A fast calibration method of traffic distribution model parameters based on multi-source data fusion

Abstract

The present invention discloses a traffic distribution model parameter fast checking method based on multi-source data fusion, which respectively adopts a bottom-up parameter checking method and a top-down parameter checking method, and parameter checking and calibration is divided into two stages The parameters obtained in the first stage are a1, b1, c1, the parameters obtained in the second stage are a3, b3, c3, and the final model parameters are the averages a4, b4, c4 of the two methods; multi-source The combined use of data and survey data; the application of heuristic algorithms can realize automatic iteration and calibration of parameters; the accuracy of the model is greatly improved, the parameter values ​​are more realistic, and the reality can be better simulated.

Description

technical field [0001] The invention belongs to the field of rail transit, and in particular relates to a method for quickly checking traffic distribution model parameters based on multi-source data fusion. Background technique [0002] Passenger flow forecasting is a key link in rail transit network planning, which directly affects the necessity and layout of rail transit network planning, and the traffic distribution model directly affects the accuracy of passenger flow forecast results. As a classic traffic distribution model, the gravity model has been widely used in various transportation modes such as high-speed railways, intercity railways, urban railways, urban rail transit, and roads. Considering that the three parameters in the gravity model are affected by multidimensional factors such as urban heterogeneity, distribution of functional areas, and job-housing balance, model parameter calibration must be carried out for specific research objects. [0003] The curre...

AI Technical Summary

Problems solved by technology

Judging from the survey data of more than 20 cities completed by our institute, such a low sampling rate leads to the lack of data in many communities, and the expansion of sampling will amplify this error, but we still use this data for model prediction

[0007] (2) In the actual comparison of TLD and OTLD data, the value of the distance segment is relatively large, and the data usually has a high degree of fitting. However, once the distance is short and the value is small, the data fitting error increases, which is mainly due to the lack of data in some traffic areas caused by

[0008] (3) At present, the development of big data is in full swing, especially mobile phone signaling data. Many cities can obtain od at the traffic center level through mobile phone data, and the sampling rate can usually reach 60% or higher. However, due to the density of base stations and mobile phone switching The ping-pong effect of the traffic area leads to insufficient accuracy of the traffic area data. It is worth studying how to apply this data to the model.

[0009] (4) The existing gravity model parameter calibration methods are mostly based on data surveys, and do not integrate big data such as mobile phones and bus swiping cards generated by residents in China; in addition, they are mostly limited to the theoretical level, and seldom consider the calibration speed of model parameters

Images