Short-term traffic flow forecasting method

Abstract

The invention relates to the field of intelligent transportation forecasting and in particular relates to a short-term traffic flow forecasting method. The short-term traffic flow forecasting method comprises the steps of 1), collecting data of flows and speeds of historical records and presetting traffic states according to combinations formed by the flows and the speeds; 2), building a forecasting model; 3), inputting the traffic states of two historical moments x(t-1) and x(t) corresponding to actual forecasting traffic states, scheduling the forecasting model for forecasting and outputting a forecasting state; and 4), recording an actual state, calculating an error between the forecasting state and the actual state, and recording and saving an error result. The short-term traffic flow forecasting method has the beneficial effects that the method is low in calculation complexity and capable of reflecting road condition changes in real time; meanwhile, the historical data of 10 minutes near the future moment are only taken into account by the algorithm; the method is good in real-time performance; a 5-minute forecasting time duration meets delay of transmission and processing and is applicable to real-time display of a traffic induction screen; the forecasting model is simple and quick to build, applicable to traffic road conditions with various complexities and high in forecasting precision.

Description

technical field [0001] The invention relates to the field of intelligent traffic prediction, in particular to a short-term traffic flow prediction method. Background technique [0002] With the development of urbanization and the increasing popularity of automobiles, the traffic environment has deteriorated, traffic congestion has increased, and traffic accidents have occurred frequently. Traffic problems are serious social problems both in developed countries and in developing countries. The loss caused by traffic problems in the United States reaches 237 billion US dollars every year, and the direct economic loss caused by traffic congestion to the society in large cities with more than one million people in my country is about 160 billion yuan, which is equivalent to 3.2% of the GDP. As we all know, the most effective way to solve traffic congestion is to increase the traffic capacity of the road network. However, the space available for construction is limited, and the...

AI Technical Summary

Problems solved by technology

However, the space available for construction is limited, and the preparation of construction funds is also a huge problem

In addition, since the transportation system is a rather complex large system, it is difficult to fundamentally solve the problem from the perspective of vehicles or roads alone.

Images