Controlling timing of transformer tap changes

Abstract

A method of generating a timing control signal (80) for controlling timing of transformer tap changes of a transformer (20) is provided. The transformer (20) comprises an on-load tap changer (30), the on-load tap changer (30) comprising a plurality of respective transformer taps (25-27). The method comprises monitoring a voltage of a primary side and / or a secondary side of the transformer (20) to obtain a monitoring signal (91), and deriving an intermediate signal (95) from the monitoring signal (91), the intermediate signal (95) being indicative of at least one zero crossing of a tap voltage at a transformer tap (25-27). Deriving the intermediate signal (95) includes estimating a phase of the monitored voltage based on the monitoring signal (91) to obtain a phase signal (94), and applying a phase correction to the phase signal for a phase shift between the monitoring signal (91) or a signal derived therefrom and the tap voltage at the transformer taps (25-27). The timing control signal (80) is derived from the intermediate signal (95).

Description

Technical Field

[0001] The present invention relates to a method of generating a timing control signal for controlling the timing of a transformer tap change of a transformer, in particular a wind turbine transformer. The invention also relates to a corresponding control system for controlling an on-load tap changer of such a transformer, and a computer program for generating such a timing control signal. Background Art

[0002] Wind turbines are typically connected to a grid, which may be a local wind farm grid or a utility grid, through corresponding wind turbine transformers. The power generation system of a wind turbine may include a generator and a power converter for a variable speed wind turbine, which may be arranged in a full converter topology or a DFIG topology. The power converter may experience voltage variations on the grid transmitted through the transformer. As a result, the wind turbine electrical power system may experience undesirable overvoltage or undervoltage conditions.

[0003] To solve this problem, document WO 2020 / 239425A1 describes the use of an electronic on-load tap changer, for example, on the primary winding of a wind turbine transformer. By changing the tap setting, voltage variations on the grid can be compensated. Figure 1 An exemplary configuration is shown in FIG, wherein a wind turbine transformer 20 may include a primary winding 21 and a secondary winding 22 provided with a tap changer 30. A switching element 31, such as a switching valve, may be connected to each tap and may be controlled for switching, for example, between the taps 25, 26 and 27. In view of the high short-circuit currents that may occur when two taps are connected via the switching element 31, a switching impedance 35, such as an inductor and / or a resistor, may be provided to reduce the short-circuit current peaks during the tap change. However, this solution is not optimal, for example due to the volume and cost of the required impedance. Another possibility is to overdesign the thyristors, install snubber circuits, etc. However, they also have similar disadvantages and increase the cost. Due to the mechanical and thermal effects of such currents, the short-circuit current peaks further affect the life of the transformer and the thyristors.

[0004] The document “Solid-state on-load transformer tap changer”, IEE Proc.: Electric Power Applications, Vol. 143, No. 6, Nov. 1996” describes the switching logic of a single-phase transformer tap changer using thyristor switching. However, there is no description of how the switching times of the thyristor switches can be derived.

[0005] Document US 10,727,755 B2 describes a forced commutation switching sequence for switching between transformer taps. However, the document does not mention how the timing of the switching can be derived.

[0006] Document US Pat. No. 3,551,789 A describes a static tap changer device in which, during a tap change, the closing of the switch associated with the new tap is delayed, thereby avoiding that both tap connections are conducting at the same time, which would result in current discontinuity on the load side.

[0007] The document ““New Controller for an Electronic Tap Changer…”Faiz J.etal.,IEEE Transactions on Power Delivery,Vol.22,No.1,Jan.2007” describes a controller design for an electronic tap changer, wherein in order to find a corresponding controller design, the following assumptions are made: only the resistive impedances in the system and the instantaneous knowledge of the amplitudes and phase angles of the waveforms of all quantities are included.

[0008] The document “Analysis, Design, and Implementation of a Fast On-Load Tap Changing Regulator”, Echavarria et al., IEEE Transactions on Power Electronics, Vol. 22, No. 2, March 2007, describes a fast on-load tap changing regulator that uses a switching inductor to limit the short-circuit current during tap changing.

[0009] It is desirable to accurately switch the switching elements of a tap changer in such a way that the short circuit current can be kept low during a tap change while not requiring a large switching impedance. Summary of the invention

[0010] There is therefore a need to alleviate at least some of the above disadvantages and to improve the timing of such on-load tap changer switching, for example for wind turbine transformers.

[0011] This need is met by the features of the independent claim.The dependent claims describe embodiments and / or examples of the invention.

[0012] According to a first aspect, a method for generating a timing control signal for controlling the timing of a transformer tap change of a transformer is provided. The transformer comprises an on-load tap changer comprising a plurality of corresponding transformer taps. The method comprises: monitoring a voltage on a primary side and / or a secondary side of the transformer to obtain a monitoring signal; deriving an intermediate signal from the monitoring signal, the intermediate signal indicating at least one zero crossing of the tap voltage at the transformer tap, wherein deriving the intermediate signal comprises estimating a phase of the monitored voltage based on the monitoring signal to obtain a phase signal, and applying a phase correction to the phase signal for a phase shift between the monitoring signal or a signal derived therefrom and the tap voltage at the transformer tap; and deriving the timing control signal from the intermediate signal.

[0013] From the monitoring signal, it can be determined when the monitored voltage has zero crossings - however, these zero crossings may be phase shifted relative to the zero crossings of the tap voltage at the transformer taps, for example, due to phase shifts between the position of the monitored voltage and the position of the transformer tap, phase shifts introduced by the transformer configuration (particularly the type and configuration of winding connections, such as delta-delta, delta-star, etc., and the phase shifts introduced respectively), and / or phase shifts (e.g., delays) introduced during the acquisition and processing of the monitoring signal. By applying a phase correction, the accuracy of the estimate of the zero crossings of the actual tap voltage can be improved. By deriving the timing control signal from this improved estimate, the timing control signal can more accurately reflect the desired timing of transformer tap switching, such as the timing of closing of a switching element coupled to the transformer tap to be subjected to a tap change. As a result, short-circuit currents that may occur during a tap change can be more effectively limited.

[0014] A phase correction may be applied, for example, by adding a (positive or negative) phase value to the phase signal.Thus, phase changes and / or delays of the monitored voltage may be taken into account in a simple and efficient manner.

[0015] The timing control signal may, for example, indicate the moment at which a zero crossing of the tap voltage occurs, or it may indicate the moment at which a switching of a switching element coupled to the transformer tap is to occur. For example, it may define a corresponding switching point, for example in the form of a peak or spike in the signal. For example, the timing control signal may be derived so as to include at least one switching point that is located within a predetermined distance to at least one zero crossing of the tap voltage, in particular before at least one zero crossing of the tap voltage. Such a distance may be a predetermined amount of time corresponding to a predetermined phase offset, for example less than 4 ms or 3 ms before a zero crossing of the tap voltage.

[0016] The phase correction may be applied concurrently with a corresponding phase offset.The intermediate signal may on the one hand indicate a zero crossing of the tap voltage and on the other hand already comprise a phase offset which shifts the switching point of the tap change to the desired timing.

[0017] The signal derived from the monitoring signal may be any signal indicating a zero crossing of the tap voltage without phase correction. The derived signal may for example be a zero crossing signal indicating a zero crossing of the monitored voltage, a phase signal or another signal derived from the monitoring signal before phase correction.

[0018] Phase shift can correspond to delay, because a certain phase difference under a certain frequency (for example, the frequency of the monitored voltage or tap voltage which is usually equal) corresponds to a certain amount of time. Therefore, the delay introduced by processing the monitoring signal can be compensated by the corresponding phase correction. The voltage being monitored can be referred to as "monitored voltage" in this article. The tap voltage can be the voltage at the connected (old) transformer tap. The tap voltage can be in particular the tap winding voltage, such as the winding voltage at the tap position of the corresponding winding. The switching point can define the time point at which the switching element connected to the new transformer tap (the tap change will occur on the new transformer tap) will be switched (closed). "Old tap" can refer to the tap that will be disconnected during the tap change, wherein "new tap" can refer to the tap that will be connected to the circuit during the tap change. In particular, by the tap change, the current flowing through the corresponding transformer winding can be changed from flowing through the old tap to flowing through the new tap.

