Fault Diagnosis Expert System Based On Cloud Servers For A Punch Automation Production Line

In order to get unified intelligent diagnosis of different components of machine,different types of devices,different enterprises about punching production line,this need a powerful data computation platform to deal with massive information which consisted of super large scale device group,the correct conclusion must be given.So cloud-based fault diagnosis expert system was designed,focused on the mechanical fault feature of thin-walled circular punching station,and the related research was studied as following:(1)Fault detection characteristic was analyzed in this study,which mainly consisted of feeding manipulator,the reverse mechanism,the pendulum-arm mechanism and material separate platform.By analyzing how the mechanism works,the location of test node in fault diagnosis was designed,the function of detecting was realized,it was applicable in other equipment.(2)Multiple fault diagnosis system was designed,including hardware design,software design and data transmission via the cloud.Hardware design included DAQ system,signal collection,communication interface.Software design included fault analysis,fault alarm.Through RDS cloud database service supported MYSQL,it was designed that the cloud database for unified management of big data from device group's fault source.The GSM module uploaded the terminal's data to the cloud.(3)The mechanism of fault reasoning of the underlying server was analyzed,the existing methods was improved in three aspects: knowledge completeness,knowledge dynamism,knowledge completeness.Dynamic adjustment of fuzzy rules was designed.For multiple faults,it introduced pivotal importance degree and pheromone in probability evaluation,the event was staged for processing,fault early warning was realized.(4)Expert system of cloud fault diagnosis was realized in fault diagnosis terminal and the client website.Through the simulation in industry field,the effectiveness of system was evaluated.In conclusion,a set of new ideas addressing the massive data of fault source was presented,which makes the overall equipment efficiency rises,the cloud-based expert system of fault diagnosis have been preliminary proved.