Short-term traffic flow prediction method and system based on convolutional neural network

Abstract

The invention discloses a short-term traffic flow prediction method and system based on a convolutional neural network. The method comprises the steps: receiving vehicle pass record data of all road intersections and generating a traffic flow dataset and a track dataset; using the track dataset as the input of a CBOW model to obtain road intersection vector expression and further obtaining the traffic flow influence relation between the road intersections by calculating the vector distances; constructing a characteristic matrix and using the characteristic matrix as the input of a prediction model; using the convolutional neural network as the prediction model, training the parameters of the prediction model, inputting the test dataset into the prediction model, calculating out and outputting the average error between a predicted value and a target value, choosing the parameter corresponding to the minimum error as the optimal parameter of the prediction model, obtaining the optimal prediction model and further output the optimal prediction value of the traffic flow. The short-term traffic flow prediction method takes the time and space relation of the traffic flow into consideration, is combined with the convolutional neural network and improves the prediction accuracy of the short-term traffic flow.

Description

technical field [0001] The invention belongs to the field of data mining, and in particular relates to a method and system for predicting short-term traffic flow based on a convolutional neural network. Background technique [0002] Since the beginning of the 21st century, the rapid development of social economy has made people more and more dependent on various means of transportation, which has made urban traffic problems more and more prominent. Accurately predicting the traffic flow of each traffic intersection can guide the driver to choose the driving route and effectively alleviate the congestion problem. At the same time, accurate traffic flow forecasting is also an important part of building a smart city. [0003] In terms of prediction methods, there are currently moving average models, k-nearest neighbor models, autoregressive models, and neural network models. When these models are applied, they only analyze the time series data of traffic flow at each intersec...

AI Technical Summary

Problems solved by technology

And the existing traffic flow prediction method, some select intersections according to the upstream and downstream of the geographical location, the problem that exists in this way is: due to factors such as traffic restriction or signal light control, the impact of traffic flow between intersections with similar geographic location is not definite; The upstream intersection may have an important impact on the current intersection traffic flow data, so the strength of the influence between intersections cannot be found, that is, the strength of the influence between intersections cannot be directly judged by geographical location; and most of the existing intersection data combinations in traffic flow forecasting are It is artificially set, so that it cannot be dynamically obtained based on actual data

However, because the statistics of traffic flow data are based on dynamic vehicle location information, there are complex direct or indirect relationships between traffic data at different intersections. The above models or methods cannot effectively deal with these relationships, and the prediction error is relatively high.

Images