[0019] In an example, a transformer tap may be provided at least on a primary side of a transformer, and a voltage may be monitored on a secondary side of the transformer, wherein applying a phase correction may include estimating a transformer phase shift between a voltage on the primary side of the transformer (primary voltage) and a voltage on the secondary side of the transformer (secondary voltage), and providing a phase correction for the transformer phase shift. Thus, even if the voltage is monitored on a side of the transformer that is different from the side on which the transformer tap is provided, accurate timing of the tap change may be achieved. This may be particularly advantageous for wind turbine transformers, wherein monitoring of the voltage may occur on a transformer secondary side, which may be coupled to a wind turbine power generation system and may be at a lower voltage level (e.g., at a low voltage (LV)), while a tap changer may be provided on a primary side, which may be coupled to a grid and may be at a higher voltage level (e.g., a medium voltage (MV) or a high voltage (HV)). The corresponding transformer phase shift may have different causes, including phase angle shift due to the transformer type (the configuration of the transformer winding connection, such as triangle-delta or triangle-star, i.e. the vector group of the transformer), and phase angle shift due to the leakage inductance of the transformer, as further explained in more detail below.

[0020] In an example, estimating the transformer phase shift may include obtaining information indicating a phase difference between a monitored voltage and a current at a secondary side of the transformer, and deriving the transformer phase shift from a transformer model using the monitored voltage and the obtained information as input. Thereby, a phase shift due to a leakage inductance of the transformer may be estimated. Estimating the transformer phase shift may additionally or alternatively include determining a phase shift due to a winding connection type, in particular a phase shift due to a vector group of the transformer. In some examples, such a phase shift due to the vector group may be considered in a model of the transformer together with a phase shift due to the leakage inductance. Thus, an accurate estimate of the transformer phase shift may be obtained in a simple manner.

[0021] The information obtained may, for example, include information about reactive and active currents of the transformer provided to the secondary side, which may, for example, include active and reactive power set points for the operation of a power generation system (e.g., a wind turbine) that provides electrical power to the secondary side of the transformer. Such active and reactive power set points may, for example, be known to a controller (e.g., a converter controller) that controls a power converter that may form part of the power generation system. From these set points, the phase relationship between the monitored voltage and the corresponding current provided to the secondary side of the transformer may be derived (the phase shift between the voltage and the current determines the power factor, which may be derived from the ratio of the active power to the apparent power).

[0022] The model of the transformer can be an equivalent circuit of the transformer, which takes into account at least the resistance and reactance of the primary side of the transformer and the resistance and reactance of the secondary side of the transformer. When the current through the transformer is known, the phase difference between the voltage on the secondary side and the voltage on the primary side (i.e., the transformer phase shift) can be estimated from such a model. For example, the current of the series branch of the primary and secondary windings of the transformer can be used to estimate the (composite) voltage drop in the vector representation, and the voltage on the primary side of the transformer can be obtained from the voltage on the secondary side of the transformer (e.g., the monitored voltage) and the (composite) voltage drop. Knowing the voltage vectors on the primary and secondary sides, the phase difference (i.e., the phase angle) between these voltages can be determined, which can correspond to the transformer phase shift. The transformer phase shift can also include the phase shift caused by the vector group (winding connection type) of the transformer. This can be considered separately from the phase shift due to leakage inductance, or it can be considered together with it, for example, in the transformer model.

[0023] The transformer phase shift may be applied as a dynamic phase correction, since it may depend on the operating point of the transformer, in particular on the phase and amplitude of the current and voltage provided at the secondary side of the transformer. For example, it may depend on the above-mentioned active and reactive power set points at which the power generation system provides electrical power to the secondary side of the transformer.

[0024] Applying the phase correction may comprise applying a fixed phase correction, wherein the phase signal may be corrected with a fixed value, and / or applying a dynamic phase correction, wherein the phase signal may be corrected with a variable value that is allowed to be changed during operation. The dynamic phase correction may in particular be allowed to be changed during execution of the method, for example during operation of the transformer. As mentioned above, some phase shifts and / or delays may be dynamically dependent on current operating conditions of the transformer and / or the system of which the transformer forms a part.

[0025] Applying phase correction can include at least partially compensating for a phase shift or delay introduced into the tap voltage compared to the phase of the monitored voltage (e.g., a phase shift caused by propagation of the voltage through the electrical system), and / or at least partially compensating for a phase shift or delay introduced during acquisition and / or processing of the monitoring signal.

[0026] For example, applying the phase correction may include one or a combination of the following: applying correction for delays introduced by acquiring the monitoring signal through digital data acquisition (e.g., through anti-aliasing filtering performed by a data acquisition board that samples the measured voltage value of the monitored voltage); applying correction for delays introduced by code execution of a processing unit that derives a timing control signal (e.g., the cycle processing time of a controller (e.g., CCU)); applying correction for delays introduced by filtering the monitored voltage or the monitoring signal (in particular, by a filter configured to remove signal distortion (e.g., harmonic distortion)); applying correction for delays introduced by electrical connections between the location of the monitored voltage and the location of a transformer (e.g., , by a cable connection between the monitoring location and the transformer); applying corrections to phase shifts introduced by circuits in the electrical path between the location of the monitored voltage and the location of the transformer (e.g. by a filtering unit connected between the power converter of the monitored voltage and the transformer, such as a harmonic filter); applying corrections to phase shifts introduced by the connection configuration of the primary and secondary windings of the transformer, in particular for the vector group of the transformer (e.g. for primary-secondary connections such as delta-star, delta-delta, star-star, etc.); or applying dynamic phase corrections to phase shifts introduced by the transformer (e.g. corrections for transformer phase shifts due to leakage inductance). The first mentioned correction may be a fixed correction, while the last correction may be a dynamic correction.

[0027] The time delay introduced into the monitoring signal may be taken into account, for example, by multiplying this delay with the detection frequency of the monitored voltage and adding the resulting phase correction to the phase signal.

[0028] In an example, the transformer may be a three-phase transformer comprising transformer windings for three electrical phases. The on-load tap changer may include a transformer tap for each of the three electrical phases of the transformer, for example at least on the primary side. A phase signal may be estimated separately for each of the three electrical phases, and / or a phase correction may be applied separately for each of the three electrical phases, and / or a timing control signal may be derived separately for each of the three electrical phases. Thus, an accurate timing of the tap change may be obtained for each of the three phases. The tap change may in particular include a change of the transformer tap for each of the three transformer windings corresponding to the three electrical phases. However, since the voltages of the three electrical phases are phase-shifted relative to each other, the timing control signal may be different for each electrical phase. The monitoring signal may include a signal for each of the three electrical phases, i.e., it may be a three-phase signal, which may form the basis for deriving a separate timing control signal for each phase.

[0029] In an example, estimating the phase of the monitored voltage may include detecting at least one zero crossing of the monitoring signal using a zero crossing detector. A zero crossing signal may thereby be obtained. For a three-phase system, a zero crossing detector may be provided for each electrical phase of the transformer, and at least one zero crossing of the monitoring signal may be detected separately for each phase (the zero crossing signal may thus be a three-phase signal).

