Foundations

The idea, where it sits next to testing and simulation, and the physical data it learns from.

• What a surrogate is: A fast, learned map from a design to its physics, from a few numbers to the full field.
• Where it fits: Tests validate simulations, simulations train the surrogate, the surrogate picks what to simulate next.
• A short history: Seventy years from response surfaces and kriging to neural fields on any geometry.
• Physical fields: Decide which quantities the model must reproduce before choosing the model.
• Surface and volume fields: Predict the field everywhere in the domain, not just on the walls: velocity, pressure, temperature in the volume.
• Representations: Voxel grids, graphs or point clouds: how geometry and fields become numbers.
• Sampling the design space: Spread a limited simulation budget evenly over the designs you care about.