In particular, the outcomes of the three ANN models are combined into a single forecast. This is done using a sequential ordered optimization procedure that establishes the weights of three models in the final output. We validate our EANN methodology using data obtained from a A2A, which is one of the major DHOs in Italy. The data pertains to a major metropolitan area in Northern Italy. We compared the performance of our EANN with the method currently used by the DHO, which is based on multiple linear regression requiring expert intervention. Furthermore, we compared our EANN with the state-of-the-art seasonal autoregressive integrated moving average and Echo State Network models. The results show that our EANN achieves better performance than the other three methods, both in terms of mean absolute percentage error (MAPE) and maximum absolute percentage error. Moreover, we demonstrate that the EANN produces good quality results for longer forecasting horizons. Finally, we note that the EANN is characterised by simplicity, as it requires little tuning of a handful of parameters. This simplicity facilitates its replicability in other cases. </p>","PeriodicalId":56141,"journal":{"name":"Optimization and Engineering","volume":null,"pages":null},"PeriodicalIF":2.0000,"publicationDate":"2024-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Optimization and Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1007/s11081-024-09883-7","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 0

Abstract

We propose an ensemble artificial neural network (EANN) methodology for predicting the day ahead energy demand of a district heating operator (DHO). Specifically, at the end of one day, we forecast the energy demand for each of the 24 h of the next day. Our methodology combines three artificial neural network (ANN) models, each capturing a different aspect of the predicted time series. In particular, the outcomes of the three ANN models are combined into a single forecast. This is done using a sequential ordered optimization procedure that establishes the weights of three models in the final output. We validate our EANN methodology using data obtained from a A2A, which is one of the major DHOs in Italy. The data pertains to a major metropolitan area in Northern Italy. We compared the performance of our EANN with the method currently used by the DHO, which is based on multiple linear regression requiring expert intervention. Furthermore, we compared our EANN with the state-of-the-art seasonal autoregressive integrated moving average and Echo State Network models. The results show that our EANN achieves better performance than the other three methods, both in terms of mean absolute percentage error (MAPE) and maximum absolute percentage error. Moreover, we demonstrate that the EANN produces good quality results for longer forecasting horizons. Finally, we note that the EANN is characterised by simplicity, as it requires little tuning of a handful of parameters. This simplicity facilitates its replicability in other cases.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

预测每小时能源需求的人工神经网络模型组合

摘要我们提出了一种集合人工神经网络 (EANN) 方法，用于预测区域供热运营商 (DHO) 未来一天的能源需求。具体来说，在一天结束时，我们预测第二天 24 小时内每一天的能源需求。我们的方法结合了三个人工神经网络（ANN）模型，每个模型捕捉预测时间序列的不同方面。具体而言，三个人工神经网络模型的结果被合并为一个预测结果。这是通过顺序有序优化程序实现的，该程序确定了三个模型在最终输出中的权重。我们使用从 A2A（意大利主要 DHO 之一）获得的数据验证了我们的 EANN 方法。数据涉及意大利北部的一个大都市地区。我们将 EANN 的性能与 DHO 目前使用的方法（基于多元线性回归，需要专家干预）进行了比较。此外，我们还将 EANN 与最先进的季节性自回归综合移动平均模型和回声状态网络模型进行了比较。结果表明，无论是从平均绝对百分比误差（MAPE）还是从最大绝对百分比误差来看，我们的 EANN 都比其他三种方法取得了更好的性能。此外，我们还证明，EANN 在更长的预测范围内也能产生高质量的结果。最后，我们注意到 EANN 的特点是简单，因为它只需调整少量参数。这种简单性有利于在其他情况下复制。

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分钟内获得全文去求助

来源期刊

Optimization and Engineering 工程技术-工程：综合

CiteScore

4.80

自引率

14.30%

发文量

审稿时长

>12 weeks

期刊介绍： Optimization and Engineering is a multidisciplinary journal; its primary goal is to promote the application of optimization methods in the general area of engineering sciences. We expect submissions to OPTE not only to make a significant optimization contribution but also to impact a specific engineering application. Topics of Interest: -Optimization: All methods and algorithms of mathematical optimization, including blackbox and derivative-free optimization, continuous optimization, discrete optimization, global optimization, linear and conic optimization, multiobjective optimization, PDE-constrained optimization & control, and stochastic optimization. Numerical and implementation issues, optimization software, benchmarking, and case studies. -Engineering Sciences: Aerospace engineering, biomedical engineering, chemical & process engineering, civil, environmental, & architectural engineering, electrical engineering, financial engineering, geosciences, healthcare engineering, industrial & systems engineering, mechanical engineering & MDO, and robotics.