Enhancing multiscale simulations for spark plasma sintering with a novel Direct FE$^2$ framework

TL;DR

This paper presents a novel multiscale simulation framework for spark plasma sintering that seamlessly couples micro- and macroscale phenomena, significantly improving accuracy and computational efficiency over traditional methods.

Contribution

The paper introduces a Direct FE$^2$ multiscale framework that enables fully coupled electro-thermal-mechanical simulations in SPS, with improved accuracy and reduced computational cost.

Findings

Achieves below 1% error in temperature and displacement compared to full FE analysis.

Reduces computational degrees of freedom by a factor of 8, accelerating simulations by 70 times.

Provides flexible modeling of diverse powder morphologies without loss of precision.

Abstract

The spark plasma sintering (SPS) process, a key technology for advanced material manufacturing, demands accurate and efficient simulation tools to capture the complex electro-thermal-mechanical interactions inherent in powder materials. This paper introduces a novel concurrent multiscale framework employing the Direct FE$^2$ method, designed for fully coupled electro-thermal-mechanical simulations in SPS. The model integrates microscale powder characteristics into a macroscopic analysis through multi-point constraints within a 3D finite element (FE) solver. This approach enables, for the first time, a direct and seamless coupling of micro- and macroscale physical phenomena, enhancing both accuracy and computational efficiency by capturing interactions across scales. The proposed method achieves a temperature and displacement error margin below 1% compared to full FE analysis while reducing computational degrees of freedom by a factor of 8, resulting in a 70-fold acceleration in simulation time. Additionally, the methodology provides robust flexibility in accommodating diverse powder morphologies without compromising precision, enabling degree-of-freedom reductions of up to 44 times. This combination of enhanced efficiency and accuracy establishes the proposed Direct FE$^2$ approach as a highly effective tool for realistic and scalable simulations of the SPS process.