Reseach on Optimal Control of Elevator Group Based upon Ant Colony Algorithm

Nowdays, people’s requirements of quality of elevators service increased, a single elevator in the building have been unable to meet the transport needs, and to install elevators become the first choice, resulting in the elevator group control system. The elevator group control system can effectively improve transportation effects, so international elevator industry give it a high degree of attention, but the the elevator control system and the group control system in domestic high-level and high-rise buildings are basically foreign elevator company’s products. Therefore, the research and development of elevator group control system is extremely urgent for domestic enterprises.As the core of the elevator control system group, the study group control algorithm is a remarkable subject. Although domestic and foreign experts and scholars has raised a variety of solutions on this issue, but they all have advantages and disadvantages. This paper was the first to use ant colony algorithms to solve elevator group control problems. Ant colony algorithm is a new type of evolutionary simulation algorithm, which was proposed in the past 10 years. It seeks the optimal answer from the colony evolution process which includes all possible answers. Ant colony algorithm directs the algorithm to the most optimal direction with combination of positive feedback and negative feedback, and keeps the searching range from stopping earlier, in this situation gets the satisfied answer in some degree. Ant algorithm is very suitable for useing in a multi-objective, non-linear, the uncertainty of the elevator group control problem.However, ant colony algorithm to be able map structure described. In light of this, a bipartite graph based on elevator landing and elevator group is designed in this paper. Here, the elevator group control issue is abstracted, so the issue of elevator group control becomes the biggest match of bipartite graph issues. The fundamental problem of elevator group control is multi-objective optimization problem, this paper will put the control objectives as passengers waiting time, riding time,and the number of passenger in elevater,energy consumption, through a combination of objective way to a weighted combination of function, This function is set to be the length of the bipartite graph, the ant colony algorithm can search for the best route.Use ant colony algorithm generates the optimal deployment of the elevaters.The paper simulated group control system in all aspects in Matlab environment to reached the algorithm simulation purposes. in this paper, the ant colony algorithms and other methods of the elevater’s deployment contracted to prove that the ant colony algorithm used in elevator group control is superioy, especially when the flow of passengers were intensive.This paper did not only study the ant algorithm used in elevator group control problem, but also provided an important opportunity for ant algorithms to be used in such multi-objective optimization issues. Therefore, ant algorithms can be used in such as economic scheduling problem, arranging schedule issues, hydro powe scheduling problems, such kinds of deployment of multi-objective optimization scheduling issuesby using this set of multi-objective optimization methods and the ideological bipartite graph model.