Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'How does the wind blow behind wind turbines and in wind farms?' taking place on Tuesday, 11 March 2014 at 16:30-18:00. The meet-the-authors will take place in the poster area.

Tommaso Morbiato treo group, Italy
Co-authors:
Tommaso Morbiato (1) F P Renato Vitaliani (2)
(1) treo group, PADOVA, Italy (2) University of Padova, Padova, Italy

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

Specialized in bridge and wind engineering, holds a PhD in Structural Dynamics, and won the financing for a road traffic to wind energy conversion research in the Progetti di Eccellenza award (Fondazione Cariparo 2010). He contributed about wind profiles generated by road traffic at IAWE (Shanghai, 2012), and pedestrian-structure interaction dynamics at IABSE (2010), national Steel Institution (2009) and ETH Zurich (2006). He also serves as PM in engineering firms in Padova (Vitaliani, 5 years), formerly working in numerours bridge engineering firms across Europe (Majowiecki, Bologna; Mimram, Paris; Arenas, Santander).

Abstract

Wind energy from road traffic: a resource estimation

Introduction

A number of patents addressing the idea of harvesting wind energy generated by traffic in motorways appeared by the end of the first decade of the millennium. None of the inventions though describe the mechanics of a brand new device dedicated to the task. The lack of a characterization of the energy resource could explain why the market does not yet acknowledge a related technology. We present an experimental activity started in 2010, whose final objective is to investigate the flow field generated by traffic in motorways eventually developing an innovative technology to cope with the first dedicated energy policies appearing worldwide [1].

Approach

Our scientific research aimed at the assessment of the selected resource, whose characteristics are quite different from atmospheric wind (to which existing turbines technology is dedicated). Though it is known that atmospheric and traffic contributions to the flow field near motorways can hardly be separated [2], it is necessary to assess to what extent an aero-generator is fed either by ambient or by traffic in order to correctly estimate the energy balance. To start with, we therefore limited all the possible applications to those presenting the simplest energetic system: we studied a pure air-vehicle interaction, i.e. the general case of ground vehicles impacting air in a straight motorway section in absence of any other obstacle modifying the flow field (i.e. tunnel exits, urban canyons).
From an economical point of view, the idea of trying and harvest wind energy generated by traffic can be seen as the search of an energy policy making development more sustainable in the transport sector. Munasinghe [3], questioning about an optimal level of renewable energy supply in general, state it as the quantity at which the costs of the marginal plant are exactly equal to the avoided costs connected with fossil generation, including environmental damage. In case of the traffic source, looking at Figure 1, the energetic rationale seems to have a double drive:
1. there will always be an optimal energy supply associated with an increment in transport demand. It is maximum nowadays as transport systems still operate with fossil fuel, and a base amount will always exist because of aerodynamic losses, in a more sustainable future when transport systems will eventually not rely on fossil generation;
2. contrary to other renewables, the transport aerodynamic losses belong to the source of costs itself, making it an extremely wise and sustainable energy source.
As power consumption of heavy duty vehicles can reach 10 times light traffic's, the energetic rationale of a harvesting concept starts from their loss repartition analysis, thus recognizing [4] that for a US Class 8 truck the corresponding aerodynamic power losses are as much as 24% of the total 382 kW in average loading conditions at 105 km/h.


