Study Of Fault Diagnosis Based On T-S Fuzzy Model And Its Application In Heat Treatment Furnace

With the rapid development of industrial technology, security and reliability requirements of the production equipment are more important than ever in the manufacturing process of modern industry. Nevertheless, there are several widespread problems existing in industrial production systems, for instance, nonlinear, strong coupling, time-varying and parameter uncertainty. Higher demands are being placed on the system to improve the technology of control and fault diagnosis.Taking the tunnel bluing furnace for heat treatment as the research object, this thesis carried out and completed the following work:1) A T-S fuzzy modeling method was proposed for a sort of nonlinear complex systems,which is based upon augmented input variables. By augmenting the input variables, then back-stepping front part structure with the consequent parameters, the input variables of the model can be efficiently determined to improve the modeling accuracy.2) On the basis of T-S fuzzy model, the reference residual sequence of every fault in the system was obtained. Taking it as the cluster center, fuzzy cluster analysis was applied for fault diagnosis quickly and accurately, thus provided reliable basis for fault early warning and exclusion. In the process of cluster analysis, the fuzzy clustering method was improved by weighting the residual of fault sample with sample standard deviation. Simulation results showed higher precision of the improved method.3) The T-S fuzzy modeling and fault diagnosis method was applied to the tunnel bluing furnace for heat treatment. The supervisory control and data acquisition system was configured through KingView6.5, which realized data acquisition, data storage, configuration monitoring,alarming, report query and other functions, improved the information level of tunnel bluing system. The modeling and fault diagnosis method was tested in the heat-treatment system, and showed receivable results.