US 12,437,130 B2
License-free surrogate model generation
Pieter Van Gils, Madrid (ES); and Can Onur, Munich (DE)
Assigned to The Boeing Company, Arlington, VA (US)
Filed by The Boeing Company, Chicago, IL (US)
Filed on Aug. 27, 2021, as Appl. No. 17/459,636.
Claims priority of application No. 20382916 (EP), filed on Oct. 21, 2020.
Prior Publication US 2022/0122207 A1, Apr. 21, 2022
Int. Cl. G06F 30/20 (2020.01); G06F 21/10 (2013.01); G06Q 40/12 (2023.01)
CPC G06F 30/20 (2020.01) [G06F 21/105 (2013.01); G06Q 40/12 (2013.12)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, by a device, a proprietary model as a functional-mockup unit (FMU);
performing, by the device, automated dataset generation on the FMU to create datasets based on design of experiments and input requirements;
determining, by the device, steady-state operational-points;
generating, by the device, a tool-agnostic surrogate model based on the datasets and the steady-state operational-points;
validating, by the device, the tool-agnostic surrogate model based on the datasets,
wherein validating the tool-agnostic surrogate model comprises:
iteratively performing, until the datasets meet a threshold requirement, one or more of:
performing the automated dataset generation,
determining the steady-state operational-points, or
generating the tool-agnostic surrogate model; and
outputting, by the device, the tool-agnostic surrogate model as an output FMU model that is free of licensing requirements of a license for a proprietary model, based on the tool-agnostic surrogate model being validated.