Based on Strohmeyer model,a general characterization method of metal material P-S-N curve considering confi-dence was proposed,and high precision P-S-N curve was obtained by nonlinear least squares(NLS)fitting.Considering the li-fespan of metal material and the dispersion of testing stress,a single lifespan model and an improved three-parameter disper-sion model were proposed.Combined with the correlation analysis of logarithmic lifespan standard deviation and stress,the data dispersion was analyzed,and the feasibility was verified with 2A12 aluminum alloy and BSM590 steel metal materials.The re-search results indicate that the proposed characterization method predicts fatigue lifetime with an error of less than 10％,the adjustment coefficient of dispersion model increases by beyond 0.4,and the fitting accuracy significantly improves,providing technical support for the refined reliability design of structural components.
维普
万方数据