The Development Of Intelligent Monitoring Expert System For Boiler Based On Knowledge

With the development of modernized wholesale manufacture and the progress of science and technology, the widespread application of boiler in domestic day by day, and the degree of technology difficulty are increasing, the structures of boiler system are getting more and more complex, and the functions are more and more perfect while the automatic degree is higher and higher. The press of the require for Fault Diagnosis (FD) technology application are increasing with days. It can rise the occupied coefficient, reduce the cost of repairing and improve benefits if guarantee the boiler equipment work safely. So it has great meanings to research the process monitor and FD for boiler.The FD technology for boiler based on Expert System (ES) is an important branch in the research of FD. According to the need situation and the research status of the FD for boiler system in domestic and abroad, construct a structure of the FD ES for boiler system based on Fuzzy Neural Network (FNN), this thesis research the study algorithm in detail and analysis its limitation deeply. And emphasis on the improvement of standard BP network, discuss the importance of combination fuzzy logic with Neural Network (NN), and the combine way. It extracted the diagnosis method with ES based on FNN, and realized the integration of the acquisition, representation, and reasoning of knowledge. Meantime, aid it with the methods of add the momentum and self-adapting regulation the learning rate. The simulation indicated that it improved the performance of network diagnosis with large scale, and solute the problem of slow convergence was, and inclined to trap in local minimum.Embedded the FNN in the ES, and developed the FD ES software for boiler system. It get better result of credence than the traditional ES. Also it has more goodness in something like how to get the diagnosis knowledge, imperfection information reasoning, and stabilization was rise in certain degree.