An urban rail short-time passenger flow prediction method based on modal decomposition and deep learning

A technology of deep learning and modal decomposition, applied in forecasting, biological neural network models, instruments, etc., can solve the problems of timing, non-stationarity and nonlinearity of short-term passenger flow data, and overcome the defects of gradient explosion , Maintain information persistence and improve stability

Pending Publication Date: 2019-06-28
SOUTHWEST JIAOTONG UNIV
View PDF3 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] To sum up, there have been a lot of research on short-term passenger flow or traffic flow prediction of urban rail transit at home and abroad, but the above-mentioned models have not fully utilized the characteristics of time series, non-stationarity and nonlinearity of short-term passenger flow data. The accuracy and robustness of short-term passenger flow forecasting are of great significance

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • An urban rail short-time passenger flow prediction method based on modal decomposition and deep learning
  • An urban rail short-time passenger flow prediction method based on modal decomposition and deep learning
  • An urban rail short-time passenger flow prediction method based on modal decomposition and deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] A short-term passenger flow prediction method for urban rail based on modal decomposition and deep learning, such as figure 1 As shown, it mainly includes the CEEMDAN decomposition stage, the Fine-to-Coarse reconstruction stage, and the GRU prediction stage. The specific content is as follows:

[0022] 1. CEEMDAN decomposition stage

[0023] At this stage, the passenger flow data of urban rail transit is regarded as a piece of data composed of continuous smooth signal and abnormal noise. Use CEEMDAN to decompose and process the original passenger flow data. For a given data x(t), EMD can decompose the data to obtain a set of intrinsic mode functions IMF from high to low frequencies i (t), (i=1, K, I), I is the number of decomposition, high frequency (short period) IMF i (t) represents the high time-varying nature in the original data, low frequency (long period) IMF i (t) represents the long-period characteristics in the original data. The specific process of EMD d...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an urban rail short-time passenger flow prediction method based on modal decomposition and deep learning. The urban rail short-time passenger flow prediction method comprises aCEEMDAN decomposition stage and a Fine-to-Colarse reconstruction stage and a GRU prediction stage. The invention provides an urban rail transit short-time traffic flow prediction model integrating aCEEMDAN and a GRU. The advantages of two models are fully combined, the CEEMDAN is utilized to decompose original passenger flow data into limited IMFs with local characteristics and different frequencies and allowances, so that the non-stable and non-linear characteristics of the data are obtained, and Fine-to-Colarse reconstruction method is introduced. The method comprises the following steps:reconstructing the IMFs by using a code reconstruction method; a decomposition result is reconstructed into a high-frequency item, a low-frequency item and a trend item, data information is fully reserved while the IMFs dimension is reduced, and finally the IMFs dimension is used as input of a GRU model to obtain time sequence characteristics of data and predict urban rail transit short-time traffic flow.

Description

technical field [0001] The invention relates to a method for predicting short-term passenger flow of urban rail based on modal decomposition and deep learning. Background technique [0002] Due to its large capacity, fast speed and low energy consumption, urban rail transit has become one of the main public transportation to alleviate traffic congestion and meet the travel needs of urban residents in metropolitan areas. Passenger flow forecast results are the main basic documents for urban rail transit operation and management. Among them, medium and long-term passenger flow forecast results are an important basis for urban rail transit planning and construction, while short-term passenger flow forecast results can support urban rail transit operation planning, passenger flow induction, and emergency management. Management and other decisions have important guiding significance for adjusting passenger travel behavior, reducing passenger congestion, and improving urban rail t...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/04G06Q50/26G06N3/04
Inventor 刘澜黄豪毛剑楠罗玥卢维科康雷雷
Owner SOUTHWEST JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap