Health Assessment Of Motorized Spindle Based On Multi-sensor Data Fusion

The motorized spindle is one of the key functional components of CNC machine tools.Its performance and health status are closely related to the overall performance of CNC machine tools.The failure of the motorized spindle may cause a series of problems such as the shutdown of the CNC machine,which affects product quality and production efficiency.Through the health assessment and preventive maintenance of the motorized spindle,the failure can be avoided and the service life of the machine can be extended.At the same time,the health state perception,fault prediction and health maintenance technology of the motorized spindle are also an important content of the intelligent development of CNC machine tools.Due to the variety of motorized spindles,complex operating conditions and failure modes,there is currently no standard parameter threshold to assess their health status.Analysis and the health degradation evaluation of the motorized spindle is of practical significance for the maintenance strategy and the realization of the condition based maintenance.In this paper,the health assessment method of the motorized spindle is proposed under the CBM framework.The main contents are as follows:1.Fault transmission path and signal feature analysis of motorized spindle.The sensor detection points are arranged according to the coupling relationship between the spindle components,and the rationality of the arrangement for detection points is verified by the analysis of the fault transmission path.Besides,the characteristic frequencies generated by common faults are analyzed as a reference for health status assessment.2.Calculation of health indicators based on single sensor data.By analyzing the typical health assessment framework and combining the characteristics of the motorized spindle,the health assessment framework for the motorized spindle is proposed.Features of the signal in time domain,frequency domain and time-frequency domain are extracted,screened and analyzed.In view of the fact that the spindle may be in the health state for a long time,a health index calculation method is proposed based on the sliding SVDD model to evaluate health status of motorized spindle.The method was verified by bearing life cycle data.3.Health assessment of motorized spindle based on multi-sensor data fusion.In this paper,the multi-sensor fusion method on decision-making layer is chosen.The sensitivity of each sensor to health status of the motorized spindle is calculated.The weighted DS evidence theory is used to fuse the sensor data to resolve the conflict between evidences and eliminate the interference of insensitive sensors.The proposed algorithm was verified by imbalance data of the motorized spindle.4.Development of software for health assessment.According to the health evaluation method of motorized spindle,the software platform is designed and developed.By combining advantages in data acquisition and human-computer interaction of Lab VIEW,data analysis of Matlab and data management of SQL Server,data collection,condition monitoring,feature extraction,health assessment,and data management are realized to achieve practical application of the proposed algorithm.The health assessment method of the motorized spindle proposed in this paper effectively solves the accuracy of multi-sensor data fusion and the problem of insufficient data in the life cycle.The software and hardware platform for health monitoring and evaluation developed in this paper provides a solid foundation for the data accumulation and quantitative analysis of motorized spindle test.