{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,30]],"date-time":"2026-03-30T11:05:29Z","timestamp":1774868729923,"version":"3.50.1"},"reference-count":36,"publisher":"Institution of Engineering and Technology (IET)","issue":"1","license":[{"start":{"date-parts":[[2026,3,30]],"date-time":"2026-03-30T00:00:00Z","timestamp":1774828800000},"content-version":"vor","delay-in-days":88,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["ietresearch.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["IET Communications"],"published-print":{"date-parts":[[2026,1]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>With the rapid development of the intelligent computing centers, multivariate time series generated from many heterogeneous devices with non\u2010independent and identically distributed (non\u2010IID) characteristic, posed a significant challenge to traditional anomaly detection models. These challenges primarily arise from the difficulty in aligning and unifying heterogeneous data and the significant reduction in model generalization due to non\u2010IID data distributions. Therefore, this paper proposes a novel anomaly detection framework for non\u2010IID data from heterogeneous devices. First, we introduce a chain of thought metric alignment and ranking mechanism based on a large language model to meet the data heterogeneity challenge. Second, we design a variational recurrent neural network model augmented with global factors to capture spatiotemporal correlation patterns across devices, effectively addressing the impact of non\u2010IID data distributions. Experiments on multiple real\u2010world datasets demonstrate that this approach achieves optimal F1\u2010scores across various heterogeneous datasets. And because of the metric ranking, the model communication efficiency and inference efficiency have been greatly\u00a0optimized.<\/jats:p>","DOI":"10.1049\/cmu2.70154","type":"journal-article","created":{"date-parts":[[2026,3,30]],"date-time":"2026-03-30T10:16:36Z","timestamp":1774865796000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Unsupervised Anomaly Detection Based on LLM for Heterogeneous Multivariate Time Series in the Intelligent Computing Center"],"prefix":"10.1049","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8757-8144","authenticated-orcid":false,"given":"Xingguo","family":"Jiang","sequence":"first","affiliation":[{"name":"China Electric Power Research Institute Beijing China"}]},{"given":"Yue","family":"Wang","sequence":"additional","affiliation":[{"name":"China Electric Power Research Institute Beijing China"}]},{"given":"Chunpeng","family":"Wu","sequence":"additional","affiliation":[{"name":"China Electric Power Research Institute Beijing China"}]},{"given":"Xiaohui","family":"Wang","sequence":"additional","affiliation":[{"name":"China Electric Power Research Institute Beijing China"}]},{"given":"Hongyin","family":"Chen","sequence":"additional","affiliation":[{"name":"China Electric Power Research Institute Beijing China"}]},{"given":"Zhenying","family":"Tai","sequence":"additional","affiliation":[{"name":"Beihang University Beijing Beijing China"}]},{"given":"Hao","family":"Yue","sequence":"additional","affiliation":[{"name":"State Grid Jibei Electric Power Co., Ltd. Beijing China"}]},{"given":"Chang","family":"Liu","sequence":"additional","affiliation":[{"name":"Beijing Electric Power Company State Grid Corporation of China Beijing China"}]}],"member":"265","published-online":{"date-parts":[[2026,3,30]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"crossref","first-page":"0006","DOI":"10.34133\/icomputing.0006","article-title":"Intelligent Computing: The Latest Advances, Challenges, and