目的 面向复杂装备故障诊断中知识分散、关系耦合复杂与结果可解释性不足等问题,提出一种基于知识图谱的故障诊断系统设计与推理方法。方法 以设备、故障、症状、原因和解决方案5类实体为核心,构建故障诊断知识图谱。采用图遍历进行候选故障检索,并结合贝叶斯置信度评估实现候选排序和路径解释。在系统实现上采用展示层、逻辑层和数据层三层架构,并通过典型测试用例对诊断效果和扩展性能进行验证。结果 在当前闭集测试条件下,系统在20个独立测试用例上的首位、前三位和前五位命中率均为100.0%,总体准确率为92.5%。在图谱规模扩展、推理深度增加和并发访问条件下,系统仍保持较好的稳定性。结论 该方法能够较好支撑复杂装备故障场景下的知识组织、候选排序与可解释诊断。
Abstract
The work aims to propose a knowledge graph-based fault diagnosis system design and reasoning method to address fragmented knowledge organization, complex relational coupling, and limited interpretability in complex equipment fault diagnosis. A diagnosis-oriented knowledge graph with five core entity types, namely equipment, fault, symptom, cause, and solution, was constructed. Graph traversal was used to retrieve candidate faults, and Bayesian confidence assessment was used to rank the candidates and explain the inferred paths. A three-layer architecture consisting of the presentation layer, logic layer and data layer was adopted. Typical test cases were used to verify diagnostic performance and scalability. Under the current closed-set testing conditions, the Top-1, Top-3, and Top-5 hit rates across 20 independent test cases were all 100.0% and the overall accuracy was 92.5%. The system remained excellent stability under conditions of knowledge graph scale expansion, increased reasoning depth, and concurrent access. This method can support knowledge organization, candidate ranking, and explainable diagnosis in complex equipment fault scenarios effectively.
关键词
知识图谱 /
复杂装备 /
故障诊断 /
图遍历 /
贝叶斯推理 /
可解释诊断 /
知识服务
Key words
knowledge graph /
complex equipment /
fault diagnosis /
graph traversal /
Bayesian inference /
explainable diagnosis /
knowledge service
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] LEI Y G, YANG B, JIANG X W, et al.Applications of Machine Learning to Machine Fault Diagnosis: A Review and Roadmap[J]. Mechanical Systems and Signal Processing, 2020, 138: 106587.
[2] CAI B P, HUANG L, XIE M.Bayesian Networks in Fault Diagnosis[J]. IEEE Transactions on Industrial Informatics, 2017, 13(5): 2227-2240.
[3] HOGAN A, BLOMQVIST E, COCHEZ M, et al.Knowledge Graphs[J]. ACM Computing Surveys, 2022, 54(4): 1-37.
[4] JI S X, PAN S R, CAMBRIA E, et al.A Survey on Knowledge Graphs: Representation, Acquisition, and Applications[J]. IEEE Transactions on Neural Networks and Learning Systems, 2022, 33(2): 494-514.
[5] NICKEL M, MURPHY K, TRESP V, et al.A Review of Relational Machine Learning for Knowledge Graphs[J]. Proceedings of the IEEE, 2016, 104(1): 11-33.
[6] 王昊奋, 漆桂林, 陈华钧. 知识图谱——方法、实践与应用[M]. 北京: 电子工业出版社, 2019.
WANG H F, QI G L, CHEN H J.Methods, Practice and Applications of Knowledge Graphs[M]. Beijing: Publishing House of Electronics Industry, 2019.
[7] 刘知远, 韩旭, 孙茂松. 知识图谱与深度学习[M]. 北京: 清华大学出版社, 2020.
LIU Z Y, HAN X, SUN M S.Knowledge Graph and Deep Learning[M]. Beijing: Tsinghua University Press, 2020.
[8] SINGHAL A. Introducing the Knowledge Graph: Things, Not Strings[EB/OL].(2012-05-16)[2026-04-09]. https://blog.google/products-and-platforms/products/search/introducing-knowledge-graph-things-not/.
