Methods and systems for fault ride through

Abstract

According to an aspect of the invention there is provided a method of operating a renewable energy power plant connected to a power network The method comprises: obtaining a frequency signal at a power plant controller, the frequency signal being indicative of a frequency level of the power network; controlling at least one renewable energy generator of the renewable energy power plant in a normal mode of operation based on the frequency level of the power network, the normal mode of operation being executed by the power plant controller via one or more local controllers of the at least one renewable energy generator; and in the event of a power network fault: controlling the at least one renewable energy generator according to a fault ride through mode of operation using the one or more local controllers; determining if the frequency signal is reliable or unreliable based on a comparison to one or more reliability conditions; preventing resumption of the normal mode of operation while the frequency signal is in an unreliable condition; and resuming the normal mode of operation once the fault has cleared and the frequency signal is in a reliable condition.

Description

[0001] METHODS AND SYSTEMS FOR FAULT RIDE THROUGH

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to methods and systems for fault ride through of a renewable energy power plant connected to a power network.

[0004] BACKGROUND

[0005] Electric transmission and / or distribution networks, herein referred to as “power networks” or "electric grids", normally operate in a voltage range around a nominal voltage value (between a lower threshold and an upper threshold). The lower threshold is, for example, in a range from about -5% to about -20% and the upper threshold is, for example in a range from about +5% to about +20% of the nominal voltage. For example, in a 110 kV network the operating range may be between 96 and 123 kV, in a 220 kV network it may be between 193 and 245 kV, and in a 380 kV network it may be between 350 and 420 kV.

[0006] A low-voltage grid fault (or low-voltage network fault) is a drop of the grid voltage below the nominal grid-operating range, which may be caused by a short-circuit, for example. Typically, such low-voltage grid faults are cleared by the network within about 100ms to 1000ms. For example, a low-voltage grid fault may be cleared by disconnecting the branch in which the fault or short-circuit occurred.

[0007] However, in many countries, renewable energy power plants are required to ride through such faults, rather than shut-down, so as to be able to resume electricity production when, or shortly after, the grid voltage has returned to the nominal grid-operating range. This is known as 'Low-Voltage Ride Through’ (LVRT), or briefly 'Fault Ride Through' (FRT). The renewable energy generators of the power plant therefore remain operational and connected to the grid during and after the low-voltage grid fault. In some cases, the renewable energy generators may even be required to support the recovery of the grid, for example by injecting additional reactive current (capacitive current) during the low-voltage grid fault.

[0008] During normal modes of operation (i.e. when the grid voltage is within the nominal gridoperating range) the individual energy generators of a power plant are typically controlled by a central power plant controller. For example, the power plant controller may provide reference values (i.e. target values or "set points") for one or more electricproduction parameters, such as the active power, the reactive power, and / or the voltage, to be produced by the individual renewable energy generators of the power plant. These reference values or setpoints may be dispatched to local controllers, associated with respective ones of the renewable energy generators, that operate the generators to implement the set points received from the power plant controller.

[0009] However, in the event of a low-voltage grid fault, the centralised control of the power plant controller is typically replaced by self-directed control carried out by the local controllers of the renewable energy generators (in order to ride through the low-voltage fault). This is also referred to as "local control". A reason for changing from central to local control may be, for example, that data communication between the central power plant controller and the local generator controllers is too slow to cope with the responsetime demand during the fault, which is typically in the order of about 5 to 20 milliseconds.

[0010] A fault may be considered to be cleared when the grid voltage has returned to the lower limit of the nominal grid-operating range, or to a lower voltage-return threshold for ensuring a hysteresis effect. Conventionally, when the fault has been cleared (and after a short transition period), the fault-ride-through operation is terminated, and normal operation is resumed under the central control of the power plant controller using the pre-fault reference values. However, when resuming the normal mode of operation, there have been occurrences of unexpected events, such as shut down of the plant.

[0011] It is an aim of the present invention to address one or more of the disadvantages associated with the prior art.

[0012] SUMMARY OF THE INVENTION

[0013] According to an aspect of the invention, there is provided a method of operating a renewable energy power plant connected to a power network. The method comprises: obtaining a frequency signal at a power plant controller, the frequency signal being indicative of a frequency level of the power network; controlling at least one renewable energy generator of the renewable energy power plant in a normal mode of operation based on the frequency level of the power network, the normal mode of operation being executed by the power plant controller via one or more local controllers of the at least one renewable energy generator; and in the event of a power network fault: controlling the at least one renewable energy generator according to a fault ride through mode of operation using the one or more local controllers; determining if the frequency signal is reliable or unreliable based on a comparison to one or more reliability conditions; preventing resumption of the normal mode of operation while the frequency signal is in an unreliable condition; and resuming the normal mode of operation once the fault has cleared and the frequency signal is in a reliable condition.

