A flywheel in a wind turbine rotor for inertia control

Transcription

1 WIND ENERGY Wind Energ. 2015; 18: Published online 18 July 2014 in Wiley Online Library (wileyonlinelibrary.com) RESEARCH ARTICLE A flywheel in a wind turbine rotor for inertia control Clemens Jauch Flensburg University of Applied Sciences, Kanzleistraße 91-93, Flensburg, Germany ABSTRACT In this paper, a flywheel energy storage that is an integral part of a wind turbine rotor is proposed. The rotor blades of a wind turbine are equipped with internal weights, which increase the inertia of the rotor. The inertia of this flywheel can be controlled by varying the position of the weights, i.e. by positioning them closer to the center of rotation (closer to the hub) or closer to the tip of the blades. The simulation model used in this study is introduced briefly. The equation system of the flywheel is set up. Finally, simulations of different scenarios show the performance of this controllable flywheel. The conclusion is that the proposed system can mitigate transients in the power output of wind turbines. Hence, it can support the frequency control in a power system by contributing to the power system inertia The Authors. Wind Energy published by John Wiley & Sons, Ltd. KEYWORDS flywheel; inertia; power control; power system frequency; wind turbine Correspondence C. Jauch, Flensburg University of Applied Sciences, Kanzleistraße 91-93, Flensburg, Germany. clemens.jauch@fh-flensburg.de This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Received 28 November 2013; Revised 24 March 2014; Accepted 18 June 2014 ABBREVIATIONS AND SYMBOLS AC alternating current c fw_damp friction (damping) coefficient of the weight positioning system (Ns m 1 ) C p aerodynamic power coefficient ( ) F force (N) F fw_act actuator force that drives the flywheel weight to the desired radial position (N) F fw_cf centrifugal force that acts on the flywheel weight (N) F fw_cntrl sum of all forces acting on the flywheel weight (N) F fw_damp friction force that acts on the flywheel weight positioning system (N) F fw_gr component of the gravitational force that interferes with the flywheel weight positioning (N) F fw_sp spring force that acts on the flywheel weight (N) g gravitational acceleration (m s 2 ) Hz hertz: unit of frequency J mass moment of inertia (kgm 2 ) J fw inertia of flywheel (kgm 2 ) J hss inertia of wind turbine drive train on high speed side (kgm 2 ) J lss inertia of wind turbine drive train on low speed side (kgm 2 ) k fw_sp spring coefficient of the spiral spring in the flywheel positioning system (N m 1 ) λ tip speed ratio ( ) L angular momentum (kgm 2 s 1 ) LUT lookup table m mass (kg) 2014 The Authors. Wind Energy published by John Wiley & Sons, Ltd. 1645

