Evidence Theory And Artificial Immune Integration Method In The Application Of Rotating Machinery Fault Diagnosis

Large rotating machinery equipments occupy more and more important status in modern industry, they usually play an important role in one plant and whether their status is good or not is directly related to whether the whole factory can safe and normal operation.Whether the rotation in the operation of the machinery and equipment in the mechanical failure to make timely and accurate judgment or not is the key to solve this problem. Thanks to modern machinery and equipment is more and more large and complicated, the chance of compound fault has been increased. Although composite failure are two failure occur at the same time, it presented the fault characteristics is not a simple superposition of two kinds of single fault characteristics, and it show the new fault features of single fault that is no. In order to solve this problem, many experts and scholars at home and abroad have put forward various fault diagnosis methods, such as, based on artificial neural network,based on wavelet analysis and the method of based on fuzzy logic, etc. And this paper is a kind of evidence theory with the combination of artificial immune integrated diagnosis method.Artificial immune is a fault diagnosis method based on bionics, mainly inspired by biological immune system, which provides a new train of thought and way to the mechanical fault diagnosis. Biological immune system has many excellent properties, such as automatic identification self-nonself, the memory function of antibody, the feature more than one corresponding of antibody and antigen and so on. Among them, the biological immune system for fault diagnosis is one of the most important revelation that it has the ability of recognition "self-nonself. This point are very similar with fault diagnosis. They are same to the operation of a state of total body whether to make a good judgment, just one for living organisms, another for inanimate machinery. In this paper, the immune detectors are generated based on the dimensionless indicators, but the usual dimensionless index number is too little, cannot well cover all fault space, especially for composite fault existing obvious flaws, so in this article cited by genetic programming to construct a new dimensionless indicators to make up for the shortage.In order to improve the accuracy of fault diagnosis, in this paper, using the theory of evidence fuses the dimensionless index diagnosis result on decision level, and regarding the final fusion result as the judgment of fault type. Verified by many times on the experimental unit, evidence theory and the artificial immune integrated diagnosis method of combining is effective.