Analytical and Computational Fluid Dynamics Models of Wells Turbines for Oscillating Water Column Systems

The sea is an important renewable energy resource for its extension and the power conveyed by waves, currents, tides, and thermal gradients. Amongst these physical phenomena, sea waves are the source with the highest energy density and may contribute to fulfilling the global increase of power demand. Despite the potential of sea waves, their harnessing is still a technological challenge. Oscillating water column systems operating with Wells turbines represent one of the most straightforward and reliable solutions for the optimal exploitation of this resource. An analytical model and computational fluid dynamics models were developed to evaluate the functioning of monoplane isolated Wells turbines. For the former modeling typology, a blade element momentum code relying on the actuator disk theory was applied, considering the rotor as a set of airfoils. For the latter modeling typology, a three-dimensional multi-block technique was implemented to create the computational domain with a fully mapped mesh composed of hexahedral elements. The employment of circumferential periodic boundary conditions allowed for the reduction of computational power and time. The models use Reynolds-averaged Navier-Stokes (RANS) or u-RANS schemes with a multiple reference frame approach or the u-RANS formulation with a sliding mesh approach. The achieved results were compared with analytical and experimental literature data for validation. All the developed models showed good agreement. The analytical model is suitable for a fast prediction of the turbine operation on a wide set of configurations during the first design stages, while the computational fluid dynamics (CFD) models are indicated for the further investigation of the selected configurations.

Exploring three-dimensional edge-based smoothed finite element method based on polyhedral mesh to study elastic mechanics problems

This paper established the three-dimensional edge-based smoothed finite element method(ES-FEM) based on polyhedral mesh, divided the smoothed domain, constructed the shape function and derived the geometric matrix and the stiffness matrix. The MATLAB software was used to prepare the corresponding computing programs, with which the paper studied the stress distribution of a hollow sphere model and a beam model under different numbers of polyhedral elements. The paper compared the calculation results from the conventional finite element methods(FEM) that use tetrahedral elements and hexahedral elements respectively in terms of stress relative error and energy relative error. The comparison results show that the three-dimensional ES-FEM based on polyhedral mesh has better precision and convergence than the conventional FEM and better adaptability to complex geometric structures.

Scale Resolving CFD Investigations of Aerothermal Field and Emissions of a Lean Burn Aeroengine Combustor

Due to the increasingly stringent international limitations in terms of NOx emissions, the development of new combustor concepts has become extremely important in order for aircraft engines to comply with these regulations. In this framework, lean-burn technology represents a promising solution and several studies and emission data from production engines have proven that it is more promising in reducing NOx emissions than rich-burn technology. Considering the drawbacks of this combustion strategy (flame stabilization, flashback or blowout or the occurrence of large pressure fluctuations causing thermo-acoustics phenomena) as well as the difficulties and the high costs related to experimental campaigns at relevant operating conditions, Computational Fluid Dynamics (CFD) plays a key role in deepening understanding of the complex phenomena that are involved in such reactive conditions. During last years, large research efforts have been devoted to develop new advanced numerical strategies for high-fidelity predictions in simulating reactive flows that feature strong unsteadiness and high levels of turbulence intensity with affordable computational resources. In this sense, hybrid RANS-LES models represent a good compromise between accurate prediction of flame behaviour and computational cost with respect to fully-LES approaches. Stress-Blended Eddy Simulation (SBES) is a new global hybrid RANS-LES methodology which ensures an improved shielding of RANS boundary layers and a more rapid RANS-LES “transition” compared to other hybrid RANS-LES formulations. In the present work, a full annular aeronautical lean-burn combustor operated at real conditions is investigated from a numerical point of view employing the new SBES approach using poly-hexcore mesh topology, which allows to adopt an isotropic grid for more accurate scale-resolving calculations by means of fully regular hexahedral elements in the main stream. The results are compared to experimental data and to previous reference numerical results obtained with Scale Adaptive Simulation formulation on a tetrahedral mesh grid in order to underline the improvements achieved with the new advanced numerical setup.

Performance metrics of the fixed stress split algorithm for multiphase poromechanics

We focus on the performance of the fixed stress splitting algorithm for the immiscible water-oil flow coupled with linear poromechanics. The two-phase flow equations are solved on general hexahedral elements using the multipoint flux mixed finite element method, whereas, the geomechanics equations are discretized using the conforming Galerkin method. The effects of the coupling parameter on the performance of the fixed stress algorithm is demonstrated for the Frio oil reservoir site.

Head model personalization: A framework for morphing lifespan brain images and brains with substantial anatomical changes

Finite element (FE) head models have emerged as a powerful tool in many fields within neuroscience, especially for studying the biomechanics of traumatic brain injury (TBI). Personalized head models are needed to account for geometric variations among subjects for more reliable predictions. However, the generation of subject-specific head models with conforming hexahedral elements suitable for studying the biomechanics of TBIs remains a significant challenge, which has been a bottleneck hindering personalized simulations. This study presents a framework capable of generating lifespan brain models and pathological brains with substantial anatomical changes, morphed from a previously developed baseline model. The framework combines hierarchical multiple feature and multimodality imaging registrations with mesh grouping, which is shown to be efficient with a heterogeneous dataset of seven brains, including a newborn, 1-year-old (1Y), 2Y, 6Y, adult, 92Y, and a hydrocephalus brain. The personalized models of the seven subjects show competitive registration accuracy, demonstrating the potential of the framework for generating personalized models for almost any brains with substantial anatomical changes. The family of head injury models generated in this study opens vast opportunities for studying age-dependent and groupwise brain injury mechanisms. The framework is equally applicable for personalizing head models in other fields, e.g., in tDCS, TMS, TUS, as an efficient approach for generating subject-specific head models than from scratch.