Abstract
Fatigue cracks pose a serious safety risk in offshore platform steel structures. Studying fatigue crack prediction is crucial for ensuring structural safety during service. Currently, a prediction method based on guided wave monitoring and particle filter prediction is widely used for predicting fatigue crack propagation. However, this method is limited by the multi-modal characteristics of guided waves, making it difficult to extract features from the crack reflection signals. Additionally, it is only suitable for single-ended fatigue cracks, which complicates direct application to double-ended fatigue cracks. In this context, this paper proposes a novel particle filter prediction algorithm based on the single-mode shear horizontal (SH) guided wave and double-ended fatigue crack growth state model. First, a Face-Shear-type Lead Zirconate Titanate(PbZrTiO₃, PZT) probe is designed to excite the single-mode SH guided wave, and an observation model is established based on the reflected signals from the cracks. Second, the state model of fatigue crack growth is constructed by obtaining crack length growth data through fatigue tests. Finally, the accuracy of the proposed method is validated through an experimental study. The experimental results show that the average prediction error in the propagation length of the fatigue crack is 1.1%.
Funding Information
  • National Natural Science Foundation of China (52475581, 52275556)
  • National Key Research and Development Program of China (2023YFF0615200)
  • Natural Science Foundation of Shandong Province (ZR2023ZD32)
  • Qingdao Natural Science Foundation (24-4-4-zrjj-164-jch)
  • State Administration for Market Regulation’s Science and Technology Plan Project (2022MK071)

This publication has 25 references indexed in Scilit: