Machine Unit Intelligent Fault Diagnosis System Based On Multi-dimensionless Immune Detectors

Due to the industrial unit has been developed in the direction of large-scale, high speed and complexity, fault probability is also constantly increasing. This paper takes the industrial unit as the studied object, and the main purpose of the study is to use artificial immune method combining with other intelligence methods to improve the accuracy ratio of fault diagnosis. The method using the negative selection algorithm of artificial immune system and dimensionless parameter to diagnosis fault is proposed. Classification feature of multiple industrial unit faults based on genetic programming is optimized firstly. Experimental result shows that the classification ability of optimum dimensionless parameter is better than that of existing ones. Then by means of the negative selection principle of artificial immune system, more types of dimensionless immune detectors were constructed. Finally the integrated diagnostic technology based on dimensionless Immune Multi-detectors provides a feasible solution to machine unit fault diagnosis is formed. The main contents of this dissertation are summarized as followings:Owing to the existing dimensionless parameter are only sensitive to some kinds of fault and the number of the existing dimensionless parameter is finite, so we need to construct some new dimensionless parameters which could overcome the deficiencies of the limited number and the diagnosis ability of conventional dimensionless parameters. Based on the research on the dimensionless parameter, a method based on the genetic programming and dimensionless parameter is proposed to construct the new dimensionless parameters. According to this method, conventional dimensionless parameters are combined and compound dimensionless parameters are formed, then fitness function is adopted to measure the performance of new generated parameters. Finally the best recognition ability of the new dimensionless parameters is obtained. Experimental result shows that the recognition ability of new dimensionless parameters is better than that of existing ones, and accurate classification of shaft fault can be achieved.Based on negative selection mechanics in immune system, dimensionless immune detector is build. As to the problem of useful fault information loss in the process of reduction and clustering, a simple, efficient and quick integrated diagnosis algorithm is presented in order to improve the accuracy of diagnosis by fusing several dimensionless immune detectors. Simulation result shows that it is effective in improving the precision of fault diagnosis.The study and simulation result shows that the intelligent fault diagnosis system in the paper based on multi-dimensionless parameters detector could satisfy the requirements of monitoring and diagnose for industry unit. The especially further research about the use of the artificial immune system to complex fault can not only make a solid foundation in this field, but also help us to discover some new research problems and possible solutions.