[0030] Detecting at least one zero crossing of the monitoring signal by the zero crossing detector may include using at least a first voltage threshold value above the zero voltage value and a second voltage threshold value below the zero voltage value, and determining the presence of a zero crossing if the monitoring signal has a value between the first voltage threshold value and the second voltage threshold value. In particular, for sampled digital signals, the voltage value may never be exactly zero, so using such two thresholds arranged near the zero value can allow for fast and efficient detection of zero crossings. The first and second voltage thresholds may be selected, for example, based on the sampling rate at which the monitored voltage is sampled to obtain the monitoring signal (i.e., the sampling rate of the monitoring signal) and / or the frequency of the monitored voltage (e.g., 50 Hz or 60 Hz). For example, compared to the voltage range of the monitored voltage, the first and second voltage thresholds may be located near the zero value of the monitored voltage. The first voltage threshold may, for example, be located at a positive amplitude of at most 20%, 10%, or 5% above zero volts, and the second voltage threshold may, for example, be located at a negative amplitude of at most 20%, 10%, or 5% below zero volts.

[0031] Monitoring the voltage may include obtaining a measurement of the monitored voltage. For example, a measurement made by a converter controller of a power generation system (e.g., a wind turbine) that provides electrical power to a transformer may be received. The obtained measurement may be filtered in the analog and / or digital domain to, for example, remove harmonic distortion, such as by applying a low pass filter.

[0032] Estimating the phase of the monitored voltage may comprise detecting the frequency of the monitored voltage using a phase locked loop (PLL), in particular by providing the monitoring signal as input to such a PLL. Thus, the frequency of the monitored voltage and therefore the frequency of the tap voltage may be accurately determined. The frequency may be derived, for example, as an angular frequency. For a three-phase transformer, the frequency may be derived from the three-phase monitoring signal corresponding to the three electrical phases of the monitored voltage. For example, a three-phase PLL may be used to derive the frequency, in particular from the positive sequence component of the three-phase monitoring signal (e.g., a positive sequence PLL may be employed).

[0033] In an example, the phase signal may be generated by an integrator that uses as inputs a detected zero crossing of the monitoring signal (eg, the zero crossing signal) and a detected frequency of the monitored voltage and gives a phase signal.

[0034] Obtaining the phase signal may, for example, include integrating the detected frequency, wherein the integration may start at a detected zero crossing of the monitoring signal. The integration may be reset at each or every second detected zero crossing of the monitoring signal. For example, the phase signal may be alternately reset to a zero value and a π value at each detected zero crossing, or may be reset to a zero value at each second detected zero crossing. In this way, the phase signal may increase linearly from a zero phase value to a 2π phase value, after which the phase signal may be reset to a zero value again. The value range of the phase signal may be limited between 0 and 2π. Thus, the integration may correspond to a counter that reaches a 2π value after each period of the detected frequency.

[0035] In a three-phase system, three separate phase signals may be derived from three separate zero-crossing signals and from a common estimate of the frequency of the monitored signal using a three-phase PLL that gives a single frequency value.

[0036] By applying a phase correction to the phase signal, a corrected phase signal may be obtained which indicates at least one zero crossing of the tap voltage. In the corrected phase signal, the zero crossings of the tap voltage may, for example, be located at phase values ​​of 0 / 2π and π. If a predetermined (and therefore known) phase offset is further applied (e.g., shifting the switching point to a time before the zero crossing), the zero crossing may be located at a correspondingly shifted phase value. The intermediate signal may correspond to the corrected phase signal or to an offset-corrected phase signal. The units of the phase signal may be different from radians; the phase signal may be provided in degrees or any other desired units.

[0037] In an example, deriving the timing control signal from the intermediate signal may include detecting an occurrence of at least one predetermined phase value in the intermediate signal, and generating the timing control signal based on the detection of the at least one predetermined phase value by generating a signal change in the timing control signal when the predetermined phase value is detected.

[0038] The predetermined phase value may correspond directly to a zero crossing of the tap voltage, or may be shifted by a phase offset from the phase value of the zero crossing, for example if such a phase offset is applied to the corrected phase signal prior to derivation, or if such an offset is applied directly to the predetermined phase value used in the detection in order to obtain the desired timing of the switching point.

[0039] In the timing control signal, a signal change, i.e. a change in the signal value, may correspond to a pulse or spike at a point in time when a predetermined phase value is detected. The signal change may indicate a zero crossing or a switching point of the tap voltage. The switching point may define a point in time at which a switching element of the tap changer is switched to effect a tap change, e.g. by triggering a thyristor of such a switching element. Thus, a switching point may be a trigger pulse of a semiconductor switch of such a switching element.

[0040] The detection of the predetermined phase value may be performed by a detector (a so-called angle zero crossing detector) which may detect the zero crossing of the tap voltage by detecting at least one predetermined phase value (in particular a phase angle) in the intermediate signal. The at least one predetermined phase value may correspond, for example, to 0, π and / or 2π (if no phase offset is applied).

[0041] Detecting the occurrence of a predetermined phase value may include providing a first phase threshold value above the corresponding predetermined phase value, providing a second phase threshold value below the corresponding predetermined phase value, and detecting the occurrence of the predetermined phase value if the intermediate signal has a value between the first and second phase threshold values. Likewise, the phase threshold value may be selected in a manner similar to the voltage threshold value described above, depending on the sampling rate and / or frequency of the intermediate signal. The first and second phase threshold values ​​may, for example, be located below 10%, 5% or 2% of the range of the phase signal above or below the predetermined phase value, respectively.

[0042] In some examples, the predetermined phase value may be π, and the first and second phase thresholds may be located above and below the value of π, respectively. Another zero crossing of the tap voltage may occur at a phase value of 0 or 2π (if no phase offset is applied). As described above, the phase signal may be reset after each 2π period, for example, after each second detected zero crossing of the monitored voltage. Therefore, there may be a discontinuity at the location where the phase signal transitions from the value 2π to the value 0.

[0043] In order to determine the switching point of such a zero crossing, after detecting the occurrence of a predetermined phase value of π, the value of the intermediate signal can be shifted by adding a phase value of π to the phase signal and keeping the signal value of the intermediate signal within the range between 0 and 2π, for example by wrapping around phase values ​​exceeding 2π. For example, if the phase value exceeds 2π, the value can be brought back into the range by subtracting 2π, for example, 1.5π can be shifted to 2.5π, 2.5π is wrapped around to 0.5π, and thus back into the range. Such a shifting of the phase signal may result in the next zero crossing of the tap voltage corresponding again to a phase value of π of the intermediate signal. Since the same predetermined phase value (e.g., π) can be repeatedly detected by using the same phase threshold, the detection of the predetermined phase value can be facilitated.

[0044] In another example, a plurality of different predetermined phase values ​​may be detected, such as a phase value of 0 / 2π, wherein the third phase threshold may be located below the 2π value and the fourth phase threshold may be located above the zero value. The presence of such a predetermined phase value may be detected if the phase signal has a value above the third phase threshold or below the fourth phase threshold.

[0045] As already mentioned, the timing control signal can be derived so as to include at least one switching point, for example a signal change such as a pulse, which is shifted from at least one zero crossing time of the tap voltage (in particular the tap winding voltage, for example the winding voltage of the corresponding transformer winding at the tap position) by a predetermined amount of time, in particular the time before the zero crossing. Thereby, the switching timing of the switching element of the tap changer can be set to a desired time point before the zero crossing of the tap voltage. The time shift can correspond to less than 40%, 30% or 20% of the tap voltage period. For example, for a frequency of 50 Hz or 60 Hz of the tap voltage, it can correspond to a shift of less than 3 ms to a position before the zero crossing of the tap voltage. The time shift can, for example, occur at a position about 2 ms before the zero crossing of the tap voltage. By such a time shift, it can be ensured that the amplitude and duration of the short-circuit current through the transformer tap that may occur during the tap change can be kept low. Note that for a tap voltage of 50 Hz, the period is 20 ms, so that 2 ms corresponds to 10% of the period.

