Monitoring of structures and defining the severity of damages that occur under loading are essential in practical applications of civil infrastructure. In this article, we analyze failure using a smart aggregate sensor–based approach. The signals captured by smart aggregate sensors mounted on the structure under loading are de-noised using wavelet de-noising technique to prevent misdirection of the event interpretation of what is happening in the material. The performance of different mother wavelets on the de-noising process was investigated and analyzed. The objective is to identify the optimal mother wavelet for assessing and potentially reducing the effects of existing noise on signal properties for structural damage detection. In addition, we propose two innovative damage indices, entropy-based dispersion and entropy-based beta, for diagnostic purposes. The proposed entropy-based dispersion damage index is based on the modified wavelet packet tree and root mean square deviation, whereas the entropy-based beta damage index is based on the modified wavelet packet tree and slope of linear regression (beta). In both damage indices, the modified wavelet packet tree uses entropy as a high-level feature. Theoretical and experimental analyses are derived by computing indices on smart aggregate–based sensor data for concrete and reinforced-concrete beams. Validity assessment of the proposed indices was addressed through a comparative analysis with root mean square deviation damage index (benchmark) and the loading history. The proposed indices recognized the cracks faster than other measures and well before major cracking incurs in the structure. This article is expected to be beneficial for smart aggregate–based structural health monitoring applications particularly when damages occurred under loading.

中文翻译：

---

基于智能骨料监测系统的混凝土和钢筋混凝土梁失效分析

在民用基础设施的实际应用中，监测结构并确定在负载下发生的损坏的严重程度是必不可少的。在本文中，我们使用基于智能聚合传感器的方法分析故障。安装在受载结构上的智能聚合传感器捕获的信号使用小波去噪技术去噪，以防止对材料中发生的事件的错误解释。研究和分析了不同母小波在去噪过程中的性能。目标是确定最佳母小波，以评估和潜在地减少现有噪声对结构损坏检测信号特性的影响。此外，我们提出了两个创新的损伤指数，基于熵的色散和基于熵的 beta，用于诊断目的。提出的基于熵的色散损伤指数基于修正的小波包树和均方根偏差，而基于熵的β损伤指数基于修正的小波包树和线性回归（β）的斜率。在两个损伤指数中，修改后的小波包树都使用熵作为高级特征。理论和实验分析是通过计算基于智能骨料的混凝土和钢筋混凝土梁传感器数据的指数得出的。通过与均方根偏差损坏指数（基准）和加载历史的比较分析，对建议指标的有效性进行了评估。建议的指数比其他措施更快地识别裂缝，并且早在结构中发生重大裂缝之前。