Designing state-of-the-art gas turbine engines with high thrust-to-weight, lower TSFC, advanced operability and improved durability, is an enormous challenge. The use of casing treatments to improve stall margin is a well-known technique for tip-limited compressors. The effectiveness of these treatments is based on the idea that for a tip-limited compressor, with a smooth casing, stall is initiated near the casing first because of the high incidence induced on the rotor blade due to the interaction with the casing boundary layer. Novel casing treatments show significant promise in improving compressor operating range, however, this improvement must not come at the expense of the compressor efficiency. To maintain a balance of operating range and efficiency the shape and positioning of the treatments must be precisely tailored to each design. Currently, traditional CPU based codes are too slow to make these tools practical within a design system with resources attainable by a design engineer. For a design and optimization process to be practical within a design cycle, the tool must be able to evaluate multiple generations of a design, with each generation consisting of multiple individual design iterations, all within a week. This is not possible with the resources available today. The focus of this proposal is to develop the fast and accurate design optimization technology needed by the industry to support the additional geometric complexity of next generation of casing treatments for compressors and fans. Building upon the commercially available GPU flow solver from ADS CFD, we will develop the physics-based tool chains to allow the design of next-generation casing treatment technology using unstructured meshes in a full wheel unsteady simulation and optimization loop.
