Prediction of Melt Pool Dimension and Residual Stress Evolution with Thermodynamically-Consistent Phase Field and Consolidation Models during Re-Melting Process of SLM

Re-melting process has been utilized to mitigate the residual stress level in the selective laser melting(SLM)process in recent years.However,the complex consolidation mechanism of powder and the different material behavior after the first laser melting hinder the direct implementation of the re-melting process.In this work,the effects of re-melting on the temperature and residual stress evolution in the SLM process are investigated using a thermo-mechanically coupled finite element model.The degree of consolidation is incorporated in the energy balance equation based on the thermodynamically-consistent phase-field approach.The drastic change of material properties due to the variation of temperature and material state is also considered.Using the proposed simulation framework,the single-track scanning is simulated first to predict the melt pool dimension and validate the proposed model with the existing experimental data.The obtained thermal histories reveal that the highest cooling rate is observed at the end of the local solidification time which acts as an important indicator for the alleviation of temperature gradient.Then,the scanning of a whole single layer that consists of multiple tracks is simulated to observe the stress evolution with several re-melting processes.After the full melting of powder material in the first scanning process,the increase of residual stress level is observed with one remelting cycle.Moreover,the predicted stress level with the re-melting process shows the variation trend attributable to the accumulated heat in the tracks.The numerical issues and the detailed implementation process are also introduced in this paper.

A Self-Learning Data-Driven Development of Failure Criteria of Unknown Anisotropic Ductile Materials with Deep Learning Neural Network

This paper first proposes a new self-learning data-driven methodology that can develop the failure criteria of unknown anisotropic ductile materials from the minimal number of experimental tests.Establishing failure criteria of anisotropic ductile materials requires time-consuming tests and manual data evaluation.The proposed method can overcome such practical challenges.The methodology is formalized by combining four ideas:1)The deep learning neural network(DLNN)-based material constitutive model,2)Self-learning inverse finite element(SELIFE)simulation,3)Algorithmic identification of failure points from the selflearned stress-strain curves and 4)Derivation of the failure criteria through symbolic regression of the genetic programming.Stress update and the algorithmic tangent operator were formulated in terms of DLNN parameters for nonlinear finite element analysis.Then,the SELIFE simulation algorithm gradually makes the DLNN model learn highly complex multi-axial stress and strain relationships,being guided by the experimental boundary measurements.Following the failure point identification,a self-learning data-driven failure criteria are eventually developed with the help of a reliable symbolic regression algorithm.The methodology and the self-learning data-driven failure criteria were verified by comparing with a reference failure criteria and simulating with different materials orientations,respectively.