Advanced Numerical Simulations - 5EUS5SNA

Goal(s)

Knowledge of CFD (Computational Fluid Dynamics) solver with a specific focus on energy systems design. Accurate use of CFD solver needs to well understand modeling of physical process and their numerical resolution (numerical method, turbulence models, multiphase flows, complex geometries, ...). First part of these courses will be a complement of previous academic years in fluid mechanic and numerical methods.
After this first common part, 3 specific courses will be available:
• Specific overview for numerical resolution of thermal hydraulic problems
• Specific overview of meshing and geometric models
• Specific overview on optimization and optimal design

Content(s)

Part 1: Computational Fluid Dynamic (CFD) for the design of energy systems (English course)
• Turbulence model
• Modeling of multiphase flows
• Complex geometries and Verification & Validation in CFD
• Project: analyze of energy systems with commercial CFD code.

Part 2: Specific overview
Option 1: thermal hydraulic problems for nuclear plants. From "systemic" approach (1D / quasi 1D modeling of all the system) to "local" approach (3D-CFD for specific component). Overall methodology will be applied on a simplify case.
Option 2: Geometric models for various application (thermal, fluid, structure, ...): data exchange, model simplification, model optimization
Option 3 « Optimization for the design of energy systems »
Some selected optimization techniques are described in detail to develop an in-depth knowledge of the user-defined parameters upon which the engineer can play in order to maximize the performance of these techniques (faster convergence to a global optimum). Since the course is oriented towards engineering and not mathematical optimization the cost of the optimization process will be carefully analyzed and practicals means to reduce this cost by the use of surrogate models will be proposed and described. Selected optimization techniques will be applied to some multi-objective multi-parameter engineering problems and analysis tools such as parallel coordinates, self-organizing maps will be described and also applied. Techniques for robust design optimization will be finally reviewed in order to avoid selecting an over-sensitive optimal design. The lectures hours will systematically alternate with lab work sessions in Matlab and python programming. Once skilled enough with the optimization alogithms, the students will apply it to the solution of a real-life optimal design problem on which they will write a full technical report.

Good knowledge in fluid mechanic (turbulent and laminar flows), heat transfer, basic in numerical simulation for fluid and basic in multiphase flows.