The Design Of Fault Diagnosis System For Industrial Explosives Production Line

The research to the fault diagnosis of Industrial emulsified explosive production line is always a quite required and significant task in the field of Industrial emulsified explosive security. As the varieties of Industrial explosives updates constantly, the equipments of industrial explosive production line become more complicated and roboticized, and the traditional method of fault diagnosis is unable to diagnose the complicated equipments accurately and real-timely. The paper does some researches to the industrial explosive production, which include two production lines as gelatinous emulsifying explosives production line and powdered emulsifying explosives line. The application of expert system and BP neural network in the industrial explosive fault diagnosis are explored and studied in this paper.This paper analyses the process flow of the two industrial explosive production lines at first, anatomizes the existing fault points of industrial explosive production process and their damage to production, and the fault points are gathered and grade disposed. Comparing expert system with neural network in the aspects of knowledge acquirement, knowledge base construction and inference ability, the whole design project of intelligent fault diagnosis system was confirmed, the expert system and neural network were combined creatively, applied in the design of intelligent fault diagnosis system was confirmed, the expert system and neural network were combined creatively, applied in the design of intelligent fault diagnosis system for those two industrial explosive production lines and engineering practice. In this way, a lot expert knowledge of failure diagnosis is scattered in neural network, and the other expert dates are decomposed as analysis databases and comprehensive database. Based on those databases, the inference engine is completed by analysis reasoning,ANN reasoning and comprehensive reasoning. At the same time, the explain mechanism is done by the comprehensive reasoning, and display the message of the diagnosis on the hum-machine interface.The diagnosis system in the paper was designed to diagnose the two production line of industrial explosive. Since two production lines occupy the independence of the resource constraints respectively, and the instruments used to be a great similarity on the productive technology, the paper adopt the design processing the data respectively and monitoring together to carry out the industrial explosive fault diagnose system. The paper adopt the DDE, a data sharing technology, to gathering the real-time data from IC computer, and the socket communication technology to send the information to the computer, where the fault diagnosis system was designed. The paper offers a method of using VC++6.O integrated development environment to design the industrial explosive intelligent fault diagnosis system, with the powerful application software SQL server 2000 and the ADO techniques to the data access, and daemonic called neural network kits of matlab. And finally realize the functions of system that monitoring and diagnosing the two industrial explosive lines.In order to test the system, set some data as those which gathered form industrial planned, and test the system whether could do the work. The result of testing indicates that the fault diagnosis system has a better ability of real-time fault diagnosis. It has a positive effect on preventing fateful production accident, and offers a valid reference to the development direction of safety monitoring and fault diagnosis for emulsified explosive production line.