# Research

Here are some examples of our cutting-edge research projects at the Aerodynamics group. For more information about our work see our publications or contact the group member invovled.

## Noise propagation from sustainable vehicle concepts

*Nicolas Pignier, Ciarán O'Reilly and Susann Boij*

Traffic noise is one of the main sources of pollution, having harmful effects on the health and impacting the economy. In order to design future sustainable transport systems, developing quieter ground vehicles is a necessity and manufacturers are pushed to develop new technologies through increasingly stricter regulations. In this project, a numerical method is developed to predict the aerodynamic sound generated by moving vehicles and its radiation in urban environment, so as to estimate its pass-by characteristics. Steps in the project include:

- computation of the airflow around a vehicle geometry using detached eddy simulations
- identification and modelling of the sound sources generated by the airflow using phased array techniques
- propagation of moving sources including reflections on large idealized buildings.

This project is part of the ECO^{2} Centre for Vehicle Design, a VinnExcellence Centre performing multi-vehicular multidisciplinary research to support a sustainable vehicle design development.

## Identification and modelling of flow sources from CFD simulations using numerical phased array methods

*Hamza Bouchouireb, Nicolas Pignier, Ciarán O'Reilly and Susann Boij*

Airframe noise in general, and landing gear noise in particular, has been identified as one of the main noise sources during the take of and landing phases of an aircraft. In an effort to reduce sound emissions, the identification and the characterization of the sound sources is of paramount importance. By extracting pressure data recorded with virtual microphone arrays inside the simulation domain, and outside of the unsteady flow region it is possible to identify and locate the corresponding sound sources. The extracted data is post-processed using standard Beamforming and a Linear Programming deconvolution algorithm. Given the novelty of the method, this project aims at benchmarking this approach on a realistic regional aircraft landing gear in order to pave the way towards numerical aircraft noise certification and eventually include aeroacoustic constraints in a multidisciplinary design optimization framework.

## Detached Eddy Simulations of aerodynamic sound generation by a bi-directional cooling fan

*Emma Alenius*

The focus of this project is state-of-the-art Computational Aero Acoustics (CAA) for the prediction of aerodynamic noise from low Mach number bi-directional fans. The application is self-ventilated railway traction motors, where the fan is mounted directly on the motor shaft, giving a robust, reliable and simple cooling principle. However, since bi-directional fans typically have straight unsymmetrical blades, they generate a significant broadband noise, and at high motor speeds the fan can become the dominant noise source on the train. The purpose of the project is to explore the possibility to numerically predict the generated near- and far-field sound. For this purpose, a mock-up of a modern railway traction motor with a radial cooling fan is used as a test case. The CAA method chosen is Direct Noise Calculations (DNC), where the flow and sound are retrieved simultaneously in one simulation. This is achieve with compressible Improved Delayed Detached Eddy Simulations (IDDES), carried out with the multipurpose commercial CFD code STAR-CCM+.

## Linearized Navier-Stokes for aero-acoustic and thermoacoustic simulations

*Wei Na, Gunilla Efraimsson and Susann Boij*

One part of this project focuses on developing a numerical methodology to simulate the acoustic properties of a hybrid liner consisting of a perforated plate, a porous layer and a Helmholtz cavity. Liners are always a standard way to reduce noise in today’s aero-engines, e.g. the fan noise can be reduced effectively through the installation of acoustic liners as wall treatments in the ducts. In order to optimize a liner in the design phase, an accurate and effcient prediction tool is of interests. Hence, a unified Linearized Navier-Stokes equations (LNSE) approach has been implemented in the thesis, combining the LNSE in frequency domain with the fluid equivalent model. The LNSE is applied in the vicinity of the perforated plate to simulate sound propagation including viscous damping effect, and the fluid equivalent model is used to model the sound propagation in the porous material including absorption. The second part of the work focuses on the prediction of thermoacoustic instabilities. Thermoacoustic instabilities arise when positive coupling occurs between the flame and the acoustics in the feedback loop, i.e. the flame acts as an amplifier of the disturbances (acoustic or fluid) at a natural frequency of the combustion system. Once the thermoacoustic instabilities occur, it will lead to extremely high noise levels within a relatively narrow frequency range, resulting in a huge damage to the structure of the combustors. Hence, a solution must be found, which breaks the link between the combustion process and the structural acoustics. The numerical prediction of thermoacoustic instabilities in the thesis is performed by two different numerical methodologies. One solves the Helmholtz equation in combination of the flame n-tau model with the low Mach number assumptions, and the other solves the Linearized Navier-Stokes equations in frequency domain with mean flow. The result show that the mean flow has a significant effect on the thermoacoustic instabilities, which is non-negligible when the Mach number reaches to 0.15.

## Aerodynamic drag reduction of trucks

*Romain Futrzynski, Gunilla Efraimsson, Julie Vernet, P.H. Alfredsson and R. Örlü *

Aerodynamic drag is of main concern when it comes to reduce energy consumption of ground vehicles. In the project we study how flow separation around an A-pillar (shown in the red ellipse above) on heavy trucks may be reduced using plasma actuators as a remedy to decrease drag. Separation control can be especially important in case of a strong side wind. A body force model that mimics the electric wind produced by plasma actuators has been calibrated against experimental data to enable Large Eddy Simulations, (LES) of the flow. A half-cylinder test case is the focus of the study as it presents a smooth curvature comparable to an A-pillar, and ensures that flow separation occurs so that the effect of actuation can be quantified.

