Search results

Surface defects reduce fatigue strength and may greatly reduce component reliability, particularly in pressure vessel weld regions, springs, and other applications. The fatigue strength of components, and thus their reliability, can be substantially increased by tensile overloading prior to use. The purpose of this paper is to investigate the effect of tensile overload on small cracks by applying a tensile overload to steel plates containing semicircular slits that simulate small surface cracks and by determining the degree of increase in the fatigue strength. The effect of tensile overload on the apparent fatigue threshold stress intensity factor range (ΔK_th) was also investigated.

A tensile overload stress of 1,000 or 1,200 MPa was applied once to all test pieces. Then, bending fatigue tests were conducted with a stress ratio R=0.1. The slit region was subjected to applied cyclic tensile stress by four‐point bending throughout the fatigue test. A test specimen to which no overload stress was applied was tested for comparison.

The improvement in ΔK_th by tensile overload is observed in the specimen with a small crack like surface defect. However, in the specimen with a small crack like surface defect, the improvement in ΔK_th by tensile overload is saturated as increasing tensile overload. The improvement rate of ΔK_th by tensile overload and the upper limits of improvement in ΔK_th were predicted. The predicted values of the improvement rate of K were well in agreement with the experimental results.

The proposed method can be applied to pressure vessels and springs.

The overload effects on fatigue strength are studied for large cracks. However, the effect is not understood at all for small cracks. This study focused the over load effects for small cracks. This is the original point of the present study.

Describes the parameter estimation procedures for the non‐linear finite element analysis using the hierarchical neural network. These procedures can be classified as the neural network based inverse analysis, which has been investigated by the authors. The optimum values of the parameters involved in the non‐linear finite element analysis are generally dependent on the configuration of the analysis model, the initial condition, the boundary condition, etc., and have been determined in a heuristic manner. The procedures to estimate such multiple parameters consist of the following three steps: a set of training data, which is produced over a number of non‐linear finite element computations, is prepared; a neural network is trained using the data set; the neural network is used as a tool for searching the appropriate values of multiple parameters of the non‐linear finite element analysis. The present procedures were tested for the parameter estimation of the augmented Lagrangian method for the steady‐state incompressible viscous flow analysis and the time step evaluation of the pseudo time‐dependent stress analysis for the incompressible inelastic structure.