Loading…
Reinforcement learning layout‐based optimal energy management in smart home: AI‐based approach
This research addresses the pressing need for enhanced energy management in smart homes, motivated by the inefficiencies of current methods in balancing power usage optimization with user comfort. By integrating reinforcement learning and a unique column‐and‐constraint generation strategy, the study...
Saved in:
Published in: | IET generation, transmission & distribution transmission & distribution, 2024-08, Vol.18 (15), p.2509-2520 |
---|---|
Main Authors: | , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | This research addresses the pressing need for enhanced energy management in smart homes, motivated by the inefficiencies of current methods in balancing power usage optimization with user comfort. By integrating reinforcement learning and a unique column‐and‐constraint generation strategy, the study aims to fill this gap and offer a comprehensive solution. Furthermore, the increasing adoption of renewable energy sources like solar panels underscores the importance of developing advanced energy management techniques, driving the exploration of innovative approaches such as the one proposed herein. The constraint coordination game (CCG) method is designed to efficiently manage the power usage of each appliance, including the charging and discharging of the energy storage system. Additionally, a deep learning model, specifically a deep neural network, is employed to forecast indoor temperatures, which significantly influence the energy demands of the air conditioning system. The synergistic combination of the CCG method with deep learning‐based indoor temperature forecasting promises significant reductions in homeowner energy expenses while maintaining optimal appliance performance and user satisfaction. Testing conducted in simulated environments demonstrates promising results, showcasing a 12% reduction in energy costs compared to conventional energy management strategies.
A machine learning‐driven smart home energy management algorithm based on reinforcement learning and artificial neural network is presented here. By planning two controllable home appliances and discharging and charging the energy storage system, the suggested algorithm minimizes the energy cost without compromising user satisfaction and ambient operations features. |
---|---|
ISSN: | 1751-8687 1751-8695 |
DOI: | 10.1049/gtd2.13203 |