Construction And Application Of Intelligent Instrument Fault Detection System

This paper uses the basic Internet of things to implement the architecture,relying on the electric spark source to apply it to the testing of the geophysical prospecting instrument.The physical parameters in the instrument are collected through the sensor and sent to the server.After the server is processed,the user is fed back to the user.The server end is connected with the FTA and the artificial neural network.The detection and diagnosis of the device fault makes the exploration of the integration of Internet of things,cloud computing and geophysical prospecting.The paper first introduces the research background of instrument fault detection,and expounds that the development direction of instrument detection is early warning.After that,through the introduction of the Internet of things strategy and the development status of the Internet of things in the developed countries in the world,it is demonstrated that the Internet of things is the next global economic growth point and will be the next global economic growth point.Growth momentum is the best strategy to maximize the benefits of other industries.Secondly,through the summary of the basic concepts of the Internet of things technology,the technical feasibility of the combination of the Internet of things and the geophysical industry is discussed,and the inevitable trend of the combination of the Internet of things and artificial intelligence and the neural network is further analyzed,as well as the possibility of combining the Internet of things with the field of geophysical exploration after the intelligence.Thirdly,through the analysis and discussion of the flow direction of the system and the functional requirements of each part,we choose the appropriate server and database implementation scheme,and combine the basic process of software design to realize the analysis and design of the basic functions of the three clients,and choose the appropriate development platform by contrast.Finally,according to the previous design,each module of the system is implemented.In all the modules,the fault diagnosis module based on artificial neural network is the most important part.In this paper,the FTA fault analysis method is used to make the types and reasons for the possible failure of the EDM in the process of use.The fault tree is established,and a suitable training sample is set up to train the neural network through the analysis of the fault tree.Then the BP neural network is selected to realize the fault diagnosis of the electric spark.By adjusting the parameters,the weights of the neural network are obtained,and the function of the module is realized.