Research On Fault Diagnosis Of Servo Stamping Motor Bearing

Servo stamping equipment has the advantages of high efficiency,high flexibility,etc.,and is widely used in the stamping workshop production lines of major auto companies.The servo stamping motor provides power for the entire stamping production line.As its key component,the rolling bearing is of great significance for health monitoring.On the one hand,the running status of the servo stamping motor rolling bearing can be understood in advance to prevent the production line from being stopped due to the failure of the rolling bearing.The occurrence of malignant incidents such as casualties;on the other hand,the obtained monitoring data has great reference value and guides equipment management personnel to regularly perform equipment operation and maintenance work.Therefore,this paper takes the rolling bearing of the servo stamping scrap motor as a key component for research,and makes an in-depth analysis of the related fault extraction technology and fault type identification technology.The specific work contents are as follows:Firstly,the structural composition of the rolling bearing of the servo stamping motor is studied,the mechanism,cause and basic form of the fault are analyzed in detail,and the natural frequency and the formula of the fault characteristic frequency are calculated from the geometric parameters of the rolling bearing.Secondly,considering that in the working process of the servo stamping equipment,due to its own or surrounding vibration,impact and other factors,it is inevitable that a large background noise will be generated.Using an improved SVD strong denoising method to preprocess the collected signals can separate the early weak fault information from the strong noise.Comparison experiments with traditional methods show that the improved method can not only increase the signal-to-noise ratio,avoid the loss of fault information through noise reduction,but also reduce the excessive dependence on the experience of researchers.The feature extraction of the pre-processed signal is carried out.In this paper,the ensemble empirical mode decomposition(EEMD)is combined with sample entropy to remove the eigenmode component(IMF)with low correlation,so that the fault characteristics are more obvious.Then calculate the sample entropy of the IMF and use it as the characteristic value.Take the different fault type data of the same working condition and the same fault type data of different working conditions for experimental analysis.The results obtained fully demonstrate the feasibility of combining the EEMD method with the sample entropy.Finally,the faults of the rolling bearing of the servo stamping scrap motor are classified,and the obtained sample entropy is used as the feature vector,and the support vector machine(SVM)model is used to identify it.In order to get better results,this article uses artificial fish swarm algorithm(AFSA)to optimize the relevant parameters of SVM to obtain the AFSA-SVM model,and improve the model by genetic algorithm(GA),get the IAFSA-SVM model.According to the experimental analysis,the fault recognition accuracy rate obtained by the AFSA-SVM model is 91.67%,and the fault recognition accuracy rate obtained by the IAFSA-SVM model is 95%.It can be seen that the classification effect of the IAFSA-SVM model is better,and the convergence speed is faster.