Active Reduction Of Aerodynamic Drag <br />Wake control: from passenger car to long haul vehicle
The objectives are the following : <br />- Carry out a physical study on aerodynamic properties of the wake for square back bluff-bodies under the influence of high-frequency perturbations due to pulsed jets in the vicinity of solid surfaces. <br />- Take advantage of advanced control strategies used in the project (extremum seeking, unsupervised learning, model based learning, machine control learning) and focus on optimized configurations (in the space of parameters) the analyses of experimental results involving high-quality measurements (high frequency rate PIV, unsteady pressure, hot films or wires, small field of views located in the region where pulsed jets interact with the main flow). Extend the validity of these results to configurations representative of industrial applications (personnal vehicles and long haul transportation) and define scaling laws for the relevant physical parameters.
Activ_ROAD project is structured in 4 tasks devoted to specific targets:
T1 – Drag reduction by high frequency pulsing. Study of turbulent wakes on 3D mock-up
A detailed study of the effects of pulsed jets on an academic model are carried out in S620 wind tunnel at Pprime. Control strategies, well identified by the partners, are applied to determine the optimal parameters and so the configurations of interest for finer analyses (PIV, unsteady pressure).
T2 Pulsed jets caracterisation and command law adaptation
An existing valve whose maximum actuation frequency is 900 Hz is used. Characteristics of the resulting pulsed jet are determined experimentally. The adjustement of the valve command laws, done by Ampere, will allow the management of the temporal evolution of vorticity flux in the flow to be controlled.
T3 Robustness and adaptative control
Innovative control strategies are to be implemented as complement of the work in T1, and on the bases of the results obtained sofar. Closed-loop strategies are targeted, so as to promote control robustness with respect to the incoming flow velocity as well as slow temporal variations of it, or modifications of the boundary layer thickness on the mock-up sides.
T4 Toward industrial applications
The control strategies previously used are adapted to the peculiarities of the targetted application (personnal car or heavy truck). Scaling laws on control parameters that allow a drag reduction will be establish. High Reynolds number studies in an industrial wind-tunnel, will concern a personnal car configuration ; a modification of the generic mock-up underhood will allow to study long-haul configurations.
The first 18 months of the project already allowed for the following results :
- Design and manufacturing of the generic mock-up (T1), which corresponds to significant technical challenges (aspect ratio modification, design of control jets, integration of control jets around the mock-up base)
- Mock-up integration in S620 wind-tunnel at Pprime, coupled with a piezometric balance (good measurement precision)
- Detailed flow characterization at the nozzle exit for control jets ; determination of pneumatic effects poorly described up to now. Use of different experimental techniques for temporal (hot-wire anemometry) and spatial (PIV 2D-2C with phase average) description.
- RANS Numerical simulations for the design of the model front part, in order to avoid flow detachment in the upstream part of model and have the benefit of well-established boundary layer on the model sides.
- Design and manufacturing of dedicated electronics for valve command, giving hand on specific parameters such as current value and current increase during the opening phase. Generalisation of this process to 32 command channels distributed over 8 boxes that allows to command the valves independantly 4 by 4.
- First test campaign with the largest side of the model parallel to the floor, after having integrated in the model the equipment concerning pneumatic, electronics, measurement tools. The first results are promising even for high values of upstream velocity targeted for personnal car applications.
The project aims at bringing a better comprehension of aerodynamic phenomena linked with flow control in the rear part of a bluff body. The link between high-frequency injection of flow perturbations near a solid surface mounted on a square-back body, shear-layer development downstream the model and mean pressure in the recirculating area is not well established and is a prerequisite for the idenfication of pertinent control strategies. Moreover, this link certainly depends on the aspect ratio of the body and on the underhood velocity, in a manner that will be precised to allow for a good description of personnal vehicles and heavy-trucks applications.
Technical prospects correspond to, in the end, the integration of active control devices in vehicles, be it personnal cars or heavy trucks, from the earliest stages of the conception of these vehicles. However, these technological evolutions will be possible only after a period of development specific to the application and concerned industry, on the basis of results obtained in the project oriented toward physical understanding and identification of the robustness of the control techniques.
Finally, on the economic side, a very significant drag gain of a few pourcent is expected on the models representative of industrial configurations. One should notice that, owing to the peculiarities of heavy-trucks exploitation for instance, a fuel consumption of 1% is completely in-line with industrial objectives. For personnal cars, or light commercial vehicles, a drag reduction between 5 and 10% and a better adaptability to cruise conditions may be targetted for the square-back configurations.
The recent test campaign and the corresponding data analyses should allow several publications in scientific journals in the next few months.
The activ_road project concerns the improvement of aerodynamic performances of road vehicles. Reduction of aerodynamic drag associated with these bluff bodies is part of the answer for major stakes in the field of energy management in link with environmental and societal challenges. In a large number of cases, the shape of the vehicle rear, for a large range of private cars as well as heavy trucks, corresponds to a square-back end and leads to the development of a turbulent wake flow linked with a recirculating low-pressure bubble. It ends up with an energy loss which increases with vehicle speed. The project aims at exploiting a specific flow control strategy to increase base pressure and thus reduce drag. The control method combines here a passive component (flaps or curved surfaces at model rear) and an active component consisting in injecting high-frequency perturbations by use of pulsed jets next to the model rear. According to preliminary experiments, this combination leads to significant modifications in the wake, namely a deflection of mean-flow streamlines favorable to a rear pressure increase, and a progressive change, with actuation frequency, in Reynolds tensor terms and kinetic energy dissipation into heat in the shear layers. The influence of these aerodynamic caracteristics on rear pressure is not clearly established, and their respective weight in the rear pressure modification is suspected to depend on the model properties (aspect ratio, underhood flow). Thus it is of utmost importance to determine the flow control effects following a well-defined protocol.
The project is based on several experimental test campaigns to evaluate, capture, model and control in a robust way the gain in aerodynamic drag. First, a generic model allowing wind tunnel tests on different aspect ratio values and underhood flow will be considered; new-generation actuators capable of generating high frequency pulsed jets and monitoring the circulation linked with the injected eddies will be integrated next to the model rear. These tests will allow the use of control strategies of growing complexity through the project, from the open-loop to robust closed-loop, to promote the understanding of the relevant physical phenomena and the development of new control strategies by use of automatic learning methods. Numerical simulations will help in designing the model and the detailed analysis of the reference flow without control. The experimental results obtained during the project are mandatory to extract characteristic parameters allowing to derive pertinent scaling laws focused on the conditions associated with drag reduction. Moreover, given the targeted industrial applications, it is mandatory to determine control robustness with respect to flow parameters such as incoming flow velocity and its slow time-variation, boundary-layer thickness at model rear and non-null yaw angle between model and main flow direction. These data make possible to estimate drag reduction provided by control, based on pneumatic technologies or other actuators technologies, on given normalized driving cycles.
Monsieur thomas castelain (Laboratoire de Mécanique des Fluides et d'Acoustique)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Pprime Institut Pprime
Insa de Lyon - Ampère Institut National des Sciences Appliquées de Lyon - Laboratoire Génie Electrique, Electromagnétisme, Microbilogie environnementale et Applications
RT RENAULT TRUCKS
PEUGEOT CITROEN AUTOMOBILES SA
LMFA Laboratoire de Mécanique des Fluides et d'Acoustique
Help of the ANR 802,217 euros
Beginning and duration of the scientific project: September 2015 - 48 Months