An Approach to Integrated Multi-Disciplinary Turbomachinery Design

Abstract

Aeroengines are designed using fractured processes. Complexity has driven the design of such machines to be subdivided by specialism, customer, and function. While this approach has worked well in the past, with component efficiencies, current material performance, and the possibilities presented by scaling existing designs for future needs becoming progressively exhausted, it is necessary to reverse this process of disintegration. Our research addresses this aim. The strategy we use has two symbiotic arms. The first is an open data architecture from which existing disparate design codes all derive their input and to which all send their output. The second is a dynamic design process management system known as “SignPosting.” Both the design codes and parameters are arranged into complementary multiple level hierarchies: fundamental to the successful implementation of our strategy is the robustness of the mechanisms we have developed to ensure consistency in this environment as the design develops over time. One of the key benefits of adopting a hierarchical structure is that it confers not only the ability to use mean-line, throughflow, and fully 3D computational fluid dynamics techniques in the same environment, but also to cross specialism boundaries and to insert mechanical, material, thermal, electrical, and structural codes, enabling exploration of the design space for multi-disciplinary nonlinear responses to design changes and their exploitation. We present results from trials of an early version of the system applied to the redesign of a generic civil aeroengine core compressor. SignPosting has allowed us to examine the hardness of design constraints across disciplines which has shown that it is far more profitable not to strive for even higher aerodynamic performance, but rather to improve the commercial performance by maintaining design and part-speed pressure ratio stability and efficiency while increasing rotor blade creep life by up to 70%.