Research On Performance Prediction Of Asphalt Pavement Based On The Construction Data

The variations of asphalt mixture parameters exist during construction and may affect the pavement performance.At the same time,the existing evaluation system on pavement performance does not fully consider the impact of the construction process.Therefore,it is necessary to analyze the impacts of the variation of asphalt mixture parameters on the performance in the construction process.At the same time,the traditional statistical indexes have certain values in the evaluation of pavement performance uniformity but exist limitations.In order to analyze the possible influence of the construction process on the performance of asphalt pavement and reasonably evaluate the uniformity of pavement performance,this study predicted the performance of pavement surface combined with the construction data and explored the method to analyze the influences of the variations of mixture parameters on the performance and uniformity of asphalt pavement.Firstly,in order to reasonably predict the service performance of asphalt pavement in the construction processes,the performance prediction models were compared and analyzed.Based on the field investigation,the prediction models of fatigue performance and permanent deformation in the specification for design of highway asphalt pavement(JTG D50-2017)were selected.The Witczak prediction model is selected for the prediction of dynamic modulus of asphalt mixture to provide a model basis for subsequent analysis.Secondly,in order to verify the feasibility of the performance prediction method,experiments were carried out in four construction sites of asphalt pavement,and the data such as asphalt mixture gradation,asphalt-aggregate ratio and aggregate density required by the above model were collected.The variation of gradation and asphalt-aggregate ratio were analyzed.It was found that the construction process,asphalt mixture type and nominal maximum aggregate size of asphalt mixture will affect the variation degree of asphalt mixture parameters Combined with the prediction model,the fatigue performance and permanent deformation were predicted and analyzed.The results show that the fluctuation of asphalt pavement performance and the change of performance uniformity will be caused.Thirdly,in order to analyze the influences of gradation and asphalt aggregate ratio variation on the fluctuation of service performance,this study selected BP neural network modeling,and expanded the amount of data through Monte Carlo method based on the verified data distribution.Then the sensitivity analysis and the influence analysis of the variations of mixture parameter were carried out.The results show that Monte Carlo method can provide data support for neural network modeling.BP neural network is feasible in analyzing the variability propagation of asphalt mixture parameters,and the model is accurate.The variations of asphalt-aggregate ratio and aggregate passing percentage 0.075 mm,2.36 mm and 4.75 mm of asphalt mixture have more significant impacts on pavement performance than other parameters.Finally,in order to reasonably evaluate the uniformity of service performance,the method of technique for order preference by similarity to an ideal solution(TOPSIS)was selected.The entropy weight method was proposed to weight different performance.The comprehensive evaluation system of pavement performance is established.The results show that the TOPSIS method based on entropy weight method is practical in the uniformity evaluation.The established evaluation system can directly reflect the uniformity of pavement performance by combining grading statistics and the variation analysis.Based on the construction data,the prediction and evaluation of uniformity for construction quality are realized.