基于关联规则分类的船用柴油机故障诊断  

Fault Diagnosis of Marine Diesel Engine Based on the Classification Base of Association

作　　者：石大亮 张毅然[1] 湛日景 林赫[1] 邱爱华 刘佳彬 Shi Daliang;Zhang Yiran;Zhan Rijing;Lin He;Qiu Aihua;Liu Jiabin(School of Mechanical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;Shanghai Qiyao Heavy Industry Company Limited,Shanghai 201108,China)

机构地区：[1]上海交通大学机械与动力工程学院,上海200240 [2]上海齐耀重工有限公司,上海201108

出　　处：《内燃机学报》2023年第4期369-375,共7页Transactions of Csice

基　　金：上海市人工智能创新发展专项资助项目.

摘　　要：鉴于数据驱动的故障诊断方法诊断过程和结果难以解释,提出了一种基于关联规则分类(CBA)的船用柴油机故障诊断和故障作用机理解释方法.通过GT-Power柴油机故障仿真试验建立了柴油机故障数据库,采用关联规则分类算法构建了故障分类器,并基于可视化方法提取了重要规则,解析故障作用机理和模型的诊断过程.结果表明:基于关联规则分类的柴油机故障诊断方法对测试集的故障识别率高达98.67%,提取的规则较好地吻合相关热力学知识,可用于故障作用机理与诊断的进一步研究.In view of the problem that it is difficult to explain the diagnosis process and results of data-driven fault diagnosis method,a method of marine diesel engine fault diagnosis and fault mechanism interpretation based on the classification base of association(CBA)was proposed.Through the GT-Power diesel engine fault simulation experiment,the diesel engine fault database was established,the fault classifier was constructed based on the classification base of association algorithm,and the important rules were extracted based on visualization method to analyze the fault mechanism and the diagnosis process of the model.The results show that the fault recognition rate of the diesel engine fault diagnosis method based on classification base of association is as high as 98.67%,and the extracted rules can better match the principle of thermodynamics related knowledge,which can be used for further research on fault mechanism analysis and fault diagnosis.

关 键 词：船用柴油机 故障诊断 关联规则分类 故障识别率 

分 类 号：TK428[动力工程及工程热物理—动力机械及工程]