2 Flywheel in wind turbine rotor C. Jauch m bl mass of rotor blade (kg) m fw mass of flywheel weight (kg) ω rotational speed (angular speed or angular frequency) (rad s 1 ) ω gen rotational speed of generator (rad s 1 ) ω ref reference rotational speed (rad s 1 or pu) ω rot rotational speed of wind turbine rotor (rad s 1 or pu) P gen electric power of the generator (W) P grid electric power fed into the grid (W or pu) P ref electric reference power [W] pu per unit r radius of a circular trajectory (m) R rotor radius (m) R fw_max maximum radius of flywheel trajectory (m) R fw_min minimum radius of flywheel trajectory (m) R var actual radius of flywheel trajectory (m) RPM revolution per minute (1 min 1 ) s distance (m) s seconds T torque (Nm) T fw flywheel torque (Nm or pu) T gen generator torque (Nm) T lss the aggregated torque that acts on the low speed shaft of the drive train (Nm) T rot wind turbine rotor torque (Nm) θ pitch angle ( ) θ dem demand pitch angle ( ) θ max maximum pitch angle ( ) θ min minimal pitch angle ( ) v fw positioning speed of the flywheel weight (m s 1 ) ambient wind speed (m s 1 ) W fw_sp energy stored in the deformed spring of the flywheel weight positioning system (Ws) W rot rotational energy, i.e. kinetic energy stored in a rotating mass (Ws) W tr translational energy [Ws] ϕ angular position of the wind turbine rotor [rad] 1. INTRODUCTION The power output of wind turbines is determined by the wind. Therefore, it is bound to fluctuate similarly to the fluctuations in the wind speed. In an AC power system, the generated power has to match the consumed power at any point in time. Differences between generation and consumption lead to a variation of the power system frequency, which should be 50 or 60 Hz, depending on the country considered. The speed with which the frequency deviates if there is a difference between generation and consumption is determined by the power system inertia. Simply put, the power system inertia is the inertia of all rotating machines that are directly connected to the power system, e.g. synchronous generators of conventional power plants. Frequency control in a power system is historically carried out by applying controllable conventional power plants. Increasing wind power penetration leads to the fact that frequency control becomes more demanding, as the power generated by wind turbines is fluctuating. In addition, the power system inertia decreases if state of the art wind turbines substitute conventional power plants. While conventional power plants are almost invariably equipped with AC connected synchronous generators, wind turbine generators are usually connected via a frequency converter. To solve these problems, flywheel storages that are used as separate units in a power system have been used in the past to outbalance variable power output of wind turbines. 1 Since the prime energy of wind turbines, i.e. wind, cannot be stored, wind turbines cannot become frequency control power plants. However, wind turbines can provide positive control power if their power output is permanently reduced. In case of dropping frequency, this retained power can then be released and fed into the grid to support the grid frequency. 2 Fixed speed wind turbines, which usually apply squirrel cage induction generators, inherently contribute to power system inertia Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd.

3 C. Jauch Flywheel in wind turbine rotor These days, however, large wind turbines are almost inevitably variable speed wind turbines. Speed variability in wind turbines is almost invariably achieved with power electronic frequency converters, which are connected between the generator of the wind turbine and the power system. Therefore, the generator of a variable speed wind turbine does not inherently contribute to power system inertia. Variable speed wind turbines can, however, contribute to power system inertia when the control strategy of the wind turbine is implemented accordingly. Different strategies have been suggested in the past, 4 and they all have in common that they take advantage of the fact that variable speed wind turbines can vary their rotor speeds in a wider range. As can be seen from Equation (1), the energy, W rot, stored in a rotating mass, e.g. a wind turbine drive train, varies with the square of the rotational speed. W rot ¼ 1 2 Jω2 (1) Hence, variable speed wind turbines can make considerable portions of the kinetic energy, which is stored in the rotating drive train, available to increase power system inertia. The downside of this method is, though, that a variation in the rotor speed invariably has an impact on the aerodynamic efficiency of the rotor. Therefore, different methods have been discussed on how this problem could be dealt with the works by Erlich and Wilch 5 and Ma and Chowdhury. 6 In this paper, the advantages of flywheel storage and variable speed wind turbines shall be combined. The flywheel, however, shall not be a separate piece of equipment, but it shall be an integral part of the wind turbine rotor. The primary purpose is to have controllability of the kinetic energy stored in the rotating drive train. This is carried out by controlling the inertia of the wind turbine rotor. The idea is not to store the largest possible amount of energy. Instead, the effect of transient wind speed variations and/or transient grid frequency variations shall be mitigated by utilizing the angular momentum of the flywheel. In the following sections, first, the wind turbine system is described briefly. Following that, the idea of the flywheel energy storage in a wind turbine rotor is introduced in detail. Subsequently, simulations demonstrate the behavior and the capabilities of the system. 2. WIND TURBINE SIMULATION MODEL The wind turbine type considered here is a variable speed, pitch to feather wind turbine. It has a conventional drive train with a gearbox and a fast spinning generator. Models of such wind turbines are described in great detail in literature Therefore, only a block diagram showing the different subsystems of the wind turbine model applied in this study is depicted in Figure 1. For a description of the symbols and abbreviations used, see Abbreviations and Symbols. LUT P ref = f( gen) P ref converter P grid rot max ref speed controller dem pitch drive aerodynamics with LUT Cp =f(, ) T rot drive train gen generator P gen = rot R/ J fw T gen LUT min = f() min flywheel T fw F fw_actu Figure 1. Block diagram of the wind turbine model. Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd. 1647

