Method for operating a wind turbine

Abstract

A method for operating a wind turbine configured to supply electrical power to an electrical power supply network via a converter comprises measuring a phase angle and measuring a voltage of the electrical power supply network. A value of at least one parameter for a phase locked loop is determined depending on the measured voltage. The phase locked loop is used to determine a corrected phase angle depending on the measured phase angle and the voltage of the electrical power supply network. The corrected phase angle is input into the converter.

Claims

1. A method for operating a wind turbine configured to supply electrical power to an electrical power supply network via a converter, the method comprising: measuring a phase angle and measuring a voltage of the electrical power supply network; determining a value of at least one parameter for a phase locked loop depending on the measured voltage; adjusting for a network error, wherein the network error comprises a voltage drop to a value of 15% of nominal value or less; using the phase locked loop to determine a corrected phase angle depending on the measured phase angle and the measured voltage of the electrical power supply network; and inputting the corrected phase angle into the converter.

2. The method according to claim 1, wherein the at least one parameter determines dynamics of the phase locked loop.

3. The method according to claim 2, wherein the at least one parameter is configured to slow the dynamics if the measured voltage is low.

4. The method according to claim 3, wherein the at least one parameter assumes a value such that the phase locked loop holds the corrected phase angle at an almost constant value if the measured voltage is below a threshold value.

5. The method according to claim 1, further comprising a proportional loop filter, wherein the at least one parameter is a proportionality constant K.sub.P, and wherein a value of the proportionality constant is determined in a voltage-dependent manner.

6. The method according to claim 5, further comprising a low-pass filter, wherein the at least one parameter is a time constant T.sub.1, and wherein a value of time constant is determined in a voltage-dependent manner.

7. The method according to claim 6, wherein dynamics of the phase locked loop are determined in a voltage-dependent manner by at least one of the proportionality constant Kp and the time constant T.sub.1.

8. The method according to claim 6, wherein the proportionality constant K.sub.P assumes a value of almost zero when there is a network error.

9. The method according to claim 8, wherein the value of the time constant T.sub.1 is considerably increased when there is a network error.

10. The method according to claim 6, further comprising a dead-time element having an adjustable dead time Tt.

11. The method according to claim 1, wherein the at least one parameter is specified as a continuous function depending on the voltage.

12. The method according to claim 11, wherein the value of the at least one parameter progresses in a voltage-dependent manner between a minimum value and a maximum value.

13. The method according to claim 1, wherein the voltage is a positive sequence voltage.

14. The method according to claim 1, wherein the voltage is a negative sequence voltage.

15. The method according to claim 1, wherein the corrected phase angle is determined in a continuous manner.

16. The method according to claim 1, further comprising setting the at least one parameter of the phase locked loop.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention is explained in greater detail below using an example. In the figures:

(2) FIG. 1 illustrates a side view of an embodiment of a wind turbine;

(3) FIG. 2 illustrates a schematic view of an embodiment of an electrical structure of the wind turbine coupled with an electrical power supply network in a wind farm;

(4) FIG. 3 illustrates a schematic view of an embodiment of a phase locked loop (PLL) for the continuous determination of the phase angle depending on the network voltage;

(5) FIG. 4 illustrates two examples of a dependency of the PLL parameters on the positive sequence voltage; and

(6) FIG. 5 illustrates an example of the temporal behavior of the PLL when there is a deep voltage error.

DETAILED DESCRIPTION OF THE INVENTION

(7) FIG. 1 shows a wind turbine 10 with a tower 12 and a nacelle 14 arranged on it. The nacelle 14 supports a rotor 16 with its rotor blades 18.

(8) FIG. 2 shows a wind farm with a plurality of wind turbines 20, 22, 24, 26, which is controlled by a wind farm controller 28. The wind farm controller 28 is thereby configured for bidirectional data exchange with the wind turbines 20, 22, 24, 26, or respectively their controllers. The wind turbines 20, 22, 24, 26 supply power to a common wind farm network, which supplies an electrical power supply network 32 via a coupling point 30. The wind turbines 20, 22, 24, 26 each supply the wind farm network via a medium-voltage transformer 34, shown for example for the wind turbine 26.

