Urban rail transit short-time passenger flow prediction method based on transfer learning

Abstract

The invention provides an urban rail transit short-time passenger flow prediction method based on transfer learning. The method comprises the following steps: establishing a passenger flow characteristic evaluation index system of each station based on historical report data of each station of the urban rail transit; clustering all stations by adopting a clustering algorithm based on the passenger flow characteristic evaluation index system of each station; performing time sequence analysis on the annual in-out passenger flow volume of the clustering center stations of each category, calculating passenger flow indexes of the clustering center stations, and obtaining an optimal passenger flow prediction model of each clustering center station; and substituting the optimal passenger flow prediction model of each clustering center station into other stations of the same category, and adjusting the optimal passenger flow prediction model through a transfer learning method to obtain the optimal passenger flow prediction model. According to the method, the subway passenger flow characteristic evaluation index system is constructed, so that the station passenger flow characteristics are effectively described. The passenger flow characteristics of different stations can be considered, and the prediction precision and prediction efficiency of the model are improved.

Description

technical field [0001] The invention relates to the technical field of rail transit passenger flow prediction, in particular to a short-term passenger flow prediction method for urban rail transit based on transfer learning. Background technique [0002] Under the development trend of intelligent and intelligent operation of urban rail transit, how to accurately monitor passenger flow and realize intelligent and dynamic operation organization is a problem that needs to be considered urgently, and realizing accurate prediction of short-term passenger flow of urban rail transit is a A key part of realizing intelligent operation. [0003] For rail transit short-term passenger flow prediction, there are many researches at home and abroad, and some achievements have been obtained. In terms of railway passenger flow forecasting, some scholars have established a fuzzy k-nearest neighbor passenger flow forecasting model to predict the short-term passenger flow of high-speed railway...

AI Technical Summary

Problems solved by technology

The object of the research prediction is only a few stations, without considering the characteristics of the subway station itself and the heterogeneity among stations, the universality of the proposed research method is limited

Images