Short-time road traffic congestion prediction method based on CS-SVR algorithm

Abstract

The invention discloses a short-time road traffic congestion prediction method based on a CS-SVR algorithm, and the algorithm belongs to the technical field of road congestion prediction. The method comprises the following steps that (1) original data is acquired and processed; (2) all parameters are initialized by adopting a Cuckoo Search (CS) algorithm; (3) a target function is built, and initial fitness calculation is carried out; (4) random walk is carried out, and a new bird nest position is calculated; (5) the optimal objective function value is updated; (6) whether a host bird changes or reserves the bird nest position or not when an external bird egg is found is judged, the global optimum position is selected, moreover, whether the number of iterations is reached or not is judged,and if the maximum iteration parameter is not reached, the operation is returned to the step S03, and iteration algebra is plus 1; and (7) if the maximum iteration algebra is reached or the precisionrequirement is met, SVR is used for carrying out traffic congestion prediction on a test sample.

Description

technical field [0001] The invention relates to a road traffic congestion flow prediction method, relates to a CS-SVR algorithm-based prediction for short-term road traffic flow, especially for traffic flow and speed prediction, and belongs to the technical field of road congestion prediction. Background technique [0002] With the development of my country's national economy and the acceleration of urbanization, more and more resources and labor force are flocking to cities to promote urban development. At the same time, the problem of traffic congestion is becoming more and more serious, especially in some big cities. Congestion not only increases the cost of urban investment and wastes a lot of energy, but also increases environmental pollution, damages people's health and brings mental distress, delays people's many things, reduces the efficiency of social activities, and causes large economic losses . Therefore, the prediction of traffic congestion, especially the predic...

AI Technical Summary

Problems solved by technology

[0002] With the development of my country's national economy and the acceleration of urbanization, more and more resources and labor are flocking to cities to promote urban development. At the same time, the problem of traffic congestion is becoming more and more serious, especially in some big cities. Congestion not only increases the cost of urban investment and wastes a lot of energy, but also increases environmental pollution, damages people's health and brings mental distress, delays people's many things, reduces the efficiency of social activities, and causes large economic losses

Images