Research Of Fault Diagnosis And Maintenance Decision-making In Engine Processing Equipment

This thesis discusses condition-based maintenance(CBM) in mechanical processingequipment, proceeds from reliability analysis, sequenced detection, fault diagnosis andmaintenance decision, and put forward the ways and means of implementingcondition-based maintenance in mechanical processing equipment.The article introduced technology and method of CBM firstly, and designed themaintenance workflow, which was contributed to the framework of the maintenance workin this article. Combining of the work of mine, gave a description relatively detailed withfault diagnosis and maintenance decision. In the part of fault diagnosis, use the expertsystem (ES) to analyze the historical date of the equipment, and derived classification ruleswhich can be used for diagnosing. Benefiting from the ID3algorithm, the rules can bederived more quickly.A maintenance decision-making model, based on Bayesian theory, was established, inorder to provide better maintenance suggestions to the maintenance staff. The model got theprobability of a more complex event from a less complex event’s probability, using theprobability reasoning ability of the Bayesian theory, this is the theoretical base. In themodel, firstly, get a priori probability according to the tools’ performance curve, obtaininformation of the current tool status by a sampling event, and use Bayesian theory tocalculate the posterior probability, which can reflect the actual degree of deterioration of thetool. Secondly, establish a matrix, representing maintenance costs, which considers variousmaintenance factors and the posterior probability, and select the optimal repair decision,based on the principle of cost minimization.The developing and designing theory on management information system is alsointroduced in this thesis. Combining with the actual project, a management informationsystem is designed for mechanical processing equipment. Database technology, networktechnology and ASP technology is used in the system. It decompound standard ofequipment’s sequenced detection in mechanical processing equipment, establish amaintenance program database, and give a maintenance program suggestion for sequenced detection staff to find fault. It realizes information management for standard and result ofsequenced detection, and offers the support on technology for sequenced detectionmanagement and CBM of plant.Finally, the thesis summarizes the studies and puts forward problems for furtherstudies.