Improving Energy Consumption Forecasting through Conditional Generative Adversarial Networks

Muhammad Fajrul Alam Ulin Nuha*, Ahmad Muklason

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Obtaining clean, high-frequency energy consumption data for households is challenging, further complicated by the presence of abnormalities in the available data. This study addresses this gap by utilizing Conditional Generative Adversarial Networks (CGANs) to generate synthetic, anomaly-free data that mirrors real consumption patterns. By incorporating time-of-day information into the CGAN model, we enhance forecasting accuracy, demonstrating a reduction in the Root Mean Square Error (RMSE) compared to traditional methods. Applied to the Electricity and Occupancy (ECO) Dataset, our approach showcases the efficacy of CGANs in addressing data quality issues in energy management. This presents a novel, cost-effective strategy for accurate energy consumption forecasting. The study compares three datasets for energy consumption forecasting: unmodified anomalies, anomalies replaced with historical mean consumption, and anomalies replaced with CGAN-generated synthetic data, with RMSE improvements observed from 255.896 to 249.336. Our findings highlight the potential of CGANs to significantly improve data quality and forecasting precision in household energy management.

Original languageEnglish
Title of host publication2024 International Seminar on Intelligent Technology and Its Applications
Subtitle of host publicationCollaborative Innovation: A Bridging from Academia to Industry towards Sustainable Strategic Partnership, ISITIA 2024 - Proceeding
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages769-774
Number of pages6
Edition2024
ISBN (Electronic)9798350378573
DOIs
Publication statusPublished - 2024
Event25th International Seminar on Intelligent Technology and Its Applications, ISITIA 2024 - Hybrid, Mataram, Indonesia
Duration: 10 Jul 202412 Jul 2024

Conference

Conference25th International Seminar on Intelligent Technology and Its Applications, ISITIA 2024
Country/TerritoryIndonesia
CityHybrid, Mataram
Period10/07/2412/07/24

Keywords

  • conditional generative adversarial network
  • energy consumption forecasting
  • synthetic data generation

Fingerprint

Dive into the research topics of 'Improving Energy Consumption Forecasting through Conditional Generative Adversarial Networks'. Together they form a unique fingerprint.

Cite this