MACSyME: Modelling, Analysis and Control for Systems with Multiple Energy sources A. Naamane N. K. M'Sirdi _

_

LSIS, CNRS UMR 6168, Domaine Univ. St Jerome, Av. Escadrille Normandie-Niemen. 13397, Marseille Cedex 20, France ( e-mail : [email protected]; [email protected])

Abstract: MACSyME* project aims to design and develop new systems combining different Energy

sources The process to control is made of three parts (production sources, the energy consumption and storage parts). The production sources (depend on wind or sunlight intensity) have stochastic behavior and are not fully controllable. Energy consumption has to be estimated and predicted. The instantaneous equilibrium of production, consumption and storage has to be maintained in an optimal level which depends on the system state, production state of charge and the demands. We have to design coupled prediction, estimation and control systems to optimize energy costs and satisfy demand. This paper describes the first work done for this project with. a specific urban wind turbine.

*MACSyME

: Modeling, Analysis and Control for Systems with Multiple Energy sources. Definition of this project has been started in the LSIS in 2007 and involve several collaborating research laboratories.

Introduction: According to IEA World Energy Oulook 2008’’The world is facing an energy and climate crisis. Globally, the energy sector emits 26 billion tonnes of CO2 each year and electricity production alone accounts for 41% of emissions. The International Energy Agency expects CO2 emissions in 2030 to have increased by 55% to reach more than 40 billion tonnes of CO2. The share of emissions coming from electricity production will increase to 44% in 2030, reaching 18 billion tonnes of CO2 ‘’The EWEA is fairly confi dent about the projections of wind power in the EU to 2030.

Figure 1:EWEA’s three wind power scenarios (in GW)

In these scenarios Urban wind turbines is an industry and technology that is still at development stage. It supposes that one can install and exploit wind in urban environment. Turbines have to be designed as part of the building itself (architecturally integrated). These turbines need compact wind devices able to supply a decentralized production, free from transport and generated losses,and is likely to make a tangible contribution to energy savings. The offshore environment may allow the relaxation of a number of constraints on turbine design, such as aesthetics and noise level. However, addressing urban conditions, noise and reliability issues create new challenges in the urban sector. This will lead to a significant modification for the development of specific urban designs in the medium and long term. So the aim of the research prioritized by our Laboratory is to develop technology that enables to deliver highly cost efficient wind turbines for individual wind turbine, by: -applying recent and advanced control methods to optimize the dynamic behavior of variable-speed wind energy conversion systems (WECS). -equipping WECS with sensors and control systems implementing supervision and data acquisition concept. The challenging open problem in WECS control is to ensure good quality electrical delivered energy while using a profoundly irregular primary source (the wind) and in most time very irregular demand.

-trying to develop new systems for control and management of power energies in the system with some following features : -Modelling subsystems to be able to optimize costs and consumption. -Analysis tools for design and supervision of multidisciplinary systems (system of systems). -Control and supervision taking into account subsystems features using prediction, forecasting and control. -Systems design using mechatronic approaches: A wind power system is a sophisticated combination of components and sub-systems that have to be designed in an interdisciplinary and integrated manner. -Management of the switching control between the different power sources in order to optimize energy consumption. - Energy sources and Energy storage have to be controlled via observers, estimators and diagnosis tools for the optimization of the whole studied system depending on weather and environment.

II- The MacSyME project The process to be controlled is composed by three parts as depicted by the scheme of figure2. It has the following features: - The energy production sources are not controllable. They have stochastic behavior as they depend on wind or sunlight intensity. - The energy consumption has also to be estimated and predicted. - The instantaneous equilibrium of production consumption and storage optimal depend on state of the previous system parts. .

Our approach consists in a first stage to develop appropriate models (for control and management) for each part of the system, and to use these models for forecasting, estimating the system state and prediction of demand targets. The next step will consist to manage the production sources, use of power and storage (for long and short time) to optimize the costs and to lead to stable behaviors. Once the power target is defined (depending on energy demand) contribution of each source (solar, wind and storage) has to be defined in correlation with forecasting (stochastic events, perturbations and weather fluctuations) and prediction of demand during a selected period (to avoid redundant switching). Several variables have to be measured: the system state variables and all parameters describing interactions between the system parts describing energy exchanges. Thus sensors can be used for most of them but measurements have to be observed or estimated. The last part to develop is the control of sources and power managements. Optimization has to consider different cost functions, the system stability and safety in the same time as the protection of the subsystems from failures and faults which may be due to bad operating points or excessively demanding operating modes components.

II.1 the studied wind turbine II.1.1 Wind turbine characterization The wind Turbine prototype used in this study has a particular shape as it is shown in figure 2.

Figur 3 : wind turbine prototype The first work carried out for this project dealt with the geometrical characterisation of the wind turbine and its behaviour under different wind speeds . We have used for this purpose CATIA (Computer Aided Three Dimensional Interactive Application) software to assist us in prototyping the wind turbine. We have also used FLUENT for CATIA which gives fluid flow analysis and provides a full generative relationship between our manufacturing-ready geometry models and the flow analysis model. In the following figures are presented the simulations results from Catia Software. These simulations allowed us to .optimize the geometrical parameters of the turbine.

Figure 3: Catia models In this part, the Catia software and Fluent were used in order to determine the difference in pressure inside and outside the blades. Different simulations with different configurations

were accomplished. As a main conclusion ,we have demonstrated that the axis of the two blades must have an angle of 15°with the central axis.

