Traffic accident prediction method based on unbiased non-homogeneous grey model and Markov model

Abstract

The invention provides a traffic accident prediction method based on an unbiased non-homogeneous grey model and a Markov model. The method includes obtaining the unbiased parameter estimation form ofthe unbiased non-homogeneous grey model by establishing the initial data of the traffic accident number and utilizing a unbiased non-homogeneous albinism differential equation, and calculating a parameter estimation value; and then taking the relative value of the fitting value and the initial value of the unbiased non-homogeneous grey model as a basis of state division, and dividing the prediction value into three states; establishing a prediction model based on the unbiased non-homogeneous grey model and the Markov model; simulating the number of the traffic accidents by utilizing the established prediction model, and predicting the future accident data values. According to the invention, the non-homogeneous exponential function is fitted through the unbiased non-homogeneous grey model,and the prediction result is corrected through the combination with the Markov chain, the improved three-step state transition probability matrix is utilized to further improve the prediction precision, and the dynamic analysis of the data is realized.

Description

technical field [0001] The invention belongs to the technical field of traffic accident prediction, in particular to a traffic accident prediction method based on an unbiased non-homogeneous gray model and a Markov model. Background technique [0002] In recent years, the total mileage of highways and the number of motor vehicles in my country have continued to increase, and traffic safety has already become an important part of national economic development and social stability. As an important task in evaluating road traffic safety, traffic accident prediction is of great significance for exploring the occurrence rules of traffic accidents, analyzing the development trend of traffic safety conditions under the existing traffic infrastructure conditions, and making scientific quantitative predictions. [0003] As a hot topic in road traffic safety, various theories have been applied to accident prediction. In the traffic system, the data presents some kind of non-stationar...

AI Technical Summary

Problems solved by technology

[0004] In order to achieve the above purpose, the present invention provides a traffic accident prediction method based on the unbiased non-homogeneous gray model and the Markov model, which solves the problems of large prediction deviation and low accuracy in the prior art

Images