Paver Of The Artificial Neural Network-based Intelligent Fault Diagnosis System

Heavy-duty engineering plants such as Paver, Transporting Car etc. became more and more complicate and roboticized nowadays. They have more and more types and forms of faults in themselves,and have latent reasons of the faults, which brings a lot of difficult problems for the fault diagnostics to solve. The traditional method of fault diagnosis is unable to diagnose the trouble in heavy-duty equipment accurately and real-timely. It is very essential to seek a more perfecting diagnosis method for the diagnostic.By theoretical researching and practising, the thesis studied deeply the main research contents and key technologies of the intelligent fault diagnosis system based on neural network which is a newly developed technology nowadays. Then we set up a experimental platform of long-range fault diagnosis center combining the computer network technology . We studied the hydraulic pressure system in the pave as our diagnosis object and developed a new method of reasoning mechanism based on a neural network with three layers. At last ,we construct a intelligent diagnosis system with two layers by the technology of neural network, expert system and fuzzy theory. Meanwhile, we studied the technology of getting the diagnosis system inbeded in Web server,and succeeded in putting up a long-range intelligent diagnosis center based on Internet.Experiment indicate that the long-range intelligent fault diagnosis system can succeed in diagnosing long-rangly the trouble of the on-the-spot equipments,and meet the required accuracy and real-time character in the diagnosis of large-scale mechanical equipment.