{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T21:26:05Z","timestamp":1790803565614,"version":"4.1.0"},"reference-count":36,"publisher":"MDPI AG","issue":"12","license":[{"start":{"date-parts":[[2026,6,18]],"date-time":"2026-06-18T00:00:00Z","timestamp":1781740800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Selective laser melting (SLM) equipment operates as a complex cyber\u2013physical system, wherein strong implicit coupling among internal subsystems presents significant challenges for condition monitoring and fault diagnosis. Existing deep learning methods often suffer from feature submersion when processing multi-source heterogeneous data and lack the capability for system-level topological causal inference. To address these issues, we propose a multisystem coupling modeling and anomaly traceability method based on a decoupled graph attention network (ST-DBGAE). Independent local spatiotemporal feature alignment modules are constructed to map heterogeneous sensory data into a unified latent space. This eliminates dimensional discrepancies while strictly maintaining the feature independence of underlying hardware subsystems, such as optical and gas circuits. A dynamic graph attention mechanism with sparse priors is subsequently introduced to adaptively capture time-varying coupling weights triggered by implicit interactions (e.g., thermal fluids), bypassing the need for predefined rigid physical connections. Furthermore, a dual-branch two-stage decoupled optimization architecture is designed. By blocking the cross-interference of global backpropagation, this architecture outputs a continuous equipment health index (HI) based on reconstruction errors and employs a topological difference matrix inference mechanism to reversely anchor the root-cause nodes responsible for cross-system cascading degradation. Experimental results based on over 310,000 real operational monitoring records from industrial SLM equipment demonstrate that the proposed model achieves a comprehensive diagnostic Macro-F1 score of 96.5% across eight operating states. The single-class detection rates (ACCs) of specific underlying anomalies are significantly improved. This method not only enables high-precision equipment health warnings but also provides a physically interpretable microscopic fault propagation mapping for predictive maintenance.<\/jats:p>","DOI":"10.3390\/s26123889","type":"journal-article","created":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T13:04:04Z","timestamp":1781874244000},"page":"3889","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Decoupled Graph Attention Modeling and Anomaly Traceability Method for Multisystem Coupling in SLM Equipment"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-6406-8763","authenticated-orcid":false,"given":"Qi","family":"Liu","sequence":"first","affiliation":[{"name":"School of Mechanical Engineering, Shenyang University of Technology, Shenyang 110870, China"},{"name":"Nanjing Zhongke Raycham Laser Technology Co., Ltd., Nanjing 210046, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weijun","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Mechanical Engineering, Shenyang University of Technology, Shenyang 110870, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongyou","family":"Bian","sequence":"additional","affiliation":[{"name":"School of Mechanical Engineering, Shenyang University of Technology, Shenyang 110870, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fei","family":"Xing","sequence":"additional","affiliation":[{"name":"Nanjing Zhongke Raycham Laser Technology Co., Ltd., Nanjing 210046, China"},{"name":"School of Mechanical Engineering and Automation, Northeastern University, Shenyang 110819, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,18]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"041101","DOI":"10.1063\/1.4935926","article-title":"Review of selective laser melting: Materials and applications","volume":"2","author":"Yap","year":"2015","journal-title":"Appl. Phys. Rev."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1041","DOI":"10.1109\/TASE.2022.3215258","article-title":"Process monitoring, diagnosis and control of additive manufacturing","volume":"21","author":"Fang","year":"2022","journal-title":"IEEE Trans. Autom. Sci. Eng."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1016\/j.promfg.2018.06.009","article-title":"Cluster identification of sensor data for predictive maintenance in a selective laser melting machine tool","volume":"24","author":"Uhlmann","year":"2018","journal-title":"Procedia Manuf."