基于Petri网的无人机可充电传感器网络优化  

Petri-Net Based Rechargeable Sensor Network Optimization with UAV

作　　者：秦怀宇[1,2] 白雪 赵不贿[1] 徐雷钧[1] QIN Huaiyu;BAI Xue;ZHAO Buhui;XU Lejun(School of Electrical and Information Engineering,Jiangsu University,Zhenjiang 212100,China;School of Electrical and Information,Jiangsu University of Science and Technology,Zhenjiang 212013,China)

机构地区：[1]江苏大学日电气信息工程学院,江苏镇江212100 [2]江苏科技大学电子信息学院,江苏镇江212013

出　　处：《控制工程》2023年第9期1575-1584,共10页Control Engineering of China

基　　金：国家自然科学基金资助项目(61874050);镇江市重点研发计划项目(GY2021006)。

摘　　要：为解决地面移动充电车难避障、易受干扰,以及数据收集受充电时间影响大的问题,提出包含一架数据收集无人机(unmanned aerial vehicle,UAV)和多架充电无人机的网络架构。首次提出广义同步连续自控Petri网系统(generalized synchronizing continuous cyber Petri net system,GSCCPNS),用于建立无人机的能量传输和运动控制模型,从可视化和数学两方面刻画网络的决策控制、能量流动和飞行轨迹。首先,利用等边三角形簇头覆盖法确定网络锚节点数量和节点总体能耗。然后,为了使无人机充电收益最大化,在传统模拟退火算法的基础上,引入局部搜索和交叉算子,并结合Petri网提出Petri模拟退火(Petri simulated annealing,Petri-SA)算法。大规模仿真表明,死节点比例降低11%~26%,飞行能耗降低7%~17%。In order to solve the problems that mobile charging vehicles are difficult to avoid obstacles,vulnerable to interference,and data collection is greatly affected by charging time,a network architecture is proposed,which includes a data collection UAV and several wireless charging UAVs.For the first time,a generalized synchronous continuous cyber Petri net system(GSCCPNS)is presented to establish the energy transfer and motion control model of UAV,and the decision control,energy flow and flight trajectory of the network are characterized visually and mathematically.Firstly,the number of anchor nodes and the total energy consumption are determined by using the equilateral triangle cluster head covering method.Then,combined with Petri net,the Petri simulated annealing algorithm(Petri-SA)is proposed by introducing local search and crossover operator to obtain the flight path to maximize the charging revenue of UAV.Finally,the large-scale simulation shows that the dead node ratio is reduced by 11% to 26%,and the flight energy consumption is reduced by 7% to 17%.

关 键 词：可充电传感器网络 无人机 PETRI网 Petri模拟退火算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]