12.12 High-Fidelity Validation

Core idea¶

High-Fidelity Validation must be treated as a system-level decision rather than an isolated technique. For a floating wind turbine modeled in OpenFAST and with reduced, LPV, and derivative-function surrogates, state what is fixed, what is optimized, what information is available, and what equations define feasibility.

The relevant quantities are high-fidelity $F_H$ , surrogate $F_L$ , design $p,c$ , and model error $e_M$ . The chapter-level formulation is

\dot x\approx\widehat f(t,x,u,p),\quad e_M=y_H-y_L.

For this section, trace how the choice changes training data, the active constraints, and the implementable engineering design. A method is useful only when its assumptions are explicit and its result answers the same system question as the baseline.

Engineering interpretation¶

Ask three questions:

Which physical, informational, computational, or economic resource changed?
Which objective component or active constraint made the change valuable?
Does the conclusion survive model, disturbance, initialization, uncertainty, and implementation checks?

A practical action is to identify reduced model. Record units and assumptions before optimization, report component objectives and margins afterward, and verify the result using an independent calculation or higher-fidelity model.

Activity 12.12: quantify high-fidelity validation¶

Activity 12.12: quantify high-fidelity validation

Compare model choices for High-Fidelity Validation in a control-oriented surrogate study.

Model	Validation RMSE	Peak-error bias	Runtime	Training points
Low fidelity	9.4%	-6%	0.03 s	0
Surrogate	5.0%	-2%	0.12 s	140
High fidelity	1.56%	0.5%	30.0 s	0

Require RMSE ≤ 5% and absolute peak bias ≤ 3%.

Identify acceptable models.
Compute runtime for 2,000 optimization evaluations.
Add a 50-run high-fidelity validation budget and recompute total time.
Increase surrogate RMSE by 40% outside its training domain and reassess.
State a quantitative trust region for using the selected model.

Chapter summary¶

The chapter connected high-fidelity simulation, training data, dynamic surrogate, optimization, high-fidelity confirmation through one system formulation. Engineering conclusions require aligned models, information, numerical accuracy, and validation.

Common mistakes¶

changing assumptions while comparing alternatives;
reporting objective improvement without verified feasibility;
hiding information, architecture, or uncertainty;
treating solver convergence as validation; and
reporting runtime without accuracy, derivatives, and tolerances.

Exercises¶

Recreate the workflow for a floating wind turbine modeled in OpenFAST and with reduced, LPV, and derivative-function surrogates.
State every variable, unit, dependency, and constraint.
Construct a common sequential or nominal baseline.
Identify active constraints and the physical bottleneck.
Design a test that could falsify the claimed benefit.

Principal sources¶

Sundarrajan and Herber on DFSMs; Bayat and coauthors on multi-fidelity wind CCD.

Open research question¶

How can surrogate trust be quantified near optimizer-driven extrapolation?