Home / Current Issue / Paper 1706377
Time Series Forecasting of Energy Demand Using Machine and Deep Learning Approach
Subject area: Science,Engineering and Technology · Area of research: Machine learning
Abstract
This study analyzes the application of machine learning (ML) and deep learning (DL) models to forecast hourly national energy consumption. As energy demand grows increasingly complex, accurate forecasting is crucial for maintaining grid stability, optimizing costs, and ensuring efficient resource management. The study evaluates various models, including Random Forest, XGBoost, Long Short-Term Memory (LSTM), and Temporal Convolutional Networks (TCN), as well as a hybrid model combining TCN with XGBoost. XGBoost had the best performance, achieving a Root Mean Squared Error (RMSE) of 393.48 and a Mean Absolute Percentage Error (MAPE) of 1.16%. The research also employed SHAP (SHapley Additive exPlanations) to interpret the model's decision-making process, highlighting the significance of features like lagged demand and temporal elements (e.g., hour of the day). These insights confirm the model's reliability for short-term energy forecasting, offering valuable tools for energy providers and operators. Using the XGBoost model, a forecast for the upcoming week was generated, demonstrating the model?s ability to maintain accuracy in short-term predictions. Residual analysis indicates the model's predictions are unbiased, further supporting its operational reliability.
Keywords
Energy Demand Forecasting, Machine Learning, Deep Learning, XGBoost, LSTM, TCN, SHAP Analysis, Time Series Analysis
References
[1] " type="xs:string"></xs:element>
[2] <xs:element name="EntityId1" type="xs:string"></xs:element>
[3] <xs:element name="EntityId2" type="xs:string"></xs:element>
[4] <xs:element name="EntityId3" type="xs:string"></xs:element>
[5] <xs:element name="EntityId4" type="xs:string"></xs:element>
[6] <xs:element name="EntityId5" type="xs:string"></xs:element>
[7] <xs:element name="Terms">
[8] <xs:complexType>
[9] <xs:element ref="pc:TermInfo" minOccurs="0" maxOccurs="unbounded"></xs:element>
[10] </xs:complexType>
[11] <xs:element name="TermInfo">
[12] <xs:complexType>
[13] <xs:element ref="pc:TermName" minOccurs="0"></xs:element>
[14] <xs:element ref="pc:TermId" minOccurs="0"></xs:element>
[15] </xs:complexType>
[16] <xs:element name="TermName" type="xs:string"></xs:element>
[17] <xs:element name="TermId" type="xs:string"></xs:element>
[18] </ct:contentTypeSchema>metaAttributes"/><ds:schemaRef ds:uri="http://www.w3.org/2001/XMLSchema"/><ds:schemaRef ds:uri="http://schemas.microsoft.com/office/2006/metadata/properties"/><ds:schemaRef ds:uri="e98a4188-4376-4a88-9eff-85460beb67b4"/><ds:schemaRef ds:uri="http://schemas.microsoft.com/office/2006/documentManagement/types"/><ds:schemaRef ds:uri="http://schemas.microsoft.com/office/infopath/2007/PartnerControls"/><ds:schemaRef ds:uri="http://schemas.openxmlformats.org/package/2006/metadata/core-properties"/><ds:schemaRef ds:uri="http://purl.org/dc/elements/1.1/"/><ds:schemaRef ds:uri="http://purl.org/dc/terms/"/><ds:schemaRef ds:uri="http://schemas.microsoft.com/internal/obd"/></ds:schemaRefs></ds:datastoreItem><?mso-contentType?><FormTemplates xmlns="http://schemas.microsoft.com/sharepoint/v3/contenttype/forms"><Display>DocumentLibraryForm</Display><Edit>DocumentLibraryForm</Edit><New>DocumentLibraryForm</New></FormTemplates><?xml version="1.0" encoding="UTF-8" standalone="no"?>
[19] <ds:datastoreItem ds:itemID="{23DEFAE7-8E6D-49CD-8F92-825FCB35CCE1}" xmlns:ds="http://schemas.openxmlformats.org/officeDocument/2006/customXml"><ds:schemaRefs><ds:schemaRef ds:uri="http://schemas.microsoft.com/sharepoint/v3/contenttype/forms"/></ds:schemaRefs></ds:datastoreItem>þÿ ÿÿÿÿ ÀF Microsoft Word 97-2003 Document MSWordDocWord.Document.8ô9²qCompObjÿÿÿÿ%rÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ
How to cite this paper
@article{1706377,
author = {Raphael Ibraimoh, Oluwakemi Shojupe},
title = {Time Series Forecasting of Energy Demand Using Machine and Deep Learning Approach},
journal = {Iconic Research And Engineering Journals},
year = {2024},
volume = {8},
number = {4},
pages = {115-127},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1706377.pdf},
abstract = {This study analyzes the application of machine learning (ML) and deep learning (DL) models to forecast hourly national energy consumption. As energy demand grows increasingly complex, accurate forecasting is crucial for maintaining grid stability, optimizing costs, and ensuring efficient resource management. The study evaluates various models, including Random Forest, XGBoost, Long Short-Term Memory (LSTM), and Temporal Convolutional Networks (TCN), as well as a hybrid model combining TCN with XGBoost. XGBoost had the best performance, achieving a Root Mean Squared Error (RMSE) of 393.48 and a Mean Absolute Percentage Error (MAPE) of 1.16%. The research also employed SHAP (SHapley Additive exPlanations) to interpret the model's decision-making process, highlighting the significance of features like lagged demand and temporal elements (e.g., hour of the day). These insights confirm the model's reliability for short-term energy forecasting, offering valuable tools for energy providers and operators. Using the XGBoost model, a forecast for the upcoming week was generated, demonstrating the model?s ability to maintain accuracy in short-term predictions. Residual analysis indicates the model's predictions are unbiased, further supporting its operational reliability.},
keywords = {Energy Demand Forecasting, Machine Learning, Deep Learning, XGBoost, LSTM, TCN, SHAP Analysis, Time Series Analysis},
month = {October},
}