The Design Of Turbine Blade Internal Cooling Systems

Research Theme: Process Management

Blade internal cooling systems play a vital role in gas turbines, and the correct design is vital for safe and efficient whole-life operations.

The blade design process is based upon heuristically-guided searches and computer simulations to verify decisions. The process is challenging, especially for innovative systems, and is resistant to automation.

This project identifies and addresses key challenges in order to accelerate the design, verification and selection of the most appropriate cooling system.

Motivation

There is a growing trend for operating cost and risk to be transferred to manufacturer, which in turn influences the design requirements. One consequence of this is that cooling designers are constantly pushed to produce improved performance whilst reducing design and manufacture costs and timescales. To meet these demands, cooling designers must be able to rapidly search the design space or options and develop confidence about a design before committing to costly physical development work.

Objectives

  • Enhance the design space exploration process and thus improve the identificationand validation suitable designs.
  • Enable designers to mature promising design concepts more quickly.
  • Use the confidence in design parameters and choices to provide decisionsupport, enabling designers to create better systems and trade-offs.

Method

This research uses observational studies, practitioner interviews and investigations into specific design tools and activities. Signposting models and maps of the design process have been constructed to bring clarity. Investigations into the software tools are undertaken using a set of industry-standard processes and software tools suitable for preliminary design.

Findings

Several key findings emerged from the Signposting model and process maps. Firstly, the design process contains bottlenecks that frustrate the work of the designer; these are the CAD-based stages that influence the ability of the designer to make and change decisions. Secondly, the decision-making process is unsupported and unguided, leading to the creation of sub-optimal designs. Further research into the decision-making process will be undertaken, to clarify the nature of suitable decision support.

Details

Although much of the design process is automated (figure 2), two key activities are unguided: identifying suitable concepts and creating CAD models that underpin verification models. Significantly, whilst many iterations are required, most iterative loops lead to unguided activities using change information derived by the designer. The CAD definition process is not aligned with creating conceptual designs; disproportionate effort is required to make or change initial choices. Success ultimately depends upon the experience and insight of the designer and the quality of attention given to downstream activities such as manufacture.

Acknowledgements

Selected Publications

  • BELL, C.P., DAWES, W.N., JARRETT, J.P. and CLARKSON, P.J. (2005) Improvingthe conceptual design of turbine rotor blade cooling systems in ICED05, 15th International Conference on Engineering Design, Melbourne, Australia.