4 Flywheel in wind turbine rotor C. Jauch 1,3 1,0 power [pu] 0,8 0,5 0,3 0,0 0,5 0,6 0,7 0,8 0,9 1,0 1,1 1,2 1,3 1,4 1,5 speed [pu] Figure 2. Power versus speed characteristic of the wind turbine. In the block diagram in Figure 1, the block flywheel is in the focus of this paper. Hence, it is discussed in detail in Section 3. The power versus speed characteristic of this variable speed wind turbine is shown in Figure 2. The drive train of the wind turbine is modeled as a two masses, spring and damper system, as is common in power system simulations of wind turbines. 8 The simulation model parameters that are relevant for the system to be introduced are listed in Table I. 3. FLYWHEEL IN WIND TURBINE ROTOR As discussed in Section 1, the concepts that have been introduced so far use the inertia of the wind turbine drive train as energy storage. This means that according to Equation (1), the rotational speed is varied in order to vary the kinetic energy that is stored in the inertia. In the system, which is proposed in this paper, the control over the stored energy is enhanced by controlling the inertia Kinetic energy A mass, m, that rotates with a certain angular speed, ω, on a circular trajectory with a certain radius, r, around a center contains kinetic energy. See Equation (1), where J is the mass moment of inertia. J is described by Equation (2). J ¼ m r 2 (2) Looking at Equation (1), it becomes obvious that the amount of energy stored in the rotating mass can be varied by varying ω. However, it can also be varied by varying J. Looking at Equation (2), it can be seen that J varies with the square of the radius, r. Therefore, r can be controlled in order to vary J and hence the kinetic energy W rot. Table I. Simulation model parameters. Parameter Value Unit Comments Rated power of wind turbine 2 MW Speed base 1500 RPM High speed side of gearbox Gearbox ratio 85 J lss 4,823,333 kgm 2 Low speed side of gearbox J hss 150 kgm 2 High speed side of gearbox R fw_min 2 m R fw_max 22 m m fw 1500 kg k fw_sp 7745 N m 1 Low speed side of gearbox c fw_damp 700 Ns m 1 Low speed side of gearbox ω ref or 1.2 pu for normal operation, 1.2 for positioning the Flywheel weight 1648 Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd.

5 C. Jauch Flywheel in wind turbine rotor 3.2. Angular momentum The angular momentum, L, describes the fact that the kinetic energy is retained in a rotating system, even if one of the variables in Equations (1) or (2) changes (Equation (3)). L ¼ J ω ¼ m r 2 ω (3) If the angular momentum changes, a torque, T, results so that the kinetic energy, W rot, remains constant. T ¼ dl dt As can be seen from Equation (3), the largest torque can be achieved if r changes, as L is a function of r 2. (4) 3.3. Flywheel system In the flywheel system proposed here, the idea is to have weights inside the rotor blades, which can be positioned along the rotor blades. These weights have a mass, and the position of these weights is a radius as the weights rotate with the rotor blades around the center of the rotor. By varying the position of these weights, i.e. the radius, a torque results that accelerates or decelerates the rotor, as can be seen from Equation (3) and Equation (4). In Figure 3, a schematic shows the principle setup of the flywheel system. In the drive train model, the inertia of the low speed shaft, J lss, is considered to be a disk. This disk is shown in more detail in Figure 3. In Figure 3, the mass of the rotor blades and the hub are depicted as aggregated masses. The blade masses are connected stiffly with the rotational mass of the hub. The weights are depicted as being mounted on the bar that connects the blade mass with the hub. The weights can vary their position on these bars by sliding along the bars. There is a spiral spring depicted between each weight and the belonging blade mass. This spring forces the weight toward the center of the rotor, against the centrifugal forces. Also shown in Figure 3 is a detailed view of one rotor blade with its weight, spiral spring and the hub. In this view, the parameter names and variable names are introduced as will be used in the following. The spiral springs are an example for an energy storage that is needed in the weight positioning system. The energy which is stored in the rotating flywheel can be increased by letting the centrifugal force pull the weight toward a larger radius. However, tapping this energy means that the weight needs to be pulled back toward the center of rotation. If moving an object a certain distance, s, requires a force, F, then translational energy, W tr is produced. W tr ¼ Fs (5) The centrifugal force, F fw_cf, in the flywheel system is described by Equation (6), where R var is the distance between the weight and the center of the rotor. F fwcf ¼ m fw R var ω rot 2 (6) Figure 3. Schematic describing the wind turbine rotor with flywheel weights in the blades. Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd. 1649