α=15°

Wind

In the following figures are represented the results obtained with fluent software. For this model, the geometry and the angle α are taken into account , we can obtain the velocity contours profiles and air flow pathlines. These simulations were carried out to appreciate the forces exerted by the air on the blades.

Figure 4 : velocity field representation

Figure 5Contours of stream function (Kg/s)

It is to be noticed that the half-cone shape, makes it possible to have an acceleration in end of the blade (like the Venturi effect). These simulations indicated us that the value of the force is tripled when the value of α passes from 0 to 15°

II.1.2 System modelling and simulation Ta partie à insérer en respectant les numerous de paragraphe et numero des figures

Figure XX

In the following figures,it is represented The block diagram of the whole system the wind turbine part and the electrotechnical part.

Figure XX Block diagram of the studied system The first experimental results achived cincern the curves of mechanical power versus Rotation velocity of the wind turbine and the torque versus Rotaional velocity. for different wind speeds.

Figure XX : simulation results :Power and Torques Vs velocity rotation

The second part, achieved was the choice and the principle of functioning of the Generator to use for a maximum efficiency. .ADoubly Fed Induction Generator (DFIG) is chosen among several asynchronuous generators It is composed of a wound rotor induction generator connected to the electric grid within rotor and stator. The stator is directly connected to the grid and the rotor is connected to the grid by the unified power flow controller (UPFC). Such system has the capacity to track the maximum of power for different wind speeds and deliver the electric power by miniminzing the Distorsin Harmonics Rate(DHR). Four phases of operation of the wind turbine at variable speed are considered: - The launching phase of the machine. The electric production starts when the mechanical speed reached approximately 70% the speed of synchronism of the generator. The electric output remains rather low - The phase of extraction of the maximum power or phase MPPT (Maximum Power Point tracking). In this zone, mechanical speed varies and can reach a value close to nominal speed. The electric output increases quickly. In this zone, The maximum power is thus obtained for each value of mechanical speed and average speeds of wind

- The phase at quasi constant mechanical speed.the electric output increases very quickly. - The phase at constant power. It is to be noticed that transfers of power in the MADA is achieved thanks to the bidirectional converters of power.In the circuit of the rotor, the MADA is able to work as a generator or engine in hypersynchronuous way or in hyposynchronuous way.

Figure 4 DFIG configuration The electrical part described above is currently simulated with PSIM software and will be tested in order to validate the approach..

Conclusions In this paper, we have presented the first results obtained in ongoing project dealing with the optimization of Energy production . An urban wind turbine prototype was presented and characterized. The first experimental results in a laboratory environment are very promising. Experimentations under real conditions that are envisaged on the roof of the Laboratory to confirm these first results. Another wind turbine of four larger times dimensions will be used.. This site would enable us to highlight the power ratio which exists between various wind turbines of different sizes. Theoretically, while multiplying by four the scale of the basic model, the power supplied by the wind turbine is 64 (4 ³) times more important compared to the same wind speed of wind. We are in the presence of a concept exploiting volume of the blades covered by the wind instead of considering a surface. REFERENCES [1] Simon Watson ‘’predicting the performance of small wind turbines in the roof-top urban environment ‘’ in the proceeding s of the Europe’s premier wind event 7-40 Mai 2007 –Milan –Italie [2] Joshua Paquette ‘’Increased strength in wind turbine blades through innovative structural design’’ in the proceeding s of the Europe’s premier wind event 7-40 Mai 2007 –Milan –Italie [3] Hannele Holttinen ‘’State of the art of design and operation of power systems with large amount of wind power’’ in the proceeding s of the Europe’s premier wind event 7-40 Mai 2007 –Milan –Italie [4] Panel 1a: ‘’Les expériences d’utilisation des sources d’énergies renouvelables dans les villes. Est-il possible de développer le potentiel des énergies renouvelables dans un environnement urbain?’’ BERLIN 2004 : La conférence sur les énergies renouvelables. 19 au 24 Janvier 2004 . [5] Van der Hoven, I., Power spectrum of horizontal wind speed in the frequency range from .0007 to900 cycles per hour, Journal of Meteorology, 14,160-164,1957 [6] B Neammanee, S Sirisumrannukul and S Chatratana. Development of a Wind Turbine Simulator for Wind Generator Testing. International Energy Journal 8 (2007) 21-28.

[7] I. Munteanu, N. A. Cutululis, A. I. Bratcu, E. Ceanga. 2004. Optimization of variable speed wind power sys tems based on a LQG approach. ELSEVIER Tran. On Control Engineering Practice. [8] L. H. Hansen, L. Helle, F. Blaabjerg, E. Ritchie, S. Munk-Nielsen, H. Bindner et al., \Conceptual survey of generators and power electronics for wind turbines," RIS_ National Laboratory, Roskilde, Denmark, Tech. Rep. RIS_-R-1205(EN), 2001. [9] S. A. De La Salle, D. Reardon, W. E. Leithead, and M. J. Grimble, \Review of wind turbine control," International Journal of Control, vol. 52, no. 6, pp. 1295{1310, 1990. [10] P. W. Carlin, S. Laxson, E. B. Muljadi, \The history and state of the art of variable-speed wind turbine technology," National Renewable Energy Laboratory, U.S.A., Tech. Rep. NREL/TP-500-28607, 2001. [11] Boukhezzar B, Lupu L, Siguerdidjane H, Hand M. Multivariable Control Strategy for Variable Speed, Variable Pitch Wind Turbines. Renewable Energy In- ternational Journal, Elsevier, 2006