[0014] In this manner, the renewable energy generators are controlled to ride through a power network fault, such as a low voltage grid fault, and the frequency signal reliability is assessed (once the fault has cleared) before the normal mode of operation is resumed. This is significant because the frequency signal can include large amounts of noise, particularly following a low voltage grid fault, compromising the reliability of the frequency signal and leading to unexpected events when the normal mode of operation is resumed. The method advantageously mitigates this issue by checking the condition of the frequency signal and preventing resumption of the normal mode of operation while the frequency signal is compromised. It is envisaged that the method will reduce or avoid unexpected shutdowns of the plant once the fault has been cleared, leading to increased overall revenue generation.

[0015] In an example, each renewable energy generator of the renewable energy power plant may be controlled in a normal mode of operation by: determining an active power reference value based, at least in part, on the frequency level of the power network; and dispatching the active power reference value from the power plant controller to the one or more local controllers. Here it shall be appreciated that the active power reference value may be determined based on the frequency level of the power network, amongst other inputs, including an (external) active power reference for the power plant that may be provided by a grid operator, a transmission system operator, or a distribution system operator for example.

[0016] The fault ride through mode of operation may, for example, comprise: freezing, upon entering the fault ride through mode, a state of the power plant controller so as to stop determining and / or dispatching the active power reference value; storing a pre-fault- operation value of at least one control variable for determining the active power reference value, and controlling the at least one renewable energy generator to ride through the fault using the one or more local controllers. The control may be resumed in the normal mode of operation using the at-least-one stored control-variable value from pre-fault operation. In this manner, the power plant controller makes advantageous use of the pre-fault control variables to determine suitable active power reference values based on the frequency signal when the normal mode of operation is resumed.

[0017] Optionally, the one or more reliability conditions may include a first frequency threshold. Determining if the frequency signal is reliable or unreliable may, for example, comprise: determining whether the frequency signal is greater than or equal to the first frequency threshold. In this manner, an unreliable condition of the frequency signal may be determined when the frequency level exceeds the first frequency threshold (due to excessive noise), precluding resumption of the normal mode of operation.

[0018] Optionally, the one or more reliability conditions may include a second frequency threshold. Determining if the frequency signal is reliable or unreliable may, for example, comprise: determining whether the frequency signal is less than or equal to the second frequency threshold. In this manner, an unreliable condition of the frequency signal may similarly be determined when the frequency level falls below the second frequency threshold (again due to dominant noise), precluding resumption of the normal mode of operation.

[0019] Optionally, the one or more reliability conditions may include a frequency rate of change threshold. Determining if the frequency signal is reliable or unreliable may, for example, comprise: determining a rate of change of the frequency signal; and determining whether the rate of change of the frequency signal is greater than or equal to the frequency rate of change threshold. Again, an unreliable condition of the frequency signal may therefore be determined when the frequency signal is rapidly fluctuating, changing at a greater rate than the threshold (due to noise effects), precluding resumption of the normal mode of operation.

[0020] In an example, the unreliable condition may be determined if: the frequency signal is greater than or equal to the first frequency threshold; the frequency signal is less than or equal to the second frequency threshold; or the rate of change of the frequency signal is greater than or equal to the frequency rate of change threshold. Any such condition may be indicative that the frequency signal is compromised due to measurement noise, such that controlling the power plant based thereon would lead to unexpected or undesirable operations.

[0021] In an example, the reliable condition may be determined if: the frequency signal is less than the first frequency threshold; the frequency signal is greater than the second frequency threshold; and the rate of change of the frequency signal is less than the frequency rate of change threshold. All three conditions may therefore be required to satisfy the reliable condition and resume the normal mode of operation.

[0022] Optionally, resumption of the normal mode of operation may be prevented until a minimum hold period has elapsed since determining that the frequency signal has changed from the unreliable condition to the reliable condition. The minimum hold period may therefore be used to ensure that the reliable condition of the frequency signal is stable and suitable for use in controlling the power plant.

[0023] In an example, the frequency signal may be determined based on a series of measurements, obtained by the power plant controller, indicative of a frequency level of the power network. The frequency signal may be determined by: limiting an indicated frequency level of the obtained series of measurements based on one or more frequency limits; and limiting a rate of change of the obtained series of measurements based on one or more frequency rate of change limits. Advantageously, the combined application of frequency saturation limit(s) and rate limit(s) provides effective noise mitigation in the determined frequency signal, facilitating more accurate and effective control of the power plant.

[0024] The one or more frequency limits may, for example, comprise: a frequency limit corresponding to the first frequency threshold; and / or a frequency limit corresponding to the second frequency threshold. As noted above, such thresholds may be suitably set for problematic noise levels.

[0025] Optionally, determining the frequency signal may further comprise filtering the obtained series of measurements using an anti-aliasing filter. The combination of filtering and conditioning the frequency signal provides for even more effective noise mitigation. According to another aspect of the invention, there is provided a power plant controller for a renewable energy power plant connected to a power network. The power plant controller is configured to execute machine readable instructions to: obtain a frequency signal indicative of a frequency level of the power network; control at least one renewable energy generator of the renewable energy power plant in a normal mode of operation based on the frequency level of the power network, the normal mode of operation being executed via one or more local controllers of the at least one renewable energy generator; and in the event of a power network fault: enter a fault ride through mode of operation, during which the at least one renewable energy generator is controlled by the one or more local controllers; determine if the frequency signal is reliable or unreliable based on a comparison to one or more reliability conditions; prevent resumption of the normal mode of operation while the frequency signal is in an unreliable condition; and resume the normal mode of operation once the fault has cleared and the frequency signal is in a reliable condition.

