A Short-term Traffic Flow Prediction Method Based on Multiple Phase Space in Spark Environment

Abstract

The invention discloses a short-time traffic flow prediction method based on multiphase space in a Spark environment. The method comprises following steps: step (1), acquiring real-time vehicle information of certain road section and sending the real-time vehicle information to a database HBase; step (2), acquiring historical traffic flow data according to historical vehicle information stored in the database HBase and storing the historical traffic flow data in a database; step (3), constructing a multiphase space model in the Spark environment, and predicting traffic flow data of a next period of time by the model in combination with the historical traffic flow data. With the adoption of the technical scheme, the multiphase space is constructed by multiple pairs of delay time and embedding dimensions, and a big data frame Spark technology is adopted, so that traffic flow of the next moment is predicted more scientifically and more accurately in real time.

Description

technical field [0001] The invention relates to the field of intelligent transportation systems, in particular to a short-term traffic real-time prediction method based on multi-phase space in a Spark environment. Background technique [0002] With the rapid development of our country's economy, the transportation industry is facing huge management pressure while bringing huge economic and social effects. A huge number of vehicles cause traffic congestion, aggravate air pollution, slow down traffic efficiency, and seriously affect people's travel. Accurate and timely use of traffic flow information for traffic flow prediction can help managers make reasonable traffic control plans and provide great reference value for people's travel decisions. The traffic flow system is a very complex nonlinear system, which conforms to the characteristics of chaos theory. Chaos theory, as a science that studies nonlinear systems over time, can better reflect the inherent laws of traffic ...

AI Technical Summary

Problems solved by technology

[0004] 2) The traditional algorithm mainly uses the shallow traffic forecasting model, but it does not consider the chaotic characteristics of the traffic system, and cannot adapt to the short-term traffic flow forecast with strong uncertainty

Although the traditional chaotic algorithm considers chaotic characteristics, it uses a single delay time and embedded dimension to make predictions, and the prediction accuracy is not ideal

Images