Numerical Modeling of Phase Change Phenomena in Particulate Flows

本研究项目的主要目的在于对重力作用下微粒流动中相变现象的行为有一个基本的了解。这种理解将有助于在冶金和应用材料工程中设计新技术，例如铸造耐高温合金，这将有助于节省能源和保护环境。众所周知，经典的多相模型（欧拉-欧拉）往往不能预测多组分材料的重力驱动凝固。考虑到这一点，另一种方法，如欧拉-拉格朗日模型，可以用来充分预测金属合金铸造过程中发生的凝固/熔化过程。在这种方法中，固体粒子的运动是用牛顿运动定律单独处理的。然而，到目前为止，基于欧拉-拉格朗日的相变相关问题模型在文献中很少受到关注，尽管它们可能为包括相变效应在内的复杂流动的可靠数值模拟提供了巨大的潜力。本文的研究主要集中在以下几个方面：1。建立并验证了具有相变效应的颗粒流动的欧拉-拉格朗日模型，其中颗粒的相变使用拉格朗日空间中的三维（3D）离散元模型（DEM）进行建模，而液相则使用欧拉空间中基于计算流体动力学（CFD）的模型进行模拟。2. 应用于相变运动粒子的界面传热传质亚网格模型的建立。这些模型作为DEM和CFD模型之间的耦合。3. 在定义明确的边界条件下，对一个确定的参考系统所进行的实验进行数值模型的验证：这是沉降在水中的“冰”颗粒的融化。本文计划利用现有“冰”实验的直接数值模拟（DNS）来验证固液相界面传热传质的亚网格模型。4. 研究液相的相分布和流动异常，如浮力和阿基米德力之间的相互作用依赖于固体的体积分数。该项目的长期愿景包括开发用于二元合金（AlSi）柱状等轴凝固数值模拟的欧拉-拉格朗日模型。为了达到这个目的，有必要了解在液相中运动的固体颗粒熔化（凝固）的基本原理。最近发表在文献中的实验研究表明，应特别注意解释重力诱导力之间相互作用的异常现象。该项目的短期目标是开发和验证一个基于特定非量纲数（如阿基米德数（Ar）、Grashof数（Gr）和Stefan数（Ste））的固化/熔化颗粒行为的亚网格半经验模型。该项目的成功将有助于提高固化基础知识和多组分材料特性的准确预测。本研究的实际意义在于建立和验证了描述相变时固液相界面传热传质的简化模型。在这个项目中开发的模型将有助于显著改进行业中使用的计算软件，以改进流程，降低加拿大先进材料的生产成本。

The main objective of this research project consists in gaining a fundamental understanding of the behaviour of phase-change phenomena in particulate flows under the influence of gravity. This understanding will help to design novel technologies in metallurgical and applied materials engineering, e.g. such as casting high-temperature resistant alloys, which go on to help save energy and protect the environment. It is well known that the classical multiphase model (Euler-Euler) often fails to predict a gravity-driven solidification of multicomponent materials. In view of this, an alternative approach such as the Euler-Lagrange model can be used to adequately forecast solidification/melting processes occurring during casting of metal alloys. In this approach the motion of the solid particles is treated individually using Newton’s law of motion. So far, however, Euler-Lagrange-based models for phase-change-related problems have met with little attention in the literature, even though they might provide great potential for reliable numerical simulations of such complex flows including phase-change effects. The proposed research focuses on the following aims: 1. Development and validation of an Euler-Lagrange model for particulate flows with phase-change effects, where the particles undergoing the phase change are modelled using the three-dimensional (3D) discrete element model (DEM) formulated in Lagrangian space, while the liquid phase is simulated using a computational fluid dynamics (CFD)-based model formulated in Eulerian space. 2. Development of subgrid models for interfacial heat and mass transfer applied to moving particles undergoing phase change. These models serve as a coupling between DEM and CFD models. 3. Validation of numerical models against experiments carried out for a defined reference system under well-defined boundary conditions: this is the melting of ‘ice’ particles settling in water. Here the Direct Numerical Simulation (DNS) of existing ‘ice’ experiments are planned to validate subgrid models for interfacial heat and mass transfer between solid and liquid phases. 4. Investigation of the phase distribution and flow anomalies in the liquid phase, such as the interplay between buoyancy and Archimedes forces in dependence on the volume fraction of solid. The vision followed by the project in the long-term consists in the development of an Euler-Lagrange model for a numerical simulation of binary alloy (AlSi) columnar-equiaxed solidification. To reach this goal it is necessary to understand the ruling principles of the melting (solidification) of solid particles moving in the liquid phase. Recent experimental studies published in the literature show that special attention should be paid to explaining anomalies relating to interaction between the gravity-induced forces. The short-term goal of this project is to develop and validate a subgrid semiempirical model for the behaviour of solidifying/melting particles depending on specific nondimensional numbers such as the Archimedes number (Ar), the Grashof number (Gr) and the Stefan number (Ste). The project's success will help advance knowledge on the fundamentals of solidification and on the accurate forecast of multicomponent material properties. The practical significance of this research consists in the development and validation of simplified models describing the interfacial heat and mass transfer between solid and liquid phases undergoing phase changes. The models developed in this project will contribute to a significant improvement in the computational software used in the industry to improve processes, reducing the production costs of advanced materials in Canada.