• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.36 no.2
• /
• pp.187-194
• /
• 2012

The strength of particle-reinforced metal matrix composites is, in general, known to be increased by the geometrically necessary dislocations punched around a particle that form during cooling after consolidation because of coefficient of thermal expansion (CTE) mismatch between the particle and the matrix. An additional strength increase may also be observed, since another type of geometrically necessary dislocation can be formed during extensive deformation as a result of the strain gradient plasticity due to the elastic-plastic mismatch between the particle and the matrix. In this paper, the magnitudes of these two types of dislocations are calculated based on the dislocation plasticity. The dislocations are then converted to the respective strengths and allocated hierarchically to the matrix around the particle in the axisymmetric finite-element unit cell model. The proposed method is shown to be very effective by performing finite-element strength analysis of $SiC_p$/Al2124-T4 composites that included ductile failure in the matrix and particlematrix decohesion. The predicted results for different particle sizes and volume fractions show that the length scale effect of the particle size obviously affects the strength and failure behavior of the particle-reinforced metal matrix composites.

• Transactions of Materials Processing
• /
• v.32 no.2
• /
• pp.74-80
• /
• 2023

A deformation behavior of commercially pure titanium (CP-Ti) is highly dependent on material and processing parameters, such as deformation temperature, deformation direction, and strain rate. This study aims to predict the multivariable and nonlinear tensile behavior of CP-Ti using machine learning based on three algorithms: artificial neural network (ANN), light gradient boosting machine (LGBM), and long short-term memory (LSTM). The predictivity for tensile behaviors at the cryogenic temperature was lower than those in the room temperature due to the larger data scattering in the train dataset used in the machine learning. Although LGBM showed the lowest value of root mean squared error, it was not the best strategy owing to the overfitting and step-function morphology different from the actual data. LSTM performed the best as it effectively learned the continuous characteristics of a flow curve as well as it spent the reduced time for machine learning, even without sufficient database and hyperparameter tuning.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.1 no.1 s.1
• /
• pp.117-125
• /
• 2001

This paper presented the analysis of stress regimes in and around inactive and active fault zones. The stress regime in the vicinity of an existing inactive fault zone is dependent on the orientation of the fault with respect to the current stress field and the contrast between the elastic properties of the faulted rock and those of the surrounding rock. In the analysis of stress regimes around an active fault zone, if the yielding stress is exceeded during loading, the localized shearing in a fault zone will result in weakness with mean stresses in the fault becoming lower than those in the surrounding rock. It can be expected that such stress gradients will induce fluid flow towards the faults zone.