METHOD FOR CONTROLLING A WIND POWER PLANT

Description

[0001] The present invention relates to a method for controlling a wind turbine and the invention relates to a corresponding wind turbine.

[0002] Wind turbines are well-known; they use a generator to produce electrical power from the wind and feed it into an electrical grid. A common topology for such a wind turbine works as follows: the generator produces alternating current (AC), this AC is rectified, and from this rectified current, which is usually provided in a DC link, an inverter generates the AC current to be fed into the grid.

[0003] Modern wind turbines are characterized by a rated output of several megawatts. To rectify such power generated by a generator, several rectifiers can be connected in parallel. If these rectifiers are actively controlled and thus also actively control the corresponding generator current they rectify, they can also be referred to as generator-side inverters. These can, at least theoretically, be identical in construction to the inverters that generate the alternating current fed into the electrical grid. To avoid confusion, the term "active rectifier" can be useful for the generator-side inverter and is therefore synonymous.

[0004] Traditionally, passive inverters have been used for inverter conversion, thus functioning as diode rectifiers. In three-phase applications, these diode rectifiers can also be referred to as B6 bridges or B6 bridge rectifiers.

[0005] As a technical improvement, especially to improve targeted control of the generator and thus to improve the control of the generator in general, it is useful or even necessary to use the aforementioned active rectifiers or generator-side inverters.

[0006] Here too, multiple inverters can be connected in parallel. This allows for the use of more cost-effective inverters. It also makes it possible to use the same inverters for generators of different sizes, i.e., different generator outputs, but to connect a varying number of inverters in parallel depending on the generator's output. Each inverter then forms a sub-inverter.

[0007] To ensure that each phase of the current to be rectified (i.e., the generator current) is evenly distributed among the generator-side sub-inverters when connected in parallel, a corresponding fraction of the generator current to be controlled for the respective phase can be specified as the setpoint current for each sub-inverter. For example, if three sub-inverters are used, each sub-inverter can control one-third of the generator current and receive a corresponding setpoint value, i.e., each receives one-third of the total setpoint current as its setpoint value.

[0008] However, these sub-inverters are connected on the generator side due to their parallel connection. Furthermore, they can also be connected via a common DC link. This creates the risk of circulating currents.

[0009] Such circulating currents can be kept low by using sufficient inductances in the generator-side inverters. Appropriate inductances connecting individual DC link circuits of the sub-inverters can also keep such circulating currents low.

[0010] The problem of circulating currents can occur particularly with inverters that use a hysteresis method for current control. In such a hysteresis method, a tolerance band with upper and lower limits is defined for the AC current to be controlled. If the generated current reaches one of the band limits, switching occurs to keep the current within the tolerance band. With circulating currents, the problem arises that in each sub-inverter, the measured partial current, which is supposed to be kept within the defined tolerance band, can contain components of a circulating current. This can negatively affect the current control.

[0011] However, such inductances can be undesirable, especially because they can be a costly component.

[0012] German patent application DE 10 2014 219 052 A1 relates to a method for generating alternating current for feeding into an electrical supply network, wherein several inverters each generate a partial current, these partial currents are superimposed to form a total current, and this superimposed total current is fed into the network. A tolerance band method is used for this purpose, which changes tolerance limits depending on the total current generated.

[0013] The Japanese patent application JP 2001-314086 A relates to an AC / DC converter in which a breakdown of a switching element in a conversion circuit is prevented by an unequal voltage generated by an unbalanced output.

[0014] European patent EP 2 262 089 B1 discloses a system for operating a wind turbine current converter. This system uses, among other things, several converters connected in parallel, which can be controlled with phase-shifted control signals in order to reduce overall switching harmonic components by canceling out phase-shifted switching waveforms.

[0015] The present invention is therefore based on the objective of addressing at least one of the aforementioned problems. In particular, a solution is to be provided in which parallel connection of inverters is possible with low inductance on the generator side, especially a parallel connection of inverters with a hysteresis method for current control. In particular, undesirable circulating currents are to be avoided. At the very least, an alternative to previously known solutions is to be proposed.

[0016] According to the invention, a method according to claim 1 is proposed. This method thus relates to the control of a wind turbine according to claim 8, which has a generator for generating generator current with one or more generator current phases. In particular, it has a generator having three generator current phases or six generator current phases. In the case of six generator current phases, these are in particular configured as two three-phase current phases.

[0017] Furthermore, an active rectifier is provided for rectifying and controlling the generator current. The active rectifier can also be referred to as a generator-side inverter, as it not only rectifies but also precisely controls the generator current. Specifically, a target current is defined for the generator current in terms of magnitude, frequency, and phase. In particular, a synchronous generator is used, and the generator current controlled is the stator current of the generator.

[0018] The active rectifier has several controllable sub-rectifiers for each generator current phase. These controllable sub-rectifiers can also be referred to as controllable generator-side sub-inverters. Each controllable sub-rectifier is characterized by a partial inductance. This partial inductance may be provided by an inductive component, but it may also result solely or additionally from the specific construction of the sub-rectifier, including the necessary connecting cables.

