Publication:
A Systematic Literature Review of Electricity Load Forecasting using Long Short-Term Memory

Simple item page
dc.contributor.authorSalleh N.S.M.en_US
dc.contributor.authorSuliman A.en_US
dc.contributor.authorJ�rgensen B.N.en_US
dc.contributor.authorid54946009300en_US
dc.contributor.authorid25825739000en_US
dc.contributor.authorid7202434812en_US
dc.date.accessioned2023-05-29T09:41:15Z
dc.date.available2023-05-29T09:41:15Z
dc.date.issued2022
dc.descriptionBrain; Deregulation; Electric load forecasting; Electric power plant loads; Electric utilities; Learning algorithms; Statistical tests; Electricity load; Electricity load forecasting; Evaluation metrics; Load predictions; Long term planning; LSTM; Machine learning algorithms; Medium-term planning; Review papers; Systematic literature review; Long short-term memoryen_US
dc.description.abstractResearch in electricity load prediction has contributed towards short-, medium-, and long-term planning for electricity power companies. One of the methods applied to perform prediction is machine learning. There are various types of dataset features, machine learning algorithms, and evaluation metrics utilised. This paper reviewed articles on electricity load prediction published in between 2019 and 2021. The review applied the systematic literature review method. In total, there were 368 articles were gathered from an online database, IEEE. The search was made based on combinations of keywords, i.e. short-term, electricity, load, demand, deep learning, forecast, time series, regression, and long short-term memory. From the collected articles, 25 articles were selected from a thorough examination of titles and abstracts. In the end, 11 complete materials were selected for final review. The review concentrated on: (i) common dataset feature and duration used, (ii) testing and validation strategies, and (iii) the evaluation metrics selected. The historical electricity load dataset was sufficient to perform electricity prediction. However, it was improved by adding independent variables into the dataset. RMSE and MAPE were the most used evaluation metrics in the reviewed articles. � 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.en_US
dc.description.natureFinalen_US
dc.identifier.doi10.1007/978-981-16-8515-6_58
dc.identifier.epage776
dc.identifier.scopus2-s2.0-85127630467
dc.identifier.spage765
dc.identifier.urihttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85127630467&doi=10.1007%2f978-981-16-8515-6_58&partnerID=40&md5=b783e922de00fa0a17f45e1175f20053
dc.identifier.urihttps://irepository.uniten.edu.my/handle/123456789/27226
dc.identifier.volume835
dc.publisherSpringer Science and Business Media Deutschland GmbHen_US
dc.sourceScopus
dc.sourcetitleLecture Notes in Electrical Engineering
dc.titleA Systematic Literature Review of Electricity Load Forecasting using Long Short-Term Memoryen_US
dc.typeConference Paperen_US
dspace.entity.typePublication
Files
Collections
SCOPUS