6 Flywheel in wind turbine rotor C. Jauch Hence, moving the weight in the flywheel against the centrifugal force from the largest radius, R fw_max, to the smallest radius, R fw_min, requires translational energy, as can be seen from Equation (7). W tr ¼ m fw R fw max þ R fwmin 2 ω 2 R fwmax R fwmin (7) This translational energy, which needs to be input, in order to move the weight from R fw_max to R fw_min, however, is identical with the rotational energy that is released due to the fact that the inertia is reduced by this relocation of the weight. Hence, assuming a lossless system, the energy that needs to be invested equals the energy that can be obtained. Therefore, an energy storage needs to be implemented that is charged up by moving the weight with the centrifugal force toward larger radiuses. When later rotational energy shall be released, this energy storage helps pushing the weight back toward smaller radiuses. In a variable speed wind turbine, the rotor speed can be varied; hence, the centrifugal force can be varied. This controlled centrifugal force is applied to move the weight against the spring force toward large radiuses. The force of the spring is applied to move the weight against the centrifugal force toward small radiuses. An actuator introduces a force, F fw_act, into the system to lock the weight in a certain position or to move it against dominant centrifugal forces or spring forces, as applicable. In this paper, the energy storage is assumed to be a spiral spring. Here, the intention is not to design the most suitable energy storage for this purpose. It shall only help illustrating the behavior of the system. The force, F fw_sp, that is required to deform the spring, and the energy, W fw_sp, that is stored in the deformed spring, are described by Equations (8) and (9), respectively. F fw sp ¼ k fw sp R var R fw min (8) W fw sp ¼ 1 2 k 2 fw sp R var R fw min (9) The spring coefficient, k fw_sp, needs to be chosen so that the centrifugal force equals the spring force at rated rotor speed and when the weight is at R fw_max. Since the rotor rotates around a horizontal axis, obviously, also the gravitational force has an impact on the positioning of the flywheel weight. The gravitational forces interfere with the weight positioning whenever the rotor blade is not in a horizontal position. In all other positions, the component of the gravitational force, which acts on the flywheel weight in the direction of the bar along which the weight can slide, is described by Equation (10). F fw gr ¼ m fw g cosðφþ (10) From Equation (10), it becomes obvious that, averaged over a whole revolution of the rotor, F fw_gr is zero. Hence, it has to be dealt with by the weight positioning system. This means primarily that the weights in all blades need to be positioned synchronously, as otherwise the rotor would suffer from imbalance. Apart from this, gravitation does not have an influence on the overall energy balance. Therefore, in the following, F fw_gr is neglected and the flywheel masses are considered one lumped mass. In order to emulate losses in the weight positioning system, a friction force, F fw_damp, is incorporated. Friction in the system is emulated by the friction coefficient, c fw_damp, which behaves like a system inbuilt damping, i.e. the friction force occurs whenever the weight varies its position with a certain speed, v fw. The sum of all forces acting on the flywheel weight is F fw_cntrl. F fw damp ¼ v fw c fw damp (11) F fw cntrl ¼ F fw cf þ F fw act F fw sp F fw damp (12) Equation (13) shows the differential equation that describes the positioning speed of the weight. dv fw dt ¼ F fw cntrl m fw (13) The position of the flywheel weight, R var, can vary between R fw_max and R fw_min, and it can be calculated from the weight positioning speed v fw Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd.