[0019] Each controllable partial rectifier controls a partial current of the generator current phase, and each generator current phase forms a total current as the sum of all partial currents of that phase. For example, three controllable partial rectifiers can be provided for each phase. Each of these three controllable partial rectifiers controls a partial current, so that a total of three partial currents are controlled. These three partial currents are added together to form the total current.

[0020] It is now planned that the active rectifier is controlled in such a way that the total current is detected for each generator current phase and each controllable partial rectifier of the respective current phase controls its partial current depending on the detected total current.

[0021] The total current can be measured by measuring each partial current and then calculating the total current from these measured partial currents. This has the particular advantage that current measurements taken at the individual rectifiers can be used for this purpose. A separate sensor for the total current is then unnecessary.

[0022] Each controllable partial rectifier of the relevant current phase now controls its partial current depending on this total current thus detected. Specifically, each controllable partial rectifier is therefore controlled by the detected total current instead of by its partial current.

[0023] In particular, the control method in which each partial rectifier has its own individual control loop to regulate its partial current (i.e., the actual value of the partial current) to a setpoint is abandoned. Specifically, a control principle is abandoned in which the desired total current (i.e., the desired generator current of the relevant phase) is divided into individual partial current setpoints, and each of these setpoints is controlled by a partial rectifier. Instead, a total current setpoint is specified only for the intended total current (i.e., the intended generator current of the relevant phase), and each individual partial rectifier is controlled based on how the measured total current behaves relative to the specified total current setpoint, thereby controlling the measured total current as a whole. In particular, it is proposed that no setpoint is specified for each partial rectifier.

[0024] It was particularly recognized here that considering the total current prevents circulating currents from adversely affecting the individual control in each sub-rectifier. Each individual sub-rectifier can also be described as a sub-rectifier with local control, and thus the proposed solution prevents circulating currents from influencing the local control. This is because the total current does not include the circulating currents.

[0025] According to one embodiment, it is proposed that the active rectifier operates according to a hysteresis method, wherein a tolerance band with an upper and a lower band limit is specified for each total current, and each partial rectifier controls its partial current depending on whether the total current reaches the upper or lower band limit.

[0026] A common tolerance band method involves checking whether the current to be controlled reaches a band limit and then switching accordingly if the band limit is reached. In previous methods, this was implemented for the partial rectifiers by each generating a partial current and checking whether this partial current falls within its specified tolerance band. Switching would then occur if this partial current reached an upper or lower band limit of its tolerance band. This variant is not used here.

[0027] Instead, the system still allows each individual rectifier to control the current and thus its own partial current by performing corresponding switching operations. However, the trigger is no longer whether the partial current reaches one of its band limits, but rather whether the total current reaches a band limit. Depending on this, each of the individual rectifiers, whose partial currents combine to form the total current, then switches. In this way, the reaching of a band limit by the measured total current is reported to all individual rectifiers, which then all react accordingly.

[0028] In particular, it is provided that each partial rectifier has at least one switching device to control its partial current by switching the switching device, and it is provided that the switching of the switching device is controlled depending on whether the total current reaches the upper or lower band limit.

[0029] The system centrally monitors the total current to see if it reaches a band limit, and if so, each individual rectifier switches accordingly. The individual partial currents continue to be generated by the rectifiers, also through switching of the switching device. However, the triggering of these switching operations is centrally controlled by monitoring the total current within the tolerance band for the total current.

[0030] A hysteresis method, also known as a tolerance band method, can be used, which connects several partial rectifiers in parallel for each generator current phase and is thus insensitive to circulating currents. The partial inductors can therefore be dimensioned small, if they are even provided as separate components. In particular, such partial inductors can potentially be used for other tasks or dimensioned for other tasks, especially for a filter function. Small partial inductors can thus lead to a cost reduction compared to larger ones.

[0031] According to one embodiment, each partial rectifier is assigned an individual delay time, and each partial rectifier additionally controls its partial current depending on this individual delay time. The assignment of this individual delay time is based, in particular, on the physical conditions. Specifically, an actual effective delay time is determined for each partial rectifier and assigned as its individual delay time. The individual delay time can, for example, depend on cable lengths or vary due to manufacturing tolerances of the components. However, it is also conceivable that the individual delay time is deliberately set to certain values ​​in order to influence the control.

[0032] In particular, it is provided here that each partial rectifier switches at least one of its switching devices after the total current has reached the upper or lower band limit and the individual delay time has elapsed. This also ensures that the partial rectifiers of the relevant phase are not switched exactly synchronously, although they are all switched depending on when the total current reaches the upper or lower band limit.

[0033] An individual delay time refers specifically to the time that elapses until a switching operation of the partial rectifier has a significant effect on the total current. For example, the individual delay time could be the time that passes after the partial rectifier switches until the resulting current reaches at least 63% of its value in the total current under consideration. Such behavior can be determined, for example, using a test signal by recording the effect of a specifically defined switching signal and relating it to that signal.

