Aerospace equipment materials demand an ultra-high level of safety and reliability, with fatigue performance being one of their core performance metrics. Traditional fatigue prediction methods rely heavily on extensive experimental tests, which are associated with high costs and long development cycles, thus failing to meet the requirements of modern aerospace engineering for efficient and accurate performance evaluation. In recent years, machine learning has exhibited remarkable potential in the fatigue life prediction of aerospace materials. This work presents a systematic review of the research progress in this field, with a focus on mainstream models and modeling workflows. It clarifies the core ideas and key research findings of both pure data-driven methods and physics-integrated approaches, and centers on the role of physical information embedding in enhancing model accuracy, credibility, and interpretability. Moreover, the paper critically discusses the existing limitations, including insufficient information mining in terms of data dimensions and complex failure mechanisms, inadequate model interpretability and low trustworthiness for engineering applications, as well as poor adaptability to complex service conditions. Finally, key research directions for addressing these limitations are highlighted, such as constructing standardized and highly reliable fatigue datasets, establishing a task-oriented automatic fusion mechanism for physical knowledge, and advancing fatigue life prediction at the level of structural components under complex service conditions.

Flaw tolerance performance of TC4 alloy was investigated for helicopter transmission systems design. The impact pit flaws and scratch flaws were investigated by fatigue specimens with same flaw depth for flaw tolerance design, and high-cycle fatigue S-N curve tests were conducted. The results show that both scratch flaws and impact pit flaws significantly reduce the fatigue limit of the TC4 titanium alloy. The scratch flaw has a higher flaw influence factor（K_flaw） of 2.29 compared to the impact pit flaw’s K_flaw value of 1.75 under the same flaw depth conditions. Fracture surface analysis provided laws of the fatigue crack initiation and propagation associated with these flaws. Scratch flaws mainly exhibit multiple sources, with crack initiated at the root of the scratch, while the characteristics of fatigue crack initiation for impact pit flaws vary with the applied stress. Multiple sources are observed under high stress lever, with crack initiated on the surface of the pit, while under low stress, the crack initiates at a subsurface of the pit.