AI Component Evaluation for Faster Turbine Design Iterations

Solution Overview

Problem

The design of gas turbine engine components is hindered by the time-consuming and costly process of producing updated temperature, deflection, stress, and strain result sets due to incremental changes in component geometry and loads, necessitating more efficient computational methods for reducing design iteration time and cost.

Innovation Solution

A component evaluation system utilizing a comparator model and analysis model to compare new component definitions with stored definitions, determining if detailed analysis is needed, and performing iterative or detailed analysis based on geometric and load similarities, thereby reducing the need for extensive computational methods.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If detailed computational analysis is performed for every incremental change in component geometry and loads, then accuracy of temperature, deflection, stress, and strain result sets is maintained, but design iteration time and cost increase significantly

Engineering Contradiction:

Improveaccuracy of result setsVSAvoiddesign iteration time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system creates simplified copies or representations of the component model that can be evaluated quickly without performing full detailed computational analysis. By using surrogate models or reduced-order models that replicate key behaviors, the system maintains acceptable accuracy while dramatically reducing computation time for design iterations.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system performs preliminary screening or preliminary analysis using simplified methods before committing to detailed computational analysis. By pre-evaluating design changes using faster approximation methods, the system identifies which changes require full detailed analysis, thereby reducing overall computation time while maintaining accuracy for critical designs.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If detailed computational analysis is performed for every incremental change in component geometry and loads, then accuracy of temperature, deflection, stress, and strain result sets is maintained, but computational cost increases significantly

Engineering Contradiction:

Improveaccuracy of result setsVSAvoidcomputational cost

Core Design Contradiction:

Measurement precisionVSLoss of energy

Solution Approach 1:

The system creates simplified copies or representations of the component model that can be evaluated quickly without performing full detailed computational analysis. By using surrogate models or reduced-order models that replicate key behaviors, the system maintains acceptable accuracy while dramatically reducing computation time for design iterations.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system performs preliminary screening or preliminary analysis using simplified methods before committing to detailed computational analysis. By pre-evaluating design changes using faster approximation methods, the system identifies which changes require full detailed analysis, thereby reducing overall computation time while maintaining accuracy for critical designs.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If comprehensive component evaluation is performed for all design changes, then reliability of component design is improved, but design iteration speed decreases

Engineering Contradiction:

Improvereliability of component designVSAvoiddesign iteration speed

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system creates simplified copies or representations of the component model that can be evaluated quickly without performing full detailed computational analysis. By using surrogate models or reduced-order models that replicate key behaviors, the system maintains acceptable accuracy while dramatically reducing computation time for design iterations.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system performs preliminary screening or preliminary analysis using simplified methods before committing to detailed computational analysis. By pre-evaluating design changes using faster approximation methods, the system identifies which changes require full detailed analysis, thereby reducing overall computation time while maintaining accuracy for critical designs.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentEP4703950A1Computer system for evaluation of components supported by artificial intelligence / machine learning

Publication Date: 2026.03.04 PRATT & WHITNEY CANADA CORP

Figure 1
Figure 2
Figure 3

AI summary

A method (100) and system (10) for evaluating a component includes receiving (102) a component definition that includes component geometry and one or more of a boundary condition set and a load set. A comparator model (34) compares (106) component geometry to one or more design envelopes (32A), and compares load sets to one or more prior load sets, each from a database of stored component definitions (32F, 32G). Based on the comparison, an analysis model (36) forms (108) a detailed analysis or an iterative analysis to produce a result set. A results module (22) generates (110) an evaluation disposition of the component definition based on the result set.