NASA · NASA-526 · record R0270

Design Optimization of Turbomachinery Rotor Blades using Neural Network Surrogate Models

National Aeronautics and Space Administration · GRC: Glenn Research Center

Overview

Development stage: PilotSource: b) Pilot – The use case has been deployed in a limited test or pilot capacity.
High-impact designation: Not high-impactSource: c) Not high-impact
Impact justification: Not reported
Start date: 2025-05-01 (day precision)Source: 05/01/2025 00:00:00
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Science
Operational functions: No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

A sample is made of a design space using Latin hypercube sampling. The geometry for these samples is then generated and evaluated using simulation tools. The result is then used to train a neural network for use as a surrogate model in design optimization.

Expected benefits

Early work shows R^2 > 0.99 (compared to FEA results) on both test and validation datasets when predicting structural performance of a rotor blade as a function of 6 design variables

System outputs

Current models provide predictions for the margin of safety associated with a rotor blade geometry, enabling faster design optimization than using FEA alone.

Technology

AI classification: Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…
System name(s): NEST (WCS) SSP
Custom-developed code: No
Public source code: Not reported

Sourcing

How it was built: Developed in-houseSource: b) Developed in-house
Vendor (as reported): Not reported
Vendors (standardized): None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: No
Privacy Impact Assessment: Not reported
Authorization to Operate: Yes
Demographic features: Not reported
Training and evaluation data: Data consists of Finite Element Analysis (FEA) simulations of rotor blades, generated with ANSYS Mechanical. Solutions were split into separate training, validation, and test datasets.
Federal Data Catalog: Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Pre-deployment testingNo answer reported

AI impact assessmentNo answer reported

Independent reviewNo answer reported

Ongoing monitoringNo answer reported

Operator trainingNo answer reported

Fail-safeNo answer reported

Appeal processNo answer reported

User and public consultationNo answer reported

Potential impacts and how they were identified

Not reported