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.

The internal displacement change around shrinkage pores, stress and strain distribution, and microstructural evolution in the defect healing zone of ZTC4 titanium alloy during hot isostatic pressing (HIP) are investigated by numerical simulation combined with experimental methods. The results demonstrate that under the high-temperature and high-pressure conditions of HIP, high stress-strain zones form around the shrinkage pores. The stress magnitude shows an inverse relationship with the distance from pore surfaces, exhibiting higher stress levels in proximity to the pore boundaries. As the shrinkage pore size decreases, the strain concentration intensifies. After HIP, a radial pore-healing zone microstructure develops at the original shrinkage pore sites. Within this healed region, heterogeneous plastic deformation occurs among distinct α/β colonies. Colonies with more readily activated dislocation slip systems experience greater deformation magnitudes, ultimately leading to the formation of equiaxed grain structures.

Thermal simulation compression tests are conducted on a new type of ultra-high strength and toughness TB17 titanium alloy using Gleeble-3500 thermal simulation testing machine under the conditions of deformation temperature ranging from 795 ℃ to 895 ℃ and strain rate of 0.001 s⁻¹ to 1.0 s⁻¹. The microstructure and plastic flow behavior of the alloy during hot deformation are analyzed, and a constitutive model of flow stress is established. The results show that the flow stress of the alloy increases rapidly with the increase of strain, then decreases slightly, and finally tends to be stable. Partial dynamic recrystallization occurs in the alloy, dominated by dynamic recovery, and the slight decrease in flow stress is related to the partial dynamic recrystallization of the alloy. Dynamic recrystallization volume fraction of the alloy is not higher than 40%, and the dynamic recrystallization mechanism is mainly dominated by the bowing mechanism. A constitutive model based on the Arrhenius equation is constructed, and the deformation activation energy Q value of the alloy at the temperature of 795-895 ℃ is obtained as 205.48 kJ/mol. The model has high prediction accuracy, with average error δ_avg of 3.987% and correlation coefficient R of 0.9972. The construction of this model provides an accurate prediction for the flow stress of TB17 titanium alloy during hot deformation and also offers a reference for the establishment of high-precision constitutive models for other alloys.

This study systematically investigates the effects of trace Gd additions of 0.5% (mass fraction, the same hereinafter) and 1.0% on the microstructure and tensile property of ZK60 magnesium alloy. The as-cast and solution-treated microstructures of ZK60, ZVK600, and ZVK610 alloys are characterized by optical microscopy, scanning electron microscopy, energy dispersive spectroscopy, differential scanning calorimetry, and X-ray diffraction. The tensile properties of the alloy specimens are measured and analyzed via room-temperature tensile tests. The results show that the as-cast ZK60 alloy has grain size of 95 μm, with coarse blocky MgZn phases and a small number of Zn₂Zr₃ particles present at grain boundaries. Trace Gd addition increases the fraction of secondary phases and transforms the MgZn phase into the Mg₃GdZn₆ phase, but does not refine the grain size. The room-temperature tensile properties of the three as-cast alloys are relatively close. The as-cast ZVK610 alloy exhibits lower yield strength and ductility, which is associated with its relatively large grain size and increased grain-boundary secondary phases. After T41 step solution treatment, the grain of ZK60 alloy become oarse, the secondary phases are nearly eliminated, and the ductility is significantly improved. However, the yield strength decreases slightly due to grain coarsening. In contrast, a small amount of grain-boundary secondary phases remains in ZVK600 and ZVK610 alloys. T42 and T43 processes, designed with prolonged high-temperature solution time or elevated solution temperature, further reduce the secondary phase fraction in the matrix but lead to additional grain coarsening, resulting in further reduced yield strength and no obvious improvement in ductility. Therefore, T41 solution heat treatment process is recommended.

Cast aluminum alloys are widely used in aerospace, automotive and other industries due to their excellent mechanical properties. However, traditional alloy design faces challenges such as vast composition space, high costs of trial-and-error experiments and difficulty in predicting the nonlinear relationship between composition and properties. This paper proposes a machine learning model that combines backpropagation neural networks, principal component analysis, and genetic algorithms for multi-objective property prediction of cast aluminum alloys. The model establishes the relationship between alloy composition and properties through the nonlinear mapping of backpropagation neural networks, reduces dimensionality via principal component analysis, and optimizes network parameters using genetic algorithms-thereby improving prediction accuracy and training efficiency. The results show that the optimized model has mean squared error of 36.28, correlation coefficient of 0.91, and mean absolute error of 2.44. In the experimental verification of ultimate strength, yield strength, and elongation after fracture, the error between experimental values and predicted values is controlled within the range of ±5%. This high prediction accuracy demonstrates the efficiency and reliability of the proposed model.

