基于Lamb波结构健康监测的压缩感知方法研究

摘要

One of the principal challenges facing the structural health monitoring is taking great capacity sets of data collect-ed from sensors and extracting information that can estimate damage effectively. Based on the analysis of sensing signals, an im-proved matching pursuit algorithm for distributed sensing network was proposed, thus reduceing the signal dimension and recon-struction time. The approach was tested on experimental Lamb wave signals obtained by acquiring acoustic emissions in an alumi-num plate with conventional piezoelectric sensors. Experimental results show that under the premise of ensuring the signal recon-struction accuracy(reconstruction error±0.25), the amount of calculation is reduced and damage location is accurate.%结构健康监测面临的一个巨大挑战就是如何从传感器收集的海量数据中提取重要信息并有效地进行损伤状况估计。通过对传感信号进行分析，提出针对分布式传感网络结构的改进匹配追踪算法，降低测量信号维数，减少重构时间。在粘贴压电传感器的铝板上获取Lamb波信号并进行该算法的验证，实验结果表明在保证重构信号精确度(重构误差±0.25)的前提下减少了计算量且损伤定位准确。