Application And Research Of Intelligent Algorithm In Intelligent Traffic Signal Control

The construction of the urban brain and the large-scale networking of the city enable the city’s managers to collect the information recorded by the massive vehicles.The location information of the vehicles in the city and the parking information of the community can be grasped in real time.The artificial intelligence deep learning can predict the traffic flow.Variety.Based on the existing models,a deep learning method in machine learning is proposed to establish a prediction model for traffic networks and intersections.Taking the traffic safety situation in China as the research object,taking the transportation network and intersection of Suzhou Industrial Park as an example,we try to use the deep learning method to scientifically analyze the original collected traffic network and intersection data,and the traffic network and The traffic at the intersection is predicted,and the green signal ratio,duration,period,and phase phase sequence of the intersection are given reasonable suggestions.LSTM avoids long-term dependency problems through deliberate design.Based on the data base of the past five units,it predicts the average running time,average speed,and total number of running vehicles of the next unit.The running time and average speed can best reflect the actual traffic capacity of a certain road,providing the most reliable reference for the vehicle to choose the road.The intersection traffic signal time division algorithm involves many traffic parameters and high applicability.It can comprehensively consider the key nodes,road conditions and phase laps,and further improve the optimization algorithm of this paper,and propose a more reasonable time division algorithm.The research data of this paper relies on the real traffic data of Suzhou Industrial Park.Some of its research results have been applied to the actual traffic management and achieved good results.For a single intersection,the data predicted by the LSTM model is used to classify the traffic signal with time division.There are a large number of intersections in the city.The intersections can be classified by rationalized classification algorithms,and then the traffic management scheme can be managed or allocated for each type of intersection to improve the efficiency of intersections.In this paper,by using the algorithm of partitioned cluster analysis,using real-time data,the 107 intersections of Suzhou Industrial Park are classified,and the similarity of intersections in clusters is measured by establishing the dissimilarity function.In the end,107 intersections in Suzhou Industrial Park are divided into three categories,each of which has a high similarity to each intersection within the ethnic group,and a large difference between the intersections outside the ethnic group.