[0026] According to yet another aspect of the invention, there is provided a method of operating a renewable energy power plant connected to a power network. The renewable energy power plant comprises one or more renewable energy generators. The method comprises: obtaining a series of measurements indicative of a frequency level of the power network; determining a frequency signal based on the obtained series of measurements, the frequency signal being determined by: limiting a frequency level of the obtained series of measurements based on one or more frequency limits; and limiting a rate of change of the obtained series of measurements based on one or more frequency rate of change limits; and controlling the one or more renewable energy generators based on the determined frequency signal.

[0027] In this manner, the method applies a combination of frequency saturation limit(s) and rate limit(s) to effectively mitigate noise in the frequency signal that is determined from the series of measurements (which may be obtained for a point of interconnection of the power plant to the power network, for example). The determined frequency signal is therefore less susceptible to variation due to measurement noise (which can be particularly prevalent following a grid fault) and the method therefore provides for more accurate and effective control of the renewable energy generators. Optionally, determining the frequency signal further comprises filtering the obtained series of frequency measurements using an anti-aliasing filter. As noted above, the combination of filtering and conditioning the frequency signal provides for even more effective noise mitigation.

[0028] According to a further aspect of the invention, there is provided a power plant controller for a renewable energy power plant connected to a power network. The power plant controller is configured to execute machine readable instructions to: receive a series of measurements indicative of a frequency level of the power network; determine a frequency signal based on the obtained series of measurements, the frequency signal being determined by: limiting a frequency level of the obtained series of measurements based on one or more frequency limits; and limiting a rate of change of the obtained series of measurements based on one or more frequency rate of change limits; and control the one or more renewable energy generators based on the determined frequency signal.

[0029] Within the scope of this invention it is expressly intended that the various aspects, embodiments, examples and alternatives set out in the preceding paragraphs, in the claims and / or in the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination. That is, all embodiments and / or features of any embodiment can be combined in any way and / or combination, unless such features are incompatible. The applicant reserves the right to change any originally filed claim or file any new claim accordingly, including the right to amend any originally filed claim to depend from and / or incorporate any feature of any other claim although not originally claimed in that manner.

[0030] BRIEF DESCRIPTION OF THE DRAWINGS

[0031] One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0032] Figure 1 schematically shows a power network connected to a renewable energy power plant that includes a power plant controller;

[0033] Figure 2 shows an exemplary control structure of the power plant of Figure 1; Figure 3 shows an exemplary control scheme of a signal quality control module of the control structure shown in Figure 2;

[0034] Figure 4 shows another exemplary control scheme of the signal quality control module shown in Figure 2; and

[0035] Figure 5 shows an exemplary method of operating the renewable energy power plant of Figure 1 in accordance with an embodiment of the invention.

[0036] DETAILED DESCRIPTION

[0037] Embodiments of the present invention relates to methods and systems for controlling a renewable energy power plant to ride through a grid fault, such as a low voltage grid fault.

[0038] In this context, the inventors have found that the grid frequency provided to the power plant controller, for example from a power meter, contains high levels of noise, particularly in the event of a low voltage grid fault. The measurement noise can be so significant as to compromise the reliability of the frequency signal, leading to various unexpected events, such as plant shutdown, when the power plant controller resumes central control of the renewable energy generators following clearance of the fault.

[0039] To address this issue, the inventors have devised methods and systems for processing the grid frequency signal to mitigate noise, as well as methods and system for evaluating the reliability of the frequency signal before a normal mode of operation can be resumed following a fault clearance.

[0040] In a first aspect, the present invention therefore relates to methods and systems for determining a frequency signal based on a series of measurements that are indicative of a frequency level of the connected power network. Advantageously, the power plant controller is configured to receive such measurements and to apply: (i) one or more frequency limits for curbing abnormal measurements due to noise, and (ii) one or more rate limits for mitigating sudden changes in the determined frequency signal due to measurement noise. In examples, the power plant controller may also be configured to further apply one or more anti-aliasing filters to the measurements to thereby combine such noise mitigation techniques with signal conditioning.

[0041] Additionally or alternatively, in a second aspect, the present invention relates to methods and systems for controlling the power plant for low voltage ride through. In particular, the renewable energy generators are typically controlled according to a ‘normal mode of operation’ until a power network fault is detected. During the normal mode of operation, the power plant controller uses the frequency signal to determine and dispatch power references (such as active power references) to the local controllers of the renewable energy generators. In turn, the local controllers are configured to control the renewable energy generators according to the set points contained within the dispatched signals to produce corresponding amounts of power I current.

[0042] However, in the event of a power network fault, such as a low voltage fault, the renewable energy generators are controlled according to a fault ride through mode of operation instead.

[0043] In particular, a state of the power plant controller that determines and dispatches the active power reference values in the normal mode of operation may be frozen (upon entering the fault ride through mode), and a pre-fault-operation value of at least one control variable may be stored while the local controllers control the renewable energy generators to ride through the fault.