[0034] Preferably, it is proposed that the delay time, i.e., the individual delay time, be determined as a function of the partial inductance of the partial rectifier. This partial inductance influences how long a partial current generated by a switching operation, or how long a change in the partial current generated by the switching operation, takes to become effective in the total current. As described, a 63% effectiveness can be assumed here instead of 100%. In this respect, the individual time constant would be defined like the time constant of a first-order delay element.

[0035] According to one embodiment, it is proposed that a deviation of each partial current from a mean partial current is detected and the partial rectifiers are controlled depending on the deviation, in particular the individual delay time is determined.

[0036] Each partial rectifier generates a partial current, and all partial currents combine to form the total current. For example, if three partial rectifiers are used, the average partial current corresponds to one-third of the total current. Ideally, each partial current corresponds to the average partial current. In the example given, under ideal conditions, each of the three partial rectifiers would generate one-third of the total current. However, due to individual variations, particularly due to different cable lengths and component variations, these partial currents can differ. This difference can be determined by comparing them to the average partial current. The difference thus determined can then be used to calculate the individual delay time. Specifically, such a difference in the partial currents is reflected in a corresponding time deviation from the average partial current.

[0037] According to one embodiment, it is proposed that a dynamic relationship between a switching operation of a partial rectifier and a resulting partial current be established, depending on a partial rectifier voltage, a current waveform associated with that partial rectifier voltage, and the generator inductance. Furthermore, it is proposed that the partial rectifier be controlled based on this dynamic relationship and the measured total current.

[0038] Such a dynamic relationship between a switching operation of a partial rectifier and a resulting partial current can be a dynamic relationship, i.e., an underlying dynamic, that characterizes a step response. This dynamic relationship can be described as a transfer function or specified as such, particularly in the context of control engineering. It should be noted, however, that a step response, or the corresponding input step, is usually assumed to be idealized. A switching operation is, in particular, the opening or closing of a semiconductor switch in the partial rectifier. The partial rectifier voltage assumed for the differential equation can then, ideally, undergo a step, either from zero to a voltage value or from a voltage value to zero.

[0039] Such a step response, to stick with this intuitive description common in control engineering, is also influenced by the generator inductance and additionally by a property of the partial rectifier, in particular by the partial inductance of the partial rectifier under consideration. A differential equation can be formulated to describe this fundamental relationship. However, specific values, or at least one specific value, namely that resulting from the partial inductance, may be unknown. The differential equation includes a voltage, which does not need to be measured but rather serves only to establish the differential equation. The differential equation can be solved to obtain the unknown property. This unknown property can be included in the differential equation as a time constant.This time constant is then determined by solving the differential equation. The differential equation is therefore solved to determine at least one previously unknown parameter, especially a previously unknown time constant.

[0040] Thus, this dynamic relationship can be described in principle by the differential equation, and can also be determined quantitatively by recording each current profile.

[0041] The result is thus the dynamic relationship, which is also known quantitatively, specifically the transfer function. The partial rectifier is then controlled based on this dynamic relationship and the measured total current. This control based on the measured total current can therefore be implemented as described above for other embodiments.

[0042] It is therefore specifically proposed that the delay time, i.e., the specific delay time of the partial rectifier under consideration, be determined based on this dynamic relationship. The partial rectifier is then controlled based on this determined delay time. Its switching operations are controlled accordingly.

[0043] The voltage measured at the partial rectifier is, in particular, the output voltage of the partial rectifier for the relevant phase. A partial rectifier voltage at the output of the partial rectifier then leads to a current waveform, specifically due to the partial inductance and the generator inductance. This relationship is described by the differential equation or by a system of differential equations, in which each current waveform is assigned to a voltage waveform, especially a voltage step.

[0044] The following example provides an illustrative explanation.

[0045] In the chosen example, a partial inductance can be present and denoted as L_GR_d, where its value can be L_GR_d = 100 µH / converter unit, i.e., 100 µH / partial rectifier. In this example, seven partial rectifiers can be connected to form a single rectifier. The generator can then have an inductance of 10 mH / partial generator system, for example, where four partial generator systems are connected to form a single generator system. Thus, with seven active rectifiers, the effective inductance is 100 µH / 7, compared to a generator inductance of 10 mH / 4. The time constant of the current change in the generator therefore allows for the aforementioned relationship between the inductances.

[0046] In the transient transition, only the behavior of the seven converters relative to each other is relevant. For example, if six partial rectifiers, which can also be called converters, carry an identical partial current of 100 A, and the seventh partial rectifier carries only 93 A, the initial state is that all switching devices are switched on and the upper limit of the total current is exceeded. The target state is that all switching devices are switched off. The goal in the transition is to eliminate the differences so that the total current (6 * 100 A + 1 * 93 A = 693 A) is again equalized by 7 * 99 A = 693 A.

