【作者】 季赛；

【导师】 袁慎芳；

【作者基本信息】 南京航空航天大学 ， 测试计量技术及仪器， 2014， 博士

【摘要】 基于无线传感器网络的结构健康监测是诸多领域的热点研究课题。工程结构监测中采用无线传感器网络能满足其多点、高效、实时的监测需求,具有快速部署、自组织成网和分布式协同计算的特点。目前,该领域的研究内容主要包括组网技术、网络管理方法和系统验证研究等方面,其中传感器网络的组网技术是结构监测的有效实施以及保证损伤识别精度的前提和基础。因此本文针对结构监测,特别是航空结构,研究无线传感器网络组网的关键技术和实现方法。本文的主要研究内容包括以下几方面:(1)针对主动结构健康监测的需求,设计实现了可工程应用的高速无线压电传感网络节点,主要由压电信号激励模块、压电信号调理电路、数据处理模块和无线通信模块组成。对节点的激励功能、高速数据采集和螺钉松动主动监测进行功能测试实验,验证节点的主动激励和高速信号采集、处理和通信的功能。(2)针对结构监测的实时、可靠和高频采样的应用需求,提出面向结构健康监测的无线传感器网络的协议栈结构,提出了一种基于两层星簇型的网络拓扑结构,针对该拓扑结构设计了一种简单有效的固定路由机制。针对结构监测中数据可靠传输的应用需求,采用双基站冗余通信机制和多级CRC校验机制,提高数据传输的可靠性。(3)在研究传感器节点时钟模型的基础上,针对结构监测领域对时间同步的特殊要求,提出一种改进的FTSP时间同步协议,克服时间同步过程中异常值干扰的问题,并在无线压电传感器节点上进行验证实验。(4)针对结构监测中传感器数量和种类众多,且采集数据量大的特点,研究基于数据压缩采样技术的压电响应数据压缩方法。提出一种面向主动结构监测的感知压缩方法:采用小波变换实现响应信号的稀疏化,选择高斯随机矩阵作为测量矩阵,提出一种Lamb波信号重构的OMP重构算法。研究面向结构监测的感知压缩技术的压缩性能、重构精度以及不同测量矩阵的重构性能。在无线高速压电传感器节点和LF-21M防锈铝板结构上进行模拟损伤测试。更多还原

【Abstract】 Structural health monitoring based on wireless sensor networks(WSN) is a hot research topic in many fields in the past years. Using wireless sensor networks technology in structures monitoring can meet the needs of multipoint and real-time efficient monitoring, which has the rapid deployment, self-organize into networks and distributed collaborative computing features. Currently, research in this area mainly includes networking technology, network management method and system validation studies and so on. Among them, network technology of wireless senor networks is the premise and foundation for structural damage identification accuracy. For structural monitoring, especially in aerospace structures, this paper carries out a deep research on the key technologies and implementation methods in networking technology.The main research contents include the following aspects:(1) For active structural health monitoring requirements, a kind of high-speed wireless piezoelectric sensor node is designed for structural health monitoring of engineering applications. The node consists of a piezoelectric excitation signal modules, piezo signal conditioning circuits, data processing module and wireless communication module. Every module in the system is tested through two experiments; the first test is the signal Stimulus and high-speed data acquisition, the second experiment is to verify the active monitoring function of screw loosening. The results show the designed wireless piezoelectric sensor node has a good performance.(2) In order to meet the application requirements of real-time, reliable, and high-frequency sampling in structural monitoring, this paper propose the protocol stack structure for wireless sensor networks in SHM. A network topology based on two-star-cluster and effective fixed routing mechanisms are also presented.(3) This paper proposed a method of redundant communication mechanism with dual base station and multi-level CRC verifi mechanism to improve the reliability of data transmission applications in SHM. To meet the special requirements of time synchronization for SHM, we proposed an improved FTSP time synchronization algorithm to exclude the disturbance of the outliers, and the results are robust.(4) Since sensor’s number and variety are large, this paper study the data compression method based on the compressive sampling technology for piezoelectric response data. A data compression method, which is called “Compressive Sensing, CS”, is proposed to apply to active structural monitoring. The processing of CS method is that: First, response signal x are measured onto the measurement data vector y through inner products with random vectors, where y is the linear projection of x and it is permitted to lose part of response signal x in the data transmissions. Next, when the base station receives the incomplete data, the response signal x can be reconstructed from the data vector y using the CS method. Finally, the test of active structural damage identification on LF-21 M aviation antirust aluminum plate is proposed. The response signal gathered from the aluminum plate is used to verify the data compression ability of the proposed method.更多还原