Research On Fuzzy Recognition Method Of Power Station Boiler Four-tube Leakage

The unplanned shutdown of the equipment caused by four-tube leakage is the main accident of power station boiler. Four-tube leakage takes a great proportion in the boiler accidents.It has been a serious threat on the safe and stable operation of the unit. Accurate and timely diagnosis on four-tube leakage is of great significance to the security and economy operation of power units. Fuzzy pattern recognition method is an important method of the fault intelligent diagnosis, and it is used widely in diagnosising such faults as leaks and jams.Fuzzy pattern recognition method based on the basic principles of fuzzy mathematics, and it mainly includes two contents:the establishment of fault knowledge library and fuzzy recognition to the fault.The establishment of the fault knowledge library involves the determination of fault set, fault symptoms set and the description of the fault symptoms. The description of the fault symptoms is most important. According to the different types of fault symptoms, the description of the fault symptoms has a variety of functions and description method. Author put forward new nonlinear functions to describe fault symptoms and new method to treatment fault symptoms comprehensively.Fault fuzzy identification generally uses the way of combining the maximum subjection principle and threshold principle. This way excludes the wrong fault symptoms by giving a threshold,then establishes membership function to recognize the right fault symptoms. According to characteristic of boiler four-tube leakage, author puts forward the method of fault fuzzy identification based on number axis and polar coordinates.This method gives the running state of boiler visual show in the number axis and polar coordinates, which is able to diagnose fault accurately and avail to reveal the contact between the various faults. On this basis, author puts forward the method of diagnosing slight leakage of four tube in boiler and leakage quantity.This topic transforms each process of the fault diagnosis into algorithm and function module based on Integrated Modular Modeling Software(IMMS). Then the boiler fault diagnosis system model is established and respectively put into models of power station boiler which burns coal powder and garbage to complete fault diagnosis. The fault diagnosis results compare precision and advantage of different method of fault diagnosis. The fault diagnosis results show that the method put forward by author is important and valuable for four-tube leakage diagnosis.