Research And Test Verification On Acceleration Method Of Random Vibration

In recent years,the rail industry and the aerospace industry have developed rapidly.During service,the random loads received by the equipment and structural parts have become more complicated,and their incentive expressions are also more diverse.The possibility of fatigue failure in the equipment and structural parts is increasing,too.Therefore,it is of great engineering significance to be able to accurately and efficiently predict the service life of equipment and the fatigue damage caused by non-Gaussian and Gaussian loads.This paper takes the vibration data of a certain type of airborne equipment mounting bracket during the flight as an example,and the acceleration test of non-Gaussian excitation and fatigue life prediction problems are explored from three aspects of data processing,test and simulation.Firstly,in order to distinguish the non-Gaussian characteristics of the input signal,the static process of the measured acceleration data needs to be judged,and the kurtosis,skewness and other parameters are calculated,so that the fatigue damage spectrum calculation methods are determined for each flight mission excitation.At the same time,the signal characteristics before and after the mining are compared,and the filtering mining can effectively avoid aliasing,but it will also change the damage characteristics and non-Gaussian.Secondly,in order to verify the accuracy of the multi-channel time-domain correlation damage editing method,the time-domain and non-Gaussian characteristics before and after the reduction were compared,and it was found that the kurtosis value remained basically unchanged,and the excitation signal was reduced from 2480 s to 2121 s,and the acceleration efficiency was 14.4%;Design reduced spectrum time-domain waveform reproduction bench test,while collecting strain response,comparing responses’ non-Gaussian characteristics and fatigue damage,and it was found that the relative error of response kurtosis and damage can be controlled at about 20%;The simulation fatigue life is calculated by dynamic transient response method and quasi-static superposition method,and compared with the test results,it was found that the damage value of the test is basically the same as the damage value of the transient response method,and the damage value calculated by the quasi-static superposition method is the largest.Then,the acceleration excitations are accelerated in frequency domain,and the singletask state fatigue damage spectrum calculation methods are compared.It was found that the rainflow counting method is suitable for non-Gaussian loads,but whose calculation speed is slower.For smooth Gaussian loads,the calculation speed of simplified Rayleigh method is the fastest which is suitable a large number of data points and the Dirlik method has the highest calculation accuracy.According to this conclusion,the accelerated synthetic test PSD spectrum is created,and the accelerated test is designed to perform the original life test and the accelerated life test respectively.It is found that the failure mechanism of the specimen will not change when the scale factor is selected 1.The single-axis harmonic response method is used to simulate and calculate the accelerated fatigue damage,and the acceleration ratio is defined.When the acceleration ratio is close to 1,the acceleration effect is the best.Finally,the results of test strain measuring point and the simulation measuring point unit damage and the acceleration ratio were basically consistent.Finally,the general method and flow of time-domain reduction and frequency-domain accelerated tests are summarized,and a detailed test flowchart is drawn.At the same time,a non-Gaussian accelerated test design interface is compiled.The design ideas and instructions are given from four aspects consist of the main interface of fatigue damage spectrum calculation,PSD calculation interface,FDS cycle coefficient calculation interface,and accelerated PSD synthesis interface.The interface greatly improves the efficiency of accelerated testing and simulation.