[0046] The method may, for example, include applying a predetermined phase offset to the phase signal before deriving the timing control signal, for example by adding the phase offset to the phase signal (for a three-phase transformer, to the phase signal of each electrical phase). This may be performed before, during or after applying the phase correction, the phase offset may, for example, be applied as part of the phase correction, or may be applied to the corrected phase signal. Such a phase offset may correspond to a desired time shift by which the switching point will be shifted from the zero crossing of the tap voltage. The intermediate signal may correspond to such an offset-corrected phase signal, which may still indicate the zero crossings of the tap voltage; however, these zero crossings no longer correspond to phase values ​​of 0 / 2pi or pi, but to offset phase values, respectively. Such a phase offset may have the advantage that the predetermined phase value and therefore the same phase threshold may be used to detect the time at which the switching point is generated in the timing control signal. By adding such a phase offset, the angle zero crossing detector may, for example, detect the predetermined phase value earlier than the actual occurrence of the zero crossing of the tap voltage, thereby shifting the switching point in the timing control signal to an earlier time.

[0047] Additionally or alternatively, such a time shift may be achieved by shifting a predetermined phase value detected in the intermediate signal, for example by shifting the first and second phase thresholds. By shifting the predetermined phase value to a lower value, the occurrence of such a shifted predetermined phase value will be detected earlier, and by shifting the predetermined phase value to a higher value, it will be detected later. The position of a switching point in the timing control signal derived from the detection of one or more predetermined phase values ​​may thus be advanced or delayed as required.

[0048] The transformer may be configured to provide a transformation between a second voltage on the secondary side of the transformer and a first voltage on the primary side of the transformer. The first voltage may be higher than the second voltage. The on-load tap changer may be provided on at least one of the primary side and the secondary side of the transformer, preferably at least on the primary side. The on-load tap changer may be an electronic on-load tap changer. It may comprise a plurality of switching elements, each switching element being coupled to a transformer tap of a winding of the transformer, wherein each switching element may comprise at least one, preferably a pair of semiconductor switches, such as a pair of thyristors.

[0049] The transformer may be a wind turbine transformer. The secondary side of the transformer may be coupled to a power generation system of the wind turbine. The primary side of the transformer may be coupled to a power grid. The transformer may be configured to provide a transformation of the electric power generated by the power generation system. The transformed electric power may be fed into the power grid. The transformer may, for example, provide a transformation of a low voltage (LV) to a medium voltage (MV) or a high voltage (HV); however, the solutions disclosed herein are not limited to this type of transformer.

[0050] According to a further aspect, a method of controlling an on-load tap changer of a transformer to perform a tap change from a first transformer tap to a second transformer tap is provided. The method may include: determining a timing control signal according to any method disclosed herein; detecting a triggering event that triggers the tap change; upon detecting the triggering event, deriving a drive signal from the timing control signal for switching (e.g., closing) a switching element coupled to the second transformer tap; and applying the drive signal to the switching element of the second transformer tap to perform the tap change.

[0051] The drive signal may in particular cause switching of the switching element at a point in time indicated by a switching point included in the timing control signal. Thus, switching may occur very precisely at a predetermined point in time before the tap voltage crosses zero. Thus, the duration and amplitude of the short-circuit current during the tap change may be kept low. The switching element may be a switching valve; for example, it may comprise two semiconductor switches, such as two thyristors connected in anti-parallel. The drive signal may close the switching element connected to the second tap to connect the second tap to the circuit on the corresponding transformer side, such as to connect the second tap of the transformer winding to the corresponding transformer terminal. The timing control signal may, for example, be used to trigger a driver, such as a thyristor driver, to generate a drive signal to close the corresponding switch of the switching element.

[0052] The method may further comprise stopping providing a drive signal to a switching element connected to the first transformer tap, for example to disconnect such a switching element. This may for example occur after expiry of an overlap period in which both switching elements are closed. It may occur sufficiently early before a zero crossing of the tap voltage to allow the switch (e.g. thyristor) of the switching element of the first tap to recover, i.e. to reach a stable blocking state. Thereby a tap change without interruption of the load current may be achieved.

[0053] During this overlap period, a winding section of the transformer winding between the first and second transformer taps may be short-circuited. The current induced in the winding section may lead to a corresponding short-circuit current. This current is phase-shifted with respect to the tap voltage and may start to rise until a zero crossing of the tap voltage occurs and may then fall again. Therefore, switching shortly before the voltage zero crossing may limit the short-circuit current.

[0054] According to another aspect of the present invention, a control system configured to control a change of a transformer tap of a transformer is provided. The transformer comprises an on-load tap changer, which comprises a plurality of corresponding transformer taps. The control system may be configured to perform any of the methods described herein; according to aspects and examples of the above methods, it may be particularly configured to determine a timing control signal for controlling the timing of the transformer tap change. The control system may comprise a processing unit and a memory, wherein the memory may store control instructions which, when executed by the processing unit, cause the control system to perform any of the disclosed methods.

[0055] According to another aspect, a wind turbine transformer may be provided, comprising an on-load tap changer and further comprising such a control system. According to a further aspect, a wind turbine may be provided, comprising a power generation system and such a wind turbine transformer coupled to the power generation system. The transformer may receive electrical power from the power generation system and may provide transformed electrical power to a power grid. The control system may be coupled to a converter controller of a power converter of the power generation system to monitor voltage, or such a converter controller may form part of the control system. The converter controller may be, for example, a converter control unit (CCU) of the power converter.

[0056] According to another aspect of the invention, there is provided a computer program for determining a timing control signal for controlling the timing of a transformer tap change of a transformer. The computer program comprises control instructions which, when executed by a processing unit of a control system for controlling an on-load tap changer of a transformer, cause the processing unit to perform any of the methods described herein. The computer program, in particular its control instructions, may be provided on a volatile or non-volatile storage medium or data carrier.

[0057] It should be clear that the method can be performed with a control system in any configuration described herein. In addition, the control system can be configured to perform any method disclosed herein.

[0058] It should be understood that the above-mentioned features and those features to be explained below can be used not only in the indicated combinations, but also in other combinations or alone, without departing from the scope of the present invention. In particular, the features of different examples and embodiments of the present invention can be combined with each other, unless otherwise indicated. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The foregoing and other features and advantages of the present invention will become further apparent from the following detailed description read in conjunction with the accompanying drawings, in which like reference numerals refer to like elements.

[0060] Figure 1is a schematic diagram showing a tap changer that employs switching impedance to limit short circuit current during tap changing.

[0061] Figure 2 is a schematic diagram showing a wind turbine according to an example of the present invention, the wind turbine including a power generation system, a transformer with an on-load tap changer, and a control system.

[0062] Figure 3 is schematically shown in Figure 2 Schematic diagram of the current flow that may occur during a tap change in a system.

[0063] Figure 4 is shown in more detail Figure 3 Schematic diagram of the section.

[0064] Figure 5 is a schematic diagram showing a control system according to an example of the present invention.

[0065] Figure 6 is a diagram showing voltage waveforms of a tap voltage and a short-circuit current occurring during a tap change, and switching points at which semiconductor switches of a tap changer can be switched.

[0066] Figure 7 is a diagram showing voltage waveforms and further short circuit currents that may occur during a tap change when switching at different times.

[0067] Figure 8 is a schematic diagram showing the generation of a timing control signal in a control system according to an example of the present invention.

[0068] Figure 9-12 is a schematic diagram showing different signals during the generation of a timing control signal according to an example of the present invention.

[0069] Fig.13 is a schematic diagram illustrating phase correction according to an example of the present invention.

[0070] Fig.14 is a schematic diagram showing a transformer model for deriving variable phase correction according to an example of the present invention.

[0071] Fig.15 It is shown Fig.14 Schematic diagram of the voltage and current vectors of the transformer model.

