Elevator Group Control Algorithm And Network Integeration

The research focus on the elevator group algorithm, and a network integration of the elevator control system is simultaneously introduced.Since it is a complex, nonlinear, uncertain, multi-objective and random decision-making problem with which the elevator group control system deals, the conception of statistic approximation is introduced. Considering synthetically the elevator running characteristics, a new evaluating function for the elevator group control system (EGCS) is suggested. A kind of fuzzy neural network is used in the elevator group algorithm. The fitness of each elevator servicing the hall call signal is deduced, so the elevator is dispatched reasonably.Based on many references, a evaluating function is proposed with performance index of hall call Waiting Time(HWT), people number in a car, energy consuming. The statistic approximation algorithm for HWT is introduced, based on the analysis of elevator traffic state, the calculation of traveling distance and stop number is explained in detail in the thesis. According to the characteristics of the elevator, a group of elevator teaching signals are constructed, by which the weight coefficients are trained according to the Widrow-Hoff rule. The algorithm is simulated by VB, the simulation results prove the algorithm is feasible.The network integration of elevator controlling system is based Ethernet on backbone network, a distributed controlling structure is adopted in the elevator control system and terminal devices are connected with Lonworks. The controlling functions are devided into several parts, which are completed by Lonworks and PLC respectively. The function of Lonworks on network , communication and control are fully exerted.