LES and RANS Analysis of the End-Wall Flow in a Linear LPT Cascade With Variable Inlet Conditions: Part II — Loss Generation

Abstract

In low-pressure-turbines (LPT) at design point around 60–70% of losses are generated in the blade boundary layers far from end-walls, while the remaining 30%–40% is controlled by the interaction of the blade profile with the end-wall boundary layer. Increasing attention is devoted to these flow regions in industrial design processes. Experimental techniques have shed light on the mechanism that controls the growth of the secondary vortices, and scale-resolving CFD have provided a detailed insight into the vorticity generation. Along these lines, this paper discusses the end-wall flow characteristics of the T106 profile with parallel end-walls at realistic LPT conditions, as described in the experimental setup of Duden and Fottner (1997) “Influence of Taper, Reynolds Number and Mach Number on the Secondary Flow Field of a Highly Loaded Turbine Cascade”, P. I. Mech. Eng. A-J. Pow., 211 (4), pp.309–320. The simulations target first the same inlet conditions as documented in the experiments, and determines the impact of the incoming boundary layer thickness by running additional cases with modified incoming boundary layers. Calculations are carried out by both RANS, due to its continuing role as the design verification workhorse, and highly-resolved LES. Part II of the paper focuses on the loss generation associated with the secondary end-wall vortices. Entropy generation and the consequent stagnation pressure losses are analyzed following the aerodynamic investigation carried out in the companion paper. The ability of classical turbulence models generally used in RANS to discern the loss contributions of the different vortical structures is discussed in detail and the attainable degree of accuracy is scrutinized with the help of LES and the available test data. The purpose is to identify the flow features that require further modelling efforts in order to improve RANS/URANS approaches and make them able to support the design of the next generation of LPTs.

Implicit Discontinuous Galerkin Solution on Unstructured Mesh for Turbine Blade Secondary Flow

“…High-fidelity turbulence eddy resolved (instead of modelled) methods, like large eddy simulation (LES), have been applied to study the endwall flows in the context of low pressure turbine (LPT) cascade [10][11][12]. Turbulence eddy resolved simulations have also been conducted more recently using the DG based methods, typically at relatively low Reynolds numbers.…”

The Current State of High-Fidelity Simulations for Main Gas Path Turbomachinery Components and Their Industrial Impact

“…The study was conducted using incoming endwall boundary layers with different momentum thicknesses and the endwall vortex characteristics and the associated loss were shown to be considerably affected. The results further revealed that RANS calculations, if performed following strict best practice rules, agreed well with LES, in particular in the prediction of the end wall vortex system, with the only significant differences observed in the loss generation of the blade wake, due to the inability of correctly predicting wake mixing [73]. Ciorciari et al [11] performed a similar investigation in presence of discrete incoming wakes generated with moving bars by running both a set of linear cascade experiments and the companion CFD simulations.…”

Machine-Learnt Turbulence Closures for Low-Pressure Turbines With Unsteady Inflow Conditions

“…5(a), when models with higher additional diffusion coefficients are tested on models with lower additional diffusion coefficients. If models developed on phases which require lower additional diffusion (phases 1-4 & 15-20) are tested on phases which require higher additional diffusion (5)(6)(7)(8)(9)(10)(11)(12)(13)(14), they still reduce the error as they contribute to a fraction of the additional diffusion coefficient required to bring about the maximum possible error reduction.…”

“…Much of the design process for LPTs is, however, conducted using Reynolds-averaged Navier-Stokes (RANS) based turbulence models. Accurate prediction of the wake mixing and other complex phenomena in LPTs is still a challenge for existing RANS models as shown by a number of studies [12][13][14]; and there is plenty of scope for improvement in these models. One of the reasons that RANS closures fail at predicting the turbulent wake mixing with the required degree of accuracy is the use of the Boussinesq approximation [15] for the stress-strain relationship.…”