Semi-supervised CWGAN-GP modeling for AHU AFDD with high-quality synthetic data filtering mechanism

IF 7.1 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Building and Environment Pub Date : 2024-11-03 DOI:10.1016/j.buildenv.2024.112265
Hua Wang , Jian Bi , Mei Hua , Ke Yan , Afshin Afshari
{"title":"Semi-supervised CWGAN-GP modeling for AHU AFDD with high-quality synthetic data filtering mechanism","authors":"Hua Wang ,&nbsp;Jian Bi ,&nbsp;Mei Hua ,&nbsp;Ke Yan ,&nbsp;Afshin Afshari","doi":"10.1016/j.buildenv.2024.112265","DOIUrl":null,"url":null,"abstract":"<div><div>Supervised learning methods demonstrated high classification accuracy for air handling unit (AHU) automated fault detection and diagnosis (FDD) scenarios with well-shaped training datasets. However, for imbalanced training datasets, i.e., much less real-world fault training data samples against an enormous amount of normal data samples, the supervised learning-based methods failed to produce satisfactory FDD results. To address the above-mentioned issue, this study proposes a semi-supervised conditional Wasserstein generative adversarial network with gradient penalty (CWGAN-GP) to generate high-quality synthetic fault training samples. The semi-supervised learning-based AHU AFDD framework is completed by identifying high-quality synthetic fault samples and inserting them into the training pool iteratively. With different numbers of real-world fault samples, comparative experiments are conducted on datasets collected by ASHRAE project RP-1312 in the summer and winter seasons. The experimental results show that the proposed AFDD method has obvious advantages over the traditional method with limited numbers of real-world fault samples. Moreover, the proposed CWGAN-GP-SSL framework achieves superior AFDD performance compared to the existing GAN-based AHU AFDD method.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"267 ","pages":"Article 112265"},"PeriodicalIF":7.1000,"publicationDate":"2024-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Building and Environment","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0360132324011077","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CONSTRUCTION & BUILDING TECHNOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Supervised learning methods demonstrated high classification accuracy for air handling unit (AHU) automated fault detection and diagnosis (FDD) scenarios with well-shaped training datasets. However, for imbalanced training datasets, i.e., much less real-world fault training data samples against an enormous amount of normal data samples, the supervised learning-based methods failed to produce satisfactory FDD results. To address the above-mentioned issue, this study proposes a semi-supervised conditional Wasserstein generative adversarial network with gradient penalty (CWGAN-GP) to generate high-quality synthetic fault training samples. The semi-supervised learning-based AHU AFDD framework is completed by identifying high-quality synthetic fault samples and inserting them into the training pool iteratively. With different numbers of real-world fault samples, comparative experiments are conducted on datasets collected by ASHRAE project RP-1312 in the summer and winter seasons. The experimental results show that the proposed AFDD method has obvious advantages over the traditional method with limited numbers of real-world fault samples. Moreover, the proposed CWGAN-GP-SSL framework achieves superior AFDD performance compared to the existing GAN-based AHU AFDD method.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
利用高质量合成数据过滤机制为 AHU AFDD 建立半监督 CWGAN-GP 模型
有监督学习方法在空气处理机组(AHU)自动故障检测和诊断(FDD)场景中,在训练数据集形状良好的情况下,显示出较高的分类准确性。然而,对于不平衡的训练数据集,即现实世界中的故障训练数据样本少于大量正常数据样本,基于监督学习的方法无法产生令人满意的故障检测与诊断结果。针对上述问题,本研究提出了一种带梯度惩罚的半监督条件瓦瑟斯坦生成对抗网络(CWGAN-GP),用于生成高质量的合成故障训练样本。基于半监督学习的 AHU AFDD 框架是通过识别高质量的合成故障样本并将其反复插入训练池来完成的。利用不同数量的真实故障样本,在 ASHRAE 项目 RP-1312 收集的夏季和冬季数据集上进行了对比实验。实验结果表明,在真实世界故障样本数量有限的情况下,所提出的 AFDD 方法与传统方法相比具有明显优势。此外,与现有的基于 GAN 的 AHU AFDD 方法相比,提出的 CWGAN-GP-SSL 框架实现了更优越的 AFDD 性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Building and Environment
Building and Environment 工程技术-工程:环境
CiteScore
12.50
自引率
23.00%
发文量
1130
审稿时长
27 days
期刊介绍: Building and Environment, an international journal, is dedicated to publishing original research papers, comprehensive review articles, editorials, and short communications in the fields of building science, urban physics, and human interaction with the indoor and outdoor built environment. The journal emphasizes innovative technologies and knowledge verified through measurement and analysis. It covers environmental performance across various spatial scales, from cities and communities to buildings and systems, fostering collaborative, multi-disciplinary research with broader significance.
期刊最新文献
A Framework for Facilitating Low-Income Net-Zero Energy Housing Delivery in Developing Countries: Insights from a Practical Case in South Africa A seasonal assessment of indoor air quality and thermal performance in naturally ventilated airtight energy-efficient dwellings A prediction of urban boundary layer using Recurrent Neural Network and reduced order modeling The hygrothermal properties of stone used in traditional construction: A Western European case study Dynamic wind patterns and indoor/outdoor pollutant dispersion in the simplified building array: Statistical and spectral analyses from scaled outdoor experiments
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1