Font Size: a A A

Study On The Method And Experimental Of Spindle Rotation Trajectory Extraction

Posted on:2022-05-18Degree:DoctorType:Dissertation
Country:ChinaCandidate:R HuangFull Text:PDF
GTID:1481306311967309Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
With the development of high-precision manufacturing technology,vibration control,precision control and specific axis trajectory control on the spindle of the machines have become important research directions,and in these active control processes,it is of great significance to the high-precision,real-time and on-line detection of spindle axis trajectory.However,in the actual test,the axis of the spindle cannot be measured directly,and the axis trajectory needs to be obtained indirectly by measuring the outer contours of the spindle.Inevitably,the roundness error of the spindle is mixed into the measurement signal,which requires the separation of the roundness error and the axis trajectory in the mixed signal in order to realize the precision control of the spindle.The main task of this paper is to study the precise measurement and separation method of spindle roundness error and axis trajectory,develop the hardware and software of the real-time online monitoring system of axis trajectory,and to lay the foundation for the active precision control of the spindle of the machine.The main work carried out is as follows,(1)The error separation technology based on the three sensors is systematically introduced,the three-point method in the frequency domain is studied,the general equation of the discrete three-point method is deduced,and the general equation of the axis trajectory of the spindle is further given under the roundness error of the spindle.By analyzing the harmonic suppression problem existing in the general three-point error separation technology(G3P method),puts forward the concept of rounding error in practical application and the resulting error separation accuracy,then a new error separation technique based on differential algorithm(D3P method)is proposed,the three-point method based on the first-order backward differential algorithm(1-D3P method)and the three-point method based on the second-order backward differential algorithm(?-D3P method)are derived.(2)In order to evaluate the error separation accuracy of each three-point method,two expressions,RRMSE and RPPE,are defined to describe the overall difference and maximum difference between the estimated value and the actual value,respectively.The separation effect of G3P method and D3P method on spindle roundness error and axis trajectory is compared.For different sampling points,the roundness error result and axis trajectory result separated by D3P method are at least 66 times and 40 times as high as the G3P method,respectively.Obviously,this is a very significant separation effect improvement.The effect of sensor installation on the error separation accuracy of G3P method,?-D3P method and ?-D3P method is further studied,and the results show that the D3P method has higher robustness to sensor installation position,and the sensor distribution position should be avoided is summarized(3)A spindle axis trajectory measurement and its active control experimental platform have been set up,and a hardware and software system for on-line measurement and error separation of axis trajectories has been developed.First of all,based on VC2010,TDM application interface and MATLAB engine,the TDM data file bulk conversion software is developed,which provides convenience for the previous error separation experiment.Then,based on ARM,AD7606,VC2010,FFTW and so on,the on-line monitoring and control hardware and software system of axis trajectory is developed,which can realize the control of axis trajectory and real-time online extraction and analysis.(4)The experimental study on spindle axis trajectory and error separation was carried out.The sensor installation position is identified by the interrelation theory.The sampling signal is filtered by moving average filter,and the error separation effect before and after filtering is compared,the results show that the error separation result after filtering is effective.Under different sensor installation angles and different sampling points,error separation experiments were carried out by using G3P method,I-D3P method and ?-D3P method,and the roundness error and axis trajectory of the spindles under the above operating conditions were obtained.The experimental results was evaluated,and the deviation formula dv between the two sequences is defined to evaluate the difference between the results of the three methods,the experimental results prove the effectiveness of the D3P method.Compared with the G3P method,with the decrease of sampling points,the error separation advantage of D3P method is more obvious,which is of practical significance for extracting the roundness error and axis trajectory of the spindle at high speed.(5)The axis trajectory extraction experiment under the active control of spindle was carried out.The thin-film restrictor of the hydrostatic bearing is controlled by a piezoelectric ceramic driver to change the oil cavity pressure of the hydrostatic bearing,thereby pushing the spindle to change the axis position,active control of spindle axis trajectories is achieved.The extraction and error separation of the active control axis trajectory at different speeds,different control frequencies and X,Y one-way control and two-way linkage control are studied.The roundness error separated by the above-mentioned operating conditions is basically the same,the characteristics of the control frequency in the spectrum of the separated axis trajectory are prominent,and the cross-coupling features increase with the speed,and become more and more obvious,which confirms the correctness of the detection system and the separation method.Active control of axis trajectory of spindle provides correct instantaneous axis position(amplitude and phase)information for active control and is of great application value.(6)Finally,the accuracy of the proposed new method and the test system developed is evaluated comprehensively.The five error sources in the error separation experiment are analyzed:sensor measurement error,sensor installation position error,sensor bracket horizontal error,sensor bracket vertical error and temperature error.The uncertainty evaluation of G3P method,?-D3P method and ?-D3P method based on Monte Carlo method is carried out.The amplitude and phase uncertainty of the harmonic components of the axis trajectory are analyzed,and the results show that using the proposed new method,the axis trajectory amplitude uncertainty was reduced by 1 times,phase uncertainty was reduced by nearly 10 times,the accuracy of error separation was improved significantly.
Keywords/Search Tags:Machine spindle, Active control, Rotation error, Axis trajectory, Error separation, Differential algorithm
PDF Full Text Request
Related items