Study On Intelligent Pavement Friction Evaluation Model And Maintenance Decision For Optimized Skid Resistance

The skid resistance of pavement affects road traffic safety,but the skid resistance of pavement will deteriorate due to the influence of traffic load,environment and other factors.Therefore,it is of great significance to monitor and maintain the skid resistance of pavement for a long time.At present,the pavement skid resistance testing equipment at home and abroad is mainly contact type,such as:pendulum friction instrument,dynamic friction coefficient tester such as fixed-point detection equipment and continuous testing equipment such as transverse force/longitudinal force friction coefficient testing vehicle.The fixed-point testing equipment is inefficient and the test results are greatly affected by human operation.The continuous testing equipment needs a lot of watering,and the wear wheel wear needs to be calibrated regularly.Therefore,in recent years,researchers began to apply high-definition camera,laser scanning and image processing technologies to achieve non-contact test of skid resistance of asphalt pavement,and put forward the corresponding skid resistance evaluation index,and establish the corresponding skid resistance evaluation model and degradation model.However,the existing pavement skid resistance evaluation indicators lack of uniform standards and are not universal.For example,most indicators do not consider the effective contact area between tire and pavement and the effective wavelength of pavement three-dimensional texture.However,the existing pavement skid resistance evaluation model and deterioration model also have some problems,such as incomplete considerations,low accuracy and weak robustness.At the same time,the problem of early deterioration of skid resistance of pavement is lack of corresponding research on maintenance decision.The thesis focuses on the above problems.Firstly,LS-40 portable three-dimensional laser surface analyzer is used to collect three-dimensional pavement texture data.The non-contact testing method effectively improves the efficiency,repeatability and accuracy of data collection.Then,an evaluation index of skid resistance of multi-scale pavement based on power spectrum was proposed:surface roughness power spectrum C_（q）,and area stratification and macro-micro texture separation were discussed to obtain the optimal contact area and macro-micro texture wavelength of different types of asphalt mixture pavement and tire.Among them,the optimal contact areas corresponding to AC-10,AC-13 and AC-16 are 40%～50%,30%～40%and 10%～20%respectively.The optimal texture wavelengths corresponding to the three are the same,that is,the optimal macro texture wavelength is5.06mm and the optimal micro texture wavelength is 0.33mm.Finally,the correlation analysis between the improved fractal dimension（FD）and the traditional height index and the friction coefficient was carried out.The results showed that the absolute value of the correlation coefficient was above 0.6,indicating the rationality of the selected evaluation index and optimizing the existing evaluation index system of skid resistance.Based on the new index system combining with the new algorithm optimize the skid resistance of existing prediction model,the paper uses the present a new algorithm of gradient hoist,give full consideration to environmental factors,the best contact area and micro-macro texture,improve the decision tree based on gradient（GBDT）pavement skid resistance prediction model,by comparing other prediction model,It is found that the R² of training and testing of the model is 0.9295 and0.8732,respectively,with high accuracy and robustness.For the research on the deterioration of pavement skid resistance,the paper uses the multi-functional pavement wear instrument independently developed by the team to carry out the indoor rut plate wear test,fully considers the influence of factors such as water,temperature,load,speed and wear times,and establishes the pavement skid resistance degradation model according to the wear data analysis.It is concluded that the cubic polynomial fitting empirical model R² reaches 0.9164,and the fitting deterioration curve is more in line with the actual experience,The long-term effective prediction of pavement skid resistance is realized.Furthermore,surface treatment technology for skid resistance of pavement was further studied on the basis of abrasion.By using time/road condition analysis method and combining with existing deterioration model,the optimal maintenance time under abrasion test condition of this study was about 230,000 abrasion times.Cost-benefit analysis was adopted for different surface treatment methods.It is concluded that the minimum equivalent annual cost（EAC）of SMA-5 ultra-thin wear layer is 6.7,which is determined as the best maintenance mode,so as to make a reasonable maintenance decision for skid resistance of pavement.