Research On Blasting Parameters Optimization And Vibration Forecast Of High Water-cut Rock Mass

With the rapid development of the national economy,as an important means of foundation engineering,blasting is widely used in various fields.However,large-scale,frequent blasting will bring security risks to the surrounding environment.Especially in the severe environment of blasting,safe and efficient blasting is particularly important.Guizhou province is located in the plateau area,there are many mountains which are mostly composed of limestone,so the typical Karst landform is formed and the traffic is underdevelopment.Because the ecological environment and vegetation have been well protected,the large annual rainfall makes the mountains in the saturated state all the time.The rocks are eroded by rain and groundwater,so fissures,gaps and karst are developed.The gaps are rich with flowing water,this makes the rocks to become a high water cut rock mass.For the blasting of this type of rock mass,the blasting effect of the current blasting parameters is poor due to the difference of the water medium relative to the air medium.The propagation law of the seismic wave in the multiphase medium is more complicated,using the traditional empirical formula to forecast are often inconsistent with the actual monitoring,and false positive are common.At present,the domestic research on high water content rock blasting parameters and vibration prediction is extremely rare.In this paper,field experiment and BP neural network are used to study the blasting parameters and vibration prediction of high water cut rock,the current blasting parameters are optimized and a more accurate vibration prediction method is obtained.The specific work and achievements are as follows,(1)A high water rock blasting construction project in Guizhou province was selected as the experimental point.Four groups of field experiments were conducted,distance between bores(a)× row spacing(b)are 3×2.5m,3.5×3m,4×3.5m,4.5×4m,respectively.The fragment size of the surface of the four groups were analyzed by software Split-Desktop3.0 and the base rates were calculated by field measurement.The block rate and the toe rate were used as the basis for the blasting effect.Theexperimental results showed that the blasting effect is the best when the hole distance(a)× row spacing(b)is 3.5 × 3m.Under this blasting parameter,the water bearing rock explosive consumption is only 0.312kg/m3,which is lower than that of the general rock blasting.(2)The blasting vibration in high water rock and ordinary rock is tested and compared.It is found that the peak and duration of blasting vibration in high water rock are larger.Therefore,the influence of water medium on blasting vibration cannot be neglected.The BP neural network model is established for the prediction of blasting vibration parameters,the water medium is also incorporated into the network model input layer.The network model was trained by field monitoring data of high water rock blasting.After comparing the predicted results with the measured results,it is found that the prediction parameters of BP neural network model are close to the measured values,which has great guiding significance for the similar engineering blasting construction.