当前位置>主页 > 期刊在线 > 计算机技术 >

计算机技术22年13期

基于知识图谱和模糊推理的机械故障诊断模型
石玉,胡瑛婷
(山东师范大学 信息科学与工程学院,山东 济南 250358)

摘  要:目前,全球制造业正快速发展,智能制造是重要的发展方向,预测性维护是重要应用场景之一。机器学习算法对数据处理有极大优势,但较难理解。基于此,对机械故障知识图谱和模糊推理的转换进行研究,将知识图谱映射到模糊本体,为数据赋予语义信息,推理得到机械故障程度。在实验数据集上测试模型,准确率为 0.84,与经典分类算法 SVM 对比分析得出模型的有效性。同时,模型具有通用性和可扩展性。


关键词:故障诊断;知识图谱;模糊推理;智能制造



DOI:10.19850/j.cnki.2096-4706.2022.013.018


基金项目: 国家自然科学基金青年项目(62002207); 山东省自然科学基金项目(ZR2020MA102);省级大学生创新创业训练计划项目(S202110445087)


中图分类号:TP277                                       文献标识码:A                                     文章编号:2096-4706(2022)13-0072-06


Mechanical Fault Diagnosis Model Based on Knowledge Graph and Fuzzy Inference

SHI Yu, HU Yingting

(School of Information Science and Engineering, Shandong Normal University, Jinan 250358, China)

Abstract: At present, the global manufacturing industry is developing rapidly, intelligent manufacturing is an important development direction, and predictive maintenance is one of the important application scenarios. Machine learning algorithms have great advantages for data processing, but they are difficult to understand. Based on this, the transformation between mechanical fault Knowledge Graph and fuzzy inference is studied, and the Knowledge Graph is mapped to fuzzy ontology, endues the data with the semantic information, and infers to obtain the degree of mechanical fault. It tests the model on the experimental data set, and the accuracy is 0.84. The validity of the model is obtained by comparing and analyzing with the classical classification algorithm SVM. At the same time, the model has universality and expansibility.

Keywords: fault diagnosis; Knowledge Graph; fuzzy inference; intelligent manufacturing


参考文献:

[1] 李廉水,石喜爱,刘军 . 中国制造业 40 年:智能化进程与展望 [J]. 中国软科学,2019(1):1-9+30.

[2] 祝旭 . 故障诊断及预测性维护在智能制造中的应用 [J]. 自动化仪表,2019,40(7):66-69.

[3] 李杰其,胡良兵 . 基于机器学习的设备预测性维护方法综述 [J]. 计算机工程与应用,2020,56(21):11-19.

[4] WLIU W B,WANG Z D,LIU X H,et al.A survey of deep neural network architectures and their applications [J]. Neurocomputing,2016,234:11-26.

[5] 袁烨,张永,丁汉 . 工业人工智能的关键技术及其在预测性维护中的应用现状 [J]. 自动化学报,2020,46(10):2013-2030.

[6] DAVID L N,MILTON B.Ontoprog:an ontology-based model for implementing prognostics health management in mechanical machines [J].Advanced Engineering Informatics,2018,38:746-759.

[7] THAKKER D,PATEL P,INTIZAR A M,et al.Combining chronicle mining and semantics for predictive maintenance in manufacturing processes [J].Semantic Web,2020,11(6):927-948.

[8] 刘鑫 . 面向故障分析的知识图谱构建技术研究 [D]. 北京:北京邮电大学,2019.

[9] YAN W,LIU H,ZANNI-MERK C,et al.Ingenioustriz: an automatic ontology-based system for solving inventive problems [J]. 

Knowledge-Based Systems,2015,75:52-65.[10] YAN W,LIU H,LIU Y S,et al.Latent semantic extraction and analysis for triz-based inventive design [J].European J. of Industrial Engineering,2018,12(5):661-681.

[11] CAO Q S,SAMET A,ZANNI-MERK C,et al.An ontology-based approach for failure classification in predictive maintenance using fuzzy c-means and swrl rules [J].Procedia Comput Sci,2019,159(C):630-639.

[12] 刘奇林 . 基于语义的工业生产设备预测性维护策略研究与实现 [D]. 重庆:重庆邮电大学,2020.

[13] HELWIG N,PIGNANELLI E,SCHUTZE A. Condition monitoring of a complex hydraulic system using multivariate statistics [C]//2015 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) Proceedings.Pisa:IEEE,2015:

210-215.[14] AVIZIENIS A,LAPRIE J,BRIAN R,et la. Basic Concepts and Taxonomy of Dependable and Secure Computing [J].IEEE Transactions on Dependable and Secure Computing,2004,1(1):11-33.

[15] 李少远,戴先中,王耀南,等 . 智能控制发展研究[C]//2010-2011 控制科学与工程学科发展报告,2011-04,北京:中国科学技术出版社,2011:45-51+217.

[16] 顾俊,张宇 . 模糊控制的应用现状与发展趋势 [J]. 化工自动化及仪表,2017,44(9):811-812+902.

[17] 丁永生,应浩,邵世煌 . 模糊系统逼近理论:现状与展望 [J]. 信息与控制,2000(2):157-163+181.


作者简介:石玉(2000—),女,汉族,山东潍坊人,本科在读,研究方向:知识图谱、预测性维护、BIM 和 GIS 融合;胡瑛婷(2000—),女,汉族,山东淄博人,本科在读,研究方向:计算机科学与技术、BIM 和 GIS 融合。