Future","volume":"2","author":"Zhu S.","year":"2023","journal-title":"Intelligent Computing"},{"key":"e_1_2_10_3_1","doi-asserted-by":"crossref","first-page":"0076","DOI":"10.34133\/icomputing.0076","article-title":"Affective Computing: Recent Advances, Challenges, and Future Trends","volume":"3","author":"Pei G.","year":"2024","journal-title":"Intelligent Computing"},{"issue":"1","key":"e_1_2_10_4_1","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1016\/j.icte.2025.01.002","article-title":"Digital Twin\u2010Empowered Intelligent Computation Offloading for Edge Computing in the Era of 5G and Beyond: A State\u2010of\u2010the\u2010Art Survey","volume":"11","author":"Tran\u2010Dang H.","year":"2025","journal-title":"ICT Express"},{"key":"e_1_2_10_5_1","doi-asserted-by":"crossref","unstructured":"K.Lv R.Wu S.Chen andP.Lan \u201cCCi\u2010YOLOv8n: Enhanced Fire Detection With Carafe and Context\u2010Guided Modules \u201d inInternational Conference on Intelligent Computing(Springer 2025) 128\u2013140.","DOI":"10.1007\/978-981-96-9901-8_11"},{"issue":"1","key":"e_1_2_10_6_1","article-title":"The Power Control Method of Data Center Based on Cloud\u2010Edge Collaboration","volume":"1754","author":"Zhao J.","year":"2021","journal-title":"Journal of Physics: Conference Series"},{"key":"e_1_2_10_7_1","article-title":"Modeling and Optimization of Collaborative Computing in Regional Multi\u2010Energy Systems for Energy Internet","volume":"39","author":"Peng Y.","year":"2023","journal-title":"Sustainable Computing: Informatics and Systems"},{"key":"e_1_2_10_8_1","doi-asserted-by":"crossref","first-page":"7736","DOI":"10.52202\/079017-0248","article-title":"MaskLLM: Learnable Semi\u2010Structured Sparsity for Large Language Models","volume":"37","author":"Fang G.","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"8","key":"e_1_2_10_9_1","doi-asserted-by":"crossref","first-page":"254","DOI":"10.32996\/jcsts.2025.7.8.30","article-title":"Self\u2010Healing Streaming Architecture: Resilience Patterns for Event\u2010Driven Microservices","volume":"7","author":"Beereddy K. R.","year":"2025","journal-title":"Journal of Computer Science and Technology Studies"},{"issue":"1","key":"e_1_2_10_10_1","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1109\/TKDE.2025.3622600","article-title":"A Comprehensive Survey on Data Augmentation","volume":"38","author":"Wang Z.","year":"2026","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"issue":"12","key":"e_1_2_10_11_1","doi-asserted-by":"crossref","first-page":"22058","DOI":"10.1109\/JIOT.2025.3549765","article-title":"Unsupervised Distributed Anomaly Detection Framework for IoT in Edge AI Network","volume":"12","author":"Jiang X.","year":"2025","journal-title":"IEEE Internet of Things Journal"},{"issue":"2","key":"e_1_2_10_12_1","doi-asserted-by":"crossref","first-page":"1883","DOI":"10.4249\/scholarpedia.1883","article-title":"K\u2010Nearest Neighbor","volume":"4","author":"Peterson L. E.","year":"2009","journal-title":"Scholarpedia"},{"issue":"2","key":"e_1_2_10_13_1","doi-asserted-by":"crossref","first-page":"395","DOI":"10.1016\/j.cie.2005.01.009","article-title":"One\u2010Class Support Vector Machines: An Application in Machine Fault Detection and Classification","volume":"48","author":"Shin H. J.","year":"2005","journal-title":"Computers & Industrial Engineering"},{"issue":"11","key":"e_1_2_10_14_1","doi-asserted-by":"crossref","first-page":"1133","DOI":"10.1109\/34.730550","article-title":"Bayesian Approaches to Gaussian Mixture Modeling","volume":"20","author":"Roberts S. J.","year":"1998","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"e_1_2_10_15_1","doi-asserted-by":"crossref","first-page":"111","DOI":"10.1007\/978-3-030-70679-1_5","volume-title":"Variational