[0072] Fig.16 is a flow chart illustrating a method of generating a timing control signal according to an example of the present invention. DETAILED DESCRIPTION

[0073] Hereinafter, embodiments and / or examples of the present invention will be described in detail with reference to the accompanying drawings. It should be understood that the following description of the embodiments is provided only for the purpose of illustration and is not intended to limit the meaning. It should be noted that the accompanying drawings are only considered as schematic representations, and the elements in the accompanying drawings are not necessarily proportional to each other. On the contrary, the representations of various elements are selected so that their functions and general purposes become clear to those skilled in the art. As used herein, the singular forms "one", "an" and "the" are intended to also include plural forms, unless the context clearly indicates otherwise. Unless otherwise stated, the terms "include", "have", "include" and "contain" should be interpreted as open terms (i.e., meaning "including, but not limited to,").

[0074] In the following, for illustrative purposes, an exemplary implementation of the disclosed solution in a wind turbine electrical power system is described; however, it should be clear that the disclosed solution may also be implemented in other environments (e.g., distribution grids, substation transformers, photovoltaic applications, railway applications, etc.) where fast and efficient tap changes are beneficial. Furthermore, examples are provided for switches in the form of semiconductor switches. However, the teachings disclosed herein may also be applied to other types of switches, such as electromechanical switches.

[0075] Figure 2 Schematically shown is a wind turbine 100 comprising a wind turbine rotor having a hub 110 and rotor blades 111. The power generation system 50 of the wind turbine may include a power converter 40 and a generator 45 and may be coupled to a wind turbine transformer 20. In the present example, the generator 45 is a doubly fed induction generator (DFIG), the stator of which is coupled to a grid 200 via the wind turbine transformer 20, while the rotor is connected to the power converter 40. In other examples, a full converter solution may be provided, wherein a generator, such as a permanent magnet synchronous generator, may be connected to the grid 200 via a power converter, through which substantially all generated power flows. The generator 45 may be coupled to the rotor hub 110 using a shaft 115 and possibly an intermediate gearbox (not shown). The wind turbine rotor may be provided with a blade pitch drive 112 controlled by a wind turbine controller 105.

[0076] A control system 10 is also provided, which may include a converter control unit (CCU) that controls the power converter 40. In other examples, the CCU may be provided separately. The wind turbine transformer 20 may include a primary winding 21, which is a higher voltage winding connected to a power grid 200, which may be a medium voltage (MV) or high voltage (HV) grid. The transformer 20 may also include a secondary winding 22, which may be a low voltage winding that receives low voltage (LV) or medium voltage (MV) electrical power from the generator 45 and / or the power converter 40. LV generally refers to a voltage range of up to 1.000V. The transformer 20 may also include an electronic on-load tap changer 30, which is connected to the power grid 200 at a higher voltage. Figure 2 In the example of , the electronic on-load tap changer 30 is provided on the primary winding 21, but in other implementations it may also be provided on the secondary winding 22. The transformer 20 may include multiple secondary windings (e.g., one connected to the generator 45 and another connected to the converter 40). The power system 50 may be a three-phase power system including a conductor for each phase, and the grid 200 may operate at a voltage frequency of 50 Hz or 60 Hz.

[0077] The control system 10 may measure transformer voltages on the LV side and / or MV / HV side of the transformer 20 to estimate the occurrence of voltage zero crossings at the taps of the tap changer and may control operation of the tap changer 30 accordingly, as explained in more detail further below.

[0078] Figure 3 A transformer winding 21 including exemplary taps 25, 26, 27 is shown. The tap changer 30 may include a switching element 31 (referred to herein as a switch) connected to each tap. Each switching element may include or consist of two semiconductor switches 32 ( Figure 4 The semiconductor switch 32 is preferably a thyristor (electrically triggered or optically triggered) connected in anti-parallel. For each tap, a certain fraction nt turns can be connected into or taken out of the current path through the transformer winding 21. As is well known, the transformation ratio of the transformer 20 depends on the turns ratio (N) of the primary and secondary windings 21, 22. P / N S ), so that by changing the tap (and thus the number of turns of the winding 21), the transformation ratio (V P / V S ) and therefore the voltage V S Can be adjusted. Figure 3 An exemplary percentage of voltage increase or voltage decrease when connecting various taps is shown. The thick dashed line indicates the primary current I occurring during the tap change between taps 25 and 26. P , this is Figure 4is shown in more detail in .

[0079] Figure 4 Shows Figure 3 , showing the situation where the switching element 31 connected to the taps 25 and 26 (i.e. switches S3 and S2) is closed during a tap change between these taps (e.g. from 26 to 25). In order not to interrupt the primary current, both taps are closed during the overlap period, which is Figure 4 The primary current I through the primary winding P The short-circuit current I flows mainly through the tap 26 and the connected switch S2, since this tap corresponds to the smaller number of turns. However, since both switches S2 and S3 are closed, and since a current is still induced in the winding section between the taps 25 and 26, the short-circuit current I CC There is a short circuit between the taps 25 and 26. Therefore, the tap 26 and the connected switching valve S2 experience I CC +I P , while the tap 25 and the connected switching valve S3 mainly experience I CC .

[0080] Preferably, no switching impedance (such as an additional resistor or inductor) is connected to any of the taps 25-27 to limit this short-circuit current I CC Instead, only the inherent tap resistance Rtap and inductance Ltap of the winding section between the taps are present in the circuit. The voltage induced across taps 25 and 26 is V P *nt,V P is the primary voltage (the voltage on the primary side), and nt is the fraction of the number of turns of the primary winding section between taps 25 and 26 (i.e., nt in this context refers to the percentage of the number of turns of the corresponding winding, i.e., the number of turns between the taps divided by the total number of turns Np of the primary winding). Therefore, since the inherent tap resistance is quite low, I CC can become very large. The instantaneous short-circuit current can be calculated by the equation nt*V P (t) = Rtap*I CC (t)+Ltap*dI CC / dt is calculated.

[0081] As above about Figure 1 As mentioned above, the previous solution uses the conversion impedance connected to the transformer tap to limit the short-circuit current. The present solution can avoid this conversion impedance and reduce the high short-circuit current I by properly controlling the switching moment of the switching element 31. CC .

[0082] Figure 5An example of a control system 10 is shown, wherein the control system 10 may include a dedicated and separate tap changer controller 11. The control system 10 may include a timing control unit 12 that generates a timing control signal according to any of the methods described herein. The unit 12 may obtain a measurement of the voltage on the secondary side of the transformer 20, because such a low voltage may be easier to measure and the controller configuration may therefore be simplified; however, the voltage may alternatively be measured on the primary side of the transformer. It should be clear that the voltage measurement need not be performed directly at the transformer terminals, but may be performed at a different location, such as in the power generation system 50, such as at the power converter 40. The unit 12 may receive a voltage measurement made on the secondary side of the transformer 20, for example, from a converter control unit 15 that controls the power converter 40. The power system is preferably a three-phase system, and the unit 12 may monitor the voltage of each of the three phases to generate a timing control signal 80 for the tap change.

[0083] The control signal generator 13 can use the timing control signal 80 to generate a control signal for switching the semiconductor switches of the switching element 31 to achieve the desired tap change. The control signal is then provided to a driver unit 14 that drives one or more semiconductor switches to perform the tap change. The driver unit 14 can be an electrical driver unit that provides an electrical drive signal (e.g., a pulse or continuous signal) to the semiconductor switches of the tap changer 30, or can be an optical driver unit that provides a drive signal in the form of light to a light-triggered semiconductor switch (e.g., a light-triggered thyristor), for example via an optical fiber. The latter implementation allows for effective separation and electrical isolation of the controller circuit from the MV / HV voltage of the primary winding 21.

[0084] Figure 5 Also shown is a converter controller unit 15, which is configured to control the power converter 40 and in particular to provide drive signals to the respective semiconductor switches of the power converter 40. To this end, it may comprise a control circuit 16 and a PLL 17. The CCU 15 may measure the voltage on the low voltage side and may accordingly determine when a tap change is required, for example when the voltage drops below a respective threshold or rises above a respective threshold. Accordingly, a respective communication connection 18 may be provided between the CCU 15 and the tap changer controller 11, via which the CCU 15 may issue a tap change command. In response, the control signal generator 13, which receives the tap change command, may generate a tap change control signal to achieve the desired tap change (see Figure 6 The switching timing in the tap change control signal may be derived from the timing control signal 80 provided by the unit 12.

