基于知识图谱和贝叶斯推理的断纸故障诊断模型  

A Paper Break Fault Diagnosis Model Based on Knowledge Graph and Bayesian Inference

作　　者：张欢欢 洪蒙纳[1] 李继庚[1] ZHANG Huanhuan;HONG Mengna;Li Jigeng(State Key Lab of Pulp and Paper Engineering South China University of Technology,Guangzhou 510640,China)

机构地区：[1]华南理工大学制浆造纸工程国家重点实验室,广东广州510640

出　　处：《造纸科学与技术》2024年第2期39-43,101,共6页Paper Science & Technology

基　　金：制浆造纸工程国家重点实验室项目(2022ZD02)。

摘　　要：纸机断纸是制约造纸企业提高产品质量和生产效益的关键原因。造纸生产过程具有高维、非线性和多变量耦合的特点,对断纸故障的预防和诊断提出了挑战。数据驱动的方法基于断纸故障历史数据建模,对断纸故障的预防起到了一定的效果。但该方法忽略了造纸工业中隐藏的机理和经验知识,无法提供对断纸原因的追根溯源。知识图谱作为一种揭示实体间关系的语义网络,可以实现断纸故障数据与知识的集成。基于本体技术的断纸知识图谱为断纸故障诊断提供了全面、可扩展的关联知识库。在此基础上,结合贝叶斯网络开发了断纸故障诊断模型,通过对某生活用纸企业断纸故障的案例分析,验证了该模型在断纸故障推理方面的有效性,断纸预测的正确率达到了85%。Paper break is the key reason that restricts papermaking mill from improving product quality and production efficiency.The papermaking process is characterized by high-dimensional,nonlinear and multi-variable coupling,which poses challenges to the prevention and diagnosis of paper break.The data-driven method is based on historical data of paper break and has a certain effect on the prediction of paper break.However,this method ignores the hidden mechanisms and empirical knowledge in the papermaking process and cannot provide traceability of paper break.As a semantic network that reveals the relationships between entities,the knowledge graph can integrate paper break data and knowledge.Based on the ontology technology,the paper break knowledge graph provides a comprehensive and scalable correlation knowledge base for paper break fault diagnosis.On this basis,a paper break fault diagnosis model is developed combined with Bayesian network.Through a case study of paper break in a tissue papermaking enterprise,the effectiveness of the model in the inference of the paper break fault is verified,the accuracy of paper break prediction reaches 85%.

关 键 词：断纸 故障诊断 知识图谱 贝叶斯网络 

分 类 号：TS733.3[轻工技术与工程—制浆造纸工程]