[0047] By switching on the rectifiers in the form 6*OFF and 1*ON for a certain period of time, an inductive series circuit of 100µH + 100µH / 6 = 116 µH is created, across which the full intermediate circuit voltage Uzw (e.g. 1160V) is now applied.

[0048] A current change of 6A (93A + 6A = 99A) must now be induced in this total inductance. Ideally, the following applies: Uzw = L * di / dt − > dt = L / Uzw * di = 116 μH / 1160 V * 6 A = 6.0000 e − 07 s = 600 ns

[0049] This is a clear example that can be adapted for other conditions. In particular, if the currents are not as similar as in this example, up to six temporarily valid equations can arise during the transition if seven partial rectifiers are present or considered. The direction of the switching operation is also relevant; in the example above, the voltage for switching on would need to be chosen so that the switch-on occurs earlier. This can be particularly relevant for the input of PI controllers for feedforward control.

[0050] According to one embodiment, a central control unit is proposed to detect the total current, generate control signals for the individual rectifiers, and transmit these signals to them in order to control the individual rectifiers. Thus, a central control unit is provided that controls the individual rectifiers based on the total current, generating and transmitting corresponding control signals to them.

[0051] In particular, individual switching time adjustments are determined for the partial rectifiers. The switching times are crucial for the partial rectifiers to generate the desired partial current. These individual switching time adjustments are provided to ensure that each partial rectifier generates the desired partial current based on the measured total current. These adjustments can include individual delay times and also take signal propagation times into account.

[0052] Optionally, these switching time adjustments can be transmitted to the sub-rectifiers, particularly via the central control unit. However, it is also possible for the switching time adjustments to be taken into account within the central control unit itself.

[0053] In particular, it is planned that, for the switching time adjustments, propagation times for the transmission of control signals from the central control unit to the respective rectifier are determined and taken into account when calculating the switching time adjustment. It was recognized that a discrete-time clock of 1 µs (clock time) can be achieved in a technically feasible manner through communication, whereby the times to be controlled can be shorter than this clock time. It is also conceivable that this clock time may sometimes be shorter than technically achievable times. Therefore, the time to be adjusted, i.e., the respective switching time adjustments, would be calculated centrally, but preferably communicated to the local system along with the vector to be converted. According to this design, all switching time adjustments are bundled in the vector to be converted and transmitted together, thus avoiding different propagation times.

[0054] It was thus recognized that not only individual delay times, which relate to the physical delays between the respective switching time of the partial rectifier and its effect in the total current, can be relevant, but also that the transmission of control signals from the central control unit to each partial rectifier must be taken into account. This can mean that such transmission times between the central control unit and each partial rectifier are the same or that deviations can be neglected, but it can also mean that there are differences that cannot be neglected. This can also depend on the chosen transmission method. Preferably, it is proposed to determine such propagation times individually between the central control unit and each partial rectifier if they are not transmitted jointly, in particular in a common vector.

[0055] In particular, it is proposed that when determining the switching time adjustments, the individual delay times of the sub-rectifiers and / or the transit times for transmitting information from the central control unit to the respective sub-rectifier be stored in a control protocol and used to control the sub-rectifiers.

[0056] One control option is for the central control unit to send an identical control signal to all sub-rectifiers pertaining to a single phase. Each sub-rectifier then adjusts its switching time based on its individual switching time adaptation. The sub-rectifier itself takes into account its individual delay time and the relevant propagation time for information transmission from the central control unit.

[0057] It is preferably proposed here to determine the individual delay times in advance, in particular by measuring them. For this purpose, an initial measurement can be carried out before commissioning the rectifier, which is proposed here as one embodiment.

[0058] Regarding the proposed storage of individual delay times and / or runtimes in the control protocol, a variant is suggested in which the central control unit takes these individual times of each sub-rectifier into account. For control purposes, the central control unit then monitors the total current. However, this can be achieved by the central control unit receiving the values ​​of each sub-current from the sub-rectifiers.

[0059] When the measured total current reaches a band limit, the individual rectifiers are controlled by the central control unit for switching. This can be achieved by the central control unit taking into account the switching time adjustment of each individual rectifier, specifically ensuring that the individual delay time and the propagation time for information are considered for each rectifier. Based on this, a control signal, i.e., a switching command, can then be sent from the central control unit to each individual rectifier at specific times. Each switching command sent by the central control unit is precisely timed so that the respective rectifier switches at the correct moment.

[0060] According to one embodiment, a control structure is proposed that provides star-shaped communication with a central clock. The communication can be designed to achieve a constant delay time with low variability, e.g., a maximum of 10 ns (maximum jitter of 10 ns).

[0061] According to this embodiment, the switching signals can therefore be transmitted simultaneously to all partial rectifiers with the same delay for all of them.

[0062] The aforementioned propagation delays can be caused particularly by electrical / optical / electrical conversion and / or by a driver stage, namely in particular a gate resistance and a gate capacitance.