Main body of abstract

To measure the wind generated from heavy traffic aerodynamic losses an experimental set-up of ultrasonic anemometers (USAs) and videocameras has been implemented in a straight motorway section. To place a conversion system the closest to energy source, we could ask what is the best place to position a device, i.e. how the losses are distributed in the surrounding space: [5] indicates on the top of the truck a significant net flow orientated in the trailing direction. Because of high turbulence in the bottom space between wheels and pavement, and because the sides are either one lane away (emergency lane) or simply moving (overtaking), the top of truck becomes the best candidate. The instruments were therefore placed at the minimum vertical headroom from road, and at higher positions, avoiding vertical alignment for adjacent instruments in order achieve not-interfering condition. Thus instruments for a typical single lane set-up, hosted by an existing VMS (variable-message sign) frame, are as in Figure 2: USAs are placed symmetrically respect to central axis of the lane, and a neutral USA is placed at 12.50 m height from road in order to get also a measure greatly influenced by atmospheric wind. To correctly estimate correlation with traffic the cameras add the synchronous traffic flow data via a stereo-tracking algorythm: an "in-the-scene" mark when one vehicle (heavy-duty coding) enters the scene, and an "out-of-scene" when exits from the scene. The data streaming of >3 months is subdivided for statistics analysis into 2 classes of 24 homogeneous sub-processes, corresponding to the repartition of a day in 1 hour slices for the weekdays class (Mon-Fri), and holydays class (Sat-Sun).
Some hints about the nature of the induced flows and the load mechanism upon an obstacle (or a device) situated in a region on top of truck arise from full scale test on road signs: both static pressure and wind speed measurements are accounted for, and it is suggested [6] that the load mechanism of a single truck is due mainly to the transient pressure field related to the potential induced flow, rather than any significant vehicle gust affecting average wind speed. In our research, the major trends of the flow related to open traffic (n vehicles) rather than a single vehicle are initially recognized from time signals observation: similarly to what is shown by our CFD U-RANS simulations [5], Figure 3 (adimensional) indicates on one side that the trailed speed prevails over the against truck motion at Y/H>1.35 (Y vertical distance from road, H=4m truck height), and on the other side that the effect of traffic arriving in clusters is not negligible when considering traffic generated wind, to the extent that a contribute to average wind can be studied in correlation with truck flow.
To proof this Figure 4a shows a particular wind speed distribution area plot where wind speeds |u^2+v^2+w^2| are the classes, and the Y-axis identifies the number of samples in which the class speed is seen. Data are aggregated in X-axis permitting direct correlation with truck flow: i.e. the samples counter that assigns wind classes is updated every hour, and associated with the corresponding amount of vehicles per hour. Figure 4b displays the same data only re-substituting the truck flows with the corresponding hour of the day.
Our resource estimation is completed by a cut-in assessment (Figure 5) for eventual wind energy conversion by a kW-range turbine: to find an estimator the variable t_cut-in,h is introduced, that is (for each h of the 24 hours) the total number of time samples in which wind speed is greater or equal to a 3m/s cut-in rate, then normalizing by 1hour. For fitting purposes to any motorway site, the contribution of the single truck is then extrapolated, also acknowledging the role of traffic clusters by a fitting range depending on traffic flow sampling time (Figure 6). To come to an actual energy conversion, a study on wind-drops in correlation with truck flow is accomplished, yet indicating how the transient behaviour of a turbine could be affected by the duration of traffic wind-drops (Figure 7). Finally, accounting for a dummy Cp=0.1 power coefficient, the t_cut-in,h resource yet reduced by transient effects can be converted [7] in valuable Wh for a typical motorway weekday (Figure 8).