[0085] In some examples, the CCU 15 may form part of the control system 10. In other examples, the CCU 15 may not be included in the control system 10. The tap changer controller 11 may be separate and distinct from the CCU 15 and may therefore be easily retrofitted into an existing wind turbine. In other examples, the tap changer controller may be integrated into the CCU 15, which may in turn form part of the control system 10. Such integration may provide fast and efficient communication between the different components of the control system 10.

[0086] exist Figure 4 In the example of , when changing from tap 26 to tap 25, this may for example include first connecting the new tap 25 by providing a corresponding drive signal to the semiconductor 32 of switch S3 (e.g., a trigger pulse or signal provided to the thyristor gate), and then disconnecting the tap 26, for example by stopping providing a corresponding drive signal to the switch S2 connected to the tap 26 (e.g., stopping providing a pulse or signal to the thyristor gate). During the overlap period, the switches S3, S2 connected to the new tap and the old tap, respectively, are closed and may be provided with corresponding drive signals from the drive unit 14, respectively.

[0087] In order to avoid high values ​​of the short-circuit current ICC, the switch connecting the new tap can be closed shortly before the zero crossing of the tap voltage (i.e. the voltage present at the currently connected transformer tap) occurs. The timing of the tap change, in particular the timing of closing the switch connected to the new tap, can be controlled by a timing control signal provided by the unit 12.

[0088] Figure 6 A diagram illustrating the tap voltage 75 and the short-circuit current I CC 76 waveforms, and including separate control Figure 4 81 and 82 of switches S2 and S3. The 50 Hz waveform of the tap voltage has a zero crossing at time t=0.01 s. To change from tap 26 to tap 25, switch S3 connected to the new tap 25 is closed at a predetermined time 71 before the zero crossing (switching point), which in this example is 1 ms (as can be obtained from control signal 81). It can also be seen that control signal 82 still causes a drive signal to be provided to switch S2 connected to the old tap 26, so that during the overlap period 73, both switches are closed and both taps are short-circuited, and the short-circuit current I CC Flow. CC is inductive and can be significantly higher than the transformer operating current. CC This causes the tap voltage to lag by about 90 degrees or π / 2. Therefore, I CCThe voltage reaches its peak value near zero crossing. By closing switch S3 shortly before zero crossing, I CC The amplitude and duration can be kept small.

[0089] After the overlap period (1 ms in this example), the switch S2 connected to the old tap is allowed to open at a second predetermined time 72 close to the zero crossing. As indicated by the control signal 82, this can be done by no longer providing a corresponding drive signal to the switch S2, for example by not providing a trigger (e.g., voltage, current, or light) to the gate of the semiconductor switch of the switch S2. However, the short-circuit current will not be interrupted immediately because the switch S2 remains on until the current through the switch S2 passes zero (e.g., the thyristor remains in the on state). As soon as this happens, the switch will be open (i.e., non-conductive) and the short-circuit current I CC is interrupted so that the old tap is no longer connected.

[0090] When the switching element 31 is implemented differently, for example by a bidirectional thyristor, an electromechanical switch or the connection of two IGBTs, IGCTs or MOSFETS, the situation may be different. For example, the short-circuit current can be interrupted directly by opening the switch. Then the second predetermined time 72 can be selected so that the switch is disconnected at the zero crossing of the short-circuit current flowing through the old tap (i.e. the tap to be disconnected). The switch S2 can be disconnected, for example, at the zero crossing of the curve 76, which occurs at about 0.011 seconds. Then, the overlapping period 73 can last from the first predetermined time 71 before the voltage zero crossing to the second predetermined time 72 at the short-circuit current zero crossing. Therefore, the current that needs to be interrupted when disconnecting the old tap can be significantly reduced.

[0091] Figure 7 The diagram shows the short-circuit current I CC , which may be caused by the different switching times between the taps. The vertical axis shows the short-circuit current I in A. CC and the tap voltage V in V P *nt. This example is for a 50Hz voltage waveform, and the thick solid line indicates that the tap voltage has zero crossings at 0, 0.01s, and 0.02s, which is consistent with Figure 6 The thinner line indicates the short-circuit current I CC , which is caused by closing the switching element connected to the new tap at different times, which is indicated in the legend. It can be seen that if the switch is closed long before the zero crossing at 0.01s (i.e. 10ms), the short-circuit current I CC The maximum short-circuit current is significantly reduced if the switch is closed shortly before the zero crossing at 10 ms (e.g., at a time of 8 ms or 9 ms). CCHow does the peak of I behave for different switching timings of switch 31 before crossing zero. Note that for I switching the switch at 0 ms (rounded off) CC The line refers to switching occurring shortly after the tap voltage crosses zero at t=0 ms, and for I switching the switch at 10 ms (rounded off) CC Lines refer to switching that occurs shortly before the tap voltage crosses zero at t=10 ms.

[0092] In order to achieve the desired reduction in short-circuit current, it is desirable to accurately derive the point in time at which switching occurs, ie, the switching point 71 before the tap voltage crosses zero. Figure 8 Shows that it can be executed Fig.16 The timing control unit 12 of the method shown in the flowchart of FIG. 9 is an exemplary implementation of the method. The monitoring signal 91 can be received by the unit 12, which corresponds to the monitored voltage V abc As described above, the monitoring signal may be monitored at the power converter 40 or at another location in the power system. Since the monitored voltage may be a three-phase voltage, the monitoring signal 91 may be a three-phase signal, as indicated by the reference numerals a, b, and c. The voltage monitoring is performed at Fig.16 The flowchart is indicated at step S10. Fig. 9 An exemplary three-phase monitoring signal 91 is shown in FIG. 1 , wherein voltage signals 91 a , 91 b , and 91 c are used for corresponding electrical phases.

[0093] A three-phase PLL 121 (particularly a positive sequence PLL) may be provided and used to detect the (angular) frequency ω (eg, in radians per second) of the monitored voltage ( Fig.16 11 in step S11). The PLL may be a synchronous reference frame PLL (SRF-PLL). The PLL may be a PLL of the CCU 15 (e.g. PLL 17), for example in an implementation in which the controller 11 is integrated in the CCU 15, or may be an additional PLL. In the present example, the angular frequency is 2π / 20ms=314.2rad / s. A voltage zero crossing detector (ZCD) 122 may also be provided and may be used to detect zero crossings in the monitoring signal 91. This may be done by comparing the instantaneous voltage value with first and second fixed voltage thresholds about zero volts for each electrical phase. These thresholds may be selected taking into account the sampling rate adopted by the unit 12 and the rated grid frequency (e.g. 50 or 60 Hz) so as to reliably detect zero crossings. The monitoring signal preferably has no higher harmonics that may cause the zero crossing detector to malfunction. The monitoring signal may, for example, be filtered to remove the corresponding harmonics. Once the value of the monitoring signal enters the range between the thresholds about zero, a signal value may be given, thereby generating a zero crossing signal 93 ( Fig.16 , step S12 in the process. Fig.10An exemplary zero-crossing signal 93 of the three phases is shown in FIG. It can be seen that the zero-crossing signal 93 includes a corresponding monitored voltage V abc A spike at a location that crosses zero (i.e., where the voltage value is between the thresholds).

[0094] An integrator 123 may also be provided, which is configured to estimate the phase of the monitored voltage by starting from a zero phase value when a zero crossing is detected and integrating the detected frequency ω. A phase signal 94 may thereby be obtained, which indicates the phase θ ( Fig.16 Each second zero crossing in each zero crossing signal 93 may reset the count of the integrator. It may be reset to a value of 0 at each second zero crossing, or may be reset alternately to a value of 0 and π at each zero crossing, so that the value of the phase signal 94 varies between 0 and 2π. Because the reset is based on the detected zero crossing, the phase signal may accurately reflect the phase of the monitored voltage. Again, this is done for each electrical phase, so that the phase signal 94 may include Fig.11 Phase values ​​of 0 / 2π and π of the phase signal may occur simultaneously with the zero crossing of the monitored voltage.