[0044] In a conventional system, the power plant controller subsequently resumes control using the stored pre-fault-operation value(s) once the fault has been cleared. Advantageously though, the methods and system of the present invention are further configured to determine if the frequency signal is reliable or unreliable based on a comparison to one or more reliability conditions, and to prevent resumption of the normal mode of operation if the frequency signal is in an unreliable condition. For example, the frequency signal may be compared to one or more reliability conditions, such as upper and / or lower frequency thresholds, in order to assess the effects of measurement noise and determine whether the frequency signal is reliable or unreliable once the fault has been cleared. The power plant controller only resumes the normal mode of operation following the fault once it is determined that the frequency signal is in a reliable condition. At that time, the power plant controller resumes providing active / reactive power control using the at- least-one stored control-variable value from pre-fault operation.

[0045] In this manner, it is envisaged that embodiments of the invention will provide fault ride through of the renewable energy power plant, and further reduce or avoid incurring unexpected shutdowns of the plant once the fault has been cleared.

[0046] Figure 1 illustrates a typical architecture in which a renewable energy power plant is connected to a main grid or wider power network. In the example shown in Figure 1 , the renewable energy power plant is a wind power plant (WPP). As will be understood by the skilled reader, a WPP comprises a plurality of wind turbine generators (WTG)s. A WTG is commonly referred to as a ‘wind turbine’. The example shown is representative only and the skilled reader will appreciate that other specific architectures are possible. In other examples, the power plant may include other renewable energy sources such as a solar power plant, a bio energy power plant, an ocean / wave / tidal energy plant, or a hybrid power plant having a combination of different types of renewable energy power plants. Thus, the invention relates to renewable energy power plants and renewable energy generators in general, rather than being specific to wind power plants and generators as in the Figures. The components of the wind power plant and power network are conventional and as such would be familiar to the skilled reader. It is expected that other known components may be incorporated in addition, or as alternatives, to the components shown and described in Figure 1. Such changes would be within the capabilities of the skilled person.

[0047] Figure 1 shows a power system 10 incorporating a WPP 12. In this example, the WPP 12 includes a plurality of WTGs 14 and a power plant controller 22, referred to hereafter as PPG 22. Each of the plurality of WTGs 14 converts wind energy into electrical energy, which is transferred from the WPP 12 to a main power network, or ‘main grid’ 16, as active power and / or current, for distribution.

[0048] Although not illustrated in Figure 1 , the WPP 12 also includes compensation equipment, such as a static synchronous compensator (STATCOM) or another type of synchronous compensator, configured to provide reactive power or reactive current support as required.

[0049] Each of the WTGs 14 is associated with a respective local WTG controller 15. In other implementations, a set of WTGs may share a single, semi-centralised WTG controller, such that there are fewer WTG controllers than WTGs. As will be understood by the skilled person, WTG controllers 15 can be considered to be local computer systems capable of operating a WTG 14 in the manner prescribed herein, and may comprise multiple modules that control individual components of the WTG or just a single controller. The computer system of the WTG controller 15 may operate according to software downloaded via a communications network or programmed onto it from a computer-readable storage medium.

[0050] The WPP 12 also includes a connecting network 18 for connecting the WPP 12 to the main grid 16 (also called the main power network). In this example, the WPP 12 and the main grid 16 are connected at a Point of Interconnection (Pol) 20, which is an interface between the WPP 12 and the main grid 16. The Pol 20 may also be referred to as the Point of Common Coupling, which may be abbreviated to ‘PCC’ or ‘PoCC’.

[0051] The Power Plant Controller (PPC) 22 is connected to the main grid 16 at a Point of Measurement (PoM) 24, such as a power meter, and is connected to each of the WTG controllers 15. For example, the PPC 22 may be configured to receive one or more measurement signals from the PoM 24 comprising measurements of the power supply from the WPP 12 to the main grid 16 and / or a frequency level of the main grid 16. The role of the PPC 22 is to act as a command and control interface between the WPP 12 and the grid 16, and more specifically, between the WPP 12 and a grid operator 26, such as a transmission system operator (TSO) or a distribution system operator (DSO). The WPP 12 is capable of altering its power or current output in reaction to set points received from the PPC 22.

[0052] The PPC 22 is a suitable computer system for carrying out the controls and commands as described herein and so may incorporate a processing module 28, a connectivity module 30, a memory module 32 and a sensing module 34, amongst others, as shown in Figure 1. The connectivity module 30, the memory module 32, and / or the sensing module 34 are configured to provide the processing module 28 with information that is indicative of a frequency level of the main grid 16, as well as power levels, current levels and / or voltage levels of the WTGs 14 and / or the WPP 12.

[0053] For example, the sensing module 34 may receive such information directly from one or more connected sensors or power meters (e.g. at the PoM 24) and communicate the information to the processing module 28. Alternatively, or additionally, the information may be determined by one or more systems that are connected to the connectivity module 30, such as the WTG controllers 15, and the information may be communicated, in turn, through the connectivity module 30 to the processing module 28. In each case, the determined information may be stored permanently, or temporarily, in the memory module 32, from which it may be recalled, on demand, by the processing module 28. The PPG 22 may also receive information regarding the grid 16 and / or local buses, substations and networks from an energy management system (not shown).

