Development and Realisation of Efficient Probabilistic Methods in the Design of Turbine Blades
Project director: | Prof. Dr.-Ing. Konrad Vogeler |
Staff: | Dipl.-Ing Thomas Bischoff |
Dipl.-Ing Matthias Voigt | |
Scientific cooperation: | AG Turbo III GuD 500 MW auf einer Welle |
Funding: | Rolls-Royce Deutschland Ltd & Co KG , MTU Aero Engines AG, ABB ALSTOM Power GmbH |
Durationt: | 11/05 - 10/08 |
Description:
A drawback of all probabilistic methods is the higher computational cost compared to deterministic calculations due to the necessary multiple calculations. At this point the theoretical part of this project is tied in. Based on the expertise of the project “Probabilistic Mechanical Design of Turbines”, which has been done within the cooperative project “GUD - Power Plant – 500MW on one Shaft”, additional probabilistic methods and techniques have to be analysed and implemented in a software tool to achieve a further reduction of the computational costs inherent in probabilistic design methods. Besides this main goal, an improved qualitative description of the result parameters (probabilities of failure, sensitivities of the results …) of a probabilistic simulation should be achieved using these new methods.
The usage of probabilistic methods by the industry in the design of highly loaded turbine blades is limited by the lack of a validated applicability prove for lifetime assessment under loads common for the components and with known distributions of the input parameters. This gap is to be closed by experiments on components and probabilistic calculations recreating these experiments with regard to the parameters
- geometry,
- temperature,
- load regime,
- environmental conditions
- multi-axle loads
decisive for component failure.