Research Of Logistics Equipment Hidden Failure Warning Model Based On Data Mining

With the rapid development of modern logistics, modern logisticsequipment is gradually to the way of automatic and intelligent, and makesinspection and maintenance to maintain its safe operation become one of themost important aspects of the logistics equipment management. The purposeof this study is to establish the hidden failure warning model of logisticsequipment based on data mining, estimate the impact of hidden fault to thehappen of functional failure, in order to monitor the trend of hidden fault andprevent the sudden fault caused by hidden fault. According to thecharacteristics of the subject, this paper will use inductive analysis tosummarize related literature at home and abroad, use the comparative analysisto verify the effectiveness of algorithm, and exemplification to verify thereliability of the Hidden Failure Warning Model.First of all, the research background and significance are detailedexpounded. The article compares and analyses current research situation ofthe domestic and foreign research from the data mining technology, faultdiagnosis technology, the using of data mining technology in fault diagnosis,respectively, pointing out the existing deficiency of fault diagnosis technology.The relevant basic knowledge of logistics equipment, fault diagnosistechnology and data mining technology are carried on the detailed elaboration,laid the theoretical foundation of this paper.Secondly, in order to improve the effectiveness of the functional failurediagnosis, this paper put forward a Weighted Association IncrementalUpdating Algorithm(WAIUA) based on the actual situation that different faultfactors have different fault contribution in the fault diagnosis process, andthis algorithm is more suitable for the real demand of equipment faultdiagnosis.Furthermore, based on the analysis of related factors of hidden failure under the failure state, this paper put forward the Hidden Failure WarningModel(HFWM) and solve the nonlinear problem with more uncertainties likehidden failure by introducing neural network method. The model not onlyrealizes the functional failure diagnosis, but also achieve the purpose ofmonitoring the hidden fault through the comprehensive analysis of faultfactors.Finally, this paper offers a brief analysis of the hydraulic system oflogistics equipment from the aspects of structure and fault factors and selectsthe representative fault data to establish the neural network of the HFWM. Atthe same time, the feasibility of the model in the process of logisticsequipment fault diagnosis has been proved.