[0054] In the absence of a fault, the WPP 12 is operated according to a normal mode of operation, during which the PPG 22 generates and sends dispatch signals to the WTG controllers 15. The dispatched signals contain active and reactive current, and / or power, set points determined by the PPG 22 based on the measurements received from the PoM 24, providing frequency and voltage support to the main grid 16. The WTG controllers 15 control the WTGs 14 according to the set points contained within the dispatch signals.

[0055] For example, the PPG 22 and the local WTG controllers 15 may each be arranged to work in a feedback mode in which they compare a reference value, e.g. from reference inputs, with a measured value, e.g. from measurement inputs, and produce a control signal based on the difference between the two input values. A bi-directional control network may be arranged between the PPG 22 and the WTG controllers 15 enabling two-way communication. For example, the uplink direction (i.e. the direction from the central PPG 22 to the local WTG controllers 15) is used to send reference values, e.g., for voltage and / or reactive power, from the PPG 22 to the local WTG controllers 15. The downlink direction may be used by the WTGs 15 to return information about their current operational state, e.g. about the amount of active power currently produced, to the central PPC 22. Such a control network may, for example, be implemented as a bus system, i.e. a CAN bus (ISO 11898) or an Ethernet bus (IEEE 802.3).

[0056] During extraordinary conditions (e.g., where the network voltage deviates from the contingency deadband), the WTG controllers 15 operate independently of the PPC 22 to fulfil predetermined network requirements. They also act to protect the WTGs 14 from any potentially harmful conditions, such as overspeed conditions in high winds.

[0057] For example, in a low-voltage event, during which the voltage level of the main grid 16 deviates from the nominal grid-operating range, the WPP 12 is operated according to a fault ride-through mode of operation instead of the normal mode of operation.

[0058] Upon entering the fault ride through mode, a state of the PPC 22 that determines and dispatches the set points in the normal mode of operation is frozen, stopping the determination and / or dispatch of active power reference values. At this point, one or more pre-fault-operation values of the control variables for determining the active power reference value may also be stored for use once the fault is cleared.

[0059] The local WTG controllers 15 therefore stop receiving updated set points form the PPC 22 and, instead, the WTG controllers 15 execute stored instruction to control the WTGs 14 to ride through the fault (under local control).

[0060] Once the fault eventually clears, the normal mode of operation is resumed by the PPC 22, which is unfrozen, and further set points are determined and dispatched to the WTG controllers 15 using the at-least-one stored control-variable value from the pre-fault operation. However, following the fault, the grid frequency signal often includes significant amounts of noise, which can dominate the signal to the extent that unexpected events occur when the normal control is resumed.

[0061] In contrast to conventional approaches, embodiments of the present invention feature a control structure that is advantageously further configured to determine whether the frequency signal is reliable or unreliable (due to measurement noise) and to prevent resumption of the normal mode of operation until a reliable condition of the frequency signal is determined. An exemplary control structure for providing such low-voltage ride through shall now be described in more detail with additional reference to Figures 2 to 4.

[0062] As shown in Figure 2, the overall controller structure of the WPP 12 comprises the central PPC 22 and the local WTG controllers 15. For simplicity, a single WTG controller 15 is shown in Figure 2, but it shall be appreciated that the PPC 22 connects to each local WTG controller 15 in a substantially identical manner. In the embodiment of Figure 2, an optional STATCOM is also provided with a STATCOM controller 33.

[0063] A measurement evaluator 34 forms part of the central PPC 22 in this example and produces signals indicative of the reactive power, Q(feedback); the grid-voltage, V(feedback); the grid frequency, f(feedback); and / or the active power, P(feedback); based on measurements received from the PoM 24. For example, the measurement evaluator 34 may be configured to receive a time-varying series of measurements indicative of one or more power characteristics at the PoM 24, such as a voltage and / or current level, and to determine the respective output signals based thereon.

[0064] The measurement evaluator 34 includes a signal quality control module 32, as shown in Figure 2, configured to determine the grid frequency signal f(feedback), along with a frequency signal condition, f(condition).

[0065] The signal quality control module 32 is configured to determine the grid frequency signal f(feedback) based on a series of the obtained measurements and applies respective frequency limits, rate of change limits, and / or filters, for conditioning the signal and removing or mitigating significant noise components. The determined grid frequency signal, f(feedback), is subsequently used in one or more active power control loops to regulate the active power output of the WPP 12.

[0066] The signal quality control module 32 is further configured to determine whether the frequency signal is reliable or unreliable (i.e. noise-comprised) in order to inform the LVRT control of the WTGs 14 (LVRT stand for "low voltage ride through"). In particular, as shall be discussed in more detail, the signal quality control module 32 is configured to determine a signal, f(condition), indicative of whether the frequency signal, f(feedback), is reliable or unreliable due to measurement noise, and the frequency condition signal, f(condition), is output to control the resumption of a normal mode of operation once the grid fault has been cleared.

