论文部分内容阅读
通常在大系统中,全局信息优化的系统,其性能要高于局部信息优化系统.全局信息优化的算法由于大系统的复杂程度往往不可行.所以通常会用分布式算法来解决此类问题.在分布式算法中,为了获得更好的系统性能,要尽可能多的采用更多的信息信息交换,然而这样会带来信息网络的负担增大.本文在预测控制性能指标中引入通信代价,并提出了一种随着系统状态变化的通信网络拓扑切换方法.文中给出了该算法在供水管网动态模型中的仿真结果,表明本方法的可行性.
Generally, in a large system, the performance of a system with global information optimization is higher than that of a local information optimization system, which is usually not feasible due to the complexity of large systems, so distributed algorithms are usually used to solve such problems. In distributed algorithm, in order to get better system performance, more information exchange should be used as much as possible, however, this will increase the burden of information network.In this paper, the communication cost is introduced in predictive control performance index, And proposes a method to switch topology of communication network with the change of system state.The simulation results of this algorithm in the dynamic model of water supply network are given and the feasibility of this method is demonstrated.