Cloud-Edge-Terminal Collaborative AI Agent for Compound Fault Detection and Diagnosis

Yiming He; Xiaoxi Hu; Weiming Shen

doi:10.1016/j.eng.2026.02.029

Engineering ›› :202602029 DOI: 10.1016/j.eng.2026.02.029

Article

research-article

Cloud-Edge-Terminal Collaborative AI Agent for Compound Fault Detection and Diagnosis

Author information +

History +

PDF

Abstract

With the continuous increase in the complexity and scale of intelligent systems, the demand for mechanical equipment maintenance for dynamic cross-domain versatility and human-computer collaborative intelligence has increased. The traditional static fault detection and diagnosis (FDD) paradigm, which is tailored to single scenarios and preset processes, has begun to exhibit inadequacy, inefficiency, and even unavailability. This paper proposes a cloud-edge-terminal collaborative artificial intelligence (CET-CAI) agent for system-level compound FDD (CFDD). The CET-CAI agent integrates the interactive and cognitive intelligence of a large language model with the deterministic reasoning capability of traditional fault diagnosis models. It can provide flexible dynamic task adaptability and efficient human-computer interaction with low communication costs. A novel distributed collaborative Transformer is designed and embedded in the edge system, which is specifically designed for multi-subsystem to reduce cross-subsystem signal interference and improve the generalization of unseen compound fault combinations. Experimental results on a multicomponent mechanical system dataset demonstrate that the proposed CET-CAI agent not only expands the dynamic task applicability of traditional diagnostic models, but also improves the robustness and readability of the diagnostic results.

Keywords

Cloud-edge-terminal collaborative / Large language model / AI agent / Compound fault detection and diagnosis / Distributed collaborative transformer

Cite this article

Download citation ▾

Yiming He, Xiaoxi Hu, Weiming Shen. Cloud-Edge-Terminal Collaborative AI Agent for Compound Fault Detection and Diagnosis. Engineering 202602029 DOI:10.1016/j.eng.2026.02.029

登录浏览全文

4963

注册一个新账户忘记密码

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Liu Y, Zhou Y, Liu Y, Xu Z, He Y. Intelligent fault diagnosis for CNC through the integration of large language models and domain knowledge graphs. Engineering 2025; 53:311-22.

[2]	Pu H, Teng S, Xiao D, Xu L, Qin Y, Luo J. Compound fault diagnosis of rotating machine through label correlation modeling via graph convolutional neural network. IEEE Trans Instrum Meas 2024; 73:1-10.

[3]	He Y, Shen W. An individual generalization framework based on independent samples towards a more reasonable fault diagnosis benchmark. Comput Ind 2025; 173:104359.

[4]	Li W, Lan H, Chen J, Feng K, Huang R. WavCapsNet: an interpretable intelligent compound fault diagnosis method by backward tracking. IEEE Trans Instrum Meas 2023; 72:1-11.

[5]	Yang Z, Baraldi P, Zio E. A method for fault detection in multi-component systems based on sparse autoencoder-based deep neural networks. Reliab Eng Syst Saf 2022; 220:108278.

[6]	Zhu Y, Zi Y, Zhang M, Xu J. LFC-DGNet: a likelihood feature compositional domain generalization network from single-fault to unseen multi-component compound fault diagnosis across machines. Adv Eng Inform 2025; 64:103037.

[7]	He Y, Shen W. MSRCN: a cross-machine diagnosis method for the CNC spindle motors with compound faults. Expert Syst Appl 2023; 233:120957.

[8]	Han S, Shao H, Cheng J, Yang X, Cai B. Convformer-NSE: a novel end-to-end gearbox fault diagnosis framework under heavy noise using joint global and local information. IEEE/ASME Trans Mechatron 2023; 28(1):340-9.

[9]	Liu B, Yan C, Liu Y, Lv M, Huang Y, Wu L. ISEANet: an interpretable subdomain enhanced adaptive network for unsupervised cross-domain fault diagnosis of rolling bearing. Adv Eng Inform 2024;62(Pt A):102610.

[10]	Pang B, Liu Q, Xu Z, Sun Z, Hao Z, Song Z. Fault vibration model driven fault-aware domain generalization framework for bearing fault diagnosis. Adv Eng Inform 2024;62(Pt A):102620.

