摘要
为了提高火灾救援的效率,消防人员逐渐使用无人机来进行火灾态势感知和监视;但无人机的费用造价高昂;一台配备了无线电中继器或视频和遥测功能的混合动力无人机预计成本约为1万美元;因此为了达到经济最大化和效率最优,文章采用多目标规划模型进行优化;该模型主要考虑经济和效率两个目标,然后设置约束条件来进行求解;遗传算法和基于数学规划的方法是国内求解帕累托前沿解的主流算法[1];应用NSGA-Ⅱ算法解决无人机排列问题;以决策变量无人机的数量组合编码作为运算对象,可以直接对集合、序列、矩阵、树、图等结构对象进行运算操作[2];这样的方式一方面有助于模拟生物的基因、染色体和遗传进化的过程,方便遗传操作算子的运用,合理且准确地给出了无人机配置方案,为有关消防部门规划提供参考,另一方面也使得遗传算法具有广泛的应用领域,如函数优化、生产调度等领域。
In order to improve the efficiency of fire rescue,firefighters are increasingly using drones for fire situational awareness and monitoring.But drones are expensive.A hybrid drone equipped with a radio repeater or video and telemetry capabilities is expected to cost about$10,000.Therefore,in order to achieve the maximum economy and efficiency,I adopted the multi-objective programming model.The model mainly considers two objectives of economy and efficiency,and then sets constraint conditions to solve.Genetic algorithm and mathematical programming-based methods are the mainstream algorithms for solving Pareto frontier solutions in China[1].NSGA-Ⅱ algorithm is applied to solve the problem of UAV alignment.Taking the number combination coding of the decision variable UAV as the operational object,it can directly operate on structural objects such as sets,sequences,matrices,graphs[2].On the one hand,this method is helpful to simulate the process of gene,chromosome and genetic evolution of organisms and facilitate the use of genetic operators.Reasonable and accurate UAV configuration scheme is given,which provides reference for the planning of relevant fire departments.On the other hand,genetic algorithm has a wide range of applications,such as function optimization,production scheduling and other fields.
作者
邱雪
王永忠
赵志
杨传军
李佳骏
QIU Xue;WANG Yongzhong;ZHAO Zhi;YANG Chuanjun;LI Jiajun(Civil Aviation Flight University of China,Guanghan 618307,China)
出处
《计算机测量与控制》
2021年第10期211-216,共6页
Computer Measurement &Control
基金
交通运输工程优势特色学科建设(D202103)
四川省大学生创新创业训练项目(S202110624214)。
关键词
森林防火
无人机组合
多目标规划
遗传算法
forest fire prevention
UAV combination
multi-objective programming
genetic algorithm
