Correlation Analysis Of Traffic Volume And Highway Road Damage Based On Big Data Of Freight Vehicle Trajectory

With the increasing traffic volume and the increasing number of large freight cars,road damage is severe,and some or even two or three years will have different levels of damage.Road maintenance has become an unavoidable problem.The key issue of highway maintenance management is to understand the damage of highway in a timely manner,and the damage of highway has a great relationship with the traffic volume of large freight vehicles.Therefore,the analysis and analysis of the correlation between the traffic volume of freight vehicles and road damage is of great significance.This paper mainly studies the correlation between the traffic volume of freight trucks and highway road losses,in order to make up for the lack of quantitative evaluation methods in the existing highway maintenance design and provide new ideas for pavement maintenance design.It mainly solves the two key problems of extracting trajectory data and analyzing the correlation between traffic volume and highway road loss.The specific research is as follows:(1)Track data preprocessing.Firstly,the trajectory data is simply cleaned,the duplicate data and drift data are eliminated,and then the improved quadtree algorithm is used to divide the trajectory data of the freight car,and the quadtree code suitable for the division method is used.There are three main types of division by this division method,which are "?" type,"I" type,and "L" type,and use different codes for each division type.Compared with the traditional quadtree method,the partitioning method used in this paper solves the problem of unbalanced data partitioning in quadtrees,and merges data blocks with no data or a small amount of data with adjacent data blocks to avoid empty areas.Appeared to improve the efficiency of data query and retrieval.(2)Correlation analysis.Simplify the axle load conversion formula in "Highway Asphalt Pavement Design Code"(JTG D50-2017),and calculate the cumulative equivalent axle times of standard axle load,and analyze the cumulative equivalent axle times of standard axle load and the road pavement condition index(PCI).This paper uses three correlation analysis methods to analyze the correlation between the cumulative equivalent axle times of standard axle load and the pavement condition index,which are the chart analysis method,the correlation coefficient method,and the time series method.The time series method uses the autoregressive integrated moving average model(ARIMA model)and autoregressive integrated moving average models with external input(ARIMAX model).The standard axle load cumulative equivalent axis times are added to the time series model for analysis,and it is found that the ARIMAX model with independent variables is more accurate.This paper uses the trajectory data of freight cars to obtain traffic volume,and adds the condition of axle load to calculate the correlation between the cumulative equivalent axle times of the standard axle load and the pavement condition index.The analysis results show that with the increase of the cumulative equivalent axle times of the standard axle load,the highway pavement condition index shows a downward trend,and the correlation coefficient of the two is-0.91,showing a negative correlation and a relatively high degree of correlation.Using the ARIMA model and ARIMAX model to analyze the correlation between the standard axle load cumulative equivalent axle times and the pavement condition index,the correlation model between the two is obtained.The experimental results show that the mean absolute percent error of the ARIMA model is 0.9%,the mean absolute percent error of the ARIMAX model is 0.3%,and the accuracy of the ARIMAX model is higher.Through the established model,the quantitative relationship between the standard axle load cumulative equivalent axle order and the highway pavement condition index is obtained,which provides a basis for highway maintenance.