| Locating the source of abnormal vibration in machinery is a key technology in mechanical fault diagnosis and repair,which can be used to quickly locate the source of abnormal vibration after the sensor has collected the abnormal vibration signal and then carry out targeted repair,avoiding blind dismantling and wasting production time and increasing production costs.The positioning of the vibration source can be obtained through the time delay between sensors,but the current positioning of the abnormal vibration source is not accurate enough,the reason is that it is difficult to obtain the exact time delay of the abnormal vibration signal and the positioning algorithm is greatly affected by the noise,to address this problem,this paper takes the vibration source to excite the plate in the formation of elastic wave propagation in the plate as the background,mainly using HHT(Hilbert-Huang transform)on the plate The paper introduces a velocity-free positioning model,improves the least-squares method by using the chan algorithm and the spectral correction iteration method to improve the positioning accuracy,analyses the influence of the plate sensor layout on the positioning results,and determines a six-sensor array with convenient arrangement and better positioning effect,and finally conducts positioning experiments on the plate with this array.The main work of this paper is as follows:(1)Combined with the theory of elastic dynamics to obtain the experimental singleaxis acceleration sensor collected vibration signal of the steel plate mainly caused by Rayleigh waves,the sensor collected vibration signal using CEEMD(Complementary Set Total Empirical Mode Decomposition)and according to the frequency to select a suitable IMF(Intrinsic Mode Function),and then according to the instantaneous frequency mutation at the endpoint to set zero part of the sequence,and finally the different sensors after processing the The IMF time series from different sensors are then subjected to generalised correlation calculations to find the time delays of the elastic wave arriving at different positions and the effectiveness of the improved method is verified in a plate time delay estimation experiment.(2)A localisation model for vibration sources that does not require pre-measurement of the wave velocity in the plate is introduced and the estimated localisation coordinates from the weighted least squares solution are corrected for the correlation of the parameters in the results of the least squares solution of the localisation model using the chan method.In view of the pathological situation of the coefficient matrix of the least squares model,the spectral correction is introduced to iteratively correct the least squares results,and the correction parameters are obtained by the L-curve method of the regularization method,and then the iteration matrix is constructed using the eigenvectors of the coefficient matrix,which reduces the number of iterations of the spectral correction and improves the accuracy of the localization,and finally the effect of the improved method is verified by simulation.(3)The relationship between the CRLB(Clamero lower bound)of the coordinate estimation results and the sensor layout was analysed.The CRLB of the model was used as a criterion to evaluate the positioning accuracy of the array,and the mean trace and mean condition number of the CRLB matrix of the model were calculated under eight commonly used sensor arrays by taking uniform points on the flat plate respectively,and the various arrays were compared to obtain the edge uniform array as a suitable A layout method.(4)Vibration source localisation experiments were carried out on a 150cm long by 50cm wide by 2mm thick steel plate made of 45 gauge steel.Nine excitation sources were selected on the plate and the time difference between sensors was obtained using the improved time delay estimation method,after which the localisation coordinates were calculated using weighted least squares,the improved chan method and the improved spectral correction iteration method respectively to verify the feasibility of vibration source localisation and the The feasibility of the vibration source localisation and the effectiveness of the improved method were verified. |