7 C. Jauch Flywheel in wind turbine rotor R var ¼ v fw dt (14) The constant mass of the weight and the variable position of the weight lead to the variable inertia of the flywheel, as shown in Equation (15). J fw ¼ R var 2 m fw (15) This flywheel inertia, J fw, is added to the inertia of the low speed shaft of the wind turbine drive train. However, if J fw changes, the flywheel produces a torque, as can be seen from Equations (3) and (4): T fw ¼ dj fw ω rot (16) dt The torque from the flywheel, T fw, contributes to the torque of the wind turbine rotor, which leads to the resulting torque, T lss, which acts on the low speed shaft of the drive train. T lss ¼ T fw þ T rot (17) With the equation system shown here, the flywheel in the wind turbine rotor can be modeled. In the following section, the system behavior is tested by means of simulations. 4. SIMULATION RESULTS In this section, different scenarios are simulated. To show the general behavior of the flywheel system, first part load operation with constant wind speed is simulated. In this scenario, only R var is varied to assess its impact on the system. In the second scenario, the turbine runs in part load when temporarily the wind speed increases to rated. For comparison, this scenario is simulated twice: with activated flywheel and with deactivated flywheel. The same is carried out in the third scenario where the turbine runs in full load operation when a temporary dip in wind speed causes the power output of the wind turbine to dip as well. The relevant simulation parameters are listed in Table I. As can be seen in Table I, the total mass of the flywheel weights is 1500 kg. Since the wind turbine considered here is a state of the art wind turbine, it has three rotor blades. Hence, the mass of a single weight in one rotor blade is 500 kg. This is less than 10% of the weight of a 2 MW wind turbine blade. Assuming that the weights are produced in the shape of a cube, this means that such a cube has a side length of 399 or 353 mm if it is made of iron or lead, respectively. Therefore, it is realistic to assume that such a weight would fit inside a rotor blade. This is even more the case if the weight is not produced in the shape of a cube but if it is adapted to the shape of the airfoil of the rotor blade Part load, constant wind, flywheel on In the scenario shown in this section, the wind turbine runs in part load, and the wind speed is constant. Figure 4 shows that the flywheel weight is driven from the smallest radius to the largest radius and back again to the smallest radius. 25 [m/s], radius [m] R var time [s] Figure 4. Wind speed,, and radius of flywheel weight, R var. Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd. 1651

8 Flywheel in wind turbine rotor C. Jauch Figure 5 shows the torque which results from the angular momentum and which the flywheel feeds into the low speed side of the drive train. The flywheel torque, T fw, is only shown in this subsection to show the general behavior. In the following subsections, diagrams of T fw are omitted as the general behavior is the same as shown in Figure 5. It can be seen from Figure 5 that the flywheel adds a decelerating torque (negative) to the drive train when the flywheel weight is moved to a larger radius. Once the flywheel weight becomes stationary, i.e. does not vary the radius of its trajectory, R var, the torque contribution is zero. When the flywheel weight is moved toward the center of rotation, i.e. R var is reduced, kinetic energy is released. Therefore, the flywheel adds an accelerating torque (positive) to the drive train. As can be seen from Figure 6, the accelerating and decelerating torque has an impact on the drive train speed of the wind turbine. Since in this scenario the wind turbine operates in part load, different rotor speeds mean different power that is fed into the grid (Figure 2) Part load, positive gust, flywheel off In the scenario shown in this subsection, the wind turbine operates in part load when temporarily the wind speed increases (Figure 7). The position of the flywheel weight is kept constant. The resulting rotor speed and electric power are shown in Figure 8, for the purpose of comparison with the results shown in the following subsection. As can be seen in Figure 8, the electric power, P grid, increases to rated power due to the increase in wind speed. It remains at rated power for as long as the rotor speed is at rated speed Part load, positive gust, flywheel on The scenario shown in this subsection is the same as in the previous subsection. Here, however, the transient increase in wind speed is not only used to increase P grid but also to charge up the flywheel energy storage by driving the flywheel weights to R var =R fw_max using the centrifugal forces (Figure 9). Comparing Figures 8 and 10 reveals that P grid does not increase quite as steeply when during the increase in wind speed the flywheel weights have to be moved to R fw_max. However, when the wind speed drops, the drop in P grid becomes a lot shallower due to the energy that is released from the flywheel. T [pu] time [s] Figure 5. Torque contribution of the flywheel T fw. T fw Figure 6. Speed reference, ω ref, rotor speed, ω rot, and electric power fed into the grid, P grid Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd.

