Research On Fault Diagnosis Of The Coal Pulverizer Based On The Information Fusion

The coal pulverizer is the core equipment used in the coal-pulverizing system of power plants, is also one of the key factors leading to the unexpected shutdown of power plant units, the running condition of the coal pulverizer directly influence the stability of boiler combustion. Fault diagnosis can effectively diagnosis and predictive failure, reduce the failure occurrence and the maintenance cost in certain degree, optimize the operating conditions of the power equipments, thereby enhance the units operation security, stability and economy.One-dimensional information has uncertainties, multi-dimensional infonnation processing approach of the information fusion can solve the uncertainties problem, would give a more accurate description of the operating condition of the coal pulverizer. Against the ambiguity of the coal pulverizer’s faults, fuzzy cluster analysis accord with the essence of the problem. But the classifying result is common relationship, is difficult to predictive failure. The fault diagnosis method based on fuzzy clustering analysis and D-S evidence theory is proposed. The basic probability assignment function of the D-S evidence theory is constructed by the equivalent matrix of the fuzzy cluster analysis that not only cares about the ambiguity of the mill failure, but also solves the subjectivity problem of the basic probability assignment function construction. The feasibility of this method is proved by the actual operating data of the MPS type medium-speed coal pulverizer.