基于WSNs节点故障定位在智能配电网研究

发布时间：2018-02-08 09:32

本文关键词： 智能配电网无线传感器网络故障定位 PSO算法 APIT算法　出处：《安徽工程大学》2017年硕士论文　论文类型：学位论文

【摘要】：智能配电网是世界电力建设的发展方向,传统的配电网发生故障造成大面积停电停产时依据经验有时能判断出故障原因却不能定位故障位置,造成经济损失严重。无线传感网的发展为智能电网故障定位提供新的发展方向。无线传感器网络因为其自动组网性、灵敏性、低能耗、成本低、可靠性较高等优势,能适应配电网复杂的网络架构,对智能配电网系统出现的故障进行精确定位,提高智能配电网定位检测故障效率。国内外研究学者正逐渐将WSNs应用在智能配电网的故障定位研究。本文针对智能配电网故障定位问题,采用无线传感网应用智能配电网对故障定位进行进一步的研究。本文总结了智能配电网电力系统的结构特点,配电终端设备、配电线路特点和通信结构,阐述了数据通信要求并分析设备终端数据通信要求和配电网电力系统结构,构建灵活的基于智能电网通信的无线传感网通信架构。增加智能配电网节点间通信的有效性和可靠性,保障智能电网的通信架构和无线传感器网络的通信架构相适应。本文针对DV-hop算法定位精度较低,分簇不均基于DV-hop算法来改进均值粒子群算法(PDV-hop,Particle DV-HopAlgorithm),在保持通信开销和硬件成本不增加的前提下,改进均值粒子群算法中粒子的速度与位移,有效地提高了分簇定位的精确度后,提高了数据的传输速率。本文对传统的PIT和APIT算法受邻节点分布疏密程度影响大,邻节点单位面积数量较小时,锚节点组成三角形数量较少,出现In-to-Out error和Out-to-In error影响定位精度,导致定位精度较低。在保持通信开销和硬件成本不增加的前提下,改进RSSI测距权重值,在保持较高精确度的情况下来确定测量次数,达到最优的测距能力,减少了数据的迭代次数。通过采用以未知节点坐标为圆心,锚节点组成的三角形求出外接球,确定外接球圆心即未知节点坐标,在确定每一个未知节点坐标后标记为已知锚节点,提高了定位节点的覆盖率,提高了定位精度,减少数据迭代次数。通过多次仿真实验结果对比分析,改进后的PDV-hoP算法和改进APIT算法,重新定位簇头节点坐标更加接近真实值,有效地提高了分簇定位的精确度后,定位节点精确度和覆盖率有显著提高。设计WSNs智能配电网硬件平台,对传感器节点硬件模块进行相应功能设计,应用Zigbee技术和STC12C5A32S2微处理芯片低功耗的优点,能够应用在智能电网完成对故障定位,信息传输等要求。将定位算法应用在建立的WSNs硬件平台模型,以温度为例模拟数据传输实验和坐标定位实验,实验结果可知设计的传感器节点硬件在WSNs智能配电网坐标定位的数据传输可靠性、定位精确度等方面都有提高。
[Abstract]:The smart distribution network is the development direction of the world electric power construction. When the traditional distribution network occurs the fault and causes the large-scale power outage and stops the production, it can judge the fault cause but can not locate the fault location according to the experience sometimes. The development of wireless sensor network provides a new development direction for fault location of smart grid. Wireless sensor network has the advantages of automatic networking, sensitivity, low energy consumption, low cost, high reliability and so on. Can adapt to the complex network architecture of the distribution network, and accurately locate the faults in the smart distribution network system. In order to improve the fault detection efficiency of smart distribution network, researchers at home and abroad are gradually applying WSNs to the fault location of smart distribution network. This paper aims at the problem of fault location in smart distribution network. The application of smart distribution network to fault location is further studied by wireless sensor network. This paper summarizes the structural characteristics of intelligent distribution network power system, distribution terminal equipment, distribution line characteristics and communication structure. The requirement of data communication and the structure of distribution network power system are analyzed. A flexible wireless sensor network communication architecture based on smart grid communication is constructed to increase the effectiveness and reliability of communication between smart distribution network nodes. The communication architecture of smart grid is suitable for wireless sensor network. In this paper, the location accuracy of DV-hop algorithm is low. Clustering unevenness is based on DV-hop algorithm to improve the mean particle swarm optimization algorithm (PDV-hopper Particle DV-Hopalgorithm). Without increasing communication overhead and hardware cost, the speed and displacement of particles in the mean particle swarm optimization algorithm are improved, and the accuracy of clustering location is improved effectively. In this paper, the traditional PIT and APIT algorithms are greatly affected by the distribution density of adjacent nodes, the number of unit area of adjacent nodes is small, the number of anchor nodes is less, and the In-to-Out error and Out-to-In error affect the positioning accuracy. Under the condition that the communication overhead and hardware cost are not increased, the weight value of RSSI ranging is improved to determine the measurement times under the condition of high accuracy, so as to reach the optimal ranging capability. The number of iterations of the data is reduced. By using a triangle with unknown node coordinates as the center and the anchor node forming a triangle to get the ball out, the outer ball center is determined as the unknown node coordinate. After determining the coordinates of each unknown node, it is marked as a known anchor node, which improves the coverage of the location node, improves the positioning accuracy and reduces the number of data iterations. After the improved PDV-hoP algorithm and improved APIT algorithm, the coordinate of cluster head node is closer to the real value, and the accuracy of clustering localization is improved effectively. The accuracy and coverage of location nodes have been improved significantly. The design of WSNs smart distribution network hardware platform, the corresponding function design of sensor node hardware module, the application of Zigbee technology and the advantages of low power consumption of STC12C5A32S2 microprocessing chip, It can be used in smart grid to fulfill the requirements of fault location, information transmission, etc. The localization algorithm is applied to the WSNs hardware platform model, and temperature is taken as an example to simulate the data transmission experiment and coordinate location experiment. The experimental results show that the sensor node hardware can improve the data transmission reliability and positioning accuracy of WSNs intelligent distribution network coordinate positioning.
【学位授予单位】：安徽工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM76;TP212.9;TN929.5

【相似文献】