Form 9.B Project Summary

Chron: 971885

Proposal Number: 06.06-2321

Project Title: Automatic Optimization of Engine Component Design Through Neural Networks and Genetic Algorithms

Technical Abstract (Limit 200 words)

The proposed work aims to automate the optimization of aeroengine components through the application of Artificial Intelligence (AI) techniques to state-of-the-art component design tools. The design of aeroengine components is a complex process requiring many steps and hundreds of minute design decisions in an iterative process. These steps range from meanline (or 1D) design through 3D viscous flow calculations (CFD) and finite element structural analysis (FEA). In the proposed work, AI will be employed to optimize the turbomachinery design process and reduce, to the degree possible, repetitive actions required by the

designer. Both neural networks and genetic algorithms will be implemented in a commercially accepted meanline design program, which is part of a complete design system for gas turbine engine components. AI will be used to create and optimize designs which will be compared to designs created and optimized manually. While initial work will be for meanline design, this work will form the foundation for adding AI to other steps in the overall component design process in Phase II. If properly applied and tested, this technology has the potential to revolutionize the design process for turbomachinery, resulting in the ability to achieve superior performance within a reduced design cycle time.

Potential Commercial Applications (Limit 200 words)

Commercialization for this technology will be very straightforward. The proposer is an established publisher of design software tools for the turbomachinery industry, with users such as Pratt&Whitney, General Electric, Allied Signal, Nuovo Pignone, General Motors, Ingersoll-Rand, Dresser-Rand, Solar Turbines, Honda, and MHI. This AI technology will first be commercialized through integration into an existing commercial meanline design program. Later work will extend it to other parts of the complete design system.. The strong client base which the proposer enjoys can be expected to warmly embrace these improvements.

Name and Address of Principal Investigator (Name,

Organization Name, Mail Address, City/State/Zip)

Dr. Peter Weitzman

Concepts ETI, Inc.

4 Billings Farm Road

White River Jct., , VT 05001

Name and Address of Offeror (Firm Name, Mail Address,

City/State/Zip)

Mr. Jonathan Stearns

Concepts ETI, Inc.

4 Billings Farm Road

White River Jct., , VT 05001