Conclusion

It is finally argued if such new cleantech might employ physical devices that highway users would perceive as an obstacle, disturbing their free run. Looking for an energy balance proof of the aerogenerator not subtracting power from running vehicles, our [9] studies about ground-vehicle induced flows in presence of a panel obstacle show how in general drag coefficient deviations from the undisturbed value get stronger with increasing the panel width, but the energy loss ratio adimensionalized to the case of absence of any obstacle is always <1.5% for a panel big as much as 15% the truck length. Therefore any energy conversion device able to harvest more than 1.5% of the aerodynamic losses (1.4 kW threshold in terms of corresponding average power losses) would provide a positive balance of the system. Our research demonstrated with Figure 8 that during daytime weekdays hours the traffic generated resource continuously allows such energy conversion.
In our scenario machines are not necessarily operating in atmospheric wind (namely an horizontal wind) thus HAWT are machines in which the rotating axis is parallel to the wind vector (HAWTs become PAWTs), whereas VAWT are machines whose axis crosses the wind vector (VAWTs become CAWTs). While PAWTs need to rotate their axis against the wind, CAWTs are inherently omnidirectional. Though the findings of the present work could fit any machine, in our work reference is made to a CAWT unit. CAWTs are divided into drag and lift operated machines: drag driven devices, named after Savonius, generally work at TSR~1, while lift driven, named after Darrieus, work at TSR>1. Darrieus type machines are more efficient than Savonius type, but lift driven devices are known for their self-starting issues, to the extent that the major part of recent CAWTs related patents aim at improving their self-starting capabilities.
Due to circadian rythm of traffic flows and due to short wind-drops between traffic clusters, the new cleantech shall imply on one side improved CAWT self-starting capabilities, on the other side a simple speed control to run with best-point power coefficient in frequent variable speed operations.



Learning objectives
The presented scenario for the innovative cleantech has been the most general one: a straight motorway section operating above a truck flow threshold. Provided variable speed operation is accepted in the wind energy system, numerous scenarios arise for motorway application (i.e. tunnel exits, urban canyons), or for transport systems in general (railway, waterway or airside flows in airports). Different flow fields apply, but the presented resource estimation approach is general and can be reproduced.


References
[1] K.M. Saqr, M.N. Musa. A Perspective Of The Malaysian Highway Energy Consumption And Future Power Supply, Energy Policy, 39:3873-3877, 2011.
[2] N. Kalthoff, D. Baumer, U. Cosmeier, M. Kohler and B. Vogel. Vehicle Induced Turbulence near a Motorway, Atmospheric Environment, 39:5737-5749, 2005.
[3] M. Munasinghe. Sustainable Development in Practice Sustainomics Methodology and Applications. Cambridge University Press, 2009.
[4] A. Delorme, D. Karbowski and P. Sharer. Evaluation of Fuel Consumption Potential of Medium and Heavy Duty Vehicles through Modeling and Simulation. Report to National Academy of Sciences, Argonne National Laboratory, Washington, DC, October 2009.
[5] A. Mattana, S. Salvadori, T. Morbiato, C. Borri. On The Ground-Vehicle Induced Flows And Obstacle Interaction. Journal of Wind Engineering and Industrial Aerodynamics, Submitted for publication INDAER-D-13-00237, August 2013.
[6] Quinn A, Backer C, Wright N. Wind and vehicle induced forces on flat plates-part 2: vehicle induced force. Journal of Wind Engineering and Industrial Aerodynamics 2001;89:831–47
[7] M. Stiebler, Wind Energy Systems for Electric Power Generation, Springer Series in Green Energy and Technology, 2008
[8] E. Simiu, R.H. Scanlan. Winds Effects on Structures: Fundamentals and Applications to Design, Wiley, Third Edition, August 1996.
[9] A. V. Oppenheim, R. W. Schafer, J. R. Buck, Discrete-Time Signal Processing (2nd Edition) (Prentice-Hall Signal Processing Series), June 2011
[10] M. Sakamaki. Tunnel Installation Type Aerogenerator. Japan Patent JP56034979A, Filed Apr. 07, 1981.
[11] T F. Wiegel, K.C. Stevens. Traffic-Driven Wind Generator. United States Patent US007098553B2, Filed Jan. 12, 2005.
[12] G.S. Fein, E. Merritt, Genedics Clean Energy, Llc. System and method for creating a networked infrastructure distribution platform of small fixed and vehicle based wind energy gathering devices along roadways. United States Patent US 7741727 B2, Filed Dec. 22, 2006.
[13] E.R. Eskridge, S.T. Rao. Measurement and prediction of traffic-induced turbulence and velocity fields near roadways, Journal of Climate and Applied Meteorology, 22:1431-1443, 1983.