2025 International Conference on Equipment Intelligent Operation and Maintenance (ICEIOM2025)

120 / 2025-05-15 21:22:14

Remaining Useful Life Prediction of Fatigue Crack Extension of TC4 Titanium Alloy at High Temperature Based on Physical Information Neural Network

Fatigue crack extension,Data-driven,Remaining life prediction,Physical information neural network,TC4 titanium alloy

In view of the traditional crack extension model directly establishes the relationship between stress ratio and crack extension rate, and pays less attention to the influence of temperature on the extension behaviour during the loading process, which leads to inaccurate prediction of the remaining life of crack extension performance of the material. This paper firstly obtains the model constants of the TC4 titanium alloy's crack extension rate through the constant amplitude fatigue crack extension test under the conditions of different temperatures and different stress ratios. Secondly, based on the physical information neural network method, we constructed the remaining life prediction model of high temperature fatigue crack extension of TC4 titanium alloy material. Secondly, based on the physical information neural network method, we constructed the remaining life prediction model of high-temperature fatigue crack extension of TC4 titanium alloy material. Combining the physical property constants of the material and the external temperature and stress ratio conditions, we established the remaining life prediction model from the ‘input end’ to the ‘output end’ of TC4 titanium alloy material. The experimental results show that the proposed method can better reflect the crack propagation characteristics of the material than the comparative methods, and the prediction deviation is lower in different individual tests. This method can be applied to the safety assessment and maintenance decision of in-service structures, and at the same time provides a theoretical basis for the prediction of the remaining life and fatigue strength of materials.