METHOD FOR OPERATING A WIND TURBINE

Abstract

A method for operating a wind turbine (WEA), which has a generator for supplying electric power to an electric supply grid, is provided. After being operated in a first operating state with a first power output and a first rotational speed, the wind turbine is set to a second operating state with a second power output and a second rotational speed. In order to set the wind turbine to the second operating state, an aerodynamic power output which is available for supply is determined, a target rotational speed is determined from the available output, and a target output to be set for the generator is specified from the target rotational speed.

Claims

1. A method for operating a wind power installation, comprising: feeding, by a generator, electrical power into an electrical supply network with a first operating state with a first power output and a first rotational speed, and controlling a change to a second operating state with a second power output and a second rotational speed, controlling change to the second operating state including: determining an aerodynamic available power for feeding, determining a setpoint rotational speed using the available power, determining a setpoint power using the setpoint rotational speed, and using the setpoint power to adjust the generator.

2. The method according to claim 1, comprising: determining the available power using a rotational speed of the generator and power fed into an intermediate electrical storage unit.

3. The method according to claim 1, comprising: determining the available power by a state observer.

4. The method according to claim 3, wherein the state observer used for determining the available power has a rotational speed of the generator and a mechanical torque of the generator as state variables to be observed.

5. The method according to claim 4, wherein the state observer used for determining the available power is defined by a structure of: $\left[\begin{array}{c}\stackrel{.}{\widehat{?}}\\ {\stackrel{.}{\hat{T}}}_{\mathrm{mech}}\end{array}\right]=\left[\begin{array}{cc}-k_{?}& \frac{1}{J}\\ -k_T& 0\end{array}\right]?\left[\begin{array}{c}\widehat{?}\\ {\hat{T}}_{\mathrm{mech}}\end{array}\right]+\left[\begin{array}{cc}{k}_{?}& -\frac{1}{J}\\ k_T& 0\end{array}\right]?\left[\begin{array}{c}?\\ T_{\mathrm{el}}\end{array}\right]$ in which J denotes a combined moment of inertia of a rotor and the generator, ? is the rotational speed of the generator, k.sub.? and k.sub.T are parameters for influencing a dynamic response of the state observer, T.sub.el is an electrical torque and is determined calculated as a ratio of a power P.sub.DC fed into the intermediate electrical storage unit and the rotational speed ?, and T.sub.mech denotes the mechanical torque of the rotor and the generator, wherein the rotational speed of the generator is an actual or observed rotational speed of the generator and the mechanical torque of the rotor is an observed mechanical torque of the rotor.

6. The method according to claim 5, comprising: determining, by a regulator, the setpoint power using the available power and the actual rotational speed of the generator.

7. The method according to claim 1, comprising: determining the setpoint rotational speed as a function of the aerodynamic available power from a rotational speed/power characteristic curve.

8. The method according to claim 5, comprising: determining, by a regulator, a difference between the setpoint rotational speed and the actual rotational speed, determining, by the regulator, a differential torque between the observed mechanical torque and the electrical torque using a regulating algorithm, determining a setpoint torque as a sum of the differential torque and the observed mechanical torque, and determining the setpoint power as a product of the setpoint torque and the actual rotational speed.

9. The method according to claim 1, comprising: temporarily increasing the first power output during the first operating state beyond a power output that the wind power installation is able to generate based on prevailing wind conditions.

10. A control unit for determining an available power of a wind power installation, comprising a generator, wherein the control unit observes a rotational speed of the generator and a mechanical torque of the generator as state variables.

11. The control unit according to claim 10, wherein the control unit is defined by a structure of: $\left[\begin{array}{c}\stackrel{.}{\widehat{?}}\\ {\stackrel{.}{\hat{T}}}_{\mathrm{mech}}\end{array}\right]=\left[\begin{array}{cc}-k_{?}& \frac{1}{J}\\ -k_T& 0\end{array}\right]?\left[\begin{array}{c}\widehat{?}\\ {\hat{T}}_{\mathrm{mech}}\end{array}\right]+\left[\begin{array}{cc}{k}_{?}& -\frac{1}{J}\\ k_T& 0\end{array}\right]?\left[\begin{array}{c}?\\ T_{\mathrm{el}}\end{array}\right]$ in which J denotes a combined moment of inertia of a rotor of the wind power installation and the generator, ? is the rotational speed of the generator, k.sub.? and k.sub.T are parameters for influencing a dynamic response of the control unit, T.sub.el is an electrical torque and is calculated as a ratio of a power (P.sub.DC) stored in an intermediate electrical storage unit and the rotational speed ?, and T.sub.mech denotes the mechanical torque of the rotor and the generator, wherein, and the control unit calculated the available power as a product of the observed rotational speed {circumflex over (?)} and the observed mechanical torque {circumflex over (T)}.sub.mech.

