Research On Dynamic Characteristic Detection System And Method Of Alternating Impact Machinery

On the basis of discussing and summarizing the current situation,trend and shortcomings of the dynamic characteristics testing methods for alternating impact machinery at home and abroad,this paper designs and implements a dynamic characteristics testing system for alternating impact machinery which is suitable for indoor test and working condition test,and verifies the feasibility of the testing system,so as to make technical preparations for the structural improvement of alternating impact machinery.In this paper,the impact velocity and displacement can be obtained by denoising and integrating the acceleration signal of the impact piston collected by the acceleration detection device,and then the impact parameters can be calculated by using these motion parameters.Firstly,according to the structure and working characteristics of alternating impact machinery and the requirements of impact acceleration detection,the hardware design of acceleration detection device is completed.Based on embedded real-time operating system and file management system,the main function software of acceleration detection device,such as data acquisition control,data storage,data processing and transmission,is realized.Then the upper computer software is developed with LabVIEW.The upper computer software realizes the online parameter setting of the signal detection device,and integrates the denoising algorithm and integration algorithm in this paper,so as to process and analyze the acceleration signal offline.In order to remove impulse,spike and circuit noise in acceleration signal,a wavelet modulus maximum denoising algorithm based on adaptive threshold is proposed.Firstly,the wavelet modulus maxima sequence of signals is preprocessed by combining the SURE threshold estimation and soft threshold shrinkage function,and then the new modulus maxima sequence is obtained by modulus maxima search.Then the wavelet coefficients are reconstructed by the piecewise cubic Hermite interpolation algorithm.Finally,the denoised signal is obtained by inverse transform of wavelet decomposition coefficients.The numerical simulation and experimental results show that the improved algorithm can not only effectively remove the noise in the acceleration signal of impact piston output by micro-electromechanical sensor,but also improve the denoising effect and operational efficiency,and retain the details of the signal more completely.The direct component and low frequency random drift in acceleration signal will cause acceleration integral error.In this paper,the integration error caused by direct component and low frequency random drift is eliminated by time domain high-pass filtering,and then the trend error caused by initial velocity or initial displacement is corrected by least squares fitting polynomial method.Finally,the high precision velocity and displacement are obtained by integration.Not only the influence of sampling frequency and sampling data length on the accuracy of numerical integration in time domain is analyzed by numerical simulation,but also the AMESim numerical simulation model of alternating impact machinery is built to get the acceleration of impact piston,and the velocity and displacement of acceleration integration are obtained to verify the effectiveness of the integral algorithm in this paper.The numerical simulation and experimental results show that the time domain integration method based on high-pass filter not only has the advantage of simplicity of calculation by using polynomial fitting directly to eliminate trend term,but also can eliminate the influence of low frequency drift,and the final integration result error is small.Finally,the validity and feasibility of the proposed dynamic performance testing system for alternating impact machinery are verified by experiments.