## Higher-order mode propagation in ducts

*Stefan Sack, Mats Åbom and Gunilla Efraimsson*

The generation and scattering behavior of fluid machines in connected duct or pipe systems is of great interest to minimize disturbing and harmful sound emission, for instance of air condition systems. In general an in duct element, e.g., fan, diaphragm and bend, can both generate sound, referred to as the active properties, and scatter sound which is referred to as the passive properties. One approach to describe the sound field created by a fluid machine or in a duct component is to apply a linear multi-port model that includes direction-depending transmission and reflection coefficients for the wave-modes present and the sound generation. The model parameters can be ascertained either experimentally or numerically. Once the multi-port data are terminated, the sound-field within the duct can be calculated for arbitrary acoustic loads. Within the framework of the European project "IdealVent", advanced methods to extract multi-port data from measurements and computations are developed. Previous work is further investigated, mainly to develop advanced procedures to find reliable setups and post-processing methods to obtain tools for gaining accurate data especially for higher order acoustic in-duct modes.

## Slipstream and flow Structures in the near wake of high-speed trains

*Tomas Muld, Gunilla Efraimsson and Dan Henningson*

Train transportation is a vital part of the transportation system of today. As the speed of the trains increase, the aerodynamic effects become more important. One aerodynamic effect that is of vital importance for passengers’ and track workers’ safety is slipstream, i.e. the induced velocities by the train. Safety requirements for slipstream are regulated in the Technical Specifications for Interoperability (TSI). Earlier experimental studies have found that for high-speed passenger trains the largest slipstream velocities occur in the wake. Therefore, in order to study slipstream of high-speed trains, the work in this project was devoted to wake flows. First a test case, a surface-mounted cube, was simulated to test the analysis methodology that was later applied to two different train geometries, the Aerodynamic Train Model (ATM) and the CRH1. The flow was simulated with Delayed-Detached Eddy Simulation (DDES) and the computed flow field was decomposed into modes with Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD). The computed modes on the surface-mounted cube compare well with prior studies, which validates the use of DDES together with POD/DMD. To ensure that enough snapshots were used to compute the POD and DMD modes, a method to investigate the convergence was proposed for each decomposition method. It was found that there is a separation bubble behind the CRH1 and two counter-rotating vortices behind the ATM. Even though the two geometries have different flow topologies, the dominant flow structure in the wake in terms of energy was the same, namely vortex shedding. Vortex shedding was also found to be the most important flow structure for slipstream, at the TSI position. In addition, three configurations of the ATM with different number of cars were simulated, in order to investigate the effect of the size of the boundary layer on the flow structures. The most dominant structure was the same for all configurations, however, the Strouhal number decreased as the momentum thickness increases. The velocity in ground fixed probes were extracted from the flow, in order to investigate the slipstream velocity defined by the TSI. A large scatter in peak position and value for the different probes were found. Investigating the mean velocity at different distances from the train side wall, indicates that wider versions of the same train will create larger slipstream velocities.

## Aerodynamic simulations of ground vehicles in unsteady crosswind

*Tristan Favre and Gunilla Efraimsson*

Ground vehicles, both on roads or on rail, are sensitive to crosswinds and the handling, travelling speeds or in some cases, safety can be affected. Full modelling of the crosswind stability of a vehicle is a demanding task as the nature of the disturbance, the wind gust, is complex and the aerodynamics, vehicle dynamics and driver reactions interact with each other. One of the objectives of this project, was to assess the aerodynamic response of simplified ground vehicles under sudden strong crosswind disturbances by using an advanced turbulence model. In the aerodynamic simulations, time-dependant boundary data was used to introduce a deterministic wind gust model into the computational domain. This project covered the implementation of such gust models into Detached- Eddy Simulations (DES) and assessed the overall accuracy. Different type of grids, numerical setups and refinements were considered. Although the overall use of DES is seen suitable, further investigations can be foreseen on more challenging geometries. Two families of vehicle models were studied. The first one, a box-like geometry, was used to characterize the influence of the radius of curvature and benefited from unsteady experimental data for comparison. The second one, the Windsor model, was used to understand the impact of the different rear designs. Noticeably, the different geometries tested have exhibited strong transients in the loads that can not be represented in pure steady crosswind conditions. The static coupling between aerodynamics and vehicle dynamics simulations enhanced the comparisons of the aerodynamic designs. Also, it showed that the motion of the centre of pressure with respect the locations of the centre of gravity and the neutral steer point, is of prime interest to design vehicles that are less crosswind sensitive.

## Linearized Navier Stokes for low Mach number internal aeroacoustics

*Axel Kierkegaard and Gunilla Efraimsson*

Traffic is a major source of environmental noise in modern day’s society. As a result, the development of new vehicles are subject to heavy governmental legislations. The major noise sources on common road vehicles are engine noise, transmission noise, tire noise and, at high speeds, wind noise. One way to reduce intake and exhaust noise is to attach mufflers to the exhaust pipes. However, to develop prototypes for the evaluation of muffler performance is a costly and time-consuming process. As a consequence, in recent years so-called virtual prototyping has emerged as an alternative. Current industrial simulation methodologies are often rather crude, normally only including one-dimensional mean flows and one-dimensional acoustic fields. Also, flow generated noise is rudimentary modeled or not included at all. Hence, improved methods are needed to fully benefit from the possibilities of virtual prototyping. This project aimed at the development of simulation methodologies suitable both as industrial tools for the prediction of the acoustic performance of flow duct systems, as well as for analyzing the governing mechanisms of duct aeroacoustics. Special focus was at investigating the possibilities to use frequency-domain linearized Navier-Stokes equations solvers, where the equations are solved either directly or as eigenvalue formulations. A frequency-domain linearized Navier-Stokes equations methodology was developed to simulate sound propagation and acoustic scattering in flow duct systems. The performance of the method was validated to experimental data and analytical solutions for several cases of in-duct area expansions and orifice plates at different flow speeds. Good agreement was generally found, suggesting that the proposed methodology is suitable for analyzing internal aeroacoustics.