The impact of non-ideal compressible flows on the fluid-dynamic design of axial turbine stages is examined. First, the classical similarity equation is revised and extended to account for the effect of flow non-ideality. Then, the influence of the most relevant design parameters is investigated through the application of a dimensionless turbine stage model embedding a first-principles loss model. The results show that compressibility effects induced by the fluid molecular complexity and the stage volumetric flow ratio produce an offset in the efficiency trends and in the optimal stage layout. Furthermore, flow non-ideality can lead to either an increase or decrease of stage efficiency up to 3-4% relative to turbines designed to operate in dilute gas state. This effect can be predicted at preliminary design phase through the evaluation of the isentropic pressure-volume exponent. 3D RANS simulations of selected test cases corroborate the trends predicted with the reduced-order turbine stage model. URANS computations provide equivalent trends, except for case study niMM1, featuring a non-monotonic variation of the generalized isentropic exponent. For such turbine stage, the efficiency is predicted to be higher than the one computed with any steady-state model based on the control volume approach.

No abstract

No abstract

The Environmental Control System (ECS) is the main consumer of non-propulsive power onboard aircraft, accounting for up to 3-5% of the total fuel consumption. The use of an electrically-driven Vapor Compression Cycle (VCC) system, in place of the conventional Air Cycle Machine (ACM), can lead to both a substantial increase of the Coefficient Of Performance (COP) at cruise conditions, and to a reduction of maintenance costs. The performance of the VCC system is highly affected by the design of its main components, namely, the compact heat exchangers and the high-speed centrifugal compressor. Therefore, the optimal system design requires the use of an integrated design methodology. This work documents the development of a data-driven compressor model based on Artificial Neural Networks (ANNs). The objective is to reduce the VCC model complexity, and the computational cost of the associated optimization problem. The model has been trained on a synthetic dataset composed of 165k unique centrifugal compressor designs generated with an in-house tool, validated with experimental data. The data-driven model has been coupled to an in-house integrated design framework for aircraft ECS, and it has been used to perform the multi-objective optimization of a VCC system for a single-aisle, short-haul aircraft, flying at cruise conditions. The results show that the number of function evaluations used to identify the Pareto front reduces by a factor of three, when leveraging the capabilities of the data-driven model. Moreover, the optimal solutions identified with the novel method cover a wider design space, due to the improved robustness of the VCC system model.

Modeling non-ideal compressible flows in the context of computational fluid-dynamics (CFD) requires the calculation of thermodynamic state properties at each step of the iterative solution process. To this purpose, the use of a built-in fundamental equation of state (EoS) in entropic form, i.e., s = s(e, ρ), can be particularly cost-effective, as all state properties can be explicitly calculated from the conservative variables of the flow solver. This approach can be especially advantageous for massively parallel computations, in which look-up table (LuT) methods can become prohibitively expensive in terms of memory usage. The goal of this research is to: i) develop a fundamental relation based on the entropy potential; ii) create a data-driven model of entropy and its first and second-order derivatives, expressed as a function of density and internal energy; iii) test the performance of the data-driven thermodynamic model on a CFD case study. Notably, two Multi-Layer Perceptron (MLP) models are trained on a synthetic dataset comprising 500k thermodynamic state points, obtained by means of the Span-Wagner EoS. The thermodynamic properties are calculated by differentiating the fundamental equation, thus ensuring thermodynamic consistency. Conversely, thermodynamic stability is properly enforced during the regression process. Albeit the method is applicable to the development of equation of state models for arbitrary fluids and thermodynamic conditions, the present work only considers siloxane MM in the single phase region. The MLP model is implemented in the open-source SU2 software [8] and is used for the numerical simulation of non-ideal compressible flows in a planar converging-diverging nozzle. Finally, the accuracy and the computational performance of the data-driven thermodynamic model are assessed by comparing the resulting flow field, the wall time and the memory requirements with those obtained with direct calls to a cubic EoS, and with a LuT method.