Study on tensile bond and shear bond failure characteristics of TSL materials

Abstract

Bond stress is a crucial indicator for assessing the strength of thin spray liner (TSL) materials used as support. The traditional approach of manually measuring the bonded residual area of TSL at the interface and defining the bond stress through threshold determination has two main drawbacks: a significant error in the manual measurement of the interface bonding area, and the unscientific selection of image threshold information. To address these issues, this paper utilizes image threshold segmentation software to recognize the interface bond area. It proposes a scientific and reasonable method for dividing image threshold information based on image data, leading to a more accurate classification of TSL bond stress values and damage modes. This approach elucidates the changes in the mechanical state of the TSL during the bearing process. The evaluation of computational thresholding and automatic thresholding segmentation was conducted using two methods, AP and IU. Results indicate that the computational thresholds obtained through the IFFT-continuous cusp mutation method offer significant advantages. From a mechanical perspective, this method provides a clear physical interpretation for each threshold value, yielding more accurate TSL mechanical measurements and a more realistic classification of damage modes, and allowing for the definition of mechanical states in the TSL bearing process. Regarding image segmentation, computed thresholding improves image segmentation effectiveness by approximately 15.6 %–32.6 % compared to automatic thresholding, demonstrating strong stability even for samples that are extremely difficult to recognize. Therefore, the threshold calculation method proposed in this paper not only enhances the study of TSL mechanical properties but also offers a new perspective for the development of image threshold segmentation.