A Short-term Traffic Flow Forecasting Method Based on Periodic Component Extraction Technology

Abstract

The invention discloses a short-term traffic flow prediction method based on a periodic component extraction technology. The method includes the steps that (A) a cyclic matrix is built for a traffic flow time series waveform at daily data intervals; (B) the largest feature value and a feature vector corresponding to the largest feature value of the daily cyclic matrix obtained in the step (A) are calculated through a PCA technology, and the feature vector is converted to a time domain waveform; (C) the traffic flow series minus the daily cycle series is a residual waveform; (D) a week cyclic matrix is built for the residual waveform at weekly data intervals, and the step (B) and the step (C) are carried out repeatedly to obtain the weekly cycle series and a final residual waveform, wherein the traffic flow series is the sum of the daily cycle series, the weekly cycle series and a random disturbance component. A prediction model is utilized for processing the traffic flow series to obtain a final prediction result, and the technical effect of predicting the traffic flow efficiently and accurately is achieved.

Description

technical field [0001] The invention relates to the field of signal processing, in particular to a short-term traffic flow prediction method based on period component extraction technology. Background technique [0002] Short-term forecasting of traffic flow is an important part of traffic planning, traffic control and intelligent transportation system. Accurate short-term forecasting of traffic flow is of great value to the operation management, safety guidance and traffic planning of intelligent transportation. However, the problem of short-term traffic flow forecasting is full of challenges, mainly because the traffic flow waveform is a time-varying, random, and complex nonlinear system, and the traffic flow waveform is affected by social rhythm, economic prosperity, personal travel willingness, The influence of many external unknown factors such as weather and emergencies has been a hot research topic for many experts and scholars in recent decades. [0003] At present,...

AI Technical Summary

Problems solved by technology

[0006] The present invention provides a short-term traffic flow prediction method based on periodic component extraction technology, which can apply the PCA method to extract the daily periodic component and weekly periodic component in the traffic flow waveform, and then predict the separated components separately to remove Obtain more accurate prediction results, solve the technical problems of low prediction accuracy and efficiency in the existing traffic flow prediction methods, and realize the technical effect of efficient and accurate traffic flow prediction

Images