[0063] According to one embodiment, it is proposed that each partial rectifier performs switching operations to generate a voltage pulse, wherein a triggering switching operation is provided to trigger a voltage pulse, and a terminating switching operation is provided to terminate a voltage pulse, and a time interval between the triggering switching operation and the terminating switching operation of the voltage pulse describes a pulse width of the voltage pulse, wherein different switching time adjustments or different and, in particular, variable delay times are provided for the triggering switching operation and the terminating switching operation in order to control the pulse width.

[0064] The partial current to be generated in each partial rectifier depends primarily on the voltage pulse and the partial inductance, as well as on the generator inductance. To increase the partial current of a first partial rectifier relative to the partial current of a second partial rectifier, the voltage pulse can be broadened. This can be achieved by appropriately selecting the switching time adjustments or delay times of the initiating and terminating switching operations. To broaden the voltage pulse, the initiating switching operation can be brought forward in time, i.e., delayed less, and / or the terminating switching operation can be postponed in time, i.e., delayed more.

[0065] The voltage pulses can be positive or negative. The triggering switching action can therefore initiate a rising or falling voltage edge, and conversely, the terminating switching action can initiate a falling or rising edge, thereby ending the positive or negative voltage pulse, respectively.

[0066] In particular, this allows the pulse width to be lengthened or shortened. The initial width can be derived from a tolerance band controller, which is generally proposed here, namely as a current controller in each sub-rectifier. The interventions that change this width are temporal in nature, so that the lengthening of the pulse width is irrelevant to the overall duration of the control signal. The fundamental behavior of the tolerance band controller is therefore not altered; only individual switching times are modified.

[0067] According to one embodiment, it is proposed that three partial rectifier arrangements, namely one partial rectifier arrangement each for the first, second, and third generator current phases, together with a grid-side partial inverter arrangement, form a partial back-to-back converter. Each partial back-to-back converter has a common DC link to which the partial rectifiers of each partial rectifier arrangement rectify and from which the partial inverter arrangement inverts. It is further proposed that the DC links of several partial back-to-back converters be coupled to allow circulating currents through these coupled DC links.

[0068] It was particularly recognized here that, in accordance with the inventive method, no circulating currents are to be intentionally controlled. The method operates in such a way that it largely eliminates or avoids circulating currents. However, it was recognized that, particularly when calculating the switching time adjustments, it can be taken into account that circulating currents, especially those intentionally generated on the grid side, are not compensated. They can then occur as balancing currents between the phases and be permitted accordingly.

[0069] 8 According to the invention, a wind energy plant according to claim 8 is also proposed and this wind energy plant comprises a generator for producing a generator current with one or more generator current phases, and an active rectifier for rectifying and controlling the generator current, wherein the rectifier has several controllable partial rectifiers for each generator current phase, each controllable partial rectifier is characterized by a partial inductance, and each controllable partial rectifier controls a partial current of the generator current phase, and each generator current phase forms a total current as the sum of all partial currents of the generator current phase concerned, wherein the wind turbine has a control unit for controlling the active rectifier, and wherein the control unit is configured to control the active rectifier such that the total current is detected for each generator current phase, and each controllable partial rectifier of the current phase concerned controls its partial current depending on the detected total current.wherein each partial rectifier (201-203) is assigned an individual delay time, and each partial rectifier (201-203) additionally controls its partial current depending on this individual delay time.

[0070] The wind turbine thus has a control unit for controlling the active rectifier. This control unit is configured to control the active rectifier in such a way that the total current is measured for each generator phase, and each controllable sub-rectifier of the respective phase controls a partial current depending on the measured total current. Specifically, the control unit can be configured for this control by implementing a corresponding sequence as a sequence code or program within the control unit. Furthermore, the control unit has appropriate interfaces for measuring the total current. It can measure the partial currents of each sub-rectifier and calculate the total current from them. The measurement of the partial currents can be configured so that the control unit receives values ​​from the individual sub-rectifiers.The corresponding measuring and / or control elements of the individual sub-rectifiers can therefore be connected to the control unit or they can be part of the control unit.

[0071] In particular, it is intended that the wind turbine, especially the control unit, is prepared to carry out a procedure according to one of the embodiments described above. For this purpose, a corresponding sequence can be implemented, in particular as a program, in the control unit.

[0072] Insofar as the procedures use a central control system, this can be part of the control unit.

[0073] The invention will now be explained in more detail below using exemplary embodiments with reference to the accompanying figures. Figure 1 shows a wind turbine in a perspective view. Figure 2 schematically shows a single-phase rectifier with three sub-rectifiers as an example. Figure 3 schematically shows a structure for controlling multiple sub-rectifiers.

