STRUCTURAL INTEGRITY & LIFETIME ESTIMATION BY MACHINE LEARNING METHODS

Many responsible structural elements are fractured because of fatigue, gradually accumulating damage and starting with a small crack that grows under loading. Fatigue crack is formed mainly in the stress concentrator, that is, the place of damage, which loosens the cross-sectional region of the material. The crack grows if the material cannot withstand the loading. Therefore, the main factors that affect the strength of structural elements are surface defects of parts, temperature, environment during service, and the character of loading and loading conditions.Therefore, it is important to study the influence of loading conditions on fatigue crack growth. Methods of strength and lifetime evaluation of the responsible structural elements often require complicated calculations. Thus, it is worth learning how to solve the tasks of fracture mechanics by machine learning methods, particularly neural networks, boosted trees, random forests, support vector machines, and the method of k-nearest neighbors, which enable the achievement of high accuracy of solutions.The monograph evaluates structural elements' strength and residual lifetime using machine learning methods.