论文总字数：31796字

摘 要

无线传感器网络目前广泛应用于通信、交通等多个领域，网络中的节点通常由能量受限的电池供电，是网络生命周期受限的因素之一。随着近年来无人机技术的发展，相较于使用静态能量发射器的传统无线充电系统，采用无人机为无线传感器网络充电，具有高机动性的优势，与网络各节点之间亦有着较为良好的视线关系，可减少网络充电配置成本并提高充电效率。因此，使用无人机为无线传感器网络节点充电成为了近年来一项热门的研究课题。

本文针对三维空间内单一无人机为无线传感器网络的多个可充电节点进行广播式充电的场景展开研究。根据各个节点在充电周期内的耗能需求，将优化问题分为无人机可/不可满足所有节点耗能需求的两类情况，并在可满足耗能需求的情况下进一步分为节点初始低电/高电量储备的两种情形。

在节点耗能需求可轻松满足的低电储备情形下，本文提出了一种单点悬停充电策略。而针对其他情形，首先在忽略无人机速度约束的条件下，通过节点耗能需求逼近或坐标范围网格划分两种方法求解无人机悬停点，继而将各悬停点的悬停时间作为变量，进行线性规划，提出了一种多点悬停的充电策略。最后考虑无人机速度约束，以TSP（Travelling Salesman Problem，TSP）问题的求解为基础，提出了一种连续悬停-飞行的充电策略。

本文通过MATLAB和LINGO实现了算法，并验证了各类情形的充电性能：（1）若无法满足所有节点的耗能需求，可最大化所有节点耗能需求满足比例中的最小值；（2）若可以满足所有节点的耗能需求，则在满足该条件的情况下，实现（a）节点低电量储备情形下，可最大化无人机的总广播输送能量值，以提高各个节点的电量储备，（b）节点高电量储备情形下，可最小化无人机充电时长以减少无人机飞行与充电耗能并减少充电能量溢出、提高有效能量传输比例。

关键词：无线传感器网络，无人机辅助充电，线性规划，MATLAB，LINGO

Abstract

WSNs (Wireless Sensor Networks, WSNs) are widely utilized in many fields such as communication and transportation. However, nodes in the network are still powered by energy-constrained batteries, which has limited the network life cycle. With the development of UAV (Unmanned Aerial Vehicle, UAV) technology in recent years, compared with the traditional wireless charging system using static energy transmitters, the UAV has the advantage of high mobility. There is also a well-preserved line of sight relationship with each node, which can reduce the network charging configuration cost and improve the charging efficiency. Therefore, the implementation of UAV to charge WSNs has become a promising research in recent years.

In this thesis, we study the scene of a single UAV wireless charging WSNs with multiple randomly distributed nodes on the ground. Each node consumes energy in the charging cycle which is called the energy consumption demand. The optimization problem is divided into two situations where the UAV can/cannot satisfy the energy consumption demands of all nodes, and the latter case can be further divided into two cases of initial low/high power reserve of the nodes. In the case of low power reserve where the energy consumption demand can be easily satisfied, this paper proposes a single-point hover charging strategy. For other cases, ignoring the UAV speed constraint, this paper firstly solves the UAV hovering point by the energy consumption demand approaching method or the coordinate range meshing method. Sets of hovering time are used as variables to perform linear programming, and a multi-point hover charging strategy is proposed. Finally, with the UAV speed constraint, a continuous hover-flight charging strategy is proposed based on the TSP (Travelling Salesman Problem, TSP) problem.

In addition, we accomplish the algorithm with MATLAB and LINGO and verify the charging performance of various cases: (1)If the consumption demands of all nodes cannot be satisfied simultaneously, the minimum ratio of the energy received among all nodes can be maximized; (2)The demand will be guaranteed if it could be. Furthermore, the designed strategy could (a)maximize the broadcast transmission energy to increase the node power reserve in the low power reserve case; (b)minimize the charging time of the UAV to decrease its own energy consumption and to reduce the energy overflow in the high power reserve case.

KEY WORDS: WSNs, UAV-enabled wireless charging, linear programming, MATLAB, LINGO

目 录

摘 要 I

Abstract II

第一章 绪论 1

1.1 引言 1

1.2 研究现状 1

1.2.1 基于节点充电需求的无线传感器网络充电策略 2

1.2.2 基于最大化无人机能量输送的无线传感器网络充电策略 2

1.3 研究内容与章节安排 3

第二章 系统模型 5

2.1 研究场景与模型构建 5

2.2 问题描述 6

2.3 本章小结 8

第三章 算法实现 9

3.1 忽略无人机速度约束下的低电情形优化问题求解 9

3.1.1 最优情况求解 9

3.1.2 次优情况的数学求解 12

3.1.3 基于节点耗能需求逼近的无人机悬停位置选择与悬停时间线性规划 14

3.1.4 基于坐标范围网格划分的悬停位置穷举与悬停时间线性规划 16

3.2 忽略无人机速度约束下的高电情形优化问题求解 18

3.3 忽略无人机速度约束下的节点耗能需求不满足情况的优化问题求解 18

3.4 考虑无人机速度约束下的连续悬停-飞行轨迹设计 19

3.5 算法小结 22

第四章 仿真与结果分析 23

4.1 低电最优情形下的单点悬停策略仿真 23

4.1.1 双充电节点场景 23

4.1.2 多充电节点场景 24

4.2 忽略无人机速度约束下的多点悬停策略仿真 25

4.2.1 基于可充电节点耗能需求逼近的无人机悬停位置选取 26

4.2.2 基于坐标范围网格划分的悬停位置穷举 28

4.3 无人机速度约束存在情况下的连续悬停-飞行策略仿真 30

4.4 高电情形仿真 32

4.5 仿真情况小结 32

第五章 结论与展望 34

5.1 结论 34

5.2 展望 34

参考文献 36

致 谢 39

第一章 绪论

1.1 引言

无线传感器网络（Wireless Sensor Networks，WSNs）由具有感知能力、通讯能力与计算能力的微型传感器节点构成，是目前在通信、交通领域有着广泛应用^[1]-[3]的一项重要技术。然而，由于无线传感器网络中各个节点仍然由能量有限的电池供电，其网络的生命周期也因此受到了限制。如何为能量受限的无线传感器网络提供持续可靠的电力补充并改进整个网络的充电效率成为了一个亟待解决的问题。

剩余内容已隐藏，请支付后下载全文，论文总字数：31796字