基于免疫机理的无线传感器网络故障诊断研究

发布时间：2018-02-22 14:26

本文关键词： 无线传感器网络免疫机理故障诊断空间特性　出处：《重庆三峡学院》2017年硕士论文　论文类型：学位论文

【摘要】：无线传感网络在信息采集、检测等方面具有强大的处理功能,在复杂问题求解方面具有最优求解能力,但其节点具有能量受限、路由状态多变、通信易受干扰、发生故障概率较大等特点,如何有效及时诊断出发生的故障已经成为无线传感器网络应用的关键问题。本文通过对空间特性下的节点故障诊断算法研究的基础上,引入对故障检测与诊断具有记忆识别、学习能力等优点的免疫机理,通过对生物免疫系统的理论、仿生机理和人工免疫系统的理解,利用免疫机理的记忆学习的优点,对故障数据库进行实时优化,本文围绕着针对节点故障诊断开展工作,提出了基于空间特性下的节点免疫故障诊断算法(Node immune fault diagnosis algorithm,NIFD算法),为无线传感网络节点故障诊断提供了一种新方法。主要研究工作如下:1.通过对无线传感网络理论进行系统性的梳理,对人工免疫系统的免疫机理分析,建立了人工免疫机理与无线传感器网络故障诊断之间的映射关系,利用人工免疫机理的记忆学习等优点对节点数据故障库进行优化。2.在节点诊断模型的基础上,通过对节点的空间相关性的研究,在基于空间特性下的节点故障诊断算法基础上,对网络节点故障进行可靠的检测,为了提高检测的准确度,引入免疫机理,建立了免疫故障诊断模型。3.通过对模型中免疫机制进行分析,提出了 NIFD算法,实现对监测区域内故障节点的有效诊断。通过仿真实验,分析该算法在故障节点的诊断精确度,虚警率和虚警概率等性能仿真,实现了对节点故障有效的检测和诊断,判别出故障类型,提高了诊断的精度。建立的故障诊断模型能够满足节点故障的检测与诊断要求,实现对故障的可靠性诊断。本文将基于空间特性的节点免疫诊断算法理论运用到无线传感网络的节点检测与诊断中,通过仿真验证该算法在节点故障检测与诊断方面具有良好的性能,为解决无线传感网络的节点故障诊断问题提供参考。
[Abstract]:Wireless sensor networks have powerful processing functions in information collection and detection, and optimal solving ability in complex problem solving. However, the nodes of wireless sensor networks are energy limited, routing state is changeable, and communication is vulnerable to interference. How to diagnose the fault effectively and timely has become the key problem in wireless sensor network application. Based on the research of node fault diagnosis algorithm based on spatial characteristics, this paper discusses how to diagnose the fault effectively and in time. This paper introduces the immune mechanism which has the advantages of memory recognition and learning ability in fault detection and diagnosis. By understanding the theory of biological immune system, bionic mechanism and artificial immune system, the advantage of memory learning of immune mechanism is utilized. To optimize the fault database in real time, this paper focuses on node fault diagnosis. A node immune fault diagnosis algorithm named Node immune fault diagnosis algorithm based on spatial characteristics is proposed, which provides a new method for node fault diagnosis in wireless sensor networks. The main research work is as follows: 1. On systematic carding, Based on the analysis of the immune mechanism of the artificial immune system, the mapping relationship between the artificial immune mechanism and the fault diagnosis of the wireless sensor network is established. The memory learning of artificial immune mechanism is used to optimize the node data fault database. 2. Based on the node diagnosis model and the research of node spatial correlation, the node fault diagnosis algorithm based on spatial characteristics is proposed. In order to improve the accuracy of detection, the immune fault diagnosis model .3. is established in order to improve the accuracy of network node fault detection. Through the analysis of immune mechanism in the model, the NIFD algorithm is proposed. Through simulation experiments, this paper analyzes the performance simulation of the algorithm in fault node diagnosis accuracy, false alarm rate and false alarm probability, and realizes the effective detection and diagnosis of node fault. The fault diagnosis model can meet the requirements of node fault detection and diagnosis. In this paper, the theory of node immune diagnosis algorithm based on spatial characteristics is applied to node detection and diagnosis of wireless sensor networks. The simulation results show that the algorithm has good performance in node fault detection and diagnosis, and provides a reference for solving the node fault diagnosis problem in wireless sensor networks.
【学位授予单位】：重庆三峡学院
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5;TP212.9

【参考文献】