(9) The wind turbine 26 is shown as a double-fed asynchronous machine. The rotor 36 drives the rotor of a generator 38 via a drive train (not shown in detail). The rotor circuit of the generator 38 is connected with the transformer 34 and the electrical power supply network 32 via a converter system 40. The stator circuit of the generator 38 is connected directly with the transformer 34 and the electrical power supply network 32. The converter system 40 has a rotor-side converter 42 and a network-side converter 44, which are interconnected via a direct current link. The converters 42 and 44 are controlled by a converter controller 46.

(10) The full wind turbine 26 is controlled via a wind turbine controller 48, which exchanges bidirectional data with the wind farm controller 28 for the control. A series of input variables are present at the converter controller 46. Measured values or values derived from the measured values can be present as input variables at the converter controller 46. The measured variables required for the control can either be measured by sensors 50 on the system side of the transformer 34 or by sensors 52 on the network side of the transformer 34. For example, the voltage and currents on the individual phases of the system are captured as measured variables. The processing of the measured values can take place by the sensors 50, 52 themselves or in the converter controller 46. The sensors 50, 52 are configured to provide suitable measured data or to provide data to the converter controller 46 for further processing. For synchronization with the electrical power supply network, at least one phase angle line and one positive sequence voltage ULV+ (Positive Sequence Voltage) are present after the processing of the measured data in the converter controller 46 as input variables for the phase locked loop (PLL) described below. The input variables can be determined in the known manner from the measured variables.

(11) FIG. 3 shows an exemplary phase locked loop (PLL) 58, which is set up to continuously determine a corrected phase angle PLL to be used for the control of the converter 40 (FIG. 2) depending on the existing positive sequence voltage U.sub.LV+ and the measured phase angle .sub.line. The phase locked loop 58 in the exemplary embodiment according to FIG. 3 is configured as part of the control software of the converter regulator 46 (FIG. 2).

(12) The measured phase angle .sub.line and the corrected phase angle .sub.PLL are present at a subtraction element 60, which is set up to form a phase angle difference from the measured phase angle .sub.line and the corrected phase angle .sub.PLL. The corrected phase angle .sub.PLL can be understood in the control-oriented sense as a setpoint value for the network-synchronous operation of the converter 40 (FIG. 2). The difference formed in the subtraction element 60 can be considered in a control-oriented sense as a control variable. The phase angle difference is present at a proportional element (P element) 62 and is amplified in a voltage-dependent manner with a factor K.sub.P. The corresponding amplification factor K.sub.P is specified in a voltage-dependent manner by a calculation block 76 and is present as an input variable at the P element 62.

(13) The measured phase angle .sub.Line is present in a parallel branch at a derivative element (D element) 66, which is set up to determine the network angular frequency from the measured phase angle .sub.Line. The network angular frequency is present as an input signal at a dead-time element (PTt element) 68, which is set up to provide the network frequency delayed for the calculation of the phase angle. The delay of the dead-time element can be optionally adjusted. The delay can preferably be specified such that, in the event of a network error, it is not a network frequency disturbed by the error but rather the network frequency before the occurrence of the error that is used for determining the phase angle. In the specified exemplary embodiment, the dead-time element continuously delays the network frequency used to calculate the phase angle by the time corresponding to three network periods. The output signal of the PTt element 68 is present at a delay element (PT.sub.1 element) 64. The PT.sub.1 element 64 has a proportional transmission behavior together with a first order delay. The corresponding parameter T.sub.1 for the delay is specified in a voltage-dependent manner by a calculation block 78 and is present as an input variable at the PT.sub.1 element 64. The output of the PT.sub.1 element 64 is present at an optional, adjustable frequency limiter 70, which limits positive and negative frequency deviations.

