{"title":"An Approach For Short Term Electricity Load Forecasting","authors":"Riya Kanwar, Shrishti Agrawal, T. Manoranjitham","doi":"10.1109/ICNWC57852.2023.10127505","DOIUrl":null,"url":null,"abstract":"Electricity load forecasting is important from both a production and utilization standpoint. Short-term electricity forecasting is particularly meaningful, as electricity usage varies significantly over longer periods, and accurate forecasts can help us to address emergency situations. A study of existing short-term electricity forecasting approaches reveals a need for further improvement. In this work, we present a new technique for short-term electricity load forecasting using LSTM (long short-term memory). The study introduces various load forecasting techniques based on the prediction time period and discusses the time series model of load forecasting. We also discuss the difficulties of predicting electricity load and the factors that affect load forecasting, as well as the drawbacks of using simple forecasting methods such as curve fitting using numerical methods. We explore the use of machine learning models, such as neural networks and backpropagation, to tackle the problem, and we discuss an approach using LSTM, a variant of recurrent neural networks. We analyze the results and discuss the advantages and drawbacks of the technique, as well as steps that can be taken to improve results and the future scope of the project.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 International Conference on Networking and Communications (ICNWC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICNWC57852.2023.10127505","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Electricity load forecasting is important from both a production and utilization standpoint. Short-term electricity forecasting is particularly meaningful, as electricity usage varies significantly over longer periods, and accurate forecasts can help us to address emergency situations. A study of existing short-term electricity forecasting approaches reveals a need for further improvement. In this work, we present a new technique for short-term electricity load forecasting using LSTM (long short-term memory). The study introduces various load forecasting techniques based on the prediction time period and discusses the time series model of load forecasting. We also discuss the difficulties of predicting electricity load and the factors that affect load forecasting, as well as the drawbacks of using simple forecasting methods such as curve fitting using numerical methods. We explore the use of machine learning models, such as neural networks and backpropagation, to tackle the problem, and we discuss an approach using LSTM, a variant of recurrent neural networks. We analyze the results and discuss the advantages and drawbacks of the technique, as well as steps that can be taken to improve results and the future scope of the project.