An Adaptive Forecasting Method of Freeway Traffic Flow Based on Machine Learning and Copula Model

Abstract

The invention discloses an adaptive prediction method for expressway traffic flow based on machine learning and a copula model. An expressway to be predicted is divided into many small sections according to the ramp entrance, and an "algorithm selector" model is designed. The model stores A variety of prediction algorithms, assigning a model to each small road section, and then using historical data to go through training and testing, the model can select the best prediction algorithm based on the latest training results, thus overcoming the bottleneck of single model prediction effect. At the same time, the present invention starts from the correlation between the predicted value and the real value of the traffic flow point prediction model, and uses the correlation theory of the copula function to obtain the conditional probability distribution of the actual value under a certain predicted value condition, and then transfer to In the conditional probability analysis of errors, the error distribution estimation is transformed into the uncertainty estimation of traffic flow prediction, so as to better capture the randomness of traffic flow and greatly improve the prediction accuracy.

Description

An adaptive prediction of highway traffic flow based on machine learning and copula model test method technical field The present invention relates to the technical field of traffic flow prediction, especially relate to a kind of based on machine learning and copula model Adaptive prediction method of highway traffic flow. Background technique For the intelligent transportation system, reliable and accurate real-time traffic flow prediction is the key to alleviating traffic congestion and realizing traffic flow. It is a necessary prerequisite for management, traffic control, traffic guidance, and improving road operation efficiency. Traffic flow prediction is to induce efficient traffic The basis for managing and alleviating traffic congestion. Traffic flow has periodicity, randomness, temporal correlation and spatial correlation. Precisely Measuring traffic flow and grasping the dynamic change trend of traffic flow are the key steps of ITS, which are important ...

AI Technical Summary

Problems solved by technology

The statistical model has the advantages of simple calculation and easy operation. However, for complex nonlinear traffic flow data, its changing characteristics cannot be fully captured, resulting in low prediction accuracy

Images