CFD Simulation Steering
You've already heard some of the advantages of using co-processing over after-the-fact post-processing, but there's more. Co-processing is an enabler of Computational Fluid Dynamics (CFD) simulation 'steering', i.e., making changes to the simulation while it is active.
Relaxation factors changed at 200 iterations
Co-processing provides immediate feedback on the state of a CFD simulation through 3D visualization and 2D plots while the simulation is running. With this feedback you can interrupt a simulation and make necessary changes to 'steer' the simulation to a more optimal outcome. For example you will often find that initially a stead-state simulation requires conservative (smaller) relaxation factors, but later you can use more aggressive (larger) relaxation factors to increase the rate of convergence per iteration. You can time such an intervention based on the feedback from immediate results updated due to co-processing.
If the immediate updated results indicate that the simulation is not progressing in the manner you'd expect (e.g., unrealistic velocity values), then you can interrupt the simulation and make the necessary changes (e.g., add extra mesh resolution) without wasting time waiting for the simulation to complete its run.
Co-processing is useful on many levels, in terms of reducing data management for transient simulations and for steering a simulation, and that's why it's a core feature of Caedium.
Recent blog posts
- Sliver Treatment Strategies for CFD
- How to Fix Small Acute Angles for CFD
- Small Feature Removal for CFD
- Fluid Visualization in Nature
- Design is Compromise
- CFD Is Not Enough
- All CFD Should Be Upfront CFD
- Speed Up CFD with Symmetry and Cyclic Conditions
- Airships Rising
- World Cup Stadium Aerodynamics with CFD