Design and implementation of modern elevator group control system

The design and implementation of a modern elevator group control system (EGCS) is introduced in this paper. The basic considerations of designing an EGCS are discussed, including related system parameters, evaluation criterions and traffic patterns. Least squares support vector machine algorithm is employed for traffic prediction. Using multi-support vector machine, the traffic pattern recognition is accomplished, then based on that recognition, the control strategies are generated. In the hall call assignment, the re-scheduling ability is achieved by the elevator suitability re-evaluation mechanism. A comparison with an older model demonstrates the effectiveness of the proposed EGCS.