Fault Diagnosis Technology Of Motorized Spindle Based On MEWT-KPCA

Motorized spindle is key functional component of CNC machine tools and the condition of motorized spindle decides the development of high-speed CNC machine tools,and even manufacturing.Research on fault diagnosis method of motorized spindle has important practical application value for potential fault mining,rapid fault location and maintenance.Meanwhile,it also provides technical support for reliability improvement,fault early warning of motorized spindle.At present,most of traditional fault diagnosis method is mainly in view of the mechanical single fault,such as bearing fault,rotor imbalance and so on,without considering the coupling between various faults of composite fault types and motorized spindle electrical fault.Failure to consider the coupling of multiple faults and the electrical fault.In this paper,a systematic study on the fault diagnosis technology of motorized spindle is carried out.On the basis of deeply understanding of motorized spindle structure,working principle,from two aspects of the traditional mechanical spindle and the spindle motor to analyze common faults and formation mechanism,excavate potential fault reasons.A fault diagnosis technology based on MEWT-KPCA is proposed,and a set of fault diagnosis system is built.From the two dimensions of fault recognition accuracy rate and efficiency(diagnosis time),the fault diagnosis method proposed in this paper is compared with the traditional fault diagnosis methods to verify the effectiveness and high efficiency.The main research results are as follows:1.The structure and working principle of the motorized spindle are introduced.It is concluded that the motorized spindle is a complex electromechanical coupling system.Through statistical analysis,the common fault location of the motorized spindle is obtained.Mechanical faults mainly occur in the bearing parts,followed by housing,rotating parts and other positions;the electrical fault occurs mainly in the stator part.Then the common failure modes and the fault formation mechanism of motorized spindle are analyzed,the potential causes are excavated.It is found that the vibration signal is more sensitive to mechanical faults.The three-phase current signal is more sensitive to electrical faults,and energy,kurtosis and root mean square value are selected to characterize the signal’s characteristic information2.The fault diagnosis technology of motorized spindle based on MEWT-KPCA is proposed,and the specific flow and operation of fault diagnosis are determined.Penalty wavelet threshold de-noising,modified empirical wavelet transform(MEWT),kernel principal component analysis(KPCA)and support vector machine optimized by genetic algorithm(GA-SVM)are discussed in detail.3.The effectiveness and efficiency of the fault diagnosis algorithm are verified by experiments.Based on the analysis of common failure modes and mechanism,mechanical faults and electrical faults experiments are designed,including single faults and compound faults.The experimental data are used to verify the effectiveness of the fault diagnosis algorithm proposed in this paper.At the same time,the fault diagnosis algorithm proposed in this paper and the traditional fault diagnosis algorithm,such as empirical mode decomposition(EMD),ensemble empirical mode decomposition,local average decomposition algorithm(LMD)and other algorithms are compared in terms of fault recognition accuracy and efficiency.The results show that the proposed algorithm performs well in both fault recognition accuracy and efficiency.4.A set of fault diagnosis system with good man-machine interface is developed based on LabVIEW and Matlab.The system can realize many functions,such as screening and preprocessing of historical data,feature extraction,dimension reduction,fault diagnosis and identification of motorized spindle.The feasibility of this system is verified by the experimental data.