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"4477","DOI":"10.1007\/s12206-024-2401-1","article-title":"In-situ defect detection in laser-directed energy deposition with machine learning and multi-sensor fusion","volume":"38","author":"Chen","year":"2024","journal-title":"J. Mech. Sci. Technol."},{"key":"ref_5","first-page":"102364","article-title":"Multi-modal sensor fusion with machine learning for data-driven process monitoring for additive manufacturing","volume":"48","author":"Petrich","year":"2021","journal-title":"Addit. Manuf."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Gurav, V., Upadhyay, A., and Sakhare, H. (2025). An explainable lightweight framework for process control and fault detection in additive manufacturing. J. Manuf. Mater. Process., 9.","DOI":"10.3390\/jmmp9120392"},{"key":"ref_7","first-page":"105","article-title":"Deep learning-based multisensor fusion for in situ defect prediction in additive manufacturing","volume":"13606","author":"Xie","year":"2025","journal-title":"Proc. SPIE"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Simsir, U. (2026). Multi-sensor process monitoring and fault diagnosis for multi-mode industrial servomotor systems with fault classification and RUL prediction: A representative case study for smart manufacturing applications. Processes, 14.","DOI":"10.3390\/pr14050772"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"e2196266","DOI":"10.1080\/17452759.2023.2196266","article-title":"Heterogeneous sensor data fusion for multiscale, shape agnostic flaw detection in laser powder bed fusion additive manufacturing","volume":"18","author":"Bevans","year":"2023","journal-title":"Virtual Phys. Prototyp."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"030006","DOI":"10.1063\/5.0234589","article-title":"Intelligent fault diagnosis of surface level defects in stainless steel plates through fused sensor data","volume":"3217","author":"Sindhu","year":"2024","journal-title":"AIP Conf. Proc."},{"key":"ref_11","first-page":"1","article-title":"A new multisensor feature fusion KAN network for autonomous underwater vehicle fault diagnosis","volume":"74","author":"Zhang","year":"2024","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2801","DOI":"10.1007\/s40964-025-01496-5","article-title":"Multimodal learning of melt pool dynamics in laser powder bed fusion","volume":"11","author":"Mojumder","year":"2026","journal-title":"Prog. Addit. Manuf."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Luo, Q., Huang, N., Bartles, D.L., and Beese, A.M. (2025). Geometry-informed multimodal variational autoencoder for real-time prediction of properties for Ti\u20136Al\u20134V fabricated using PBF-LB. J. Intell. Manuf., 1\u201320.","DOI":"10.1007\/s10845-025-02693-3"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Nie, Q., Geng, J., and Liu, C. (2026). A review of fault diagnosis methods: From traditional machine learning to large language model fusion paradigm. Sensors, 26.","DOI":"10.3390\/s26020702"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"03007","DOI":"10.1051\/itmconf\/20258003007","article-title":"Development of an online defect detection system for additive manufacturing based on multi-sensor fusion","volume":"80","author":"Wang","year":"2025","journal-title":"ITM Web Conf."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"122426","DOI":"10.1109\/ACCESS.2024.3452314","article-title":"Physics-embedded recurrent graph neural network for fault diagnosis of complex systems","volume":"12","author":"Gao","year":"2024","journal-title":"IEEE Access"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"110328","DOI":"10.1016\/j.ress.2024.110328","article-title":"Causal intervention graph neural network for fault diagnosis of complex industrial processes","volume":"251","author":"Liu","year":"2024","journal-title":"Reliab. Eng. Syst. Saf."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Xing, K., Wang, Y., Lu, H., and Yun, D. (2025, January 4\u20136). Research on fault diagnosis and predictive maintenance algorithm of complex industrial system based on graph neural network. Proceedings of the 2025 Asia Conference on Energy Conversion Systems and Power Electronics (AECSPE), Kuala Lumpur, Malaysia.","DOI":"10.1109\/AECSPE66597.2025.00150"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Ma, J., Huang, J., Liu, S., Luo, J., and Jing, L. (2024). A self-attention Legendre graph convolution network for rotating machinery fault diagnosis. Sensors, 24.","DOI":"10.3390\/s24175475"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"104528","DOI":"10.1016\/j.dsp.2024.104528","article-title":"Semi-supervised few-shot fault diagnosis driven by multi-head dynamic graph attention network under speed fluctuations","volume":"151","author":"Jiang","year":"2024","journal-title":"Digit. Signal Process."