[0067] In order to determine such grid faults, the measurement evaluator 34 may also comprise an LVRT detector 38, as shown in Figure 2, arranged to detect voltage drops. If the LVRT detector 38 ascertains that the voltage drops below a fault threshold (and persists there for a minimum-time interval), the LVRT detector 38 may produce an LVRT flag 39 indicating that a fault has happened. That is, the LVTR flag 39 "On" is activated upon detecting a low voltage condition; e.g. if the voltage at the PoM 24 is below a low-voltage threshold (e.g. the lower level of the nominal grid-operating range). This acts as an interrupt and the LVRT flag 39 therefore affects the subsequent operation of the PPG 22 and the control of the WPP 12.

[0068] In this respect, the central PPG 22 is shown to further include: an LVRT control module 35, a reactive- power controller 36 (briefly "Q controller"), and an active-power controller 37 (briefly "P controller").

[0069] The LVRT control module 35 is configured to receive the LVRT flag 39 and the frequency signal condition, f(condition), and to control the functionalities of the Q controller 36 and the P controller 37 accordingly. For example, in the event of a grid fault indicated by the LVRT flag 39, the LVRT control module 35 is configured to execute instructions to switch from a normal mode of operation to a fault ride through mode of operation. Meanwhile, when the LVRT flag 39 indicates that the grid fault is cleared, the LVRT control module 35 is configured to check whether the frequency condition signal, f(condition), further indicates a reliable condition of the frequency signal, f(feedback) before executing instructions to resume the normal mode of operation.

[0070] It shall be appreciated that embodiments of the invention are therefore principally concerned with the conditions for switching from the LVRT mode of operation to the normal mode of operation. Nonetheless, an exemplary normal mode of operation is described briefly below to establish the operational changes that may be implemented when a fault is detected. In this respect, it shall be appreciated that the Q controller 36 and the P controller 37 are not described in detail here to avoid obscuring the invention. During normal operation of the WPP 12, the Q controller 36 receives the signals produced by the measurement evaluator 34 which are relevant for Q control, e.g. Q(feedback) and / or V(feedback). The Q controller 36 may also receive external control inputs for active power, reactive power, and / or voltage control to be achieved at the PoM 24, and / or the slope of a droop function defining the mapping of the measured voltage to reactive power to be produced. In order to determine reference values for controlling the WTGs 14, the Q controller 36 comprises a V-Q-PF controller 40 and a Q set-point dispatcher 41. The V-Q-PF controller 40 produces an internal overall reference value, Qref, for the reactive power to be produced by the WPP 12, based on the inputs to the Q controller 36. "V" stands for voltage, "Q" for reactive power, and "PF" for power factor; "V-Q-PF" thereby indicates that the V-Q-PF controller 40 is enabled to receive V, Q, and / or PF, as an external reference, and to produce the internal reference value, Q ref, based on such external reference. The Q set point dispatcher 41 splits the overall Qref into individual reactive power setpoints, Qset, for the local WTG controllers 15 of the WTGs 14, and the STATCOM controller 33, if applicable. The Q set point dispatcher 41 also receives signals, Qavail, from the local WTG controllers 15 and, if applicable, the STATCOM controller 33 indicative of the available reactive power, i.e. that amount of reactive power that could currently be produced at maximum by the individual WTGs 14, and by the STATCOM, if applicable. Signalling of the available reactive power, Qavail, enables the Q set point dispatcher 41 to split up the overall, Qref, to be produced according to the abilities of the individual WTGs 14 (and the STATCOM, if applicable) without having to limit the overall P production of the WPP 12.

[0071] As indicated at the outset, the notion of "reactive power" includes other related parameters, such as reactive current, power factor, etc. Therefore, the reference value, Qref, may also indicate an amount of reactive current to be produced by the by the WPP 12, and Qset may indicate reactive current setpoints. Thus, the WTGs 14 and / or the STATCOM may also be commanded with reactive-current setpoints instead of reactivepower (and, optionally, active-power) setpoints.

[0072] Turning to the P controller 37, during normal operation, the P controller 37 similarly receives the signals produced by the measurement evaluator 34 which are relevant for P control, such as P(feedback) and f(feedback). The P controller 37 may also receive signals, Pavail, from the local WTG controllers 15 indicative of the available active power, i.e. that amount of active power that could currently be produced at maximum by the individual WTGs 14. In some embodiments, the P controller 37also receives external control inputs, e.g. a curtailment signal from the grid provider by which the grid provider can prescribe to what extent P production of the WPP 12 shall be curtailed. The curtailment prescription may be expressed in absolute terms (e.g. in MW), or in relative terms, e.g. as a percentage of the plant's nominal active power or the active power currently producible by the WPP 12, that is the sum of the Pavail. In order to determine reference values for controlling the WTGs 14, the P controller 37 comprises a frequency (f) and active-power (P) controller 42 and a P set-point dispatcher 43. The f-and-P controller 42 produces an internal overall reference value, Pref, for the active power to be produced by the WPP 12, based on the inputs to the P controller 36 from the measurement evaluator 34, f(feedback) and P(feedback). For example, when the grid frequency f(feedback) is above a frequency threshold, the f-and-P controller 42 may generate a reduced value of Pref. If the WPP 12 operates in a curtailed mode, the f-and- P controller 42 may also generate an increased value of Pref when the grid frequency, f(feedback), is below a frequency threshold, in order to contribute to frequency control of the grid. In embodiments with an external curtailment-functionality, the external curtailment signal also influences the evaluation of Pref. For example, if the signal demands that the active power produced shall be x % of the active power currently producible by the WPP 12, Pref will be set to x % of the sum of all Pavail. The P set point dispatcher 43 splits the overall Pref into individual active power set- points, Pset, for the local WTG controllers 15, using the Pavail information. The splitting does not have to be uniform, for example WTGs 14 of the first row subjected to higher load could be curtailed more than those with lower load.

