A real-time detection method for expressway abnormal events based on mobile phone data

Abstract

The invention discloses a mobile-phone-data-based real-time detection method for an abnormal event of a highway. The method comprises: a real-vehicle experiment is carried out to determine a switching point position of a highway and the highway is divided into a plurality of sections; mobile phone access amounts of section base stations in an abnormal event state and a normal state are extracted respectively; according to the mobile phone access amounts in the normal state, a time sequence model is constructed and numbers of mobile phone getting access to base stations in an abnormal event state are predicted; an abnormal event occurrence threshold value is calculated; and on the basis of the time sequence model, a number of mobile phone access to a base station at a future time is predicted in real time and an abnormal event determination index is calculated, and the abnormal event determination index is compared with the abnormal event occurrence threshold value to determine whether an abnormal event occurs. With the method, an abnormal traffic event can be identified accurately and a reliable decision-making basis is provided for the management staff.

Description

technical field [0001] The invention specifically relates to a method for real-time detection of abnormal events on expressways, in particular to a method for detecting abnormal events on expressways in real time using mobile phone data, and belongs to the technical field of intelligent transportation. Background technique [0002] With the rapid development of our country's economy, the traffic volume of expressways has been increasing, and a series of traffic problems have arisen, including: environmental pollution, traffic accidents, traffic congestion and so on. Abnormal traffic events will reduce the traffic capacity of local road sections, thereby affecting the overall operating efficiency of the road. Rapid and accurate identification of traffic events can provide event information for the traffic operation management center, facilitate the formulation of corresponding traffic management and control measures, and improve the operating efficiency and service level of e...

AI Technical Summary

Problems solved by technology

[0003] At present, the traffic operation and management center mainly detects abnormal events through manual inspection and fixed detectors, which is time-consuming and laborious.

Moreover, the layout of highway detectors is relatively sparse, and it is difficult to accurately identify abnormal events that occur at a long distance.

Images