Journal of Computer Applications ›› 0, Vol. ›› Issue (): 72-78.DOI: 10.11772/j.issn.1001-9081.2024040484

• Artificial intelligence • Previous Articles     Next Articles

Device fault diagnosis method based on knowledge graph and multi-task learning

Haigang JIANG1(), Min ZHU2, Xiaoqiang JIANG3   

  1. 1.Operation and Maintenance Center,Shanghai Installation Engineering Group Company Limited,Shanghai 200080,China
    2.Commissioning and Testing Center,Shanghai Installation Engineering Group Company Limited,Shanghai 200439,China
    3.School of Urban Operation Management,Shanghai Urban Construction Vocational College,Shanghai 200438,China
  • Received:2024-04-23 Revised:2024-07-09 Accepted:2024-07-10 Online:2025-01-24 Published:2024-12-31
  • Contact: Haigang JIANG

基于知识图谱和多任务学习的设备故障诊断方法

蒋海刚1(), 朱敏2, 蒋小强3   

  1. 1.上海市安装工程集团有限公司 运维中心,上海 200080
    2.上海市安装工程集团有限公司 调试检测中心,上海 200439
    3.上海城建职业学院 城市运营管理学院,上海 200438
  • 通讯作者: 蒋海刚
  • 作者简介:蒋海刚(1975—),男,江苏宜兴人,高级工程师,CCF会员,主要研究方向:机器学习、非结构化数据处理、设备故障诊断
    朱敏(1993—),女,陕西咸阳人,工程师,硕士研究生,主要研究方向:建筑智能化、建筑节能
    蒋小强(1980—),男,湖南岳阳人,教授,博士,主要研究方向:建筑设备、BIM运维。
  • 基金资助:
    上海建工集团股份有限公司重点课题项目(23JCSF?21)

Abstract:

To address the issue of insufficient intelligent analysis and autonomous decision-making capabilities as well as low fault diagnosis efficiency in building equipment operation and maintenance processes, a novel fault diagnosis method based on knowledge graphs and multi-task learning was proposed. Firstly, an operation and maintenance-oriented knowledge graph was constructed, and multi-source heterogeneous data were extracted from building equipment systems by using natural language processing and entity linking techniques, thereby obtaining rich knowledge representation. Secondly, in the case of few-shot labeling, multi-source symptom associated identification was explored. And unlabeled data were used to optimize the model parameters iteratively through self-training and co-training strategies, so as to improve generalization capability of the model. Finally, when designing fault root cause localization technology based on deep knowledge reasoning, a probabilistic graphical model was utilized to trace the fault propagation paths in complex equipment systems, thereby enhancing the accuracy and interpretability of fault analysis. Simultaneously, fusion mechanism was introduced into multi-task learning framework, thereby improving performance of the proposed method on fault diagnosis tasks. Experimental results demonstrate that the proposed method achieves a fault diagnosis accuracy of 92% with an average diagnosis time of 6.5 seconds per case, outperforming the comparison models in evaluation metrics such as accuracy, precision, and recall.

Key words: knowledge graph, building equipment, fault diagnosis, multi-task learning, operation and maintenance optimization

摘要:

针对建筑设备运维过程中的智能分析与自主决策能力不足、故障诊断效率低等问题,提出一种基于知识图谱和多任务学习的设备故障诊断方法。首先,构建面向运维的知识图谱,利用自然语言处理和实体链接技术提取建筑设备系统的多源异构数据,从而获取丰富的知识表示。其次,在小样本标注的情况下,探索多源症状关联识别,并把未标注数据通过自训练和协同训练策略迭代优化模型参数,提高模型泛化能力。最后,在设计基于深度知识推理的故障根因定位技术时,借助概率图模型追溯复杂设备系统的故障传播路径,提高故障分析的准确性和可解释性。同时,引入多任务学习框架融合机制,提升所提方法在故障诊断任务上的性能。实验结果显示,所提方法的故障诊断准确率达92%,平均每条记录诊断时间达6.5 s,在准确率、精确率和召回率等评估指标上均优于对比模型。

关键词: 知识图谱, 建筑设备, 故障诊断, 多任务学习, 运维优化

CLC Number: