ABSTRACT

Structural health monitoring (SHM), which allows the detection of defects at an early stage by attaching sensors to the target, is an effective method of enhancing the reliability and the safety of important engineering structures. One of the practical difficulties of SHM is that usually a large area must be monitored using a limited number of sensors fixed at certain locations. And the sensor placement is a decisive contributor to the detection capability of SHM because measured signals generally depend on the location of a defect with respect to a sensor. In order to quantify the detection sensitivity more reasonably, this study proposes an analytical method based on a closed-form probability density function and a numerical method based on Monte Carlo simulation to quantify the detection sensitivity, taking into account the randomness of sensor location. The effectiveness of the proposed detection sensitivity analysis model has been examined using simulated inspection signals of low frequency electromagnetic monitoring for detecting full circumferential pipe wall thinning.

摘要

结构健康监测（SHM）通过将传感器安装到目标上，可以在早期阶段检测缺陷，是提高重要工程结构的可靠性和安全性的有效方法。SHM的实际困难之一是通常必须使用固定在某些位置的有限数量的传感器来监视大区域。传感器的位置是SHM检测能力的决定性因素，因为测量的信号通常取决于缺陷相对于传感器的位置。为了更合理地量化检测灵敏度，本研究提出了一种基于封闭形式概率密度函数的分析方法和一种基于蒙特卡洛模拟的数值方法来量化检测灵敏度，考虑到传感器位置的随机性。所提出的检测灵敏度分析模型的有效性已通过使用低频电磁监控的模拟检查信号检测全周管壁变薄的方法进行了检验。