Research On The Stability Of Road Slope Base On RBF Neural Network

Xiabi road is located in northeast of Biru county in Tibet. With the developmet of Tibet economy,to perfect infrastructure of road in Tibet has already become a necessity,but the geological disaster by slope project has become more and more serious, especially, it is easier to cause geological disaster by building road in mountainous district of complexed the geological condition at high altitude areas. Analysis and assessment of slope stability is an important question which is not avoid in road construction.But its complexity,and uncertainty brings infinite difficulty to the analyses.factor that influence slope stability are anfractuous.and they are not linear but nonlinear.so it is hard to express it in accurate mathematical expression;while artificial neural network technology is fuzzy theory arithmetic which has strong nonlinear disposal processing.so using it in slope stability analyses is very significant.This paper takes the K85 slope of Xiabi road in Tibet as main research object, expounding the geological setting of the slope area,analyzing differential features and influential factors of the slope., trying to utilize method of artificial neural network to evaluate the stability of the K85 slope in Xiabi road in Tibet, Three research findings are as below:1. The main factors influencing the stability of the K85 slope in Xiabi road in Tibet included lithology,rock mass structure the geological structure,topography,influences of weathering,the action of water,action of vibration and human engineering activities.2. Using method of unitary to process the real sample data,we may improve learning efficiency and the forecasting precision.3. Through comparing the result with the method of artificial neural network,Bishop and the transmit coefficient method,we can find that the method of RBF artificial neural network model is practicable to avaluate slope stability.