[0073] The active and reactive power set points, Pset and Qset, are dispatched to the respective local WTG controllers 15 during the normal mode of operation, which control the respective WTGs 14 accordingly to produce corresponding amounts of active and / or reactive power.

[0074] For this purpose, the local WTG controllers 15 include a local Q controller 44 and a local P controller 45, which shall also be described briefly to avoid obscuring the invention. The STATCOM controller 33 may also have a local Q controller 44, if applicable. The STATCOM is used inside the WPP 12 to regulate the reactive power according to the references sent by the central PPG 22. Regarding reactive-power production, the STATCOM can be considered as another "renewable energy generator"; therefore the same control configuration will apply to it as for the WTGs 14.

[0075] During the normal mode of operation, the local Q controllers 44 receive their individual Qset value as an input. Another input is a signal indicative of the reactive power measured at the output of the respective WTG 14 (or the STACOM), referred to as "Q(feedback)" in Figure 2. The local Q controllers 44 produce control signals for the WTGs 14 (or the STATCOM), particularly the converters thereof, which cause the respective WTGs 14 to produce an amount of reactive power according to the individual value of Qset. A function of the local WTG controllers 15 referred to as "Qavail" determines the amount of reactive power that could be produced at maximum, e.g. by evaluating the current margin which is left for reactive power production for the WTG 14 under the prevailing operation conditions. It produces the Qavail signal, which is fed back to the PPC 22.

[0076] The local P controllers 45 similarly receive their individual Pset value as an input during the normal mode of operation. Another input is a signal indicative of the active power measured at the output of the respective WTG 14, referred to as "P(feedback)" in Figure 2. The local P controllers 45 produce control signals for the WTGs 14, particularly the converters thereof, to produce an amount of active power according to the individual value of Pset. A function of the local WTG controllers 15 referred to as "Pavail" determines the amount of reactive power that could be produced at maximum. For example, each WTG controller 15 may measure and / or determine the currently prevailing wind speed and calculate, for example based on a power curve of the WTG 14, what amount of active power could be produced under the prevailing wind condition. It produces the Pavail signal, which is fed back to the PPC 22.

[0077] Now the configuration of the control structure shall be described for providing fault ride through, when a fault, such as a low voltage grid fault, is detected.

[0078] As mentioned, if a low-voltage condition is detected by the LVRT detector 38 of the measurement evaluator 34, the measurement evaluator 34 will send a signal to the LVRT control module 35 of the PPC 22, e.g. sending the signal LVTR flag 39 "On". Upon receiving the LVRT flag 39 “On”, the LVRT control module 35 is configured to interrupt the normal-operation software of the PPG 22 and to enter the LVRT mode. A state of the PPG 22 is therefore frozen upon entering the LVRT mode, stopping the determination and / or dispatch of active / reactive power reference values I set points, and one or more control variables are stored in a memory of the PPG 22. For example, the pre-fault values for the integral components in frequency and active power controllers may be saved. The values saved can be the current ones; in other embodiments values of the integral components recorded at the pervious cycle step are saved. Using the values from previous sample may more precisely reflect the pre-fault values. Along with the saving of the integral components, the current values of active power references are saved to enable these references to be frozen to these saved values later.

[0079] At this time, in the absence of updated set points from the PPG 22 (or receipt of zero power set points), the WTG controllers 15 (and the STATCOM controller, if applicable) take over local control to ride through the fault. For example, the local WTG controllers 15 may each calculate the amount of current to be injected, according to their own local low-voltage measurement determined at the output of that WTG 14, and control the respective WTG 14 accordingly.

[0080] This manner of operation continues while the LVRT ON signal remains. The LVRT flag 39 "Off” may be activated by the LVRT detector 38 upon determining a normal voltage condition at the PoM 24. For example, the normal voltage condition may be determined when the voltage at the PoM 24 is greater than or equal to a voltage-return threshold, such as a lower threshold of the nominal grid-operating range. However, even once the normal voltage condition is received, the LVRT control module 35 continues to operate the WPP 12 in the fault ride through mode of operation until it is further confirmed that the frequency signal, f(frequency), is in a reliable condition.