12. A wind power installation comprising: a generator for feeding electrical power into an electrical supply network, wherein the wind power installation is configured to operate in, a first operating state with a first power output and a first rotational speed, and after the first operating state, the wind power installation is controlled to transition into a second operating state with a second power output and a second rotational speed; and a control unit for determining an available power of the wind power installation based on an observed rotational speed of the generator and a mechanical torque of the generator.

13. (canceled)

14. The wind power installation according to claim 12, wherein the control unit is defined by a structure of: $\left[\begin{array}{c}\stackrel{.}{\widehat{?}}\\ {\stackrel{.}{\hat{T}}}_{\mathrm{mech}}\end{array}\right]=\left[\begin{array}{cc}-k_{?}& \frac{1}{J}\\ -k_T& 0\end{array}\right]?\left[\begin{array}{c}\widehat{?}\\ {\hat{T}}_{\mathrm{mech}}\end{array}\right]+\left[\begin{array}{cc}{k}_{?}& -\frac{1}{J}\\ k_T& 0\end{array}\right]?\left[\begin{array}{c}?\\ T_{\mathrm{el}}\end{array}\right]$ in which J denotes a combined moment of inertia of a rotor of the wind power installation and the generator, ? is the rotational speed of the generator, k.sub.? and k.sub.T are parameters for influencing a dynamic response of the control unit, T.sub.el is an electrical torque and is calculated as a ratio of a power (P.sub.DC) stored in an intermediate electrical storage unit and the rotational speed ?, and T.sub.mech denotes the mechanical torque of the rotor and the generator.

15. The wind power installation according to claim 14, wherein the control unit is configured to determine the available power as a product of the observed rotational speed {circumflex over (?)} and the observed mechanical torque {circumflex over (T)}.sub.mech.

16. The method according to claim 2, wherein the intermediate electrical storage unit is a DC voltage intermediate circuit.

17. The method according to claim 5, comprising: determining the available power as a product of the observed rotational speed {circumflex over (?)} and the observed mechanical torque {circumflex over (T)}.sub.mech.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0050] The invention will be explained in more detail below by way of example with the aid of exemplary embodiments with reference to the appended figures.

[0051] FIG. 1 shows a wind power installation in a perspective representation.

[0052] FIG. 2 shows a structure for specifying the setpoint power to illustrate a proposed method.

[0053] FIG. 3 illustrates a rotational speed regulating structure in a block diagram.

[0054] FIG. 4 shows a plurality of diagrams illustrating a process of a frequency dip with the sequence and effect of support by a wind power installation.

[0055] FIG. 5 shows, in a similar way to FIG. 4, diagrams to illustrate the sequence of a frequency dip together with support by a wind power installation, in which in contrast to FIG. 4 a variable setpoint value can be adjusted for the supporting power.

[0056] FIG. 6 shows simulated power profiles for the return of the power to the normal power and therefore to the normal operating point according to one embodiment and in comparison with another embodiment.

DETAILED DESCRIPTION

[0057] FIG. 1 shows a wind power installation 100 having a tower 102 and a nacelle 104. A rotor 106 having three rotor blades 108 and a spinner 110 is arranged on the nacelle 104. During operation, the rotor 106 is set in a rotational movement by the wind and thereby drives a generator in the nacelle 104.

[0058] FIG. 2 shows a block diagram 2 for the specification of a setpoint power P.sub.setpoint, which is intended to be provided to the wind power installation as a power setpoint value in order to return the wind power installation to a normal operating point on the basis thereof. Block diagram 2 schematically shows the sequence in an overview.

[0059] As input variables, a power P.sub.DC fed into a DC electrical voltage intermediate circuit and the rotational speed ? of the generator are fed into the power observer 4 at its input 6. This power observer 4 is a state observer, which can have the rotational speed of the generator and the mechanical torque of the generator as observed state variables. As a result, the available power {circumflex over (P)}.sub.avaDC is obtained, which is also referred to as the aerodynamic power available to be fed in. The index of this observed power {circumflex over (P)}.sub.avaDC also indicates that it relates to the DC voltage intermediate circuit. Specifically, this power is that which, although it depends on the prevailing wind speed and also on the current operating state of the wind power installation, i.e., it depends on how much power can be drawn from the wind, nevertheless also already takes into account that a part of this power that can be drawn from the wind is being used in another way, in particular for excitation of the generator or for covering the losses during the energy conversion in the generator. This available power {circumflex over (P)}.sub.avaDC therefore describes a power which is actually available and could be fed into the electrical supply network by the power inverter.

