多传感器数据融合在结构健康监测中的应用.doc
文本预览下载声明
多传感器数据融合在结构健康监测中的应用
Aaa
摘要:随着我国经济的高速发展和科学技术的进步,许多大型复杂结构都安装了长期健康监测系统。然而,大型结构具有较多的结构冗余度和环境荷载的不确定性;此外,来自监测系统的海量数据也包含大量的噪声和不确定性。因此,如何合理有效地处理这些来自结构和健康监测系统的海量不确定测量数据与信息,进而对结构的健康状况进行评价成为国内外同行关注的热点和难点。基于此,本文从从多传感器数据融合的概念、基本原理出发,探讨了数据融合技术在结构健康监测与诊断中应用的可行性,重点研究了基于人工智能和Kalman滤波的数据融合技术在结构健康监测及诊断中的应用方法,并由此说明了基于多传感器数据融合的健康监测与诊断是可行的、有效的。
关键词:结构健康监测;数据融合;特征提取;神经网络
Abstract:With the progress of our country economy and the rapid development of science and technology, many large complicated structures are installed long-term health monitoring system. However, the large structure has more structural redundancy and environmental load’s uncertainty; in addition, massive data from the monitoring system also contains a lot of noise and uncertainty.Therefore, how to reasonably and effectively handle a amount of uncertain data and information from massive structure and health monitoring system so that make evaluation on the structural health that has been a hot and difficult topic in the domestic and foreign counterparts.Based on this, from the concept and the basic principle of multi-sensor data fusion,this paper discussed the feasibility of the data fusion technology which is applied in structural health monitoring and diagnosis,focused on the application method of data fusion technology based on artificial intelligence and Kalman filter in structure health monitoring and diagnosis, and this explained the health monitoring and diagnosis based on multi-sensor data fusion is feasible and effective.
Keywords:structure health monitoring;date fusion;feature extraction;neural network
0 引言
随着科技进步和社会发展,地球上的超大型、复杂建筑越来越多,而这些工程中,由于种种隐患的存在,不可避免地导致不同形式的结构和系统的损伤积累和抗力衰减,从而抵抗自然灾害、甚至正常环境作用的能力下降,极端情况下引发灾难性的突发事故。为了保障建筑物结构的安全性、完整性、适用性与耐久性,已建成使用的许多重大工程结构和基础设施急需采用有效的手段监测和评定其安全状况、修复和控制损伤。如果能够在灾难到来之前对其进行健康监测、诊断、安全评估及灾难预警,以趋利避害,将有助于从根本上消除隐患、避免灾害事故的发生。因此,对工程结构进行健康监测与诊断,成为当前国际社会重点研究的课题,具有重要的理论意义和实用价值[1-5]。
结构健康监测与诊断技术经历了三个发展阶段:第一阶段是以领域专家的经验为基础的经验诊断技术,对诊断信息只能作简单的数据处理;第二阶段是以传感器技术和动态测试技术为手段,以信号处理和建模处理为基础的现代诊断技术;第三发展阶段是
显示全部