Nonlinear System Identification for Real-Time Structural Health Monitoring in Composite Materials

9360372新的软纤维/树脂复合材料制造技术可以生产高强度、低重量的结构部件，而光纤传感技术提供了一种方法，当这些传感器安装在材料结构上或嵌入材料结构中时，可以传感各种结构特性，如应力、应变和弹性。复合材料结构可能会因制造不当、滞后磨损、冲击和腐蚀而退化。虽然嵌入式传感器提供了一种收集与复合材料结构和动力历史有关的海量数据的方法，但如果嵌入式传感器要用于确定结构完整性，使用能够处理分布式和非线性问题的方法，则有必要对这些信息进行处理、分类和解释。这项研究将利用先进的基于神经网络的技术来建模和分析动态结构信息，以及使用神经网络分类器来执行实时传感器解释。这将产生一个处理、集成和解释嵌入式传感器数据的智能系统，以实时、稳健、准确地测量厚截面复合材料结构的结构完整性。该项目涉及与McDonnell Douglas在嵌入式传感器信号处理和基于神经网络的复合材料结构分类领域的合作。***

9360372 Sofge Fiber/resin composite fabrication technologies new can produce high-strength, low weight structural components, while fiber-optic sensing technology provides a means for sensing various structural properties such as stress, strain, and elasticity when these sensors are mounted on, or embedded within, a material structure. Composite structures may degrade due to improper manufacture, hysteretic wear, impacts and corrosion. While embedded sensors provide a means for gathering vast quantities of data relating to the structural and dynamic history of composite structures, it is necessary to process, itegrate, and interpret this information if the embedded sensors are to be useful in determining structural integrity, using methods that can deal with the distributed and nonlinear nature of the problem. This research will utilize advance neural network based techniques for modeling and analyzing dynamic structural information, as well as neural network classifiers for performing real-time sensor interpretation. This will yield an intelligent system for processing , integrating, and interpreting embedded sensor data to give a real-time, robust, accurate measure of structural integrity for thick-section composite structures. This project involves cooperation with McDonnell Douglas based on earlier work with them in the area of embedded sensor signal processing and neural network based classification for composite structures. ***