分布式电梯群控系统多目标调度算法  

Multi-objective scheduling algorithm for distributed elevator group control system

作　　者：王遵彤[1] 孙栋[1] 乔非[1] 吴启迪[1] 

机构地区：[1]同济大学电子与信息工程学院,上海201804

出　　处：《控制理论与应用》2010年第5期602-608,共7页Control Theory & Applications

基　　金：国家自然科学基金资助项目(60674076)

摘　　要：基于多代理系统(MAS)的分布式电梯群控系统将电梯及群控器映射为具有不同功能的代理(agent),呼梯信号的分派通过各agent协商解决,使梯群调度算法的计算工作量分散到各agent.基于拟市场模型,分布式群控算法主要包括较厢代理(C-agent)报价算法及呼梯信号代理(HC-agent)电梯分派算法等.设计实现了旨在同时降低平均候梯时间、平均乘梯时间及长候梯率的多目标分布式群控算法DMO.仿真结果表明,基于MAS的分布式电梯群控系统是可行的,所设计的分布式群控算法能够使平均候梯时间、平均乘梯时间及长候梯率同时得到优化.The distributed elevator group control system is based on the multi-agent system（MAS）.The elevators and the group-controller are mapped respectively to different agents with specific functions;the hall calls are dispatched to elevators through negotiations among agents,thus the computation of scheduling algorithm is decentralized to different agents.Based on the market-like model,the distributed group-control algorithm includes mainly the quotation-generation algorithm of C-agents and the elevator dispatching algorithm of HC-agent.A distributed multi-objective group-control algorithm（DMO） is designed to optimize the average waiting time（AWT）,the average riding time（ART）,and the long waiting percentage（LPT）.Simulation results show that the proposed MAS-based distributed elevator group control system is feasible,and the DMO algorithm has high performance in optimizing AWT,ART,and LPT.

关 键 词：电梯群控 多代理系统(MAS) 分布式调度 多目标优化 

分 类 号：TU857[建筑科学]