Methods for Machine Learning with Applications to Deep Networks","author":"Pinheiro Cinelli L.","year":"2021"},{"key":"e_1_2_10_16_1","volume-title":"ICLR (Poster)","author":"Higgins I.","year":"2017"},{"issue":"12","key":"e_1_2_10_17_1","doi-asserted-by":"crossref","first-page":"3306","DOI":"10.1109\/TPDS.2022.3151853","article-title":"A Bayesian Approach to Distributed Anomaly Detection in Edge AI Networks","volume":"33","author":"Odiathevar M.","year":"2022","journal-title":"IEEE Transactions on Parallel and Distributed Systems"},{"issue":"12","key":"e_1_2_10_18_1","doi-asserted-by":"crossref","first-page":"5431","DOI":"10.14778\/3750601.3750689","article-title":"EasyAD: A Demonstration of Automated Solutions for Time\u2010Series Anomaly Detection","volume":"18","author":"Liu Q.","year":"2025","journal-title":"Proceedings of the VLDB Endowment"},{"key":"e_1_2_10_19_1","unstructured":"L.Li W.Chen W.Zhang S.Yuan andH.Guo \u201cPhysDiff: A Physically\u2010Guided Diffusion Model for Multivariate Time Series Anomaly Detection \u201d inThe Thirty\u2010Ninth Annual Conference on Neural Information Processing Systems(Curran Associates Inc. 2025)."},{"key":"e_1_2_10_20_1","doi-asserted-by":"crossref","unstructured":"Z.Chen Z.Wu W. K.Cheung H.\u2010N.Dai B.Choi andJ.Liu \u201cMSHTrans: Multi\u2010Scale Hypergraph Transformer With Time\u2010Series Decomposition for Temporal Anomaly Detection \u201d inProceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining(ACM 2025) 274\u2013285.","DOI":"10.1145\/3711896.3737057"},{"issue":"24","key":"e_1_2_10_21_1","doi-asserted-by":"crossref","first-page":"7611","DOI":"10.3390\/s25247611","article-title":"Meta Variational Memory Transformer for Anomaly Detection of Multivariate Time Series","volume":"25","author":"Qin K.","year":"2025","journal-title":"Sensors"},{"key":"e_1_2_10_22_1","doi-asserted-by":"crossref","DOI":"10.1016\/j.neucom.2025.132134","article-title":"MST\u2010MFGAT: An Adaptive Temporal Convolution With Feature Graph Attention Network for Time Series Anomaly Detection","volume":"664","author":"Wang D.","year":"2025","journal-title":"Neurocomputing"},{"key":"e_1_2_10_23_1","first-page":"1273","volume-title":"Artificial Intelligence and Statistics","author":"McMahan B.","year":"2017"},{"issue":"6","key":"e_1_2_10_24_1","doi-asserted-by":"crossref","first-page":"862","DOI":"10.1109\/TPDS.2024.3379905","article-title":"Taking Advantage of the Mistakes: Rethinking Clustered Federated Learning for IoT Anomaly Detection","volume":"35","author":"Fan J.","year":"2024","journal-title":"IEEE Transactions on Parallel and Distributed Systems"},{"issue":"4","key":"e_1_2_10_25_1","doi-asserted-by":"crossref","DOI":"10.1016\/j.ipm.2025.104107","article-title":"Effectively Detecting and Diagnosing Distributed Multivariate Time Series Anomalies via Unsupervised Federated Hypernetwork","volume":"62","author":"Hao J.","year":"2025","journal-title":"Information Processing & Management"},{"key":"e_1_2_10_26_1","doi-asserted-by":"crossref","unstructured":"H.Wang L.Mu\u00f1oz\u2010Gonz\u00e1lez D.Eklund andS.Raza \u201cNon\u2010IID Data Re\u2010Balancing at IoT Edge With Peer\u2010to\u2010Peer Federated Learning for Anomaly Detection \u201d inProceedings of the 14th ACM Conference on Security and Privacy in Wireless and Mobile Networks(ACM 2021) 153\u2013163.","DOI":"10.1145\/3448300.3467827"},{"issue":"13","key":"e_1_2_10_27_1","doi-asserted-by":"crossref","first-page":"4014","DOI":"10.3390\/s25134014","article-title":"FedSW\u2010TSAD: SWGAN\u2010Based Federated Time Series Anomaly Detection","volume":"25","author":"Zhang