大数据驱动的舰船电子设备故障智能诊断研究  

Research on intelligent fault diagnosis of ship electronic equipment driven by big data

作　　者：康迪[1] 刘娇[1] 唐磊[1] KANG Di;LIU Jiao;TANG Lei(Shijiazhuang Information Engineering Vocational College,Shijiazhuang 050000,China)

机构地区：[1]石家庄信息工程职业学院,河北石家庄050000

出　　处：《舰船科学技术》2021年第14期160-162,共3页Ship Science and Technology

基　　金：石家庄市高等教育科学研究项目(20171024)

摘　　要：舰船电子设备故障诊断主要依靠故障分类器对故障特征的识别,因此分类器识别结果的准确性尤为重要。在分类器识别过程中,会存在部分未被滤除的噪声信号对其进行干扰,影响分类器识别结果,降低其诊断结果准确率。为了解决此类问题引入大数据驱动,提出大数据驱动的舰船电子设备故障智能诊断研究。依托大数据信息流,完成对故障特征的提取。采用数据特征免疫算法,对特征数据进行免疫计算,在免疫特征数据基础上完成对噪声数据的滤除计算。最后,对分类器识别特征量进行更新,实现分类器对故障目标的智能诊断。通过与传统诊断方法的10组差异化数据对比表明:提出的诊断方法,具有诊断结果稳定性好、适用性强、准确率高的特点。Fault diagnosis of ship electronic equipment mainly depends on fault classifier to identify fault features,so the accuracy of classifier recognition results is particularly important.In the process of classifier recognition,there will be some unfiltered noise signals to interfere with it,which will affect the classifier recognition results and the accuracy of the diagnostic results.In order to solve this problem,the intelligent fault diagnosis of ship electronic equipment driven by big data is proposed.First,relying on the big data information flow to complete the fault feature extraction and then,using the data feature immune algorithm,the feature data is immunized,and then the noise data is filtered and calculated on the basis of the immune feature data.Finally,the classifier is updated to realize the intelligent diagnosis of fault target.Compared with10 groups of differential data of traditional diagnostic methods,the proposed diagnostic method has the characteristics of good stability,strong applicability and high accuracy.

关 键 词：大数据驱动 舰船 电子设备 故障智能诊断 

分 类 号：U664[交通运输工程—船舶及航道工程]