Research Of The Prediction Model Of Traffic Flow Based On Lsvr

The transportation system is the infrastructure and "the circulatory system" for supporting the social economy development and holds the important status in the social economy system. Whether transportation questions are solved well or not will affect the development of the national economy and the enhancement of the lives quality directly. Prompt and accurate traffic flow prediction is the foundation of realizing the traffic flow control and induction. With forecast time span reducing, its non-linearity, time-dependent nature and uncertainty become more and more stronger. The orthodox prediction models such as History Average Model, Time-Series Model, Nonparametric Regressive Model, Kalman Filtering Model, neural Network Model and Combination Prediction Model can't predict well in effect and precision. This article proposes a forecast model using Lagrange Support Voter Regression (LSVR) algorithm which demonstrates more precise and quicker than BP Neutral Network Model through the simulation experiment.In the first chapter, we mainly introduces the development conditions of municipal transportation in our country and its question countermeasures, contents and the key question of Intelligence Transportation System, the research goal and the significance of this paper.In the second chapter, correlative knowledge of the traffic flow theory and current traffic flow forecast methods are introduced.In the next chapter, we study the traffic flow forecast model for BP Neutral Network. First, the base theory of Neutral Network is introduced, and then we study the algorithm of BP Neutral Network as well as the improved algorithm, and finally, traffic flow prediction model is established.In the fourth chapter, we study the traffic flow forecast model for Lagrange Support Vector Regression Machine. In this chapter, we introduce the mathematics rationale of SupportVector Machine (SVM)--Statistical Study Theory and norm SVM algorithm, and thenpromote it to Lagrange Support Vector Regression algorithm and establish forecast model. Finally, we carry through comprehensively and systematically contrasts between Lagrange Support Vector Regression and BP Neutral Network.