[0074] Figure 1Figure 1 shows a schematic representation of a wind turbine according to the invention. The wind turbine 100 has a tower 102 and a nacelle 104 on the tower 102. An aerodynamic rotor 106 with three rotor blades 108 and a spinner 110 is provided on the nacelle 104. During operation of the wind turbine, the aerodynamic rotor 106 is set into rotation by the wind and thus also rotates an electrodynamic rotor or rotor of a generator, which is directly or indirectly coupled to the aerodynamic rotor 106. The electric generator is arranged in the nacelle 104 and generates electrical energy. The pitch angles of the rotor blades 108 can be changed by pitch motors at the rotor blade roots 109 of the respective rotor blades 108.

[0075] The wind turbine 100 has an electric generator 101, which is indicated in the nacelle 104. Electrical power can be generated by means of the generator 101. A feed-in unit 105 is provided for feeding electrical power into the grid; this unit can be specifically designed as an inverter. This allows a three-phase feed-in current and / or a three-phase feed-in voltage with amplitude, frequency, and phase to be generated for feeding into a grid connection point (PCC). This can be done directly or in conjunction with other wind turbines in a wind farm. A plant control unit 103 is provided for controlling the wind turbine 100 and also the feed-in unit 105. The plant control unit 103 can also receive setpoint values ​​from external sources, in particular from a central park computer. An active rectifier, which can be part of the feed-in unit 105, is connected to the generator 101.

[0076] Figure 2Figure 1 illustrates a partial rectifier arrangement 200 of one phase, comprising several partial rectifiers 201-203. Several such partial rectifier arrangements 200 of one phase can then together form a complete active rectifier, which is then used to rectify and control the generator current of a wind turbine generator. However, only one partial rectifier arrangement of one phase is considered here. Corresponding partial rectifier arrangements are provided for other phases.

[0077] The partial rectifier arrangement 200 of a phase thus comprises three partial rectifiers 201 - 203. The third partial rectifier 203 can also be considered partial rectifier N, representing all other partial rectifiers that together form the partial rectifier arrangement 200 of a phase.

[0078] In this Figure 2The basic structure of this partial rectifier arrangement 200, consisting of several partial rectifiers 201-203 in one phase, will be explained, specifically on the generator side. Each partial rectifier 201-203 can have a DC output 211-213. Each partial rectifier 201-203 can also be configured as part of a partial converter arrangement. In this case, an internal DC link would be provided, and an AC output would be provided instead of the DC output 211-213.

[0079] Each partial rectifier 201-203 has a generator-side output 221-223. Furthermore, each partial rectifier 201-203 has a partial inductance 231-233. Each partial inductance 231-233 is in the Figure 2 It symbolizes a component connected to the generator-side output 221-223. However, it also represents inductances that may arise, for example, from the supply line, or takes these into account.

[0080] A control unit 241-243 is provided to control each partial rectifier 201-203. The control unit can receive control signals and use them to control semiconductor switches of the partial rectifier, thereby generating a pulsed voltage signal that results in a modulated sinusoidal current. A switching voltage US1, US2, or USN is established directly at the generator-side output of the inverter, which essentially alternates between a positive value, a negative value, and zero, depending on the corresponding switch positions. At the generator-side output of the partial inductor 231, 232, or 233, a generator-side voltage U1, U2, or UN, and a generator-side current i1, i2, or iN are established. Each of these generator-side currents i1, i2, or iN constitutes a partial current of the generator current of the respective phase. This generator current of the respective phase thus forms the total current of this generator phase.This is represented as the total stream i S in . Figure 2 The current is drawn and flows through a generator inductance 234. The generator 250 is shown here in the schematic representation of the Figure 2 , which can also be viewed as an equivalent circuit diagram, is considered as a voltage source with the voltage UG.

[0081] The generator-side voltages U1, U2, and UN and the partial currents i1, i2, and iN can be detected at a sensing point 261, 262, or 263 and transmitted to the respective control unit 241–243, or the respective control unit 241–243 detects the respective voltage and partial current at the sensing point. The control units 241–243 can then transmit the detected values ​​to a central control unit and / or a control unit.

[0082] Figure 3 This schematically illustrates a possible control concept in a simplified manner.

[0083] This uses Figure 3a partial rectifier arrangement 200 of one phase, as used in Figure 2 was explained. In the Figure 3 For the sake of clarity, only a part of the partial rectifier arrangement 200 is shown in one phase. Figure 2 The diagram shows the control units 241-243 for each partial rectifier 201-203. These control units 241-243 detect the generator-side voltages U1, U2, and UN, as well as the partial currents i1, i2, and iN. These values ​​are transmitted to a central control unit 300, which is schematically delimited here by a dashed outline.

[0084] Of these values, the partial currents i1, i2 - iN are summed in a summing element 302, resulting in the total current iΣ. This total current iΣ can be compared to the total current iS of the Figure 2 correspond to, or ideally should correspond to, this. The total flow i S of the Figure 2However, this refers to the actual total flow, whereas the total flow i Σ of the Figure 3 The calculated value is the sum of the partial currents i₁, i₂ - iₙ. If measurement errors or inaccuracies are disregarded, the calculated total current iΣ corresponds to the actual total current iS. Figure 2 .