9 C. Jauch Flywheel in wind turbine rotor 25 [m/s], radius [m] R var time [s] Figure 7. Wind speed,, and radius of flywheel weight, R var. Figure 8. Speed reference, ω ref, rotor speed, ω rot, and electric power fed into the grid, P grid. 25 [m/s], radius [m] R var time [s] Figure 9. Wind speed,, and radius of flywheel weight, R var. Figure 10. Speed reference, ω ref, rotor speed, ω rot, and electric power fed into the grid, P grid. Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd. 1653

10 Flywheel in wind turbine rotor C. Jauch With a different control strategy, which postpones the charging up of the flywheel until P grid has reached its rated value, the increase in P grid in Figure 10 could become as steep as in Figure 8. The downside of this strategy, however, would be that in case of shorter gusts the time might not suffice for driving the flywheel weights to R fw_max Full load, negative gust, flywheel off In the scenario shown in this subsection, the wind turbine operates in full load when temporarily the wind speed drops below rated wind speed (Figure 11). The position of the flywheel weight is kept constant. The resulting rotor speed and electric power are shown in Figure 12 for the purpose of comparison with the results shown in the following subsection. Figure 12 shows that P grid temporarily drops below rated power, and subsequently recovers to rated power, when the wind speed has recovered to its initial value Full load, negative gust, flywheel on The scenario shown in this subsection is the same as in the previous subsection. Here, however, the transient drop in P grid, caused by the transient drop in wind speed, shall be mitigated by releasing energy from the flywheel. Figure 13 shows that the flywheel weight is initially at R fw_max, as the wind turbine is in full load operation. Full load operation means that there has been sufficient energy in the wind to charge up the flywheel energy storage, i.e. to drive the flywheel weight to R fw_max with the centrifugal force. Figure 14 shows the power and the rotor speed. Comparing Figures 14 with 12 reveals that the trough in power is mitigated by the energy released from the flywheel. The flywheel energy storage can be recharged once the wind speed has recovered and P grid has returned to its rated value (Figure 14). In order to drive the flywheel weight back to R fw_max, the speed of the rotor has to be increased slightly. In Figure 14, it can be seen that the speed setpoint, ω ref, is increased. The wind turbine considered here is a speed controlled variable speed wind turbine. Therefore, in full load operation, the pitch system would reduce the aerodynamic power once ω rot has reached ω ref. As can be seen from Figure 2, the power of the wind turbine is 1 pu when ω rot reaches its rated value of pu. Increasing ω ref and hence ω rot to 1.2 pu still leads to P grid = 1 pu, but the additional aerodynamic power can be used to drive the flywheel weights to R fw_max. 25 [m/s], radius [m] R var time [s] Figure 11. Wind speed,, and radius of flywheel weight, R var. Figure 12. Speed reference, ω ref, rotor speed, ω rot, and electric power fed into the grid, P grid Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd.

