A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load Forecasting

Multi-Energy Systems (MES) allow optimal interactions between different energy sources. Accurate load forecasting for such intricate systems would greatly enhance the performance and economic incentive to employ them. This article proposes a state-of-the-art deep learning based architecture to forec...

Full description

Saved in:
Bibliographic Details
Main Authors: Zain Ahmed, Mohsin Jamil, Ashraf Ali Khan
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Open Access Journal of Power and Energy
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10964383/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849713248087048192
author Zain Ahmed
Mohsin Jamil
Ashraf Ali Khan
author_facet Zain Ahmed
Mohsin Jamil
Ashraf Ali Khan
author_sort Zain Ahmed
collection DOAJ
description Multi-Energy Systems (MES) allow optimal interactions between different energy sources. Accurate load forecasting for such intricate systems would greatly enhance the performance and economic incentive to employ them. This article proposes a state-of-the-art deep learning based architecture to forecast multiple loads. The algorithm utilizes load correlations to select optimal input parameters. These optimal inputs are fed to D-TCNet (Deep – Temporal Convolution Network). This network uses multi-layer perceptrons (MLP) to encode the spatial relationship among exogenous variables which is fed to a Temporal Convolutional Network (TCN). The TCN resolves temporal information in the multi-load time series which is used for forecasting these loads for fixed output horizon. The proposed novel method is used on the energy consumption data for multi energy system of University of Austin Tempe Campus. The proposed method shows improved performance across all three energy types as well as all four seasons.
format Article
id doaj-art-acbcfed15ff247fa82397591a9ce358e
institution DOAJ
issn 2687-7910
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Open Access Journal of Power and Energy
spelling doaj-art-acbcfed15ff247fa82397591a9ce358e2025-08-20T03:14:01ZengIEEEIEEE Open Access Journal of Power and Energy2687-79102025-01-011220921910.1109/OAJPE.2025.355933610964383A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load ForecastingZain Ahmed0Mohsin Jamil1https://orcid.org/0000-0002-8835-2451Ashraf Ali Khan2https://orcid.org/0000-0002-5601-0887Department of Electrical and Computer Engineering, Faculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, NL, CanadaDepartment of Electrical and Computer Engineering, Faculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, NL, CanadaDepartment of Electrical and Computer Engineering, Faculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, NL, CanadaMulti-Energy Systems (MES) allow optimal interactions between different energy sources. Accurate load forecasting for such intricate systems would greatly enhance the performance and economic incentive to employ them. This article proposes a state-of-the-art deep learning based architecture to forecast multiple loads. The algorithm utilizes load correlations to select optimal input parameters. These optimal inputs are fed to D-TCNet (Deep – Temporal Convolution Network). This network uses multi-layer perceptrons (MLP) to encode the spatial relationship among exogenous variables which is fed to a Temporal Convolutional Network (TCN). The TCN resolves temporal information in the multi-load time series which is used for forecasting these loads for fixed output horizon. The proposed novel method is used on the energy consumption data for multi energy system of University of Austin Tempe Campus. The proposed method shows improved performance across all three energy types as well as all four seasons.https://ieeexplore.ieee.org/document/10964383/Multi-energy systemsmulti-task learningtemporal convolutional network
spellingShingle Zain Ahmed
Mohsin Jamil
Ashraf Ali Khan
A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load Forecasting
IEEE Open Access Journal of Power and Energy
Multi-energy systems
multi-task learning
temporal convolutional network
title A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load Forecasting
title_full A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load Forecasting
title_fullStr A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load Forecasting
title_full_unstemmed A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load Forecasting
title_short A Novel Multi-Task Learning-Based Approach to Multi-Energy System Load Forecasting
title_sort novel multi task learning based approach to multi energy system load forecasting
topic Multi-energy systems
multi-task learning
temporal convolutional network
url https://ieeexplore.ieee.org/document/10964383/
work_keys_str_mv AT zainahmed anovelmultitasklearningbasedapproachtomultienergysystemloadforecasting
AT mohsinjamil anovelmultitasklearningbasedapproachtomultienergysystemloadforecasting
AT ashrafalikhan anovelmultitasklearningbasedapproachtomultienergysystemloadforecasting
AT zainahmed novelmultitasklearningbasedapproachtomultienergysystemloadforecasting
AT mohsinjamil novelmultitasklearningbasedapproachtomultienergysystemloadforecasting
AT ashrafalikhan novelmultitasklearningbasedapproachtomultienergysystemloadforecasting