This paper focuses on optimizing algorithms of signal timing of urban intersecti on based on real-time traffic volume under relatively fixed periods. Firstly o ptimizing function is made to minimize total delay of all approaches of all phas es. Then non-linear programming model with restrictions of saturation and minim um green light time is constructed. And then genetic algorithm is applied to sol ve this model and the best scheme of signal timing can be obtained. The process of constructing and solving the function is showed in a simulation experiment. I n this experiment the optimization with no regard of the restriction of saturati on and the optimization with the target of minimum total delay of all key approa ches of all phases are compared. The result shows that the restriction of satura tion should be considered and the goal function should be constructed from the t arget of minimum total delay of all approaches of all phases in order to ensure correctness and rationality of optimization.
Computer and Communications