Urban area passenger flow volume prediction method based on mobile phone signaling data

Abstract

The invention discloses an urban area passenger flow volume prediction method based on mobile phone signaling data. The method comprises the following steps: S1, urban area division; S2, preprocessingof the mobile phone signaling data; S3, residence data processing according to a user track based on the mobile phone signaling data; and S4, urban area passenger flow prediction based on an urban passenger flow prediction model fusing the graph convolutional neural network of the M historical monitoring time periods and external influence factors. The track sequence of the user is obtained through the mobile phone signaling data, so that the change of the passenger flow volume in the urban area is obtained according to the track sequence of the user, and compared with a method for obtaininguser information through GPS data in the prior art, more complete and effective position information can be obtained, and a good foundation is provided for prediction of the passenger flow volume; andsecondly, the spatial correlation of the grid region is extracted by adopting a graph convolutional neural network, so that the accuracy and practicability of passenger flow volume prediction are further improved.

Description

technical field [0001] The invention relates to the technical field of urban traffic planning, and more specifically relates to a method for predicting passenger flow in urban areas based on mobile phone signaling data. Background technique [0002] In recent years, with the rapid development of my country's economy, the process of urbanization and automobileization has accelerated, and the number of motor vehicles has increased rapidly. In cities, traffic congestion has become a bottleneck restricting urban economic development. [0003] In order to improve the efficiency of traffic network use and solve traffic congestion and traffic safety problems, it is a very efficient method to monitor and estimate passenger flow in urban areas. In the prior art, GPS data with accurate positioning is generally used as the data basis, and regression model and machine learning models to predict the flow of people. However, in terms of data usage, compared with traditional location acqui...

AI Technical Summary

Problems solved by technology

[0003] In order to improve the efficiency of traffic network use and solve traffic congestion and traffic safety problems, it is a very efficient method to monitor and estimate passenger flow in urban areas. In the prior art, GPS data with accurate positioning is generally used as the data basis, and regression model and machine learning models to predict the flow of people. However, in terms of data usage, compared with traditional location acquisition methods, GPS data has greatly improved accuracy and coverage, but it is more restrictive. For example, in smart In mobile devices, only when the user uses an application program (APP) that requires GPS positioning will the positioning be activated. In addition, in terms of model selection, traditional regression models and machine learning models are difficult to extract the complex spatio-temporal relationship of passenger flow in urban areas. Cannot play a role in accurate prediction

Images