[9] QU H Y, ZHANG Y Y, LIANG K, et al.A Knowledge- Graph-Driven Method for Intelligent Decision Making on Power Communication Equipment Faults[J]. Electronics, 2023, 12(18): 3939.
[10] CHEN Q, LI Q, WU J Y, et al.Application of Knowledge Graph in Power System Fault Diagnosis and Disposal: A Critical Review and Perspectives[J]. Frontiers in Energy Research, 2022, 10: 988280.
[11] CAI C, JIANG Z Y, WU H, et al.Research on Knowledge Graph-Driven Equipment Fault Diagnosis Method for Intelligent Manufacturing[J]. The International Journal of Advanced Manufacturing Technology, 2024, 130(9): 4649-4662.
[12] HAGBERG A A, SCHULT D A, SWART P J. Exploring Network Structure, Dynamics, and Function Using NetworkX[C]//Proceedings of the 7th Python in Science Conference. Pasadena:[s. n.], 2008.
[13] MCKINNEY W.Data Structures for Statistical Computing in Python[C]//Proceedings of the 9th Python in Science Conference. Austin: SCIPY, 2010.
[14] HARRIS C R, MILLMAN K J, VAN DER WALT S J, et al. Array Programming with NumPy[J]. Nature, 2020, 585(7825): 357-362.
[15] HUNTER J D.Matplotlib: A 2D Graphics Environment[J]. Computing in Science & Engineering, 2007, 9(3): 90-95.
[16] 张帜. Neo4j权威指南[M]. 北京: 清华大学出版社, 2017.
ZHANG Z.Secondary Authoritative Guide[M]. Beijing: Tsinghua University Press, 2017.
[17] 牟文斌. Django开发入门与项目实战[M]. 北京: 电子工业出版社, 2021.
MOU W B.Django Development Introduction and Project Actual Combat[M]. Beijing: Publishing House of Electronics Industry, 2021.
[18] 明日科技. Python树莓派开发从入门到精通[M]. 北京: 清华大学出版社, 2021.
Mingri Keji.Python Raspberry Pie Development from Entry to Mastery[M]. Beijing: Tsinghua University Press, 2021.
[19] ZHENG Y, ZHAO F, WANG Z.Fault Diagnosis System of Bridge Crane Equipment Based on Fault Tree and Bayesian Network[J]. The International Journal of Advanced Manufacturing Technology, 2019, 105(9): 3605-3618.
[20] ADEMUJIMI T, PRABHU V.Fusion-Learning of Bayesian Network Models for Fault Diagnostics[J]. Sensors, 2021, 21(22): 7633.
[21] QIAN G, ZHENG S G, CAO L H.Bayesian Network Based on a Fault Tree and Its Application in Diesel Engine Fault Diagnosis[C]//Proceedings of International Conference on Mechatronics and Information Technology: Control Systems and Robotics.[s. l.] : SPIE, 2005.
[22] VENKATASUBRAMANIAN V, RENGASWAMY R, YIN K W, et al.A Review of Process Fault Detection and Diagnosis Part I: Quantitative Model-Based Methods[J]. Computers & Chemical Engineering, 2003, 27(3): 293-311.
[23] VENKATASUBRAMANIAN V, RENGASWAMY R, KAVURI S N.A Review of Process Fault Detection and Diagnosis Part II: Qualitative Models and Search Strategies[J]. Computers & Chemical Engineering, 2003, 27(3): 313-326.
[24] VENKATASUBRAMANIAN V, RENGASWAMY R, KAVURI S N, et al.A Review of Process Fault Detection and Diagnosis Part III: Process History Based Methods[J]. Computers & Chemical Engineering, 2003, 27(3): 327-346.
[25] ISERMANN R.Fault-Diagnosis Systems: An Introduction from Fault Detection to Fault Tolerance[M]. Berlin: Springer, 2006.
基金
某部装备质量技术基础项目（2026-JSZBJCXX-F001）
PDF(2446 KB)
