Research On The Influence Of Sampling Time Length On Damping Identification

Damping is an important parameter to characterize the energy dissipation in the vibration system.The accurate identification of damping value has direct impact on the application of damping in the field of structural damage diagnosis,vibration reduction and noise reduction.Improving the damping identification method has been an important topic of concern to scholars.Among all kinds of damping identification methods,the research on the sampling time length is not enough.In this paper,the effect of sampling time length on damping identification is studied.In this paper,the influence of the sampling time length on the accuracy of damping recognition was analyzed without noise interference.Taking the half power bandwidth method and the inner product method as an example,the simulation of 500?5000 and 40000 sampling points with the damping ratio of 0.006?0.03 and 0.1 was given.The simulation results show that the two damping identification methods have higher sampling time and higher accuracy of damping recognition.The noise disturbance is unavoidable in practical engineering,while the effect of the sampling time length on the accuracy of the damping was discussed in this condition.The relationship between the spectral amplitude of response signal and the number of sampling points was deduced respectively,so as to the spectral amplitude of noise signal.Theoretical analysis indicated that the sampling time should not be too long,otherwise the response signal would be masked by the noise signal and damping recognition error would be increased.In this paper,the critical sampling length was defined as the equation of response signal spectrum and noise signal spectrum.The critical sampling length expression was obtained by theoretical derivation,which provided a reference for selecting the reasonable sampling length.As the long sampling time would bring the error damping identification problem,the research of simultaneous equations method with short data was used to identify the signal damping.The signal damping ratio was obtained by constructing the system of equations with different Fourier transform basis functions.For the integration time of the system,the arbitrary integration time and specific integration time T(28)kp w(k is positive integer)were discussed.It was pointed out that the selection of specific integration time could reduce the negative frequency interference and improve the accuracy of damping recognition.The simultaneous equation method had the advantages of low computational complexity,fast recognition speed and strong anti-noise ability.The applicability of simultaneous equations to MDOF response signal damping recognition was studied.For the non-dense mode signal,the identified order modes were subtracted from the original signal in turn,and the MDOF response signal damping was simplified to multiple SDOF response signal damping identification.For the dense mode signal,an iterative algorithm was introduced to eliminate the mutual interference of each order mode and identify the order modal damping.The simultaneous equation method was applied on the gear box cover damping identification experiment.It was found that the response signal of the gear box cover had four order modes at 208.824 Hz,363.913 Hz,636.561 Hz and 940.817 Hz while the fourth order modal damping ratios were 0.004798,0.004509,0.001933 and 0.001674 respectively.The four-order modal signal was subtracted from the original signal,and the peak value was not observed in the frequency domain of the residual signal.The practical application of this method was verified.