(14) The output of the parallel branch with the members 64, 66, 68 and 70 is present together with the output of the P element 62 at a summation element 72. The sum of the amplified measured phase angle and the filtered network angular frequency is integrated into the corrected phase angle .sub.PLL via an integration element (I element) 74. An integration constant KI is specified for this in the I element. The phase angle .sub.PLL is provided to control the switching elements of the converter 40 (FIG. 2).

(15) The calculation blocks 76 and 78 of the PLL 58 shown in FIG. 3 are configured to determine the PLL parameters K.sub.P and T.sub.1 depending on the positive sequence voltage U.sub.LV+, which is present as an input variable at the two calculation blocks 76, 78. In order to determine the PLL parameters, functions or respectively characteristic curves are to be specified in the calculation blocks 76, 78 depending on the positive sequence voltage U.sub.LV+. The two functions or respectively characteristic curves are shown symbolically in FIG. 3 by the curves f(U.sub.LV+). The characteristic curves are specified as functions depending on the network voltage. Alternatively, the characteristic curves can also be specified in the form of a look-up table, which specifies the characteristic curve's progression through a plurality of value pairs of the network voltage and of the respective PLL parameter. It can be provided for the functions f(U.sub.LV+) as a linear function or a non-linear function. The explanation for FIG. 4 covers this in greater detail.

(16) The functions f are substantially constant values for the control parameters K.sub.P and T.sub.1 are present in an upper range around values for the positive sequence voltage of 100%, with respect to the nominal value of the positive sequence voltage U.sub.LV+. The upper range is defined for example by values of 90% to 115% of the nominal value of the positive sequence voltage. The upper limit of this range can correspond for example to a maximum value for the positive sequence voltage, in the case of which the wind turbine can still be operated at least temporarily in the event of a network voltage failure. The functions are further preferably specified such that substantially constant values for the control parameters K.sub.P and T.sub.1 are present in a lower range around values for the positive sequence voltage of 0%, with respect to the nominal value of the positive sequence voltage. The lower range can range for example from percentage values of 0 to 15%. Depending on the electrical power supply network, to which the wind turbine should supply power, the lower range can also be selected broader, e.g. from 0 to 30%, or narrower, e.g. from 0 to 5%. The curve progression between the lower range and the upper range can be specified depending on properties of the incoming power network. However, the correlation should at least be such that the functions are continuous over the entire definition range and fall or rise in a monotone manner over the entire relevant range. A smooth progression, i.e. a constant first deviation of the functions, is also preferably advantageous.

(17) In an embodiment, the functions are specified depending on further network variables such as a network impedance, a network short-circuit current, frequency gradients or low-frequency voltage harmonics, which are taken into consideration as variables in the functions and which exist instead as input variables at the calculation blocks 76 and 78. In another embodiment, a plurality of characteristic curves and/or functions for determining the PLL parameters K.sub.P and T.sub.1 are specified in the calculation blocks 76 and 78, from which suitable characteristic curves can be selected for determining the PLL parameters K.sub.P and T.sub.1 depending on the control signals present at the blocks. For this, the plurality of characteristic curves and/or functions are each assigned to a corresponding control signal value for the pending input signals. The calculation blocks 76 and 78 can have a characteristic line selector for selecting the characteristic curves/functions, which is set up to select both a characteristic curve for determining the PLL parameter K.sub.P and a characteristic curve for determining the PLL parameter T.sub.1 depending on the control signals and their signal values according to the assignment. The control signals are thereby preferably provided by a higher-level controller than the converter controller 46 (FIG. 2). For example, the wind turbine controller 48 (FIG. 2) or the wind farm controller 28 (FIG. 2) can be set up to provide corresponding control signals. The provision can thus be adjusted for example for the actual reactive power supply of the wind turbines or respectively of the wind farm and the actual error pattern or respectively the temporal progression of the voltage during the network error. The higher-level controller can thus adjust the behavior of the phase locked loop according to different phases of the network error, whereby a dynamic adjustment of the PLL parameters is possible over a broad range through the voltage dependency of the functions or respectively characteristic curves.

