Pin-on-disc friction experiments were carried out to investigate the tribocorrosion of AISI 316L in seawater, demonstrated that the tribocorrosion of AISI 316L was dominated by the formation and corrosion of strain-induced-martensite with austenite.

The existing methods for thermal barrier coating (TBC) life prediction rely mainly on experience and formula derivation and are inefficient and inaccurate. By introducing deep learning into TBC life analyses, a convolutional neural network (CNN) is used to extract the TBC interface morphology and analyze its life information, which can achieve a high-efficiency accurate judgment of the TBC life. In this thesis, an Adap–Alex algorithm is proposed to overcome the problems related to the large training time, over-fitting, and low accuracy in the existing CNN training of TBC images with complex tissue morphologies. The method adjusts the receptive field size, stride length, and other parameter settings and combines training epochs with a sigmoid function to realize adaptive pooling. TBC data are obtained by thermal vibration experiments, a TBC dataset is constructed, and then the Adap–Alex algorithm is used to analyze the generated TBC dataset. The average training time of the Adap–Alex method is significantly smaller than those of VGG-Net and Alex-Net by 125 and 685 s, respectively. For a fixed number of thermal vibrations, the test accuracy of the Adap–Alex algorithm is higher than those of Alex-Net and VGG-Net, which facilitates the TBC identification. When the number of thermal vibrations is 300, the accuracy reaches 93%, and the performance is highest.

Wheel rail rolling contact fatigue is a very common form of damage, which can lead to uneven rail treads, railhead nuclear damage, etc. Therefore, ANSYS software was used to establish a three-dimensional wheel–rail contact model and analyze the effects of several main characteristics, such as the rail crack length and crack propagation angle, on the fatigue crack intensity factor during crack propagation. The main findings were as follows: (1) With the rail crack length increasing, the position where the crack propagated by mode I moved from the inner edge of the wheel–rail contact spot to the outer edge. When the crack propagated to 0.3–0.5 mm, it propagated to the rail surface, causing the rail material to peel or fall off and other damage. (2) When the crack propagation angle was less than 30°, the cracks were mainly mode II cracks. When the angle was between 30 and 70°, the cracks were mode I–II cracks. When the angle was more than 70°, the cracks were mainly mode I cracks. When the crack propagation angle was 60°, the equivalent stress intensity factor reached the maximum, and the rail cracks propagated the fastest.

While selecting the most suitable infrared thermal imaging detection scheme for online inspection during laser cladding processing, this paper designs the RespathU-net semantic segmentation defect detection network for cladding coating defects in infrared images. The network is based on the U-net network framework. It is optimized and improved by redesigning the coding network structure, expanding the network perceptual field, and connecting the paths of residuals, thus enhancing the segmentation effect on the defective areas of the melt coating by addressing the problems that the original network cannot realize the end-to-end output and the poor segmentation effect on the complex objects. The generalization performance test and defect detection experiment of the RespathU-net network were conducted using the Kolektor SDD dataset and the infrared dataset constructed in this paper. The designed network is compared with fully convolutional networks, SegNet, U-net, and DeepLab_V3 in terms of average exchange ratio, similarity coefficient, and running time. The results show that the proposed RespathU-net achieves good multi-size feature recognition, and the effect is much better than other semantic segmentation networks. The via experimental results verify that the actual defect detection accuracy of the designed network is 87.01% .

A piezoelectric polyvinylidene fluoride film stress sensor was developed to measure the contact pressure on the sheet near the cutting edge in the fine-blanking process. Two cases with and without cracks in the fine-blanking surface were studied. The relationship between the contact pressure and crack width was investigated using the Kriging methodology at different punching speeds. The maximum contact pressure appeared at the beginning of sheet punching, but not at the end of V-ring indention. For the case with and without cracks in the fine-blanking surface, one and two fluctuations of contact pressure are observed, respectively. Using the Kriging model, a fine-blanked surface with a reduced crack width can be obtained using the optimal values of contact pressure and punching speed.