Study On Intelligent Optimization Of Blasting Parameters And Evaluation Of Blasting Effect In An Underground Mine

With the increase of ore demand and the popularity of environmental protection and efficient production,the blasting,an important means of mine production,has been put forward with higher requirements.Due to the complexity of ore and rock structure,reasonable blasting parameters should be selected in order to achieve good blasting effect.It is reliable to obtain blasting parameters through field tests,but the cost is high and the efficiency is low.Therefore,intelligent algorithm and numerical simulation adopted to optimize relevant parameters,which improves the research efficiency and conforms to the current research trend of blasting parameter optimization.Blasting is regarded as the main means of exploitation and mining in an underground mine in China.With the expansion of production scale,it is necessary to optimize and evaluate the existing blasting technology,so as to provide technical reference for the subsequent production.In order to solve these problems in a scientific and effective way,relevant literatures are reviewed,and on this basis,intelligent algorithm and numerical simulation are applied to optimize relevant parameters.The specific research includes the following aspects:(1)Based on 56 groups of mine rock explosibility statistical data,rough set theory(RS)is adopted to simplify the data,and the result shows that"lumpiness average pass rate"is a redundant attribute.Using 51 sets of data as training samples and 5 sets of data as test samples,the average error between the predicted explosibility index and the actual value is 9.68%using the BP neural network(BPNN)before attribute reduction,while the average error of the BP neural network(rs-bpnn)after attribute reduction is 7.29%.the same,using the attribute reduction in front of the support vector machine(SVM)to predict the ore-bearing rock explosive index and the actual value of the average error is 1.84%,and after the attributes reduction of support vector machine(RS-SVM)the average error is 1.71%,so choose RS-the SVM prediction model,get the mine ore body and wall rock explosive sex are?level grade,belong to"difficult to burst"level.(2)According to the existing blasting model and the actual situation of mine,the optimization model of tunnel blasting cost is established.The whale algorithm was used to optimize the mathematical model constructed,and the blasting parameter combination under the condition of the minimum cost was obtained.the resistance line of surrounding holes was 0.72m.the spacing of surrounding holes was 0.58m.the resistance line of caving holes was 0.91m.the spacing of caving holes was 0.87m.the unit consumption of explosives was 1.13Kg/m~3,and the unit comprehensive cost was 30.10 yuan/m~3.(3)According to the rock mechanics and mining and other related parameters of the mine,the critical caving span of the goaf roof is determined to be 39.80m by using the theoretical formula.On this basis,the roof blasting induced caving model is established by using the finite element analysis software(ANSYS/LS-DYNA).Due to the difference in differential time,the blasting effect will be different.Therefore,three different schemes of 20ms,35ms and 50ms between pre-split hole and caving hole are respectively set in the numerical simulation,and the blasting process of the three blasting schemes is simulated by software.By comparing and analyzing the equivalent stress of the three observation points with the compressive strength of the ore and rock,it is considered that the effect of induced caving is the best when the interval is 20ms.(4)Through collecting the relevant data of mine,the blasting effect evaluation system is established,the weight of six indexes is determined by ahp,and then the expected value of blasting effect influencing factor is calculated.The heap shape(score)is 4.60.The lumpiness distribution(score)is 4.31.The explosive unit consumption is 0.57kg/t.The blasting volume per meter is 6.12t.The impact on the surrounding rock(score)is 4.57,and the impact on the surface(score)is 4.60.On the basis of evaluation and classification,the index data are processed by cloud model,and the determination of"good"blasting effect is 0.25,and the determination of"better"is 0.14.Finally,the grading grade of the blasting effect of the mine stope is determined to be"good".