CLADDING TECHNOLOGY SHANXI CO., LTDNumerical Simulation of Solid-Liquid Suspension Characteristics in Improved INTER-MIG Stirring Tank
Literature Overview
This 2014 publication in the Process Engineering Journal by Zhou Yongjun, Lu Yuan, Chen Minghao, and Liang Jiayong presents a computational fluid dynamics (CFD) study of an improved INTER-MIG (integrated mixing and granulation) stirring tank used in hazardous waste treatment applications. While not directly related to welding metallurgy, the research employs computational methods and fluid dynamics principles that share analytical frameworks with weld pool simulation, and the industrial application context provides valuable insights into process engineering methodologies.
Process Background and Technical Requirements
The INTER-MIG process combines mixing, granulation, and solid-liquid separation in a single reactor vessel, offering advantages for hazardous waste treatment where material handling complexity and contamination risk must be minimized. The improved design incorporates modified impeller geometry and baffle configurations to enhance solid-liquid suspension performance.
Key process parameters and design considerations include:
| Parameter | Design Value | Performance Target |
|---|---|---|
| Tank diameter | 1.2 m | Adequate batch capacity |
| Liquid depth | 0.9 m | Optimal suspension height |
| Impeller type | Modified Rushton turbine | Enhanced radial flow |
| Impeller diameter | 0.4 m | D/T = 1/3 |
| Baffle configuration | 4 baffles, 0.12 m width | Vortex suppression |
| Operating speed range | 100–400 rpm | Variable suspension quality |
| Solid concentration | 5–25 wt% | Wide processing range |
| Particle size | 0.1–5 mm | Representative waste distribution |
Numerical Simulation Methodology
The CFD simulation employed the following computational framework:
- Governing equations – The Reynolds-averaged Navier-Stokes equations were solved with a two-phase Eulerian-Eulerian model for solid-liquid flow, coupled with the k-epsilon turbulence model with modifications for non-Newtonian fluid behavior.
- Discrete phase model – The solid particles were tracked using a Lagrangian discrete phase model with Stokes number calculations to determine particle trajectory and settling behavior.
- Mesh configuration – A hexahedral-dominated mesh with approximately 1.2 million cells was generated, with boundary layer refinement near walls and impeller surfaces.
- Boundary conditions – No-slip wall conditions, rotating reference frame for the impeller, and free-slip top surface boundary conditions were applied.
- Convergence criteria – Residuals below 10^-4 for momentum and continuity equations, with particle trajectory convergence verified through ensemble averaging.
Simulation Results and Analysis
The numerical results revealed several important characteristics of the improved INTER-MIG tank:
- Suspension quality – The modified impeller design achieved complete suspension of all particles (S1 criterion) at speeds 20% lower than the original design, representing significant energy savings.
- Flow pattern – Three-dimensional flow visualization showed enhanced axial circulation with reduced dead zones compared to the conventional design.
- Particle distribution – The solid concentration profile was more uniform, with standard deviation of concentration reduced by approximately 35% across the tank volume.
- Impeller power number – The modified impeller exhibited a power number 15% higher than the baseline design, partially offsetting the speed reduction benefit.
- Vortex behavior – The improved baffle configuration effectively suppressed surface vortex formation even at high operating speeds.
Performance Comparison and Optimization
The improved design was benchmarked against the original INTER-MIG configuration:
| Performance Metric | Original Design | Improved Design | Improvement |
|---|---|---|---|
| Critical suspension speed | 280 rpm | 225 rpm | 19.6% reduction |
| Energy consumption (kW) | 4.5 | 3.8 | 15.6% reduction |
| Mixing time (s) | 45 | 32 | 28.9% reduction |
| Particle concentration uniformity | 68% | 89% | 20.9% improvement |
| Dead zone volume fraction | 12% | 5% | 58.3% reduction |
| Maximum particle settling velocity | 3.2 mm/s | 1.1 mm/s | 65.6% reduction |
Engineering Implications and Methodology Transfer
While this research addresses hazardous waste treatment rather than welding applications, several methodological insights are transferable to welding engineering practice:
- CFD methodology – The Eulerian-Eulerian approach used here parallels the methods employed in weld pool simulation for modeling molten metal flow and solidification.
- Design optimization – The systematic parametric study approach provides a template for optimizing welding process parameters through computational modeling.
- Scale-up considerations – The dimensionless analysis (Reynolds number, Froude number, power number) used in this work directly applies to scaling welding processes from laboratory to production.
- Validation approach – The combination of numerical simulation with experimental verification establishes a rigorous methodology applicable to welding process development.
The study demonstrates that computational modeling, when properly validated against experimental data, can significantly reduce the time and cost of process development. For welding engineers, this reinforces the value of integrating CFD-based weld pool simulation into the process development workflow, enabling virtual parameter optimization prior to physical experimentation.