Abstract
Pitting corrosion on the surface of an aluminum alloy plate will reduce its structural strength and affect structural safety. In this paper, a quantitative identification method of corrosion degree based on the combination of multiple damage characteristics and sensor path optimization is proposed. First, finite element (FE) analysis is used to simulate the effect of different degrees of corrosion on Lamb waves. Nine parameters were analyzed for their correlation with corrosion degree, from which three damage indices (DIs) based on the amplitude and energy of the scattered signals were ultimately selected. Correlation functions between three damage indices and corrosion degree were then established for quantitative identification. Additionally, the influence of the number of paths on the results was studied, and the optimal number of paths for identification was determined. And multiple simulation experiments were conducted involving different corrosion locations and degrees, resulting in the mean absolute error (MAE) of less than 46 mm³. Finally, an experimental system was built to conduct corrosion identification experiments, achieving an MAE of less than 50 mm³, closely matching the simulation results. These research findings demonstrate that the correlation functions derived from simulation exhibits good accuracy and stability in the detection of corrosion with regular shapes.
Funding Information
  • National Key Research and Development Project (2023YFB3709605)
  • National Natural Science Foundation of China (62573259)
  • National Natural Science Foundation of China (62273206)
  • National Natural Science Foundation of China (62373219)
  • National Natural Science Foundation of China (62373215)
  • National Natural Science Foundation of China (62503500)
  • Natural Science Foundation of Shandong Province (ZR2025QC1541)

This publication has 38 references indexed in Scilit: