Construction Of PEMD Method And Its Application In Engineering Blasting

The control of damage caused by blasting vibration is always involved in engineering blasting activities,and the analysis of blasting vibration signal is the basis of studying the control of blasting vibration hazard.Based on the Natural Science Foundation of Jiangxi Province(20192BAB206017),taking blasting vibration signal as the research object,and aiming at the modal mixing problem in the analysis of blasting vibration signals by empirical mode decomposition method,utilizes the orthonormality of Principal Component Analysis,a completely orthogonal empirical mode decomposition method is proposed.And based on the experiments of noise reduction,delay identification and blind gun detection of blasting vibration signals,the application of PEMD method in engineering blasting is realized,and the following conclusions are obtained:(1)In order to study the effectiveness of PEMD method in removing modal aliasing,a simulation test of simulated signals was designed.Studies have shown that the intrinsic mode function decomposed by EMD has multiple dominant frequencies and mode mixing phenomenon,while the IMF component decomposed by PEMD has a single dominant frequency,which eliminates the mode mixing phenomenon.(2)The noise reduction of blasting vibration signal is studied by using PEMD method.Studies have shown that: compared with EMD and Ensemble Empirical Mode Decomposition,the signal-to-noise ratio of noise reduction signals is increased by 1.15 d B and 0.38 d B respectively,with the smallest root mean square error.In addition,PEMD has the highest sensitivity to the frequency identification of noiseless signals and the best noise filtering effect.In the blasting vibration signal de-noising experiment,PEMD has an ideal effect of removing noise burr,and PEMD has the best effect of preserving medium and low frequency vibration signals at 0?300 Hz and filtering out high frequency noise above 300 Hz.(3)PEMD method is used to identify the time delay of each detonator in delayed blasting.Studies have shown that in the blasting delay identification experiment of similar physical model,the EMD is affected by the interference of the mode aliasing factor,the identification errors of 10 ms,16 ms and 21 ms at the same elevation fluctuated within 0?10 ms,and the identification errors of 16 ms at different elevations fluctuated within 0?16 ms.The identification errors of PEMD were all 0 ms.Therefore,compared with EMD,the identification accuracy of PEMD is not affected by factors such as delay time and elevation,and the identification accuracy is effectively improved.The results show that the recognition rate of EMD method fluctuates in the range of 74%?91%,while the recognition rate of PEMD method is stable over 90%.(4)Aiming at the selection of principal IMF components in PEMD delay identification process,a principal component screening model was constructed by using cross-correlation function and applied to blasting misfire detection.Studies have shown that the detection rate of PEMD to the hole in the open-air step blasting experiment is 100%.In the tunnel blasting misfire experiment,PEMD detected 1,3,5 and 1 blasting misfire at the lift hole,two around hole,bottom hole and peripheral hole respectively,and the blasting misfire detection results of the monitoring data on the left and right sides of the tunnel were the same.The specific location of the blasting misfire was further determined by combining with the layout of the gun holes,and it was consistent with the distribution of the blasting misfire after blast.Finally,compared with other blasting misfire detection methods,it is found that the accuracy of the wavelet time-energy density method is unstable due to the choice of wavelet basis,and the blasting misfire misjudgment phenomenon appears.At the same time,PEMD is superior to high-precision magnetic method,transient electromagnetic method and frequency division multiple access method in the indicators such as hole size,detection distance,geological conditions and operating cost by constructing evaluation indexes of blasting misfire detection.