Study On Fault Immune Diagnosis Technology Based On Probability Density Function

Large-scale mechanical equipment of industrial units is significant in modern industry.The operation condition for industrial units is directly bearing on the safety of the production safety and stable operation.Therefore,the long-term stable and safe operation of industrial unit equipment under high load is always the goal of equipment managers.Equipment maintenance personnel hope to know the real-time operation status of the system,timely warn of early failure of equipment,adjust equipment load,reasonably arrange time for shutdown and maintenance,improve the reliability of the system,and avoid economic losses and casualties caused by sudden major safety accidents.Therefore,it is of great significance to carry out fault diagnosis for mechanical equipment of industrial units.More than half of the rotating mechanical equipment faults of large industrial units are caused by the non-designed external vibration caused by the operation of the equipment.Therefore,monitoring the device with vibration index is the most widely used method among the many feasible methods at present,and also the method with the highest reliability.On the basis of studying a large number of literatures and combining the relevant research background of the subject,this paper proposes an artificial immune fault diagnosis method based on the stable Alpha distribution model.The specific work is as follows:Firstly,a binary coding method based on multiple parameters and an improved weighted piecewise immune matching algorithm suitable for the artificial immune algorithm model were proposed,which solved the problem that the multi-parameter model could not reasonably consider the independence between parameters and the internal relation between parameters when using the immune algorithm to match antigens and antibodies.Secondly,an immune fault diagnosis method based on Alpha stable distribution was proposed.Alpha stable distribution can fit both non-gaussian fault signals and gaussian normal signals,so it can build a precise model of the vibration signal,and the fitting parameters can be used for single fault classification.Therefore,using Alpha stable distribution parameter fitting to modeling normal signal and fault signal.Combined with the artificial immune algorithm model,to diagnose the fault.Finally,six different working states were tested on the combined rotary machinery fault diagnosis experiment device in the large rotary machinery fault diagnosis experiment platform.Experiments show the effectiveness of this method for single fault diagnosis.At the same time,the experimental results show that this method has good performance of fault detection.In summary,this paper combines the Alpha stable distribution model with the artificial immune model,and designs an immune fault diagnosis method based on the Alpha stable distribution according to the properties of the probability density function,which is a supplement to the research on the probability density function fitting and artificial immune fault diagnosis technology,and also a new exploration of the industrial unit equipment fault immune diagnosis technology.