Fatigue damage modelling and life prediction of engineering components under variable amplitude loadings are critical for ensuring their operational reliability and structural integrity. In this paper, five typical nonlinear fatigue damage accumulation models are evaluated and compared by considering the influence of load sequence and interaction on fatigue life of P355NL1 steels. Moreover, a new nonlinear fatigue damage accumulation model is proposed to account for these two effects. Experimental datasets of pressure vessel steel P355NL1 and four other materials under two‐block loadings are used for model comparative study. Results indicate that the proposed model yields more accurate fatigue life predictions for the five materials than the other models.

For engineering components subjected to complex multiaxial loadings, critical plane approaches like Fatemi-Socie criterion have been commonly utilized for life prediction of these components. Within the Fatemi-Socie criterion, the normal stress sensitivity parameter k is usually fitted from additional experimental data, which introduces inconvenience for practice especially under limited testing data conditions. In this regard, a simple critical plane-based damage parameter is put forward with no additional material constants, which attempts to provide a robust method for multiaxial fatigue analysis of turbine disk alloys. Using experimental datasets of TC4 and GH4169 alloys under different loadings, the proposed model provides better correlations with fatigue life of the two alloys than the models of Smith-Watson-Topper and Wang-Brown.

For engine components under complex loadings, multiaxial fatigue life prediction is critical for ensuring their structural integrity and reliability. Combining the critical plane method with the virtual strain energy concept, a new multiaxial fatigue damage parameter is proposed to characterize the influence of both shear/normal mean stress and non-proportional hardening on fatigue life. Particularly, no extra material constants are needed for model application. Experimental data of TC4 and GH4169 alloys under various loading paths are utilized to evaluate and validate the proposed damage parameter as well as four other models. Results show that the proposed damage parameter yields a higher accuracy on multiaxial fatigue life prediction than others.

No abstract