Vibration Research On The Label Sewing Machine Based On Operational Transfer Path Analysis

With the rapid development of industry, higher requirements are put forward to the packaging machinery. Currently most label sewing machines have been found poor performance in vibration control.In order to effectively enhance the vibration control performance of the label sewing machine, vibration mechanism of a label sewing machine were studied in this paper,by testing and analyzing the vibration signals with several method like OPA(Operational Transfer Path Analysis),modal analysis and some simulation ways.First of all, the mechanical system and vibration characteristics were preliminarily studied. The mechanical system were introduced and some potential excitation ways were presented. To find out how the framework influences vibration of label sewing machine,ANSYS Workbench was used to solve the natural frequencies and mode shapes of the framework of label sewing machine.An experimental modal analysis model was also set up based on Pulse vibration analysis system,and modal parameter of each order was obtained.After comparing the natural frequencies and mode shapes analyzed in numerical and experimental way, the reliability of the finite element model was verified,and it was also found that the first natural frequency of the framework was 121 Hz,which is far higher than the excitation frequency, the framework is less likely to cause sympathetic vibration. Vibration signals were tested on the machine. Spectral analysis of the signals indicated periodic excitation exist in the mechanical system,while vibration intensity analysis found that vibration of A2(X direction) and A21(Z direction) were typical and need further study.OPA theory and its shortcomings were introduced in chapter 3, and existing improved method was also illuminated.2 aspects of improvement were made to fit serious coupled input signal of the machine: multiple coherence analysis based on partial coherence theory was introduced to prevent omission situation of transfer paths, and frequency response function matrix was used to solve the cross couple problem of input signal.OPA models OPA(A), OPA(B) was established, respectively, using the signal from A2( X direction) and A21(Z direction) as system output. After verifying the modals with improved multinational coherence analysis, six-input models,which consists X and Z direction of path 1,2,3,was found a proper modal to accurately describe the real mechanical system. On the other hand, after solving the cross couple problem of input signal, vibration output of target point was calculated with OTPA modal and was compared with measured output, which indicated that relative error of calculated result was significantly reduced.Modal OPA(A), OPA(B) were analyzed, indicating that path 1X contributes most for the vibration of X direction of A2,which is mainly excited by sewing structure. While path 2Z contributes most for the vibration of Z direction of A21,which is mainly excited by the electromotor.Dynamic property of sewing structure was then analyzed by simulation, and the structure was improved according to the simulation results, by replace a lighter material and optimizing the cam contour.The innovation of this paper is: improved the OPA to fit the application of label sewing machine by using a improved multiple coherence analysis based on partial coherence theory and frequency response function matrix to solve the cross couple problem of input signal. Combined the signal testing and some analyzing method like OPA, finite element simulation and so on, to study the vibration mechanism of label sewing machine,which is more comprehensive and accurate and more suitable for practical application.