11 C. Jauch Flywheel in wind turbine rotor Figure 13. Wind speed,, and radius of flywheel weight, R var. Figure 14. Speed reference, ω ref, rotor speed, ω rot, and electric power fed into the grid, P grid. 5. CONCLUSION AND FUTURE WORK In this paper, a flywheel energy storage inside a wind turbine rotor is proposed. The advantage of using the rotor of a wind turbine for this purpose is that the inertia of a flywheel increases with the square of the radius of the flywheel. State of the art wind turbine rotors of the multimegawatt class invariably have long rotor blades. Hence, positioning flywheel weights inside these rotor blades leads to large inertias. The system proposed here allows varying the position of the flywheel weights inside the rotor blades. Hence, energy can be stored and released by varying the position of these weights. The simulations shown here reveal that with the proposed flywheel system temporary changes in the power output of a wind turbine can be mitigated. A variable speed wind turbine that combines both a flywheel in the rotor, as proposed in this paper, and a speed control strategy for controlling the energy in the drive train could compensate even larger troughs in the power output. The primary purpose of the flywheel energy storage will be to increase the inertia of the power system. Wind turbines will never be able to act as primary reserve power plants, as the amount of energy that can be stored in a wind turbine is not sufficient for that. However, with a flywheel in the rotor, wind turbines can contribute substantially to power system inertia and by doing so ease the task of frequency control by other power plants. In future, the design of the mechanical structure of such a flywheel wind turbine rotor has to be worked out in detail. The weight positioning system, i.e. the actuator, the energy storage and the rail system, on which the flywheel weight slides, have to be designed in detail. The flywheel weights and the positioning system will obviously have an impact on the mechanical design of the rotor blades. Hence, it needs to be revised to allow for housing the flywheel system internally. The variable inertia of the wind turbine rotor inherently brings about the fact that the resonance frequencies of the wind turbine are no longer fixed but that they become variable too. Hence, the designs of different components of the wind turbine have to be fitted to the resulting band of resonance frequencies. Above all, the controller of the wind turbine needs to be designed so that it minimizes the mechanical loads that arise from the flywheel and from the variable resonance frequencies. Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd. 1655

12 Flywheel in wind turbine rotor C. Jauch REFERENCES 1. ENERCON GmbH. Enercon wind-diesel and stand-alone systems. ENERCON_Stand-alone_en.pdf. (Accessed 07 July 2014). 2. EirGrid. EirGrid code. Version 3.3, 30/01/2009, (Accessed 07 July 2014). 3. Koch FW, Erlich I, Shewarega F, Bachmann U. Dynamic interaction of large offshore wind farms with the electric power system. IEEE Bologna PowerTech, 2003 Bologna, Italy. 4. Yingcheng X, Nengling T. Review of contribution to frequency control through variable speed wind turbine. Renewable Energy 2011; 36(6): Erlich I, Wilch M. Primary frequency control by wind turbines. IEEE Power and Energy Society General Meeting, 2010 Minneapolis, USA. 6. Ma HT, Chowdhury BH. Working towards frequency regulation with wind plants: combined control approaches. IET Renewable Power Generation 2010; 4(4): Hansen MH, Hansen A, Larsen TJ, Øje S, Sørensen P, Fuglsang P. Control Design for a Pitch-Regulated, Variable Speed Wind Turbine. vol. Risø-R-1500(EN). Risø National Laboratory: Roskilde, Denmark, Hansen A, Jauch C, Sørensen P, Iov F, Blaabjerg F. Dynamic Wind Turbine Models in Power System Simulation Tool DIgSILENT. vol. Risø-R-1400(EN). Risø National Laboratory: Roskilde, Denmark, Sørensen P, Hansen A, Janosi L, Bech J, Bak-Jensen B. Simulation of Interaction Between Wind Farm and Power System. vol. Risø-R-1281(EN). Risø National Laboratory: Roskilde, Denmark, Sørensen P, Hansen A, Christensen P, Meritz M, Bech J, Bak-Jensen B, Nielsen H. Simulation and Verification of Transient Events in Large Wind Power Installations. vol. Risø-R-1331(EN). Risø National Laboratory: Roskilde, Denmark, Wind Energ. 2015; 18: The Authors. Wind Energy published by John Wiley & Sons, Ltd.