Soft computing models for forecasting day-ahead energy consumption. (2022)

• Record Type: Journal Article
• Authors: Lydia, M.; Edwin Prem Kumar, G.
• Is Part Of: Materials today. Volume 58:Part 1(2022)
• Page Start: 473
• Page End: 477
• Publication Date: 2022
• Subjects: Forecasting -- Energy consumption -- Time-series -- Linear Regression -- SMO -- Bagging; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2022.03.003
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: The massive increase in automation, innovative technologies and standard of living has witnessed an equivalent rise in energy consumption. The depletion of conventional energy sources and the challenges faced in harnessing the non-conventional sources have reinforced the necessity for optimized energy consumption. Forecasting energy consumption will serve as a great tool for meticulous planning, judicious usage and outage prevention. In this paper, time series forecasting models have been used to forecast day-ahead energy consumption. Soft computing approaches have been employed for building single step and multiple step forecasting using five different datasets. The performance of the models has been evaluated using appropriate performance criteria.