Micropores are one of the critical factors affecting materials’ performance and service life. As the need for a deeper understanding of micropore evolution and damage mechanisms grows, assessing the mechanical properties of materials containing micropores and predicting the lifespan of related metal structural components becomes increasingly complex. This paper focuses on the evolution process, regularities, and research methods of micropores in metal materials. Based on recent research and practical applications, the key stages of micropore evolution are discussed, encompassing nucleation, growth, coalescence, collapse, interaction, and the influence of other microstructures. Firstly, the advantages and limitations of commonly used characterization methods such as scanning electron microscopy, transmission electron microscopy, and X-ray computed tomography are introduced in the study of micropore evolution. Subsequently, critical theoretical models for micropore evolution, such as the Gurson model and its extensions, are summarized. By using a multiscale approach combining the crystal plasticity finite element method, dislocation dynamics, and molecular dynamics, the factors influencing the micropore evolution, such as external stress conditions, internal microstructures, and micropore characteristics, are specifically elaborated, and the basic physical mechanisms of micropore evolution are analyzed. Finally, a comprehensive review and summary of current research trends and key findings are provided, and a forward-looking perspective on future research directions is presented.

Rolling mill multi-row bearings are subjected to axial loads, which cause damage of rolling elements and cages, so the axial vibration signal contains rich fault character information. The vertical shock caused by the failure is weakened because multiple rows of bearings are subjected to radial forces together. Considering the special characters of rolling mill bearing vibration signals, a fault diagnosis method combining Adaptive Multivariate Variational Mode Decomposition (AMVMD) and Multi-channel One-dimensional Convolution Neural Network (MC1DCNN) is proposed to improve the diagnosis accuracy. Additionally, Deep Convolutional Generative Adversarial Network (DCGAN) is embedded in models to solve the problem of fault data scarcity. DCGAN is used to generate AMVMD reconstruction data to supplement the unbalanced dataset, and the MC1DCNN model is trained by the dataset to diagnose the real data. The proposed method is compared with a variety of diagnostic models, and the experimental results show that the method can effectively improve the diagnosis accuracy of rolling mill multi-row bearing under unbalanced dataset conditions. It is an important guide to the current problem of insufficient data and low diagnosis accuracy faced in the fault diagnosis of multi-row bearings such as rolling mills.