A New Multivariate Regression Model For Pavement Friction Prediction Based On Different Pavement Texture Types

With the rapid development of modern society,transportation plays an vitally important role in maintaining the.operation of the whole city.The safety of transportation is related to people's life and property safety,so it has always been the focus of research in the field of transportation.Research shows that road skid resistance performance is an important factor affecting traffic safety.Due to the lack of skid resistance performance of the road,accidents such as rolled-up vehicles,rear-end collision and out of control are common.In this paper,the skid resistance performance of asphalt pavement is studied respectively from two aspects of laboratory asphalt rutting board and actual asphalt pavement.In the aspect of asphalt rutting board,changes in the law of road skid resistance under different working conditions is studied.In the aspect of outdoor asphalt pavement,asphalt pavement with different textures is selected as the research object,and the required parameters and indicators are collected.Firstly,by examining the correlation between the transverse and longitudinal pendulum values,a fitting equation is established.Secondly,a multivariate regression friction model of pavement with different textures is established to predict the skid resistance of the road.Finally,BP neural network and RBF neural network is used to establish friction prediction models based on rutting board data and actual asphalt pavement data,the accuracy of both prediction methods is more than 95%.The analysis shows that the prediction accuracy and generalization ability of the artificial neural network model are higher than the multiple regression friction model,which proves that the artificial neural network model is better in predicting the friction data.This paper verifies that there is a large correlation between the transverse and longitudinal friction coefficients,and also provides its own research results in analyzing the influencing factors of road skid resistance performance and constructing the friction prediction model.