Rock Slope Angle Prediction Based On SMR Rock Mass Classification Method

Aiming at the problems existing in the rock slope angle design,this paper collects the geological data of the built slope,and trains the BP neural network prediction model for the sample.The input layer influence factor of the prediction model adopts the parameters of SMR rock mass classification method,and the slope geological data is quantified according to the scoring standard of the parameters.For the problem of multiple joints in rock slope,the initial parameter values are weighted adjusted,and the adjusted parameter values are input into the BP neural network prediction model to predict the slope angle.The predicted values of the adjusted parameters,the predicted values without adjustment and the actual values of the constructed slopes are compared and analyzed.The main achievements of this paper are as follows:(1)The classification results of SMR rock mass classification under weighted adjustment are closer to the actual values than those without adjustment.(2)The BP neural network prediction value is compared with the actual value of the constructed slope.The results show that it is feasible to predict the slope angle of rock slope by using BP neural network.(3)The values of rock mass classification method are adjusted,and the predicted value of rock slope angle obtained by BP neural network is closest to the actual value of the built slope.