Deep learning with LSTM networks for medium to long range demand forecasting: Comparison with machine learning benchmark model

Salah Bouktif, Ali Fiaz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Accurate load forecasting has become increasingly important as it can help power companies for better scheduling to reduce excessive electricity production. This study presents an approach that uses long short-term memory (LSTM) neural network to construct a forecasting model for medium to long term aggregate load forecasting. An array of linear and non-linear machine learning models are also built on electric load dataset and the best performing model is used as benchmark for comparison with deep learning approach. Weather related variables temperature, humidity and wind speed are used as exogenous inputs. The proposed methods are applied on a publicly available dataset of electric load consumption in metropolitan France. The goal is to ascertain with what accuracy can aggregated load values in Megawatts can be predicted using machine learning and deep learning approaches. Results show that LSTM based prediction model using only time lags gives low forecast errors and outperformed machine learning models optimized with hyperparameter tuning.

Original languageEnglish
Title of host publicationECOS 2018 - Proceedings of the 31st International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems
PublisherUniversity of Minho
ISBN (Electronic)9789729959646
Publication statusPublished - 2018
Externally publishedYes
Event31st International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, ECOS 2018 - Guimaraes, Portugal
Duration: Jun 17 2018Jun 21 2018

Publication series

NameECOS 2018 - Proceedings of the 31st International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems

Conference

Conference31st International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, ECOS 2018
Country/TerritoryPortugal
CityGuimaraes
Period6/17/186/21/18

Keywords

  • Deep learning
  • Long short-term memory (LSTM)
  • Machine learning
  • Medium to long-term load forecasting
  • Recurrent neural network

ASJC Scopus subject areas

  • General Environmental Science
  • General Energy

Fingerprint

Dive into the research topics of 'Deep learning with LSTM networks for medium to long range demand forecasting: Comparison with machine learning benchmark model'. Together they form a unique fingerprint.

Cite this