(18) The left curve of FIG. 4 shows potential characteristic curve progressions for the PLL parameter T.sub.1 of the PT.sub.1 element 64 depending on the positive sequence voltage U.sub.LV+, as they can be specified in the block 78, and the right curve shows potential characteristic curve progressions for the PLL parameter K.sub.P of the P element 62 depending on the positive sequence voltage U.sub.LV+, as they can be specified in the block 76. The positive sequence voltage is thereby applied as a percentage value referring to its nominal value U.sub.N. For the PLL parameter T.sub.1, a percentage representation referring to a specified parameter T.sub.1,N was selected, which is scaled by means of a stored function. For the PLL parameter K.sub.P, a percentage representation referring to a specified parameter K.sub.P,N is selected, which is also scaled by means of a stored function.

(19) Referring to the left curve of FIG. 4, the PLL parameter T.sub.1 of the PT.sub.1 element 64 only has a slight slope and substantially constant, very high values in the range from 0 to 20% of the positive sequence voltage. For a positive sequence voltage from 100 to 120%, the parameter also only has a slight slope and substantially a constant value of approximately T.sub.1,N. In a range of approximately 20 to 50% of the positive sequence voltage, the characteristic curves have a steep gradient. The solid characteristic curve changes the behavior of the PT.sub.1 element only for deeper voltage errors than the dotted characteristic line.

(20) Referring to the right curve of FIG. 4, the PLL parameter K.sub.P of the P element 62 also only shows a slight slope and a substantially constant value of approximately K.sub.P in the range from 0% to approximately 20% and from 100% to 120% of the positive sequence voltage. In a transition area between 20% and 100% of the positive sequence voltage, the characteristic curve progressions are selected such that they have an inflection point. The rise of the characteristic curves in the inflection point characterizes the behavior of the P element in the transition area. The progression of the curves is smooth and monotone. Depending on the network properties, the ranges with different rises can deviate in their width and in their course.

(21) FIG. 5 shows the temporal behavior of the PLL 58 depending on the positive sequence voltage broken down to the PLL elements. The case in which the positive sequence voltage U.sub.LV+ falls from a value with 100% of the nominal voltage U.sub.N for a period of time determined by the network error simulated here to a value of 10%. The positive sequence voltage of 10% then jumps back to the nominal value with 100%.

(22) The proportional element 62 accesses the error variable from .sub.Line and .sub.PLL. If the PLL is in a steady state before a drop in the positive sequence voltage, then the output variable of the P element 62 is zero. In the event of a voltage drop to 10%, the parameter K.sub.P is set to zero or a very small value. The output of the P element 62 P.sub.OUT thereby shows the progression shown as an example in FIG. 5, where a reduction in the P value with a slight overshoot results before the output variable 0 is output again. A similar temporal behavior shows the P element 62 during the return of the positive sequence voltage and its jump from 10% to 100%.

(23) The output variable of the PT.sub.1 element 64 PT.sub.1,OUT also shows a very similar progression where the voltage changes also result in slight overshoots.

(24) The output variable of the integrator 74 I.sub.OUT is of particular interest here. It shows that the phase angle continues linearly with time during the voltage drop. Only the two phase jumps lead to a change in the phase in the subsequent cycle.

(25) The above exemplary embodiment is based exclusively on an assessment of the voltage in the positive sequence system. An examination of the negative sequence system would also generally be possible. An array of curves would be saved in the controller for the negative sequence voltage, which determines a curve depending on a further variable, which takes into consideration the state of the power network. Thus, for example, a network impedance and/or the inductive or capacitive portion of the fed-in current or power could be taken into consideration. A rate of change of the peak or effective value of the network voltage could also be applied.