The hot deformation behavior of 9310 gear steel is investigated by hot compression test within temperature range from 800 ℃ to 1100 ℃ and strain rate range from 0.001 s⁻¹ to 10 s⁻¹. The peak stress constitutive relation model and processing maps under different strains are constructed and the process parameters are optimized. The results show that the flow stress of 9310 gear steel increases obviously with the decrease of deformation temperature and the increase of strain rate. Under the condition of high temperature (900-1100 ℃) and low strain rate (0.001-0.1 s⁻¹), the flow stress curve primarily exhibits strain softening characteristics. The constitutive relation model demonstrates high predictive accuracy, with a correlation coefficient of approximately 0.998 and an average absolute relative error of about 4.724%. The obtained better deformation process parameters range from 1010 ℃ to 1100 ℃ and from 0.05 s⁻¹ to 1.41 s⁻¹, among which the optimal process parameters are in the range of deformation temperature 1070 ℃ and strain rate 0.05-1 s⁻¹, corresponding to the main plastic deformation mechanism of dynamic recrystallization. The range of unstable deformation process parameters is approximately 800-925 ℃ and 0.04-10 s⁻¹, where flow localization is the primary plastic deformation mechanism. The microstructure verification results are in good agreement with the prediction and optimization results of the processing map.

The high-temperature chemical stability and wettability of EC95 powder surface layer, corundum powder surface layer, and zirconium powder surface layer of ceramic shell to DZ125 directional solidification superalloy are comparatively studied by the sessile-drop experiment. The morphology of ceramic surfaces as well as the morphologies and compositions of the interfacial reaction zones between the ceramics and the alloy melt are observed and analysed. The contact angles between the alloy melt and the ceramics are calculated, and the influence of interfacial reactions and wettability on the mechanical penetration defect of the casting surface are discussed. The results show that the surface roughness values of EC95 powder surface layer, corundum powder surface layer, and zirconia powder surface layer ceramics are 3.987, 3.391 μm, and 2.085 μm, respectively. The interfacial reaction products between the alloy and the three types of ceramics are mainly composed of Hf oxides, accompanied by a small amount of alloying components. For the corundum powder surface layer ceramics, the alloy surface is almost completely covered by a layer of Hf oxide after the experiment. This oxide layer effectively inhibits the further interfacial reaction between the alloy and the ceramic surface layer, making it more suitable for the investment casting of DZ125 superalloy. The contact angles between DZ125 superalloy and EC95 powder surface layer, corundum powder surface layer, and zirconia powder surface layer are 84.95°, 75.71°, and 132.96°, respectively. Due to the influence of interface reactions and ceramic surface roughness, the wettability between corundum surface layer shell and DZ125 alloy is better than other surface layers.

A numerical simulation study is conducted on SiC_f/SiC ceramic matrix composite turbine guide vanes to investigate the effects of variations in temperature ratio and flow ratio on the overall temperature and cooling performance of the vanes. A total of 16 operating conditions with variable temperature ratios and flow ratios are selected for simulation calculations. The results show that with the increase of both flow ratio and temperature ratio, the maximum temperature, minimum temperature and average temperature of the vanes all decrease, and high temperature occurs at the leading edge of the lower platform. The cooling efficiency of the vane suction surface increases gradually from the leading edge to the trailing edge, while that of the vane pressure surface first decreases and then increases from the leading edge to the trailing edge. For the local cooling efficiency, when the flow ratios are 5.28%, 7.54%, and 8.44%, respectively, the average cooling efficiency first decreases and then increases with the rise of temperature ratio. When the flow ratio is 3.69%, the average cooling efficiency decreases gradually as the temperature ratio increases. Meanwhile, the average cooling efficiency of the mid-section increases with the rise of flow ratio.

NiCrAlY is a commonly used metallic bond coat material for thermal barrier coating in gas turbines. This study investigates the effect of two NiCrAlY powders with different aluminum contents for gas turbine fabrication on the thermal cycling behavior of HVOF-NiCrAlY+APS-nanostructured YSZ (nYSZ) thermal barrier coatings (TBCs) within the temperature range from room temperature to 1150 ℃. The results show that the growth rate of the Al₂O₃ thermally grown oxide (TGO) on the surface of HVOF-Ni25Cr5Al0.5Y is lower than that of HVOF-Ni22Cr10Al1Y. Similar to the microstructured YSZ (mYSZ)/mYSZ interface, the nYSZ/mYSZ interface can also act as a crack initiation site, leading to the formation of a local crack network in the nYSZ layer. The failure mechanism of two HVOF-NiCrAlY+APS-nYSZ TBCs is consistent with that of the traditional APS/HVOF-MCrAlY (M=Ni and Co)+APS-mYSZ system, which is mainly attributed to the propagation and coalescence of cracks in the nYSZ layer adjacent to the HVOF-NiCrAlY/APS-nYSZ interface. The thermal cycling lifetime of the Ni25Cr5Al0.5Y+APS-nYSZ coating is slightly longer than that of the Ni22Cr10Al1Y+APS-nYSZ coating. Meanwhile, it can effectively improve the thermal cycling life of HVOF-MCrAlY+APS-YSZ TBCs to increase the bonding strength of the YSZ/YSZ interface in the APS-YSZ layer and avoid cracking at the YSZ/YSZ interface and the outer surface of the APS-YSZ.