How to construct a prediction model for highway traffic accidents under extreme weather conditions

Road traffic accident prediction refers to the estimation and speculation of the future situation of traffic accidents. It achieves the description process of the future state of traffic accidents by systematically exploring the past and present states of traffic accidents, and considering the changes in related factors. Extreme weather conditions are the high incidence period of traffic accidents, and accurately predicting traffic accidents with extreme weather conditions is of great research significance. This paper studies how to build a highway traffic accident prediction model in extreme weather based on the gray Mixture model. Firstly, the k-means clustering algorithm is used to obtain sample data and perform preprocessing. Then, grey correlation analysis is used to analyze the correlation between sample data sequences and predict the number of traffic accidents. Finally, residual correction is applied to the prediction model to achieve the construction of a highway traffic accident prediction model under extreme weather conditions. The experimental results show that the Root-mean-square deviation of the prediction results of the model built in this paper is below 0.3, the Pearson correlation coefficient is between 0.90 and 0.98, and the prediction time is within 8s. The prediction accuracy is high and the prediction effect is good.