Optimal Durability and Reliability Testing of Gas Turbine Components

Through the traditional design and development cycles, prototype hardware components are built, tested, and the design is then modified for unforeseen shortcomings revealed in testing. These cycles are both too time consuming and costly. The cost of physical tests is rising while the cost of computer cycles is plummeting. It seems only common sense to replace the former with the latter. This Phase I will demonstrate feasibility of developing a integrated framework to demonstrate the durability and reliability of turbine engines. This will be accomplished through the use of physics based probabilistic simulations, scaled turbine engine testing and statistical evaluation. The combination of scaled testing and Weibull-based probabilistic analysis, full scale testing can be dramatically reduced to a highly cost effective function unlike today. The modeling approach proposed herein will highlight issues and problems well before any actual full-scale testing takes place. The envisioned framework will consider the inherent uncertainty that exists in real world systems through the combination of multiple engine tests to provide a sound statistical database of component durability and Weibull analysis and Bayesian techniques to minimize uncertainty at the system level.