Proceedings of GT2010 ASME Turbo Expo 2010: Power for Land, Sea and Air June 14-18, 2010, Glasgow, Scotland GT2010-22228 FRAMEWORK FOR MULTIDISCIPLINARY OPTIMIZATION OF TURBOMACHINERY Mark G. Turner ∗ Kevin Park Kiran Siddappaji Soumitr Dey David P. Gutzwiller Department of Aerospace Engineering University of Cincinnati Cincinnati, OH, USA Ali Merchant CADNexus Burlington, MA, USA Dario Bruna NASA Glenn Research Center Cleveland, OH, USA ABSTRACT A multidisciplinary optimization framework is presented for turbomachinery that looks at weight and efficiency as multiple objective functions. Both the blades and disks are considered in a multi-level optimization approach. An axisymmetric solver with loss models is used for the flowpath and blade design, and opti- mized disks are created at each step of the process. Constraints include temperature dependent strength requirements for many common materials. The other constraint limits the work done by the component. A genetic algorithm is used to find the pareto front for the multi-objective functions. Optimization of the 10 stage GE EEE compressor is presented to demonstrate the framework. Detailed parameter based CAD models are also produced so these can be used as a starting point for higher fidelity opti- mization. INTRODUCTION In the 2007 IGTI Scholar Lecture, Professor Edward Gre- itzer [1] described the importance of multidisciplinary ap- proaches and collaboration in the design of aircraft engines as well as in research. This is also true in the role of optimization. A single discipline optimization will not give a system-wide opti- mum because of the balance of issues related to jet engine design. ∗ Address all correspondence to this author. Email: mark.turner@uc.edu Performance and weight are extremely important, but ultimately the components must not fail, and cost is always an issue for the designer. Acoustics, aeromechanics, controls, heat transfer and material science must be part of the system design. Mainte- nance, manufacturability, aircraft integration, and business mod- eling must also be addressed in a successful aircraft engine or gas turbine design. There have been several recent presentations of optimiza- tion in turbomachinery. Several papers describe high-fidelity op- timization of a single blade row with a single objective func- tion [2–5]. A multidisciplinary approach connected to CAD was presented by Staubach [6]. One paper by Kipouros et al. [7] de- scribes a multi-objective approach for looking at blockage and loss for the optimization of a single blade row. This paper presents a potential framework for multidisci- plinary design optimization that also addresses multi-objective functions, and multi-fidelity. Figure 1 shows the flowchart for the proposed process for optimizing turbomachinery components within an overall engine cycle optimization. This approach to multi-fidelity allows for an extremely broad design space to be explored with appropriately low-level and very fast tools, and to narrow the design space with higher fidelity tools. There are sev- eral key points about this process: 1. The engine cycle code NPSS drives the overall optimization [8–10] 2. The turbomachinery component design is started with mean- 1 Copyright c  2010 by ASME