[0085] In any case, this calculated total current iΣ is input into a modulation block 304. Modulation block 304 uses a tolerance band method to modulate a target current itarget. For this purpose, a tolerance band is applied around this predetermined current itarget, which is specified in terms of magnitude, frequency, and phase, and thus defined as a sinusoidal current. Depending on whether the total current iΣ touches an upper or lower tolerance limit, a switching signal between 0 and 1, or between 0 and -1, is output. This also depends on whether the current to be generated is positive or negative, to put it simply.

[0086] The result of the modulation block, i.e., the tolerance band method implemented in modulation block 304, is therefore a switching signal that is fundamentally intended for each partial rectifier. The partial rectifier is to switch the switching voltage US1, US2, or USN according to the switching signal, namely to the negative value, the positive value, or to zero.

[0087] Thus, this switching signal output by modulation block 304 can switch the switch position in each partial rectifier 201, 202, or 203. However, any circulating currents that affect, for example, partial currents I1 and I2, have no effect on the switching signal generated by the tolerance band method in modulation block 304.

[0088] This already achieves an important goal, namely the generation of the total current by parallel-connected partial rectifiers essentially independent of circulating currents. Furthermore, it is proposed to also consider time differences between the individual partial rectifiers 201–203. While such consideration could be performed centrally in a common computing block, for example, for the purpose of illustration this is shown in Figure 3 However, each sub-rectifier 201-203 is shown individually. The basic operating principle explained below is the same for all sub-rectifiers 201-203. Therefore, this will be explained below for sub-rectifier 201.

[0089] The partial rectifier 201, which can also be referred to as the first partial rectifier, transmits the generator-side voltage U1 and its partial current i1 to the central control unit, and these values ​​are also passed to a first transit-time detection block 311. In the transit-time detection block, a transit time is recorded, and a switching time adjustment is determined based on this. The recorded switching time adjustment is passed to the adjustment block 321. The adjustment block 321 essentially delays the switching signal S. The result is an adjusted switching signal S1. The adjusted switching signal S1 is still a switching signal that can have the values ​​1, 0, or -1, which can also be encoded differently. However, the adjusted switching signal S1 is delayed compared to the unmodified switching signal S.

[0090] The time-of-flight detection block 311 also receives this adapted switching signal S1, as well as the generator-side voltage i1 and the partial current i1 of the first partial rectifier. By considering the adapted switching signal S1, the time-of-flight detection block 311 can then determine precisely when a switching command was transmitted. A switching command can be one in which the switching signal changes from 0 to 1 or back, or from 0 to -1 or back. This exact time is then known in the time-of-flight detection block 311 and can be compared with the resulting generator-side voltage U1 and partial current i1 generated by the first inverter 201. Thus, it is possible to identify precisely which signal results from this adapted switching signal. The temporal behavior of the resulting signals is taken into account, as well as their amplitude and amplitude profile.It is possible to consider only one of the two signals, i.e., only the voltage or only the partial current.

[0091] Furthermore, the runtime measurement block 311 takes into account the unchanged switching signal S. From this, it can be deduced what the overall desired signal should look like.

[0092] Alternatively, or in addition, an average of the partial currents i₁, i₂, and iₙ can be used. To calculate this average, simply divide the calculated total current iΣ by the number of partial rectifiers, i.e., N. This is illustrated by quotient block 306.

[0093] In this respect, the propagation delay control block 311 calculates time delays for switching time adjustment, by which the switching signal S is adjusted to obtain the adjusted switching signal S1. These time delays can differ for a rising edge from 0 to 1 compared to the falling edge from 1 to 0. The same applies to the edge from 0 to -1 and back from -1 to 0. This allows not only for delays to compensate for propagation delays but also for changes to pulse widths. For example, a rising edge from 0 to 1 and a falling edge back from 1 to 0 result in a voltage pulse with a pulse width. By specifying different delay times for the rising edge from 0 to 1 and the falling edge from 1 to 0, the pulse width can be modified.

[0094] In this sense, the delay blocks 312 and 313 and the matching blocks 322 and 323 are used analogously for the further sub-rectifiers 202 and 203. The result is that a matching switching signal S1-S3 is generated for each sub-rectifier 201-203. Each matching switching signal S1-S3 can take different delays into account and also generate pulses of different widths.

