Colloquium: Tharun Raj Gujjala (FPP)
28 October 2019 14:30 - Location: Lecture Room D, Faculty of Aerospace Engineering, Kluyverweg 1, Delft.
Conceptual Robust Design of Turbomachinery Operating with Non-Conventional Fluids
Over the years, applications for the turbomachinery usage and the operating range during its lifetime have widened. The single operating point optimized design leads to a problematic phenomenon called ‘local optimization’ which means that the design has a very narrow range of better performance around that point and deteriorates significantly at other points and thus requires the machine to be robust enough with stable and maximum average performance over the operating conditions range. The principal aim of the thesis is to develop a design model for the conceptual design of a robust multi-stage axial turbine for the given operating conditions range and distribution by employing an optimization approach along with developing a procedure to incorporate the number of stages decision in this framework and quantify the efficiency gain of the robust machine over a single design point optimized machine The two methodologies i.e. RO (Robust Optimization) and MO (Multipoint Optimization) to achieve robustness in any application are implemented in the model and are compared in terms of computational expense and ease of implementation. In MO methodology, two sampling algorithms i.e. FSA (Fourier transform-based Sampling Algorithm) and SSA (Slope-based Sampling Algorithm) were developed for the selection of the most appropriate points from the given distribution. To check the effectiveness of the developed model, two different test cases have been chosen i.e. one operating with the near-ideal fluid and the other one with an organic liquid. From the acquired results, the overall performance of the robust machine design obtained from the model is better than the single point optimized machine design. The obtained gain values for both the test cases can be very significant if the time window of the variation of the operating conditions is lower. The model was validated with the results of efficiency values and its trend obtained by using an opensource code for turbomachinery design i.e. Multall by performing CFD simulations.