[11]	Wu F, Shen T, Bäck T, Chen J, Huang G, Jin Y, et al. Knowledge-empowered, collaborative, and co-evolving AI models: the post-LLM roadmap. Engineering 2025; 44:87-100.

[12]	Jiang P, Sonne C, Li W, You F, You S. Preventing the immense increase in the life-cycle energy and carbon footprints of LLM-powered intelligent chatbots. Engineering 2024; 40:202-10.

[13]	You Y, Tan K, Jiang Z, Zhang L. Developing a predictive platform for salmonella antimicrobial resistance based on a large language model and quantum computing. Engineering 2025; 48:174-84.

[14]	Ma Y, Zheng S, Yang Z, Pan H, Hong J. A knowledge-graph enhanced large language model-based fault diagnostic reasoning and maintenance decision support pipeline towards industry 5.0. Int J Prod Res 2025; 0:1-22.

[15]	Peng H, Liu J, Du J, Gao J, Wang W. BearLLM: a prior knowledge-enhanced bearing health management framework with unified vibration signal representation. Proc Conf AAAI Artif Intell 2025; 39(19):19866-74.

[16]	Lin L, Zhang S, Fu S, Liu Y. FD-LLM: large language model for fault diagnosis of complex equipment. Adv Eng Inform 2025;65(Pt A):103208.

[17]	Wang J, Li T, Yang Y, Chen S, Zhai W. DiagLLM: multimodal reasoning with large language model for explainable bearing fault diagnosis. Sci China Inf Sci 2025; 68(6):160103.

[18]	Fabre W, Haroun K, Lorrain V, Lepecq M, Sicard G. From near-sensor to in-sensor: a state-of-the-art review of embedded ai vision systems. Sensors 2024; 24(6):5446.

[19]	Asghari A, Sohrabi MK. Server placement in mobile cloud computing: a comprehensive survey for edge computing, fog computing and cloudlet. Comput Sci Rev 2024; 51:100616.

[20]	Huaranga-Junco E, González-Gerpe S, Castillo-Cara M, Cimmino A, García-Castro R. From cloud and fog computing to federated-fog computing: a comparative analysis of computational resources in real-time iot applications based on semantic interoperability. Future Gener Comput Syst 2024; 159:134-50.

[21]	He Y, Shen W. CFSPT: a lightweight cross-machine model for compound fault diagnosis of machine-level motors. Inf Fusion 2024; 111:102490.

[22]	Qin Y, Wang Y, Li Zhaojun S, Wang B, Ding A, Wang C, et al. An in-depth tutorial on BJTU-RAO bogie datasets for fault diagnosis. IEEE Access 2025; 13:60879-88.

[23]	Ding A, Qin Y, Wang B, Guo L, Jia L, Cheng X. Evolvable graph neural network for system-level incremental fault diagnosis of train transmission systems. Mech Syst Signal Process 2024; 210:111175.

[24]	Zhang W, Peng G, Li C, Chen Y, Zhang Z. A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals. Sensors 2017; 17(2):425.

[25]	He Y, Zhao C, Shen W. Cross-domain compound fault diagnosis of machine-level motors via time-frequency self-contrastive learning. IEEE Trans Ind Inf 2024; 20(7):9692-701.

[26]	He Y, Chen J, Zhou X, Huang S. In-situ fault diagnosis for the harmonic reducer of industrial robots via multi-scale mixed convolutional neural networks. J Manuf Syst 2023; 66:233-47.

[27]	Zhou H, Yang J, Zhu Q, Chen J. Cross-speed spindle motor bearings fault diagnosis combined with multi-space variable scale adaptive filter and feedforward hybrid strategy. ISA Trans 2025; 157:368-80.

[28]	Li T, Zhou Z, Li S, Sun C, Yan R, Chen X. The emerging graph neural networks for intelligent fault diagnostics and prognostics: a guideline and a benchmark study. Mech Syst Signal Process 2022; 168:108653.

[29]	Zhang Y, Zhang S, Zhu Y, Ke W. Cross-domain bearing fault diagnosis using dual-path convolutional neural networks and multi-parallel graph convolutional networks. ISA Trans 2024; 152:129-42.