[0095] However, these zero crossings may not correspond to zero crossings of the tap voltage at the transformer taps because phase shifts / delays may exist in the electrical path between the monitoring location and the transformer taps (e.g., the path between the power converter 40 and the tap changer 30), and because delays / phase shifts may be introduced by signal processing. For example, the data acquisition board may include an analog anti-aliasing filter to adapt the input signal to the sampling rate of the electronic equipment, and the monitoring signal may be delayed by such filtering. In addition, digital filtering to remove the above-mentioned high frequency components (harmonics) may introduce delays. Therefore, a phase correction unit 124 may be provided, which applies a phase correction ( Fig.16 Step S14). Thus, a corrected phase signal 95 can be obtained, having a corrected phase θ corr . The phase correction unit 124 may also receive the detected frequency ω to convert any delay given in time units into a corresponding phase shift. The total time delay may be determined and may be multiplied by the angular frequency and the resulting phase shift may be added to the phase value of the phase signal 94. Note that by adding this phase correction, a phase value of π may appear earlier in the corrected phase signal so that the time at which this phase value appears may correspond very accurately to the time at which the actual zero crossing of the tap voltage occurs, even in the presence of a delay. When the phase correction is added, phase values ​​exceeding 2π may wrap around to the phase value range of 0 to 2π. Thus, a phase jump from 2π to zero may effectively appear earlier in the corrected phase signal, such that adding the phase correction corresponds to shifting the phase signal to an earlier time. Fig.11 Corrected phase signals 95a, 95b and 95c are shown for three electrical phases.

[0096] Fig.13 The phase correction is shown in more detail. It may include a phase shift or delay corresponding to a fixed phase shift or delay Δt FIX The fixed phase correction 132 and / or corresponding variable phase shift or delay Δt VAR Dynamic phase correction 131, variable phase shift or delay Δt VAR The fixed phase correction 132 may, for example, include corrections for delays in signal processing of the monitoring signal as described above, and / or may include phase shifts introduced by electrical connections between the monitoring location and the transformer tap (e.g., if the transformer 20 is placed on the ground of the wind turbine) or other circuits connected between the monitoring location and the transformer tap (e.g., a filter placed between the power converter 40 and the transformer 20 when monitoring is performed at the power converter 40 (e.g., by the CCU 15)).

[0097] For example, if monitoring takes place on the secondary side of the transformer and the transformer tap is provided on the primary winding 21 of the transformer (or vice versa), the transformer 20 may for example introduce a dynamic phase shift. Such a phase shift may depend on the active and reactive power set points of the power generation system 50 of the wind turbine. Fig.14 A model 135 of the transformer 20 is shown, in particular an equivalent circuit which can also be referred to as a T model. The model 135 includes a leakage inductance (X σ1 , X σ2 ) and the winding resistance (R1, R2), and the magnetizing inductance X m and the resistance R simulating the iron loss m The presence of the series branch causes a voltage drop in the transformer when current I1 I2 flows. Depending on the active and reactive currents generated by the power generation system 50, this voltage drop may cause a phase shift (referred to herein as transformer phase shift) between the primary and secondary side voltages U1, U2. The monitoring signal (voltage V abc ), active and reactive power set points (P and Q, respectively), and a model 135 of the transformer 20 (e.g. Fig.13 The transformer phase shift is estimated by using X, R in the figure.

[0098] exist Fig.14 In , U2 is the voltage on the secondary side 22 where the monitoring signal is obtained, and U1 is the voltage on the primary side 21 where the transformer tap is provided. For the estimation, the magnetizing branch (parallel branch) can be neglected, so I = I1 = I2. The phase difference between U1 and U2 can be obtained from Fig.15The voltage drop across the series branch corresponds to ΔU = I*(R1+R2)+j(X σ1 +X σ2 )). Based on the voltage drop ΔU and the measured voltage U2, it can be Fig.15 The voltage U1 is derived from the vector diagram (U1 = U2 - ΔU). Therefore, the variable transformer phase shift Δθ var It can be derived from the vector diagram, in particular from the angle between the voltage vectors U1 and U2. This can be done separately for each electrical phase. It can be appreciated that when the active and / or reactive power setpoint of the power generation system 50 changes, the vectors I, U2 and their phase relationship can change, resulting in a change in the transformer phase shift.

[0099] This variable phase shift or delay is determined by Fig.13 The dynamic phase correction 131 shown in FIG. Fig.13 As shown in , the fixed and dynamic phase corrections 132 , 131 can each provide a correction value in delayed form, which can be multiplied by the detected frequency ω to obtain a phase correction value, which is added to the phase signal 94 to obtain a corrected phase signal 95 .

[0100] The corrected phase signal 95 may be an intermediate signal indicating zero crossings of the tap voltage at phase values ​​of 0 / 2π and π. A detector 125 may be provided to detect predetermined phase values ​​in the corrected phase signal 95 to generate the timing control signal 80. The detector 125 may be referred to as an "angle zero crossing detector" (ZCD) because it may detect the phase angle corresponding to the zero crossing of the tap voltage. It may operate similarly to the voltage ZCD 122 by providing first and second phase thresholds near the predetermined phase value to be detected and comparing the instantaneous phase value of the corrected phase signal 95 with these thresholds. If the phase value is between the thresholds, a change in the output signal (i.e., the timing control signal), such as a pulse or spike, may be generated. Again, this may be performed separately for each electrical phase. Fig.12 Shows from Fig.11 The corrected phase signal 95 is a timing control signal 80 derived from three electrical phases. It can be seen that by considering the delay, Fig.10 The pulse indicating the zero crossing of the tap voltage is shifted to an earlier time than the zero crossing indicated in the zero crossing signal 93 of FIG.

[0101] The predetermined phase values ​​detected by the detector 125 may correspond to 0 / 2π and π. To detect a zero crossing at a phase value of 0 / 2π, where there is a discontinuity in the corrected phase signal 95, the value of π may simply be added to the corrected phase signal 95, and values ​​exceeding 2π may be wrapped around again to the range of values ​​between 0 and 2π. This effectively shifts the signal so that a zero crossing may be detected again at a phase value of π. Thus, the same phase threshold may be used to detect zero crossings at phase values ​​of 0 / 2π and π.

[0102] exist Fig.12 In the timing control signal 80 shown by way of example in FIG. 1 , the pulses correspond to the actual zero crossing of the tap voltage (e.g. at the connected transformer tap 26). The signal can be provided to the control signal generator 13 to generate a control signal so that the tap change is performed at the desired switching time. The signal 80 can in particular provide precise knowledge of the zero crossing of the tap voltage so that the switching of the switching element 31 can occur at a desired time point before the zero crossing.

[0103] In other examples, the timing control signal 80 may be derived such that signal changes in the timing control signal 80 may correspond to predetermined times at which switching should occur, i.e., they may correspond to respective switching points. Such a timing control signal 80 may be generated by applying a phase offset to the phase signal 94 or the corrected phase signal 95 ( Fig.16 The resulting offset phase signal may be an intermediate signal that still indicates a zero crossing of the tap voltage (however, not at a phase value of 0 / 2π or π, but at a correspondingly shifted phase value). The phase offset may correspond to a predetermined time 71 before the zero crossing, which may be determined, for example, by multiplying the desired amount of time by the detected frequency. For example, a phase offset may be applied when applying a phase correction in unit 124. When summing the variable and fixed delays 131, 132, the desired time before the zero crossing may be added, for example, as an additional fixed delay Fig.13 . Thus, the phase value of π in the corrected phase signal may appear earlier than the actual zero crossing of the tap voltage, so that the pulses in the timing control signal 80 are generated at an earlier point in time. Thus, the timing control signal 80 provided to the control signal generator 13 may directly indicate the point in time at which the switching of the switching element 31 will occur during the tap change. Thus, the generation of the signal for switching the semiconductor switch to achieve the tap change may be facilitated, and more precise timing may be achieved. The timing control signal 80 may in particular include a series of pulses, which may be used for triggering the thyristors of the switching elements of each electrical phase.