[0060] To this extent, although this available power is that which is available at the intermediate circuit, it nevertheless differs from the power fed into the intermediate circuit in that it is continuously available because of continuous replenishment from the wind and would not lead to a change in the rotational speed of the wind power installation.

[0061] From this available power {circumflex over (P)}.sub.avaDC, a setpoint rotational speed ?.sub.setpoint is then determined in a characteristic curve block 8 in which a rotational speed/power characteristic curve is stored. This setpoint rotational speed is an input variable for the rotational speed regulator in the rotational speed regulator block 10, which will be explained further in FIG. 3.

[0062] The available power {circumflex over (P)}.sub.avaDC is likewise an input variable for the rotational speed regulator in block 10, and the rotational speed is also a further input variable for this rotational speed regulator in the rotational speed regulator block 10. In this case, the measured rotational speed or the rotational speed present as an estimated state variable in the power observer 4 may be used.

[0063] The rotational speed regulator in block 10 then generates the setpoint power for the wind power installation at its regulator output 12 in a dynamic process.

[0064] FIG. 3 illustrates a rotational speed regulator in the sense of the rotational speed regulator block 10. Correspondingly, the setpoint rotational speed ?.sub.setpoint, the actual rotational speed ? and, indirectly, the available power are also taken into account here as input variables. The available power is not indicated in FIG. 3, but the available mechanical torque {circumflex over (T)}.sub.mech is related thereto and indirectly forms a corresponding input variable.

[0065] The setpoint rotational speed may initially be fed through a digital filter 30, which may for example be a first-order low-pass filter. A regulator structure with two degrees of freedom can be formed in this way, which allows an improvement of the robustness of the regulating loop in relation to modelling inaccuracies of the regulating path for consistently good setpoint value tracking performance. The converse case, i.e., that there is good robustness with good modelling accuracy but not so good setpoint value performance, should also be mentioned here.

[0066] A difference between the setpoint and actual values of the rotational speed is then formed in the difference block 32. This difference is sent to the regulating block 34, which determines the corresponding differential torque therefrom by means of a regulating algorithm. The regulating algorithm may, in particular, comprise a P structure, a PI structure or a PID structure.

[0067] The differential torque determined in this way is added to the observed mechanical torque {circumflex over (T)}.sub.mech at the adder component 36 so that a setpoint torque is obtained as T.sub.setpoint, which can be transmitted as an input variable and specified variable to the control of the wind power installation. This is represented by the wind power installation block 38. The calculation of the setpoint power P.sub.setpoint by multiplying the setpoint torque T.sub.setpoint by the rotational speed ? may also be carried out in the wind power installation block 38.

[0068] Correspondingly, the wind power installation is thus regulated in the wind power installation block 38, and the wind power installation per se is also contained therein and ultimately delivers as output the physical variable of its rotational speed, i.e., the rotational speed ? of the generator, which is fed back to the difference component 32. The difference component 32 may naturally also be referred to as an adder component 32 with negative input.

[0069] The field of application is concentrated on the use of wind power installations for network support in the event of frequency dips, for example in the event of power station outages. The power observer may in this context also be used as a basis for various other regulating algorithms for a wind power installation, for example for reserved power regulation or optimized regulation of the working point during normal operation of the wind power installation.

[0070] The conduct of a network support by power increase is made even more compatible than before, and in particular the reduction of the effective power after the end of the power increase is kept as small as possible. In this way, a further disruption of the energy system power balance is restricted to a minimum amount.

[0071] A return to an optimal working point after the end of a power increase is schematically represented in FIG. 4, which shows a typical behavior.

[0072] FIG. 4 represents as a function of time t the wind speed v, the network frequency f, the rotational speed n, which may also be denoted as ?, and the power P fed in. At time t.sub.0, the frequency dip occurs, or is detected. The frequency f therefore dips, the power P is set far above the previous value, and kinetic energy is used for this, with the result that the rotational speed n decreases slowly. It is assumed that the wind speed is essentially constant over the entire period of time studied.

