Scholarlink

Abstract <p>Large keyed dry joints are widely used in precast segmental bridges due to their ease of construction and superior shear performance. However, their shear damage mechanisms and full-process signal evolution remain insufficiently understood. In this study, four large keyed dry joint specimens, each consisting of three segments, were tested under direct shear loading to simulate the joint stress conditions of segmental cap beams. Acoustic emission (AE) monitoring was employed to capture real-time signal streams during damage development. An automatic stage identification method was proposed by integrating the damage cumulative index (DCI) with the energy derivative, enabling objective recognition of multiple characteristic stages throughout the loading process. In addition, a data-driven classification framework combining principal component analysis (PCA) and <i>K</i>-Means clustering was applied, demonstrating that AE signals can be partitioned into separable groups in multi-parameter feature space. The experimental results confirmed that variations in key-tooth parameters significantly affected specimen stiffness, load-bearing capacity, and signal distribution patterns. These findings highlight the effectiveness of multiparameter AE analysis in tracking the evolution of AE signal streams and provide a methodological basis and experimental data support for damage identification and early warning in large keyed dry joint structures.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Damage Stage Segmentation and Signal Clustering of Large Keyed Dry Joints Based on Acoustic Emission Monitoring

  • 期刊论文

  • 发表日期：

    2025年11月1日

查看全文
  • Ye Tian,
  • Duo Liu,
  • Xudong Chen,
  • Jiandong Zhang

摘要

Abstract

Large keyed dry joints are widely used in precast segmental bridges due to their ease of construction and superior shear performance. However, their shear damage mechanisms and full-process signal evolution remain insufficiently understood. In this study, four large keyed dry joint specimens, each consisting of three segments, were tested under direct shear loading to simulate the joint stress conditions of segmental cap beams. Acoustic emission (AE) monitoring was employed to capture real-time signal streams during damage development. An automatic stage identification method was proposed by integrating the damage cumulative index (DCI) with the energy derivative, enabling objective recognition of multiple characteristic stages throughout the loading process. In addition, a data-driven classification framework combining principal component analysis (PCA) and K-Means clustering was applied, demonstrating that AE signals can be partitioned into separable groups in multi-parameter feature space. The experimental results confirmed that variations in key-tooth parameters significantly affected specimen stiffness, load-bearing capacity, and signal distribution patterns. These findings highlight the effectiveness of multiparameter AE analysis in tracking the evolution of AE signal streams and provide a methodological basis and experimental data support for damage identification and early warning in large keyed dry joint structures.