Claims

1. A method for controlling a wind power installation (100) and the wind power installation comprises - a generator (101) for generating a generator current with one or more generator current phases, and - an active rectifier for rectifying and controlling the generator current, wherein for each generator current phase the rectifier - has a plurality of controllable sub-rectifiers (201-203), - each controllable sub-rectifier (201-203) has a partial inductance (231-233), and - each controllable sub-rectifier (201-203) controls a partial current of the generator current phase and each generator current phase forms a summation current as a sum of all the partial currents of the relevant generator current phase, characterized in that - the active rectifier operates according to a hysteresis method and is controlled so that for each generator current phase - the summation current is detected and - a tolerance band with an upper and a lower band limit is prescribed for each summation current and - each sub-rectifier (201-203) controls the partial current thereof depending on whether the summation current reaches the upper or lower band limit, wherein - each sub-rectifier (201-203) has at least one switching means in order to control the partial current thereof by way of switching the switching means and - each sub-rectifier (201-203) is assigned an individual delay time, and - each sub-rectifier (201-203) controls the partial current thereof additionally depending on said individual delay time, and that - each sub-rectifier (201-203) switches at least one of the switching means thereof, - after the summation current has reached an or the upper or lower band limit, and - the individual delay time has elapsed.

2. The method as claimed in one of the preceding claims, characterized in that - a or the delay time is determined depending on the partial inductance of the sub-rectifier (201-203).

3. The method as claimed in one of the preceding claims, characterized in that - a deviation of each partial current from an average partial current is detected, and - the sub-rectifier (201-203) is controlled, in particular a or the individual delay time is determined, depending on the deviation.

4. The method as claimed in one of the preceding claims, characterized in that - depending on - a sub-rectifier voltage, - a current profile associated with the sub-rectifier voltage and - a detected generator inductance of the generator (101) - a dynamic correlation between a switching process of a sub-rectifier and a resulting partial current is established, in particular as a transmission function, and - the sub-rectifier (201-203) is controlled depending on this dynamic correlation or the transmission function and depending on the detected summation current, in particular in that - a or the delay time is determined depending on this dynamic correlation or the transmission function and the sub-rectifier (201-203) controls the switching processes thereof depending on the delay time determined in this way.

5. The method as claimed in one of the preceding claims, characterized in that - a central control system (300) - detects the summation current, - generates control signals for the sub-rectifiers (201-203) and - transmits same to the sub-rectifiers (201-203) in order to control the sub-rectifiers, wherein - in particular individual switching time adjustments for the sub-rectifiers (201-203) are determined and optionally transmitted to the sub-rectifiers (201-203), in particular by way of the central control system (300), and for the switching time adjustments - propagation times for the transmission of control signals of the central control system (300) to the respective sub-rectifier (201-203) are determined and taken into account in the determination of the switching time adjustments, wherein in particular - the individual delay times of the sub-rectifiers (201-203) and / or - the propagation times for the transmission of information from the central control system (300) to the respective sub-rectifier (201-203) - are stored in a control protocol and used to control the sub-rectifiers (201-203).

6. The method as claimed in one of the preceding claims, characterized in that - each sub-rectifier (201-203) performs switching processes in order to generate a voltage pulse, wherein in each case - a triggering switching process is provided to trigger a voltage pulse, and - a terminating switching process is provided to terminate a voltage pulse, and - a time interval between the triggering switching process and the terminating switching process of the voltage pulse describes a pulse width of the voltage pulse, wherein - different switching time adjustments and different and in particular variable delay times are provided for the triggering switching process and the terminating switching process in order to control the pulse width as a result.

7. The method as claimed in one of the preceding claims, characterized in that - in each case three sub-rectifier arrangements (200), namely in each case a sub-rectifier arrangement of a first, second and third generator current phase, together with a network-based sub-inverter arrangement form a back-to-back sub-converter, - each back-to-back sub-converter has a common DC link to which the sub-rectifiers of a respective sub-rectifier arrangement rectify and from which the sub-inverter arrangement inverts, and - the DC links of a plurality of back-to-back sub-converters are coupled in order to permit circulating currents via said coupled DC links.

8. A wind power installation (100) comprising - a generator (101) for generating a generator current with one or more generator current phases, and - an active rectifier for rectifying and controlling the generator current, wherein for each generator current phase the rectifier - has a plurality of controllable sub-rectifiers (201-203), - each controllable sub-rectifier (201-203) has a partial inductance (231-233), and - each controllable sub-rectifier (201-203) controls a partial current of the generator current phase and each generator current phase forms a summation current as a sum of all the partial currents of the relevant generator current phase, characterized in that - the wind power installation (100) has a control unit (300) for controlling the active rectifier, and wherein the control unit (300) is set up in such a way that - the active rectifier operates according to a hysteresis method and is controlled so that for each generator current phase - the summation current is detected and - a tolerance band with an upper and a lower band limit is prescribed for each summation current and - each sub-rectifier (201-203) controls the partial current thereof depending on whether the summation current reaches the upper or lower band limit, wherein - each sub-rectifier (201-203) has at least one switching means in order to control the partial current thereof by way of switching the switching means and - each sub-rectifier (201-203) is assigned an individual delay time, and - each sub-rectifier (201-203) controls the partial current thereof additionally depending on said individual delay time, and that - each sub-rectifier (201-203) switches at least one of the switching means thereof, - after the summation current has reached an or the upper or lower band limit, and - the individual delay time has elapsed.