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Wen, W., Qin, J., and Chang, Q. (2026). Semi-supervised graph attention network for screw pump fault diagnosis: Revealing the dynamic coupling of multi-source information. Entropy, 28.","DOI":"10.3390\/e28030338"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Wang, B., and Zhao, S. (2025). MTAGCN: Multi-task graph-guided convolutional network with attention mechanism for intelligent fault diagnosis of rotating machinery. Machines, 13.","DOI":"10.3390\/machines13050347"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"4175","DOI":"10.1007\/s00170-025-15773-8","article-title":"Graph attention-based knowledge reasoning for mechanical performance prediction of L-PBF printing parts","volume":"138","author":"Xiao","year":"2025","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Liu, Q., Liu, W., Bian, H., and Xing, F. (2026). Deep spatiotemporal condition monitoring and subsystem fault classification for selective laser melting equipment. Coatings, 16.","DOI":"10.3390\/coatings16050517"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"2724","DOI":"10.1021\/ie503530v","article-title":"Modified independent component analysis and Bayesian network-based two-stage fault diagnosis of process operations","volume":"54","author":"Yu","year":"2015","journal-title":"Ind. Eng. Chem. Res."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Duan, H., Chen, G., Yu, Y., Du, C., Bao, Z., and Ma, D. (2025). DyGAT-FTNet: A dynamic graph attention network for multi-sensor fault diagnosis and time\u2013frequency data fusion. Sensors, 25.","DOI":"10.3390\/s25030810"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Li, C., Ma, Z., Zeng, Y., Yang, Z., Li, J., Yang, Z., Xiong, J., Niu, S., Wang, Z., and Zhao, H. (2026). Machine learning-driven alignment architecture of heterogeneous data with transient varying semantics. Nat. Commun.","DOI":"10.1038\/s41467-026-72377-w"},{"key":"ref_28","first-page":"1","article-title":"Application of a multimodal deep learning model based on recursive fusion feature map with transformer-TCN for complex fault diagnosis of flying wing UAV actuators","volume":"74","author":"Zhang","year":"2025","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_29","first-page":"1","article-title":"Disentangled feature representation based on multiscale content learning in industrial heterogeneous nonstationary fault detection","volume":"74","author":"Yu","year":"2025","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Aryal, B., Modekwe, G., and Lu, Q. (2026). Multi-level temporal graph networks with local-global fusion for industrial fault diagnosis. arXiv.","DOI":"10.1088\/1361-6501\/ae739a"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"551","DOI":"10.1021\/acs.iecr.4c02832","article-title":"Improved Pearson correlation coefficient-based graph neural network for dynamic soft sensor of polypropylene industries","volume":"64","author":"Han","year":"2024","journal-title":"Ind. Eng. Chem. Res."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"14479","DOI":"10.1109\/TII.2024.3454065","article-title":"Dynamic graph-driven rotating machine fault diagnosis: An adaptively updating cross-domain relationship information","volume":"20","author":"Yang","year":"2024","journal-title":"IEEE Trans. Ind. Inform."},{"key":"ref_33","first-page":"1","article-title":"A two-stage fault diagnosis method with rough and fine classifiers for phased array radar transceivers","volume":"73","author":"Chen","year":"2024","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Komorska, I., and Puchalski, A. (2024). Condition monitoring using a latent space of variational autoencoder trained only on a healthy machine. Sensors, 24.","DOI":"10.3390\/s24216825"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Zhao, S., Zhang, H., Sun, B., and Wang, Y. (2026). Spatio-temporal graph neural networks for anomaly detection in complex industrial processes. Sensors, 26.","DOI":"10.3390\/s26051597"},{"key":"ref_36","unstructured":"Shan, X. (2025). Fault Identification and Localization in Distribution Grids Based on an Attention-Hybrid Graph Neural Network. [Ph.D. Dissertation, University of Kentucky]. Available online: https:\/\/uknowledge.uky.edu\/ece_etds\/216."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/12\/3889\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T04:48:58Z","timestamp":1782449338000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/12\/3889"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,18]]},"references-count":36,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["s26123889"],"URL":"https:\/\/doi.org\/10.3390\/s26123889","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,18]]}}}