Font Size: a A A

Failure Prognostic Method And System For Motorized Spindle Based On Fusion Of Multi-Signal

Posted on:2018-08-22Degree:MasterType:Thesis
Country:ChinaCandidate:G F LiuFull Text:PDF
GTID:2311330515476009Subject:Mechanical Manufacturing and Automation
Abstract/Summary:
As one of the key parts of high-grade CNC machine tools,motorized spindle directly affects the reliability of whole machine,and impacts the operating rate and availability of production line.So it is of great theoretical and applicable value to study failure prognostic method and failure prognostic system which is beneficial to improve the reliability of motorized spindle and machine tool,and innovate maintenance strategy,and implement the intelligentize of the key components and the whole machine.Traditional method of condition monitoring and failure prognostic for single part usually extracts the characteristic quantity of the single signal to predict future degradation rule and residual life of part,but for complex hydraulic and electromechanical coupling system-motorized spindle,single signal is not comprehensive and accurate enough to reflect the condition of the fault,and the research work on failure prognostic method and system of motorized spindle based on fusion of multi-signal has higher theory and use value.Combining with the demand of current motorized spindle industry and hot spot and difficult problem of research,in the understanding of structure,working principle of motorized spindle and failure forms,the cause and the main and secondary characteristic signal of common fault,this paper proposed a complete set of failure prognostic method for motorized spindle based on fusion of multi-signal using the integration,which made the failure prognostic algorithm more accurate and comprehensive.Using Lab VIEW and MATLAB mixed programming method to form failure prognostic software system,and combining with related hardware system this paper formed a complete set of motorized spindle system.Three experiments have been carried out to verify the validity of the failure prognostic method and system.The main content of this paper are as follows:1.Fault analysis of motorized spindle.Through the analysis of the structure and working principle of motorized spindle and the relationship between each subsystem,illustrated that the motorized spindle is a complex hydraulic and electromechanical coupling system.Through detailed analysis of failure form,failure reason and primary and secondary characteristic signal of common faults of motorized spindle mechanical system and electrical system,confirmed the vibration and noise signals can reflect most of faults of the motorized spindle.2.Research on method for failure prognostic of Motorized spindle.Based on the study of specific principles,advantages and disadvantages of feature extraction,signal analysis,trend prediction and residual life assessment algorithm,put forward prognostic method of motorized spindle based on fusion of multi-signal,this method used the wavelet packet decomposition and reconstruction(WPD)to achieve noise reduction and the empirical mode decomposition(EMD)to extract the fault characteristic,and put forward fusion of multi-signal method based on the self-organizing map(SOM)and kernel principal component analysis(KPCA),and Gaussian process regression(GPR)was applied to implement health trend prediction and residual life evaluation to ensure the accuracy and integrity of prognostic method of motorized spindle.3.Designing an overall scheme for the failure prognostic system of motorized spindle.This paper presented a scheme for failure prognostic system,including hardware system and software system solutions.In the aspect of hardware system,this paper briefly illustrated the parameters,working principle and advantages and disadvantages of the loading system,PXI high-speed synchronous acquisition system and sensors.In the aspect of software system,this paper introduced Lab VIEW,MATLAB and its toolkits,and finally determined using Lab VIEW and MATLAB mixed programming method to build failure prognostic system.4.Development of failure prognostic system of motorized spindle and its experimental analysis.A system for failure prognostic of motorized spindle was developed,and its structures,background program block diagram and operations were given,and according to the module its working principle,structure,function and program code were introduced in detail.Meanwhile,three typical fault experiments including motorized spindle rotor unbalance,rotor scratches and stator interturn short circuit were used to verify the effectiveness of the failure prognostic method and system proposed.In the study of this article,a failure prognostic method and system of motorized spindle which is overall and practical is developed,providing the technical support for improving maintenance strategy,realizing the intelligent failure prognostic and fault diagnosis and having higher theoretical and engineering application value.
Keywords/Search Tags:Motorized spindle, Failure prognostic, Fusion of Multi-Signal, LabVIEW
Related items