<p>A multi-physics multi-scale model of multi-layer laser powder bed fusion (LPBF) process in additive manufacturing is developed. In macro-scale, the Shan-Chen-type lattice Boltzmann method and a modified cellular automata method are fully coupled to reproduce the molten pool dynamics and grain growth. In meso-scale, the multi-orientation dendritic growth is computed in a domain around the molten pool boundary using the phase field method based on the macro-scale computed temperature gradient, solidification velocity, and crystallographic orientation. LPBF experiments with different process parameters are carried out, and the characterization results agree well with the simulation results in both scales. Both experimental and numerical results show the porosity increases with an increase in laser scanning speed and a decrease in laser power. Periodic pore distribution and serrated grain boundary are observed in both results, and the formation mechanisms are discussed based on the simulation results. In addition, both experimental and meso-scale numerical results show that Laves phase tends to form a long-chain morphology at a lower cooling rate and a discrete morphology at a higher cooling rate. Polycrystalline solidification simulation shows that the Laves phase exhibits a bright long-chain morphology at the diverging grain boundary, while it shows a discrete band morphology at the converging grain boundary. The meso-scale Laves phase formation mechanisms are also discussed.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Multi-Scale Numerical Modeling and Experimental Study of Molten Pool Dynamics and Solidification During the Additive Manufacturing of Inconel 718 Superalloy

  • Ling Shi,
  • Songzhe Xu,
  • Yaqing Hou,
  • Tao Hu,
  • Yuequn Wu,
  • Chaoyue Chen,
  • Jiang Wang,
  • Zhongming Ren

摘要

A multi-physics multi-scale model of multi-layer laser powder bed fusion (LPBF) process in additive manufacturing is developed. In macro-scale, the Shan-Chen-type lattice Boltzmann method and a modified cellular automata method are fully coupled to reproduce the molten pool dynamics and grain growth. In meso-scale, the multi-orientation dendritic growth is computed in a domain around the molten pool boundary using the phase field method based on the macro-scale computed temperature gradient, solidification velocity, and crystallographic orientation. LPBF experiments with different process parameters are carried out, and the characterization results agree well with the simulation results in both scales. Both experimental and numerical results show the porosity increases with an increase in laser scanning speed and a decrease in laser power. Periodic pore distribution and serrated grain boundary are observed in both results, and the formation mechanisms are discussed based on the simulation results. In addition, both experimental and meso-scale numerical results show that Laves phase tends to form a long-chain morphology at a lower cooling rate and a discrete morphology at a higher cooling rate. Polycrystalline solidification simulation shows that the Laves phase exhibits a bright long-chain morphology at the diverging grain boundary, while it shows a discrete band morphology at the converging grain boundary. The meso-scale Laves phase formation mechanisms are also discussed.