摘要
根据目的层预约电梯系统的新型客流分配模式,分析并归纳出目的层预约情况下电梯群控系统的特征属性,提出一种多目标模糊神经网络电梯群控策略.把目的层预约电梯系统的特征属性作为多目标模糊神经网络的输入,建立并训练模糊神经网络,最后利用该模糊神经网络进行派梯计算.仿真试验证明该群控策略可得到满意的效果.与传统电梯群控策略以及其他目的层预约电梯群控策略的对比试验证明,该方法在提高电梯平均系统时间、长候梯率、长乘梯率及节约能耗等方面都体现出较好的性能优化效果,对高峰交通流模式下的电梯群控系统效果显著.
This paper analyzes and sums up the characteristic attributes of elevator group-control system with destination registration according to the new flow distribution modes of the elevator system, and proposes a new elevator group-control policy based on multi-objective fuzzy neural network. The proposed policy regards the characteristic attributes of the elevator system as the inputs to establish a multi-objective fuzzy neural network, which is then trained and used as the elevator group-control calculator. This novel elevator group-control calculator can be used to dispatch elevators. Simulated results show that the pared with the traditional elevator group-control policy tion registration, the ding ratio and power proposed policy behaves better in consumption, especially in down policy is of excellent performance. Moreover, as comand some other elevator group-control policies with destinaaverage waiting and riding time, long-waiting ratio, long-ripeak trafiqc pattern.
出处
《华南理工大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2007年第1期13-18,共6页
Journal of South China University of Technology(Natural Science Edition)
基金
广东省自然科学基金资助项目(06025724)
关键词
目的层预约
多目标
模糊神经网络
电梯群控系统
destination registration
multiple objective
fuzzy neural network
elevator group-control system