[0104] The control system 10 may include a processing unit 61, which may include a microprocessor, an application specific integrated circuit, a programmable logic device (PLD) such as a field programmable gate array (FPGA), a digital signal processor, etc. It may also include a memory 62, which may include volatile and non-volatile memory, in particular a hard disk drive, a flash memory, a RAM, a ROM, an EEPROM, etc. The memory 62 may store control instructions, which, when executed by the processing unit 61, may cause the processing unit 61 to perform any of the methods disclosed herein. The control instructions, in particular the corresponding software code, may, for example, implement Figure 8 The control system 10 may include further components not explicitly shown, such as input / output interfaces for data sources and / or components to be controlled, a user interface including a display and input units, and other components common to computer systems.

[0105] By generating the timing control signal in the above manner, the main error sources that may cause zero-crossing erroneous detection can be considered and compensated. For example, both the fixed time delay and the variable phase shift caused by the voltage drop in the transformer can be compensated. The generation of the timing control signal can be based on PLL, and high accuracy can be provided in detecting the actual zero-crossing of the tap voltage. This can make it possible to switch the switching element at the most appropriate moment to reduce the short-circuit current during the tap change process without using a conversion impedance. In addition, the solution can provide information about the various zero-crossing moments of the tap voltage of each electrical phase. Therefore, for each electrical phase, the accuracy of zero-crossing detection can be improved.

[0106] Although specific embodiments are disclosed herein, various changes and modifications may be made without departing from the scope of the invention. The present embodiments are to be considered in all respects as illustrative and not restrictive, and all changes coming within the meaning and equivalency range of the appended claims are intended to be embraced therein.

Claims

1. A method of generating a timing control signal (80) for controlling the timing of a transformer tap change of a transformer (20), wherein the transformer (20) comprises an on-load tap changer (30), the on-load tap changer (30) comprising a plurality of corresponding transformer taps (25-27), wherein the method include: - monitoring the voltage on the primary side and / or the secondary side of the transformer (20) to obtain a monitoring signal (91); - deriving an intermediate signal (95) from the monitoring signal (91), the intermediate signal (95) indicating at least one zero crossing of the tap voltage at the transformer tap (25-27), wherein deriving the intermediate signal (95) comprises estimating the phase of the monitored voltage based on the monitoring signal (91) to obtain a phase signal (94), and applying to the phase signal a phase correction for a phase shift between the monitoring signal (91) or a signal derived therefrom and the tap voltage at the transformer tap (25-27); and - deriving said timing control signal (80) from said intermediate signal (95).

2. The method according to claim 1, wherein the transformer taps (25-27) are provided at least on the primary side of the transformer (20), and wherein the voltage is monitored on the secondary side of the transformer (20), wherein applying the phase correction comprises estimating a transformer phase shift (Δθ ) between the voltage on the primary side of the transformer and the voltage on the secondary side of the transformer VAR ), and applying phase correction to the transformer phase shift.

3. The method according to claim 2, wherein estimating the transformer phase shift (Δθ VAR ) comprises obtaining information indicating a phase difference between the monitored voltage and a current on the secondary side of the transformer (20), and deriving the transformer phase shift from a model (135) of the transformer (20) using the monitored voltage and the obtained information as input, and / or comprises determining a phase shift caused by a winding connection type of the transformer, in particular a vector group of the transformer.

4. A method according to any of the preceding claims, wherein applying a phase correction comprises applying a fixed phase correction (132) and / or applying a dynamic phase correction (131), wherein the fixed phase correction (132) corrects the phase signal by a fixed value and the dynamic phase correction (131) corrects the phase signal by a variable value that is allowed to change during operation.

5. The method of any one of the preceding claims, wherein applying phase correction comprises one or a combination of the following operations: - applying a correction for the delay introduced by acquiring said monitoring signal (91) by digital data acquisition; - applying corrections for delays introduced by code execution by a processing unit deriving said timing control signal; - applying correction for a delay introduced by filtering the monitored voltage or the monitoring signal (91); - applying a correction for the phase shift introduced by the electrical connection between the location where the voltage is monitored and the location of said transformer (20); - applying a correction for a phase shift introduced by a circuit in the electrical path between the location where the voltage is monitored and the location of said transformer (20); - applying a correction for the phase shift introduced by the connection configuration of the primary and secondary windings of the transformer, or - Applying dynamic phase correction to the phase shift introduced by the transformer (20).

6. A method according to any one of the preceding claims, wherein the transformer (20) is a three-phase transformer comprising transformer windings for three electrical phases, and wherein the on-load tap changer comprises a transformer tap (25-27) for each of the three electrical phases of the transformer (20), wherein the phase signal (94) is estimated separately for each of the three electrical phases, and / or wherein the phase correction is applied separately for each of the three electrical phases, and / or wherein the timing control signal (80) is derived separately for each of the three electrical phases.

7. A method according to any one of the preceding claims, wherein estimating the phase of the monitored voltage to obtain a phase signal (94) comprises detecting at least one zero crossing (93) of the monitoring signal (91) using a zero crossing detector (122).

8. A method according to claim 7, wherein detecting at least one zero crossing of the monitoring signal (91) by the zero crossing detector (122) includes using at least a first voltage threshold above the zero voltage value and a second voltage threshold below the zero voltage value, and determining the presence of a zero crossing if the monitoring signal (91) has a value between the first voltage threshold and the second voltage threshold.

9. The method according to any one of the preceding claims, wherein estimating the phase of the monitored voltage to obtain the phase signal comprises detecting the frequency (92) of the monitored voltage using a phase locked loop (PLL) (121).

10. A method according to claim 9 and according to claim 7 or 8, wherein the phase signal (94) is generated by an integrator (123) which uses as input the detected zero crossings (93) of the monitoring signal and the detected frequency (92) of the monitored voltage.

11. The method according to any of the preceding claims, wherein the timing control signal (80) is derived from the intermediate signal (95) include: - detecting the presence of at least one predetermined phase value in said intermediate signal (95); as well as - generating the timing control signal (80) based on the detection of the at least one predetermined phase value by generating a signal change in the timing control signal (80) at the moment when the predetermined phase value is detected.

12. A method according to any one of the preceding claims, wherein the method further comprises applying a predetermined phase offset to the phase signal (94, 95) before deriving the timing control signal (80).

13. A method of controlling an on-load tap changer (30) of a transformer (20) to perform a tap change from a first transformer tap (26) to a second transformer tap (25), wherein the method include: - Determining a timing control signal (80) according to the method of any one of claims 1 to 12; - detecting a triggering event triggering said tap change; - upon detection of the trigger event, deriving a drive signal (81) for switching a switching element (31) coupled to the second transformer tap (25) based on the timing control signal (80); as well as - applying said drive signal (81) to said switching element (31) of said second transformer tap (25) to perform said tap change.

14. A control system configured to control changing of transformer taps of a transformer, wherein the transformer (20) comprises an on-load tap changer (30), the on-load tap changer (30) comprising a plurality of respective transformer taps (25-27), wherein the control system (10) is configured to perform the method according to any one of the preceding claims.

15. A computer program for generating a timing control signal for controlling the timing of a change of a transformer tap (25-27) of a transformer (20), wherein the computer program comprises control instructions which, when executed by a processing unit (61) of a control system (10) for controlling an on-load tap changer (30) of the transformer (20), cause the processing unit (61) to perform a method according to any one of claims 1 to 13.

Images