| Southwest and other mountainous landslides occur frequently,sudden,random,often to the impact of the road within the scope of the house to bring great damage,and even endanger the scope of the peopleās lives within the scope of security.In order to adapt to the rapid economic development,often need to carry out the fall rock control work,and the collapse of rock movement characteristics is to carry out the implementation of protective design of the important indicators.Based on the field investigation of the collapse of Dafu Village in Longquan Yushanquan Town,this paper studies the characteristics of the collapse disaster and analyzes the influencing factors of the rockfall movement characteristics.The results show that the fluctuating model of the falling rock,the slope height,the slope slope and the slope material as the variable design are used to study the characteristics of the rock-Distance,offset ratio,maximum bounce height,maximum motion speed,impact area.At the same time,a fall prediction model,a dumping prediction model and a slump prediction model are established based on the experimental data based on wavelet neural network.Predicting the travel distance of rockfall,and converting the predicted value into the same scale as the actual value of the collapsed site.Finally,based on the results of indoor experiments on the mountain town of Dafu 5 group collapse proposed some prevention and treatment recommendations.The main contents and achievements of this paper are as follows:(1)Different instability modes,the height of the collapse source,the slope and the material composition of the slope have different effects on the movement characteristics of the rock.The height of the collapse source is not sensitive to the level of movement of the falling stone,the offset ratio,the maximum bounce height;the impact of the falling and dumping level on the level of the falling rock,the offset ratio,the maximum bounce height is stronger than the slump,The speed of movement is not great;when the slope of 50-60 °,the level of rock movement is relatively large,when the slope of 50 ° to reach the maximum.When the slope is 40-60 °,it is easy to produce a larger lateral offset.When the slope is 40-70 °,the maximum bounce height of the falling stone is positively correlated with the slope,and the maximum velocity is always positively correlated with the slope.The turf slope can effectively reduce the bounce height of the falling stone.(2)Different modes of motion correspond to different slope thresholds.When the slope is 20-40 °,a jump,slide,roll,or a combination of motion occurs.Slope of 40-50 °,the movement of the performance of the jump and slide or between the two combined movement,rolling rarely occur.Slope of 50-70 °,the occurrence of pure sliding and pure jump,sliding + jump combination of movement rarely occur,basically do not roll.(3)Based on the wavelet neural network,the predicting model of rockfall migration is proposed based on the fall prediction model,the dumping prediction model and the slump prediction model.The model can predict the migration distance of the rockfall in the engineering example.(4)Based on the results of laboratory tests,some suggestions on the prevention and control of rockfall are put forward.According to the experimental results and the location of the dwellings,the different energy levels and different protection heights of the passive protection system are set.The "human" Wall,excavated in the highway near the slope of the excavation of rock fall,for the bulk of the distribution of dangerous rock slope active protection system. |