Analysis Of Blasting Vibration Signal And Prediction Of Vibration In Hongling Lead-zinc Mine

The influence of the structure vibration caused by blasting has attracted people's attention.Because of the randomness,fuzziness and uncertainty of blasting seismic wave,there are still many problems to be solved and perfected in the research of blasting vibration.Hongling lead-zinc mine has left a large number of goaf in the Pre exploitation process;the influence of blasting vibration for surface and underground structures is very complex during the subsequent pillar recovery process.In view of the above questions,based on the blasting vibration test data in Hongling lead-zinc mine,this paper has studied deeply and systematically on the blasting vibration problems from the blasting vibration analysis and blasting vibration prediction.The main results are as follows:(1)The energy distribution of the frequency band of the blasting vibration signal under different blasting conditions and position conditions is analyzed.And the influence of different blasting conditions and position conditions on the energy distribution of the blasting vibration signal frequency band is discussed in detail.The inherent law of the characteristic parameters of the blasting vibration was revealed.(2)Taking the maximum charge and distance as the input,the prediction of peak vibration velocity of blasting vibration was carried out by several empirical formulas and compared the results of the predicted results with the actual results to choose the best empirical formula for predicting the peak vibration velocity in Hongling lead-zinc mine.Finally the formula for predicting peak vibration velocity of the blasting vibration in Hongling lead-zinc mine is obtained.(3)Taking the maximum charge,the total dose,the number of initiating segments,the horizontal distance and elevation difference as the input and the peak vibration velocity,frequency and duration of blasting vibration as the output,the prediction model of blasting vibration based on generalized regression neural network was established.The prediction of blasting vibration based on GRNN was carried out.(4)Based on the support vector machine,the blasting vibration prediction model is established and the peak vibration velocity,frequency and duration of blasting vibration are predicted.Comparing the predicted results of blasting vibration based on SVM with the results based on GRNN and empirical formula to analyze the advantages and disadvantages of the three models and select the best prediction model.(5)Analyzing the blasting vibration monitoring results of the canteen,garage and office buildings in Hongling lead-zinc mine by using the response energy criterion method.And the energy of the blasting vibration response of the controlled structure in Hongling lead-zinc mine was predicted by using the support vector machine to prove the performance of support vector machine in blasting vibration prediction once again.