基于自适应在线遗传PID的监测网络拓扑控制研究  

The topology control of work-face monitor network based on self-adapting online genetic PID

作　　者：张晓光[1,2] 阮殿旭[1,2] 刘旭东 

机构地区：[1]中国矿业大学机电工程学院,江苏徐州221116 [2]西安电子科技大学综合业务网理论国家重点实验室,陕西西安710071 [3]枣庄矿业集团,山东枣庄277100

出　　处：《中国矿业大学学报》2012年第1期95-101,共7页Journal of China University of Mining & Technology

基　　金：综合业务网理论及关键技术国家重点实验室开放课题项目(ISN10-10);中国博士后科学基金项目项目(20060390277)

摘　　要：针对煤矿井下工作面环境复杂、无线传感器网络节点能量有限和通信易受干扰等实际特点,为了解决网络连通性、链路可靠性和能耗的问题,提出了一种基于自适应在线遗传PID的井下工作面无线传感器网络的拓扑控制算法.在局部平均算法的基础上,将控制算法和生物智能算法引入到WSN的拓扑控制中,可以克服现有的拓扑控制算法存在的收敛速度慢、算法不稳定等缺点,有效地提高能耗有效性和收敛速度.结果表明:将控制思想和人工智能引入到拓扑控制优化,与局部平均算法对比,节点平均启动能耗降低了84%,启动网络所有节点消耗的能量降低了60%～70%;启动时耗提高9.2%～12.7%,提高了收敛性和能效性.In an allusion to the actual characteristics of the complexity of underground environment, the energy limitation of wireless nodes and the interruption of communication, and in order to solve the problems of network connectivity, link reliability and energy cost, a topology control algorithm for wireless sensor network（WSN） based on the self-adapting online genetic PID （PID of self-adapting online genetic algorithm, SAOGA-PID） in the underground workface was proposed. Based on the local mean algorithm （LMA）, we introduced biological intelligence algorithm and closed-loop control theory to overcome the shortcomings of slow convergence and instability of the existing topology control algorithm, and to improve the energy efficiency and convergence rate. The results show that, compared with the LMA through the close-loop con- trol theory and biological intelligence algorithm, the average start-up energy consumption of the nodes and the start-up energy consumption of all nodes were reduced by 84% and 60% 70%respectively. The time consumption was increased by 9.2%-12.7%, and the conver gence and the energy-effeciency were improved as well.

关 键 词：自适应 遗传算法 无线传感器网络 拓扑控制 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]