Residual fatigue life prediction based on a novel improved Manson-Halford model considering loading interaction...

Area of Science:

• Materials Science
• Mechanical Engineering
• Fatigue Analysis

Background:

• Classical fatigue cumulative damage models like Manson-Halford do not account for load interaction effects.
• Existing correction models often have complex parameter determination or rely on single parameters, leading to inaccurate life predictions.
• There is a need for improved fatigue life prediction models that are accurate and easy to apply.

Purpose of the Study:

• To develop a novel, improved fatigue cumulative damage model.
• To overcome the limitations of classical models by incorporating load interaction effects.
• To establish a model that is solely based on S-N curve parameters for simpler application.

Main Methods:

• A new correction method was developed to dynamically modify the classical fatigue model.
• The improved model considers the relationship between adjacent loads and varying fatigue life under different stress states.
• Multi-level fatigue tests were conducted on 300M steel and other materials to validate the model.

Main Results:

• The new improved model accurately predicted the remaining fatigue life of 300M steel, reducing prediction error by 11.75% compared to the classical model.
• Validation with diverse material fatigue data showed the improved model achieved the highest prediction accuracy in most cases.
• The improved model demonstrated a minimum relative prediction error of 3.79% and a maximum reduction in relative prediction error of 28.73% compared to classical models.

Conclusions:

• The newly developed improved fatigue life prediction model effectively addresses the shortcomings of classical models.
• The model's reliance on S-N curve parameters simplifies application without compromising accuracy.
• The model shows significant potential for accurate and reliable fatigue life prediction across various materials and loading conditions.