Poly-dimensional gas turbine system modelling and simulation

• Main Author: Bala, Arjun
• Other Authors: Pilidis, P.
• Published: Cranfield University 2007
• Subjects: 621.43
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.695262

The intense global competition in the commercial aviation and power generation industry is placing a significant pressure on minimizing cost while meeting challenging goals in-terms of performance, efficiency, emissions and reliability. During recent years an opportunity has been identified and is currently being focused on, for reducing the design and development cost by largely replacing the larger scale and expensive hardware based testing, with multi-fidelity and cross-domain predictive simulation platforms. A greater use of such predictive simulations not only save some costs directly associated with hardware design and testing but also enables engine design trade-offs and component interactions, to be studied in detail earlier on before a commitment is made towards the final hardware design. Experts have estimated a reduction of 30 to 40% in development time and cost when such dynamic simulation techniques are implemented. Furthermore, to keep the engine development on going, joining forces with various gas turbine industrial manufacturers, research centres and universities especially within the European Union is of utmost importance because tomorrow's advanced engine configurations can no longer be developed with today's simulation tools in the way they are currently used. The research work, presented within this thesis has been conducted under the flagship of "VIVACE-ECP" (Value Improvement through a Virtual Aeronautical Collaborative Enterprise - European Cycle Program) an European Union sponsored collaborative research project, geared towards the development of an advanced gas turbine performance modeling and simulation platform with cross domain analysis capability. The research work undertaken by the author within the scope of this thesis and the project, fundamentally encompasses around the two distinct aspects; 1) development of a new and modern (0-0) gas turbine performance simulation industrial core tool called as PROOSIS and 2) development in the form of a prototype demonstrator a multi-fidelity simulation technology, fundamentally aiming to reduce engine development cost and time. The new and modern PROOSIS application framework conforms to an 00 programming schema giving the tool features in terms of flexibility, extensibility, robustance, etc. Although, PROOSIS has been envisaged as a long term development process, several of its current capabilities and component modelling philosophies have been discussed in detail. The prototype (3-D) integrated Aerodynamic Component Zooming Framework makes the optimal use of two different simulation platforms at different fidelity levels, thus allowing for variable' complexity analysis to be performed as required. In order to demonstrate the prototype (3-D) integrated Aerodynamic Component Zooming Framework a case-study has been developed. The case study is to study the effect of VSV on a single stage compressor (or fan) during part speed performance and which was successfully completed. The integrated component zooming technique has been performed using a custom developed workflow management tool referred to as "Integrated Workflow Controller" making use of a distributed computing architecture. The key contribution of the author within the scope the project and the thesis has been the development of the modern object oriented GT (0-0) cycle code PROOSIS framework and the development of the modern (3-D) integrated aerodynamic compressor zooming framework. Within this thesis, full and comprehensive information on the research work undertaken by the author in order to achieve the above discussed goals, along with suitable results have been presented. Also discussed in detail are results generated as a part of the software testing, verification and validation of both 1) PROOSIS and 2) (3-D) Integrated Aerodynamic Component Zooming Framework. In an effort to reduce engine development cost and time as discussed earlier, the research work undertaken by the author part of the CU team has made an extensive and optimal use of modern, sophisticated and cross domain, numerical simulation technology readily available and affordable, at different fidelity levels. Additionally, the collaborative effort which has been another key aspect of the project in creating a standard and a modem GT simulation framework (with a prototype component zooming capability) for the advanced gas turbine systems in future has also been achieved. This has been possible by mutually sharing the technical expertise between all participating GT industrial manufacturers, research centres and universities within the European Union. It is the author's opinion that both of the above highlighted developments form a strong foundation for future technological developments leading to an even more sophisticated and capable, multi-disciplinary and multi-fidelity simulation environment which will lead to a significant reduction in engine development cost and time.