Experimental studies have been conducted on a modified T106 low pressure turbine (LPT) profile in an annular 1.5 stage axial turbine rig at the Chair of Thermal Turbomachines and Aeroengines, Ruhr-Universität Bochum. The rig setup allows the highly resolved measurement of unsteady wake–stator flow interaction in both space and time. Incoming wakes are generated by a variable-speed driven rotor equipped with cylindrical bars. In the present paper, an experimental approach to the investigation of unsteady phenomena is proposed. Time-averaged and instantaneous measurement data from 2D flow field traverses at the stator exit are provided for the analysis of the periodically unsteady vortex formation, displacement, and suppression. Additional time-accurate blade pressure data are used to study the relationship between the flow structures downstream of the stator row and the immediate intermittent wake impact on the blades. Bar wake kinematics is also discussed in relation to the observations.
We present a mathematically well-founded approach for the synthetic modeling of turbulent flows using generative adversarial networks (GAN). Based on the analysis of chaotic, deterministic systems in terms of ergodicity, we outline a mathematical proof that GAN can actually learn to sample state snapshots from the invariant measure of the chaotic system. Based on this analysis, we study a hierarchy of chaotic systems starting with the Lorenz attractor and then carry on to the modeling of turbulent flows with GAN. As training data, we use fields of velocity fluctuations obtained from large-eddy simulations (LES). Two architectures are investigated in detail: we use a deep, convolutional GAN (DCGAN) to synthesize the turbulent flow around a cylinder. We furthermore simulate the flow around a low-pressure turbine stator using the pix2pixHD architecture for a conditional DCGAN being conditioned on the position of a rotating wake in front of the stator. The settings of adversarial training and the effects of using specific GAN architectures are explained. We thereby show that GAN are efficient in simulating turbulence in technically challenging flow problems on the basis of a moderate amount of training data. GAN training and inference times significantly fall short when compared with classical numerical methods, in particular, LES, while still providing turbulent flows in high resolution. We furthermore analyze the statistical properties of the synthesized and LES flow fields, which agree excellently. We also show the ability of the conditional GAN to generalize over changes of geometry by generating turbulent flow fields for positions of the wake that are not included in the training data.
The experimental results reported in this contribution address the time-dependent impact of periodically unsteady wakes on the development of profile and end wall boundary layers and consequently on the secondary flow system. Experimental investigations are conducted on an annular 1.5 stage axial turbine rig at Ruhr-Universität Bochum’s Chair of Thermal Turbomachines and Aeroengines. The object under investigation is a modified T106 profile low-pressure turbine (LPT) stator row at a representative exit flow Reynolds number of 200,000. By making use of an annular geometry instead of a linear cascade, the influence of curvilinear end walls, nonuniform, increasing pitch across the span and radial flow migration can be represented. Incoming wakes are generated by a variable-speed driven rotor equipped with cylindrical bars. Special emphasis is put on the wake-induced recurrent formation, suppression, weakening, and displacement of individual vortices and separated flow regimes. For this, based on a comprehensive set of time-resolved measurement data, the interaction of impinging bar wakes and boundary layer flow and thus separation and its periodic manipulation along the passage end walls and on the blade suction surface are studied within the frequency domain.
In this work we present the results of the numerical investigations concerning interaction effects between periodically incoming wakes, secondary flow development and boundary layer flow inside a low pressure turbine (LPT) equipped with modified T106-profile blades. The numerical predictions obtained using 3D Unsteady Reynolds-Averaged Navier-Stokes (URANS) simulations are compared against stationary and time resolved experimental results conducted in the same configuration and discussed in part 1 of this 2-part work. For the investigation of the spatial topologies and the dynamics of the coherent flow structures under periodical wake impact, a frequency domain analysis as well as proper orthogonal decomposition (POD) of the unsteady data is proposed and discussed.
In this study the unsteady behavior of the boundary layers developing on a LPT stator profile and their effect on secondary flow patterns in a 1.5-stage turbine configuration are investigated under the influence of periodic inflow perturbations. The experimental setup previously employed to analyze the unsteady secondary flow in the stator wake has been enhanced by hotfilm sensor arrays placed on the stator profiles at different blade height positions to provide time-resolved data from within the passage. The turbine inflow is perturbed by periodically passing circular bars and a modified T106-profile has been considered for the blading. The modified profile, labeled as T106RUB, was developed for matching the transition and separation characteristics of the original T106 profile at low flow speeds, thus facilitating measurements to be taken in a large-scale test rig with its improved accessibility. The transition phenomena occurring in the profile boundary layers are investigated under both unperturbed and periodically perturbed inflow by means of spectral analysis, the semi-quantitative characterization of the wall-stress system and an evaluation of the statistic quantities. In particular, the periodic changes of the suction side boundary layer flow region towards the trailing edge are studied in detail. Furthermore, time-resolved hot-film measurements at different blade height positions facilitate a detailed comparison of the quasi two-dimensional mid-span profile flow and the near end wall profile flow which is subject to influence of secondary flow structures. These information are employed to assess to which extent the additional turbulence originating from the wakes affects the blade boundary layers and thus the secondary flow structures. Furthermore, the role of the perturbation frequency on the coupled system of boundary layers and secondary flow structures is evaluated.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.