Causal Mechanism Transfer Network for Time Series Domain Adaptation in Mechanical Systems

Author:

Li Zijian1,Cai Ruichu1ORCID,Ng Hong Wei2,Winslett Marianne3,Fu Tom Z. J.4,Xu Boyan1,Yang Xiaoyan5,Zhang Zhenjie6

Affiliation:

1. School of Computer, Guangdong University of Technology, Guangzhou, China

2. Advanced Digital Science Center, Singapore

3. University of Illinois, Urbana Champaign

4. Bigo Technology Pte. Ltd.

5. YITU Pte. Ltd.

6. PVoice Technology

Abstract

Data-driven models are becoming essential parts in modern mechanical systems, commonly used to capture the behavior of various equipment and varying environmental characteristics. Despite the advantages of these data-driven models on excellent adaptivity to high dynamics and aging equipment, they are usually hungry for massive labels, mostly contributed by human engineers at a high cost. Fortunately, domain adaptation enhances the model generalization by utilizing the labeled source data and the unlabeled target data. However, the mainstream domain adaptation methods cannot achieve ideal performance on time series data, since they assume that the conditional distributions are equal. This assumption works well in the static data but is inapplicable for the time series data. Even the first-order Markov dependence assumption requires the dependence between any two consecutive time steps. In this article, we assume that the causal mechanism is invariant and present our Causal Mechanism Transfer Network (CMTN) for time series domain adaptation. By capturing causal mechanisms of time series data, CMTN allows the data-driven models to exploit existing data and labels from similar systems, such that the resulting model on a new system is highly reliable even with limited data. We report our empirical results and lessons learned from two real-world case studies, on chiller plant energy optimization and boiler fault detection, which outperform the existing state-of-the-art method.

Funder

Natural Science Foundation of China

China Postdoctoral Science Foundation

NSFC-Guangdong Joint Found

Publisher

Association for Computing Machinery (ACM)

Subject

Artificial Intelligence,Theoretical Computer Science

Cited by 10 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Financial Market Volatility Forecasting Based on Domain Adaptation;2024 International Joint Conference on Neural Networks (IJCNN);2024-06-30

2. Graph Domain Adaptation: A Generative View;ACM Transactions on Knowledge Discovery from Data;2024-01-12

3. Second-order Confidence Network for Early Classification of Time Series;ACM Transactions on Intelligent Systems and Technology;2023-12-19

4. Life-cycle modeling driven by coupling competition degradation for remaining useful life prediction;Reliability Engineering & System Safety;2023-10

5. Sourcing celadons with EDXRF and LA-ICP-MS from the Xunyang city burial complex, 202 B.C–907 A.D.;Heritage Science;2023-02-08

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3