ICME at GE: Accelerating the Insertion of New Materials and Processes

JOM, Nov 2006 by Backman, Daniel G, Wei, Daniel Y, Whitis, Deborah D, Buczek, Matthew B, Et al

The accelerated insertion of materials (AIM) initiative provides the opportunity to reduce the materials development cycle time by up to 50% and thereby lessen the lead time required for new materials and processes. The program was founded to revolutionize the way designers and materials engineers interact, to achieve a leap forward in the application of computational materials science and integration with design engineering tools, and to create an environment where the design/materials team can learn from and build on previous developments. The centerpiece of the AIM system is the designer knowledge base, which provides a framework for managing experimental data, executing linked models describing processing, microstructure, properties, and producibility, and calculating confidence bounds for system predictions.

INTRODUCTION

New materials and processes enable, and often pace, developments in industry. This has certainly been the case for the gas turbine, where a new blade alloy or new melting process can change the face of modern military and commercial aviation. A relentless pursuit continues for "better, faster, cheaper," driven by the market forces that are shaped by customerexpectations for improvement. The reality is thai materials are now the pacing element in achieving further signilieant improvements in engine technology. Design engineers are running out of strategies to work around materials limitations, and engine programs are disrupted when new materials are not ready on time. The accelerated insertion of materials (AIM) initiative was conceived, sponsored, and monitored by the U.S. Defense Advanced Research Projects Agency (DARPA) and the U.S. Air Force Research Laboratory (AFRL) to combat lhis problem.

The AIM initiative provides the opportunity to reduce the materials development cycle time hy up Io .50% and thereby lessen the lead time required for new materials and processes. This methodology can be used not only for new alloy introductions, but also for new part or new vendor qualifications. process modilications. or even material nonconformance issues. The AIM program jump-started the concept of integrated computational materials engineering (ICME). being the first to address the growing need in the aviation industry for a coordinated strategy of linking the computational materials tools already available Io explore material chemistry and process design space forihc optimum alloy composition and thermomechanical processing for a given application.

From 2001-2004, General Electric (GE Aviation conducted re search under the AIM program in collaboration with academia and other partners from industry. While AIM is generically extensible to other materials systems, the GE team used the nickel-based siipcralloy iurbine disk alloy Ren� 88 DT to develop, validate, and demonstrate a system that embodies the AIM methodolog).

The program was founded upon bringing about three systemic changes: to revolutionize the way designers and materials engineers interact, to achieve a leap forward in the application of computational materials science and integration with design engineering tools, and to create an environment where lhe design/ materials team can learn from and build on previous developments. The centerpiece of the AIM system is the designer knowledge base (DKB). which provides a framework for managing experimental data, executing linked models describing processing, microstructure, properties, and producibility and calculating confidence hounds for system predictions. The DKB informs design engineers about material performance and producibility as well us transfers materials information to a design engineering trade study tool. The GE learn established models. uncertainty methods, and use-cases for integration within the DKB.

Materials Models

Acceleration from the integrated AIM system depends on models with sufficient fidelity and robustness to confidently predict processing, microstructure, and mechanical properties. The GE team employed available and newly developed models to predict the effects of key process parameters on microsiriiclure and properties. Modeling efforts have focused on those that capture the physics governing material-system behavior. Some of the commercially available modeling tools include Thermo-Calc (thermoduiamics). DICTRA (kinetics), and DEFORM� large-strain deformation and heat-treat thermal modeling).

Uncertainty

The design of a complex system, such as an aircraft engine, must account for uncertainty that is inherent in materials behavior ami manufacturing processes. The GE AIM team has evaluated data Lind modeling uncertainty and applied Monte Carlo analysis to calculate the uncertainly that is produced by variations in processing history, microstructural features, measurement emirs, and the inadequacy of physically based models.

Use cases

The GE team recognized early that models, digitization utilities, and an integration framework are necessary ingredients of an AlM system. However, purpose and benefit are delivered by identifying important problems whose solutions both accelerale development and reduce risk. Use cases, which codify an accelerating methodology and describe the steps toward solving such problems, are the key to success-they provide direction for AIM tool development and ensure that the individual pieces are coordinated to reap a benefit in the end. The GF team established use cases to aid development and testing of the DKB. These use cases included methodologies to design a heat treatment to better balance properties, identify optimum parameters for a dual heat-treatment process, and calculate heat transfer using inverse methods.