[0073] The network support by the increased power input ends at the time t.sub.1, at which the power P then falls slowly, namely greatly below the previous value and therefore also greatly below the amount possible because of the existing wind speed.

[0074] At time t.sub.2, the power then gradually increases again. The rotational speed n then increases again gradually as well.

[0075] At time t.sub.3, the conditions have been normalized again and the operating point is a normal operating point, with a power P fed in and a rotational speed n as before the frequency dip. In the example, however, the frequency f has already recovered significantly earlier.

[0076] As represented in FIG. 4, the power setpoint value of the wind power installation (WPI) is therefore reduced after the end of the adjusted power increase duration over the period of time t.sub.inertia, lead, back. After the end of this period of time, i.e., for instance at the time t.sub.2, the power setpoint value is at the value which is assigned to the current rotational speed according to a firmly preadjusted rotational speed/power characteristic curve. Since, however, the rotational speed is too low relative to an optimal value for the current wind conditions because of the braking effect of the previous power increase, this power setpoint value does not correspond to the aerodynamic power currently being caused by the wind. Instead, the power setpoint value often lies significantly below the effective power level delivered before the initiation of the network support, to which References [1, 2] also relate. Because of the great reduction of the effective power, the wind power installation (WPI) accelerates within a few seconds back to a higher rotational speed and returns to a normal operating behavior. From the point of view of the energy system, however, such a great reduction of the effective power output is to be evaluated as a new disruption, and is therefore to be reduced as far as possible; Reference [1] also relates to this.

[0077] A power increase with a variable setpoint value is represented in FIG. 5.

[0078] The profiles and conditions of FIG. 5 correspond to those of FIG. 4, and reference is therefore made to FIG. 4 for explanations. The essential difference of the behavior according to FIG. 5 is that in the case of the frequency dip at to, the power fed in is initially increased very gradually to a starting power, but the starting power is not maintained but lies approximately by a constant differential value above a fictitious power assigned to the decreasing rotational speed n. This fictitious power is indicated in FIG. 5, as well as in FIG. 4, respectively by dashes in the lower diagram.

[0079] As a result, however, the power dip behaves here very similarly as explained with reference to FIG. 4.

[0080] One published method is that which is referred to as Dynamic Inertia Regulation by Repower Systems according to Reference [3]. In this method, control of the return of the wind power installation (WPI) from a working point with a reduced rotational speed to an optimal working point, which may also be referred to as a normal working point, is carried out by the specification of a setpoint rotational speed gradient, or a gradient corridor defined by minimum and maximum gradients. This accordingly leads to a constant difference between the aerodynamic and electrical torques during the period of the return. One disadvantage of this method is the use of a rotational speed gradient in the context of the regulating method. This requires a high-quality measurement of the rotational speed signal, so that a numerical derivation is not unsuitable because of an excessively high noise level for the regulation.

[0081] Otherwise, strong filtering of the rotational speed signal or, similarly to the method, the use of a state observer for the rotational speed and/or its derivation, could be a solution. This, however, is not described in Reference [3].

[0082] FIG. 6 shows a comparison of the simulated power profiles for a known method P.sub.old and for a method proposed herein P.sub.new, which is parameterized for a return duration of about 60 s with a wind speed v.sub.w=8.5 m/s. It can be seen that the power P.sub.new according to the method proposed decreases much less after the end of the power increase at t.sub.1 than the power P.sub.old according to the comparative method.

[0083] Provided is a reduction in the power of the wind power installation (WPI) after the end of the power increase during the network support and controllability of the duration of the return to the optimal working point of the wind power installation (WPI).

[0084] A particular intention is to achieve controllability of the depth of the power dip after the end of the network supporting power increase phase by adjustability of a setpoint value filter in the regulating method, and, with suitable parameterization of the filter, i.e., the state observer, significant reduction of the power dip compared with previously known methods and therefore a reduction of destruction of the power balance nonequilibrium in the energy system in question.

[0085] Referring to the embodiments and results explained in detail, but also in a way which may be generalized, the following will also be added for explanation.

[0086] Provided is a method for network support by wind power installations, which are abbreviated to WPI, in the event of a significant dip of the network frequency. The possibility of temporarily increasing the delivered effective power of a wind power installation in such a case is known, and will be referred to here as support.

[0087] The effective power increase in response to a network frequency dip may be carried out over an adjustable period of time and with a particular established increase relative to the effective power at the time when the support is initiated.