X.","year":"2025","journal-title":"Sensors"},{"key":"e_1_2_10_28_1","doi-asserted-by":"crossref","DOI":"10.1016\/j.jnca.2025.104259","article-title":"Integrated Probabilistic Clustering and Deep Reinforcement Learning for Bias Mitigation and Device Heterogeneity of Federated Learning in Edge Networks","volume":"242","author":"Singh N.","year":"2025","journal-title":"Journal of Network and Computer Applications"},{"issue":"32","key":"e_1_2_10_29_1","doi-asserted-by":"crossref","first-page":"27475","DOI":"10.1007\/s00521-025-11602-x","article-title":"INQMAD: Incremental Streaming Anomaly Detection With Density Matrices, Quantum Measurement, and Density Estimation","volume":"37","author":"Gallego J.","year":"2025","journal-title":"Neural Computing and Applications"},{"issue":"1","key":"e_1_2_10_30_1","first-page":"1","article-title":"Using MI\u2010LASSO to Study Populist Radical Right Voting in Times of Pandemic","volume":"11","author":"Chan K. M.","year":"2024","journal-title":"Research & Politics"},{"issue":"1","key":"e_1_2_10_31_1","doi-asserted-by":"crossref","first-page":"31","DOI":"10.1109\/TNB.2009.2035284","article-title":"SVM\u2010RFE with MRMR Filter for Gene Selection","volume":"9","author":"Mundra P. A.","year":"2009","journal-title":"IEEE Transactions on Nanobioscience"},{"key":"e_1_2_10_32_1","unstructured":"M. F.Bal\u0131n A.Abid andJ.Zou \u201cConcrete Autoencoders: Differentiable Feature Selection and Reconstruction \u201d inInternational Conference on Machine Learning(PMLR 2019) 444\u2013453."},{"key":"e_1_2_10_33_1","volume-title":"NeurIPS Foundation Models for Decision Making Workshop","author":"Choi K.","year":"2022"},{"key":"e_1_2_10_34_1","unstructured":"J.Yan B.Zheng H.Xu et\u00a0al. \u201cMaking Pre\u2010Trained Language Models Great on Tabular Prediction \u201darXiv preprint arXiv:2403.01841(2024)."},{"key":"e_1_2_10_35_1","unstructured":"M.Lukasik H.Narasimhan A. K.Menon F. X.Yu andS.Kumar \u201cMetric\u2010Aware LLM Inference \u201darXiv:2403.04182v1(2024)."},{"key":"e_1_2_10_36_1","volume-title":"Resampling\u2010Based Multiple Testing: Examples and Methods for p\u2010Value Adjustment","author":"Westfall P. H.","year":"1993"},{"key":"e_1_2_10_37_1","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1109\/ISSRE5003.2020.00014","volume-title":"2020 IEEE 31st International Symposium on Software Reliability Engineering (ISSRE)","author":"Liu P.","year":"2020"}],"container-title":["IET Communications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/ietresearch.onlinelibrary.wiley.com\/doi\/pdf\/10.1049\/cmu2.70154","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/ietresearch.onlinelibrary.wiley.com\/doi\/full-xml\/10.1049\/cmu2.70154","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/ietresearch.onlinelibrary.wiley.com\/doi\/pdf\/10.1049\/cmu2.70154","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,30]],"date-time":"2026-03-30T10:16:44Z","timestamp":1774865804000},"score":1,"resource":{"primary":{"URL":"https:\/\/ietresearch.onlinelibrary.wiley.com\/doi\/10.1049\/cmu2.70154"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1]]},"references-count":36,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1]]}},"alternative-id":["10.1049\/cmu2.70154"],"URL":"https:\/\/doi.org\/10.1049\/cmu2.70154","archive":["Portico"],"relation":{},"ISSN":["1751-8628","1751-8636"],"issn-type":[{"value":"1751-8628","type":"print"},{"value":"1751-8636","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1]]},"assertion":[{"value":"2025-12-16","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-18","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-30","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70154"}}