Research On Detection Equipment Layout Method For Highway Traffic Accident Risk Assessment

In recent years,the rapid rise of intelligent assisted vehicles has led to a sharp increase in car ownership.The unbearable carrying capacity of roads and the frequent occurrence of traffic accidents pose a serious threat to people’s lives and property safety.Therefore,research on traffic accident safety has become a common task for countries around the world.The research on traffic accident warning at home and abroad mainly focuses on traffic flow prediction,traffic safety analysis,traffic accident warning models,and traffic accident warning systems.By utilizing machine learning,data mining,and deep learning technologies,traffic accident warning models and systems are established to identify potential locations and times of traffic accidents in advance.On the basis of the gradual maturity of early warning technology,technical personnel often have a strong personal desire for the hardware layout of accident early warning and detection equipment.This study will provide a novel theoretical support for technical personnel to carry out scientific and reasonable layout.(1)Based on the data support of existing research topics and through statistical analysis of accident occurrence patterns,it was found that incomplete traffic signs and markings,lack of roadside protection,and poor driving sight distance are the main causes of traffic accidents;Unlike traditional least squares methods,the Geographically Weighted Regression(GWR)model that considers spatial factors is used to analyze the risk factors of accidents and identify important influencing factors that induce traffic accidents.Research has shown that there is a strong correlation between street rate,incomplete traffic markings,and the number of lanes and the number of accidents,which can be used as data support for selecting node importance indicators in the future.(2)Topologize the road network,use nodes to describe the location of the deployment of early warning detection equipment,and establish an alternative node set;Establish a multi-dimensional objective layer and use Analytic Hierarchy Process(AHP)to construct a node importance evaluation model.The obtained results will be used to support the data of the required objective function in the multi-objective optimization model for constructing accident warning point layout.This model has multifaceted and accurate evaluation of network node importance.(3)A multi-objective optimization model for accident warning and detection equipment layout based on construction cost,total vehicle mileage of covered sections,and node importance is constructed.The "priority aggregation operator" theory is extended to apply to multi-objective weight assignment,avoiding the drawbacks of traditional weighting methods,strengthening the important relationship between targets,and the designed scoring function is more reliable;Using simulated annealing in intelligent optimization algorithms to seek the global optimal solution of the model,the practical application effect of the model was evaluated through practical cases.When verifying the feasibility of the model on national and provincial roads in Zhangqiu District,the iterative results tended to be stable.After comparing multiple schemes,the final result had high road section coverage,total vehicle mileage,and high importance considering economic costs.