[0088] The consequence of an increased effective power output is typically a reduction of the rotational speed of the WPI compared with the moment before initiation of the support. This may be represented illustratively with the aid of the acceleration equation of a rotating one-mass system:

[00003]$J.Math..Math.\stackrel{.}{?}=T_{\mathrm{mech}}-T_{\mathrm{el}}=\frac{1}{?}.Math.\left(P_{\mathrm{mech}}-P_{\mathrm{el}}\right)$

[0089] In this formula, J denotes the moment of inertia of the rotating masses of the WPI, ? denotes the rotational speed of the WPI, T.sub.mech and P.sub.mech respectively denote the torque and the power occurring on the shaft because of the wind, and T.sub.el/P.sub.el denote the electromechanical torque, or the power, of the generator. If a generator power which exceeds the mechanical power is drawn over a period of time, braking of the WPI takes place.

[0090] This inequality of the power balance occurs with an approximately constant wind speed during a support event. The effect is further exacerbated because the aerodynamic efficiency of the WPI may deteriorate with a decreasing rotational speed and constant wind speed relative to an optimal working point before the start of the support function. An optimal working point is achieved with a particular tip-speed ratio, i.e., the ratio of the speed of revolution of the blade tip to the wind speed. The decrease of the tip-speed ratio during the initiation of support typically leads to a decreasing aerodynamic power with a constant wind speed. After the end of the power increase, the WPI is thus often at a suboptimal working point in terms of rotational speed and wind speed, i.e., the rotational speed is too low in relation to the wind speed, so that the tip-speed ratio is not optimal. The problem thus arises of subsequently reaccelerating the WPI in order to reach a higher rotational speed and, in association therewith, higher aerodynamic efficiency. To this end, various options may be selected: [0091] A strong reduction of the effective power output after the end of the supporting power increase, which leads to a rapid acceleration of the WPI and therefore a rapid return to an optimal working point. Reference [2] relates to this. [0092] A small reduction of the effective power output after the end of the supporting power increase, which leads to a slow acceleration or, in the event of too little reduction, even further braking of the WPI.

[0093] In terms of frequency stability in an energy system, the latter option is advantageous, which Reference [1] also relates to. A strong reduction of the effective power by a large number of WPIs in an energy system is equivalent to another disruption of the power balance of this energy system. Often, it is precisely such a disruption of the power balance, which is caused for example by a failure of a power station or a line, that was responsible for a frequency dip and therefore for the initiation of the support function. It is therefore appropriate to limit the power reduction to a minimum extent, so that the rotational speed of the WPI does not decrease further but can be increased in a controlled way and over a more prolonged period of time back to the optimal rotational speed.

[0094] Described herein is returning the WPI to an optimal operating point after the end of a supporting power increase with the least possible power reduction in relation to the effective power delivered by the initiation of support. To this end, a method which is graphically represented in FIG. 2 is provided.

[0095] For each measurement, the values of rotational speed (?) and power in the DC intermediate circuit (P.sub.DC) are recorded. From these, with the aid of a state observer, an available aerodynamic power relating to the intermediate circuit ({circumflex over (P)}.sub.avaDC) is calculated, i.e., it corresponds to the aerodynamic power less the generator losses. For this available power, a setpoint rotational speed (?.sub.setpoint) is calculated by means of a characteristic curve. By using the calculated available power {circumflex over (P)}.sub.avaDC, a power setpoint value is finally output by a rotational speed regulator. This is converted by the existing power regulating software of the WPI into a control signal for the electrically excited generator, which leads to the WPI reaching the setpoint rotational speed over a parameterizable period of time. The individual method components will now be described separately:

[0096] For the power observer and on the basis of Equation (1), the following state space model of the WPI can be set up, with the state variables rotational speed (?) and mechanical torque (T.sub.mech):

[00004]$\left[\begin{array}{c}\stackrel{.}{?}\\ {\stackrel{.}{T}}_{\mathrm{mec}}\end{array}\right]=\left[\begin{array}{cc}0& \frac{1}{J}\\ 0& 0\end{array}\right]?\left[\begin{array}{c}?\\ T_{\mathrm{mech}}\end{array}\right]+\left[\begin{array}{c}-\frac{1}{J}\\ 0\end{array}\right].Math.T_{\mathrm{el}}$

[0097] Of the variables used here, the mechanical torque T.sub.mech cannot be measured and therefore needs to be calculated from the measurement data by means of a state observer. Since the rotational speed signal is furthermore often measured only with a low resolution and with a low sampling rate, a state observation is also proposed for this value. A suitable observer structure may be formulated as follows, reference being made to Reference [4] for further explanation:

[00005]$\left[\begin{array}{c}\stackrel{.}{\widehat{?}}\\ {\stackrel{.}{\hat{T}}}_{\mathrm{mech}}\end{array}\right]=\left[\begin{array}{cc}-k_{?}& \frac{1}{J}\\ -k_T& 0\end{array}\right]?\left[\begin{array}{c}\widehat{?}\\ {\hat{T}}_{\mathrm{mech}}\end{array}\right]+\left[\begin{array}{cc}{k}_{?}& -\frac{1}{J}\\ k_T& 0\end{array}\right]?\left[\begin{array}{c}?\\ T_{\mathrm{el}}\end{array}\right]$

[0098] Here, in contrast to the measured input variables of rotational speed ? and the electrical torque

[00006]$T_{\mathrm{el}}=\frac{P_{D.Math..Math.C}}{?},$

observed variables are denoted by ?. The two parameters k.sub.? and k.sub.T influence the dynamic behavior and, with time-discrete implementation, also the stability of the state observer, and must be selected while taking these aspects into account.

[0099] The calculation of the aerodynamic power {circumflex over (P)}.sub.avaDC is carried out by means of the product of rotational speed and torque.

[0100] Regarding the characteristic curve, it should be explained that this component assigns a setpoint rotational speed ?.sub.setpoint to each calculated available power {circumflex over (P)}.sub.avaDC, for example by means of a static characteristic curve.

[0101] The rotational speed regulator must achieve a return of the WPI to the calculated setpoint rotational speed while taking into account an effective power reduction which is as small as possible. Its structure is represented in FIG. 3 in the form of a simplified block diagram. The rotational speed error is calculated, and a differential torque between the observed mechanical torque and the electrical torque is calculated therefrom by means of a P/PI/PID regulator C. This is added to the observed mechanical torque, and a setpoint torque, or after multiplication by the current rotational speed a setpoint electrical power, is therefore calculated.

[0102] Preferably, the setpoint rotational speed is filtered by a digital filter F, for example a first-order low-pass filter, before calculation of the rotational speed error. This gives rise to a regulator structure with two degrees of freedom, which allows an improvement of the robustness of the regulating loop in relation to modelling inaccuracies of the regulating path for consistently good setpoint value tracking performance or vice versa. The duration of the return of the WPI from the reduced rotational speed to the setpoint rotational speed may be adjusted according to the desired requirements through the selection of a parameter of the filter F. For example, a stabilization time of the regulating loop of 60 s is a preferred selection.

[0103] Besides the reduction of the power dip in the recovery phase, the described power observer also allows a further option for the increase of the power after the detection of a frequency dip. In this case, the power may be increased by a fixed value relative to the time-variable normal rotational speed-dependent power setpoint value. Here, with the power observer, there is the possibility for this of using the calculated aerodynamic power, rather than as previously a rotational speed-dependent power setpoint value, as a reference for the power increase.

[0104] The invention, at least according to one embodiment, has the object of making the conduct of a network support by support after a frequency dip particularly compatible for the energy system, and in particular to keep the reduction of the effective power after the end of the power increase as small as possible. In this way, a further disruption of the energy system power balance is restricted to a minimum amount.

REFERENCES

[0105] [1] Asmine, C.-?. Langlois: Field Measurements for the Assessment of Inertial Response for Wind Power Plants based on Hydro-Qu?bec Trans?nergie Requirements. Proceedings of the 13.sup.th International Workshop on Large-Scale Integration of Wind Power into Power Systems, Berlin, October 2014. [0106] [2] M. Fischer, S. Engelken, N. Mihov, A. Mendonca: Operational Experiences with inertial Response Provided by Type 4 Wind Turbines. Proceedings of the 13.sup.th International Workshop on Large-Scale Integration of Wind Power into Power Systems, Berlin, October 2014. [0107] [3] T. Kr?ger, J. Geisler, S. Schr?der (Repower Systems AG): Dynamic Inertia Regulation. Published international patent application, publication number WO 2011/124696 A2. [0108] [4] C. M. Verrelli, A. Savoia, M. Mengoni, R. Marino, P. Tomei, L. Zarri: On-Line Identification of Winding Resistances and Load Torque in Induction Machines. IEEE Transactions on Control Systems Technology, Vol. 22(4), July 2014.