To model the maximum temperature at three points of interest (POIs) during an inductive process, we faced a classic challenge: achieving high-fidelity accuracy (fine grid and temporal resolution) while minimizing computational effort. Each simulation depends on three parameters across three segments (current, inductor y-offset, and velocity), leading to a complex, high-dimensional design space.
The challenge
A fully high-fidelity model would need a large number of expensive simulations. A purely low-fidelity model is faster and unreliable. Neither one is a real option.
The solution: multifidelity modeling
Our multifidelity algorithm combined 20 high-fidelity and 69 low-fidelity simulations, 89 samples in total, into one global model that predicts high-fidelity temperature outcomes at all POIs.
An adaptive design of experiments: start from a small, correlated batch of low- and high-fidelity samples, build a data-driven surrogate, then keep proposing new points where high-fidelity insight is worth the most. Only the essential high-fidelity runs get made.
The results
- The multifidelity metamodel reproduces the high-fidelity system behavior at an acceptable error rate.
- A standard machine learning model trained on the same limited high-fidelity dataset performs significantly worse.
- The approach holds: low-fidelity data explores high-fidelity behavior efficiently and reliably.
