基于非合作博弈的网络化控制系统传感器采样周期调度研究  

Sensor sampling period scheduling research of networked control systems based on non-cooperative game

作　　者：付丽霞[1] 王成功[1] 张勇[1] 毛剑琳[1] Fu Lixia Wang Chenggong Zhang Yong Mao Jianlin(Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, Chin)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南昆明650500

出　　处：《计算机与应用化学》2017年第7期555-558,共4页Computers and Applied Chemistry

基　　金：国家自然科学基金资助项目(61163051);云南省教育厅科学研究基金项目(2015Y071)

摘　　要：针对采用交换式以太网的网络化控制系统的数据传输和控制性能优化问题,提出了控制回路根据非合作博弈的战略式表述调整传感器采样周期的调度优化方法。首先,将非合作博弈理论引入网络化控制系统,将不同控制回路对网络资源和同一控制器资源的竞争关系描述为非合作博弈,建立了基于非合作博弈的网络化控制系统模型。然后,使用True Time2.0工具箱搭建了基于非合作博弈的网络化控制系统仿真平台,对单控制回路和双控制回路情况下控制回路的传感器选择不同的采样周期时控制回路的实际控制性能进行量化,得到了相应的战略式表述。最后,在此仿真平台上对本文所提方法进行了仿真研究,并在此基础上加入了帕累托优化策略进行进一步优化,仿真结果表明:不同的控制回路能够根据非合作博弈的战略式表述令传感器节点选择最合理的采样周期以提升自身的控制性能,且此过程也优化了系统的数据传输,而帕累托优化能够在博弈达到纳什均衡后进一步提升系统的控制性能。此方法可为实际的工程应用提供一定的参考价值。Aiming at the problem of data transmission and control performance optimization of the Networked Control System which using Switching Ethemet, the paper proposed a scheduling optimization method which control-loop using strategic form representation of non-cooperative game to adjust the sensor sampling period. Firstly, the non-cooperative game theory is introduced into the Networked Control System, and the competition relationship of the different control loop to the network resource and the same controller resource is described as the non-cooperative game, a Networked Control System model based on the non-cooperative game is established. Then, using the TmeTime2.0 toolbox to build a simulation platform of Networked Control System which based on non-cooperative game, to quantify the actual control performance of control loop when the sensor in the control loop select different sampling period for single control loop and double control loop, then get the corresponding strategic form representation. Finally, make simulation of the method proposed in this paper on this simulation platform, and joined the Pareto Improvement strategy for further optimization. The simulation results show that different control loops can make the sensor nodes to select the most reasonable sampling period based on the non-cooperative game strategic form representation so that to enhance their control performance, and this process also optimized the system data transmission, and the Pareto Improvement can further enhance the control performance of the system when it achieved the game to Nash equilibrium. This method can offer some reference value for practical engineering application.

关 键 词：TRUETIME 非合作博弈 网络化控制系统 调度优化 帕累托优化 

分 类 号：TP272[自动化与计算机技术—检测技术与自动化装置]