A hybrid control for elevator group system

In order to increase the elevators running efficiency and quality of service, the optimizing control strategy of elevators is studied. In this paper a new hybrid control method which optimizes passenger service in an elevator group is described. It is capable of optimizing the neural-controller based on Particle Swarm Optimization (PSO) of an elevator group controller. Starting from the operation characteristics of elevator group control system, the architecture and the traffic pattern of an elevator group control system are described, and the optimization cost criterion function is proposed. The PSO algorithm is used to optimize the weights and biases of the neural network. Some weighted parameters of the Radial Basis Function (RBF) neural network can be modified based on the PSO, so that different weight settings and their influence on the elevator supervisory group control (ESGC) performance can be tested. It can reduce the passenger's average waiting time by allocating an appropriate number of elevator cars to the lobby floor. The results prove that the hybrid method is effective.

[1]  Yoshiaki Ichikawa,et al.  A parameter tuning method for an elevator group control system using a genetic algorithm , 1998 .

[2]  Andrew G. Barto,et al.  Elevator Group Control Using Multiple Reinforcement Learning Agents , 1998, Machine Learning.

[3]  I. Parry,et al.  A parameter identification self-adaptive control system , 1970 .

[4]  Li-Chen Fu,et al.  Dynamic scheduling approach to group control of elevator systems with learning ability , 2000, Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No.00CH37065).

[5]  Shinji Kusumoto,et al.  An approach to safety verification of object-oriented design specification for an elevator control system , 1997, Proceedings Third International Workshop on Object-Oriented Real-Time Dependable Systems.

[6]  H. Nguyen,et al.  A comparative evaluation of fine regenerative and nonregenerative vector controlled drives for AC gearless elevators , 2000, Conference Record of the 2000 IEEE Industry Applications Conference. Thirty-Fifth IAS Annual Meeting and World Conference on Industrial Applications of Electrical Energy (Cat. No.00CH37129).

[7]  Kotaro Hirasawa,et al.  Elevator Group Control Using Multiagent Task-Oriented Reinforcement Learning , 2005 .

[8]  Christian Magele,et al.  Particle swarm optimisation for Pareto optimal solutions in electromagnetic shape design , 2004 .

[9]  Sandor Markon,et al.  Basic study of optimal design of linear motor for rope-less elevator , 2008 .

[10]  U. Baumgartner,et al.  Particle swarm optimization - mass-spring system analogon , 2002 .