Condenser fault diagnosis base on grey multiple attribute fusion

Abstract

Grey multiple attribute fusion method is put forward in this paper and applied to condenser fault recognition in which grey relation analysis is combined with multiple attribute decision making. First the state parameters of condenser are fuzzed and use traditional relation analysis method to calculate the relation coefficients between pattern samples and the samples to be diagnosed. The classical domain and joint domain of the fault types are determined by extension interval. The weight value is calculated by proportion coefficient method to be introduced into traditional grey relation index calculation. The best relation index is obtained by optimizing resolution ratio to ensure the stability of relation index space. It is verified that the method provided in this paper can improve the accuracy and reliability of condenser fault recognition by the simulation examples.