[0081] An exemplary implementation of the interrupt can therefore be as follows: the LVTR flag 39 "On" signal is an LVRT ON trigger; i.e. activates an interrupt in the plant-controller software, and a second, or static LVTR flag 39 "On" signal will be latched internally in the LVRT control module 35. This static signal may hold the value, even if the LVRT flag “Off” is received, until the reliable condition of the frequency signal is detected. At that time, the LVRT control module 35 may generate a Pause-Release command, and the normal mode of operation is resumed. When the PPC 22 transitions to the normal mode of operation, a locking mechanism activated to lock the normal-operation references during the LVRT mode of operation stage may be unlocked, and the normal-operation reference values are re-established. The active power references to the WTGs 14 are therefore released from freezing, for example along with the integral component of the P / f controller 42. Reference values relating to active-power set-points provided by the PPC 22, such as curtailment production or frequency control, are now changed towards reference values corresponding to normal operation. In this manner, the PPC 22 is operable is resume determining and dispatching active power set points to the WTG controllers 15 for centralised control, where the active power set points are determined based, at least in part, on the grid frequency signal, f(frequency), which has been checked for reliability.

[0082] As shall be appreciated from the above, the present invention is particularly concerned with improving the quality of the grid frequency signal, f(feedback), and the conditions for issuing the pause-release command when a fault condition has cleared. In this exemplary embodiment, such functions are performed by the signal quality control module 32, which shall now be discussed in more detailed with additional reference to Figures 3 and 4.

[0083] As shown in Figure 3, the signal quality control module 32 may include a saturation block 102, a rate limiter block 104, and / or an anti-aliasing filter block 106, for determining the frequency signal. That is, in the described example, the signal quality control module 32 includes three major functional elements, units or modules for determining the frequency signal, f(frequency). Each of these units or modules may be provided by suitable software running on any suitable computing substrate using conventional or custom processors and memory.

[0084] The saturation block 102 receives an input of a first signal, FrqOut, that is indicative of the frequency level of the main grid 16, derived from the series of measurements obtained from the PoM 24 (i.e. derived from the voltage and / or current measurements). For example, the current and / or voltage measurements may be obtained by the measurement evaluator 34 at a particular sampling rate, and the measurement evaluator 34 may determine a corresponding frequency level of the power network 16 based on the obtained series of measurements, which forms the first signal, FrqOut. In general, the frequency level should remain within an acceptable operating range around a nominal frequency of the grid, such as 50Hz + / - 0.1 Hz. However, the indicated frequency level may vary significantly due to noise, particularly in the event of a low voltage fault. For example, the measured noise may be relatively large compared to the voltage of the power network during a low voltage fault, introducing high and / or low frequency oscillations that dominate the first signal, FrqOut, provided as an input to the saturation block 102.

[0085] The saturation block 102 is therefore configured to remove or mitigate the relatively high or low frequency oscillations from the first signal, FrqOut, which arise due to the measured noise. This may be achieved by limiting the frequency level of the first signal, FrqOut, according to upper and lower frequency limits, Fsat_Upper and Fsat_Lower, as shown in Figure 3. For this purpose, the saturation block 102 may therefore also receive the upper and / or lower frequency limits as respective inputs configured to remove measurement noise. Such inputs may be provided by an operator, such as the TSO 28 for example, or from another control input. The upper and lower frequency limits may be + / - 6% respectively of the nominal frequency, for example. Accordingly, for a 50 Hz nominal grid frequency, the upper frequency limit may be an upper saturation limit of 53 Hz, removing higher frequency noise. The lower frequency limit may be a lower saturation limit of 47 Hz, for example, for removing lower frequency noise.

[0086] The output of the saturation block 102 is therefore a second signal, Frq_sat, which is indicative of the frequency level of the main grid 16 and limited to frequency values between the upper and lower frequency limits. The second signal, Frq_sat, is provided as an input to the rate limiter block 104.

[0087] The rate limiter block 104 is configured to remove or mitigate sudden changes or fluctuations of the indicated frequency level that arise due to the noise, for example due to spikes or troughs in the measured frequency level due to noise. The rate limiter block 104 is therefore configured to limit the rate of change of the frequency level of the second signal, Frq_sat, using upper and lower rate limits, Upper_RateLimiter and Lower_RateLimiter, as shown in Figure 3. For this purpose, the rate limiter block 104 may therefore also receive the upper and / or lower rate limits as respective inputs configured to remove or mitigate the changes due to measurement noise. Again, such inputs may be provided by an operator, such as the TSO 28 for example, or from another control input. The upper rate limit may therefore be a rate of change of 6 Hz / s, for example. The lower rate limit may be a rate of change of -6 Hz / s, i.e. a decrease of 6 Hz / s.

[0088] The output of the rate limiter block 104 is therefore a third signal, Frq_Sat+Rate, which is indicative of the frequency level of the main grid 16, limited to frequency values between the upper and lower frequency limits, and limited to a rate of change of frequency between the upper and lower rate limits. In this example, the third signal, , Frq_Sat+Rate, is provided as an input to a filter, specifically an anti-aliasing filter 106.

[0089] The anti-aliasing filter block 106 is configured to remove or mitigate aliasing effects in the third signal, Frq_Sat+Rate, due to the sampling of the voltage and / or current measurements. For this purpose, the anti-aliasing filter block 106 may therefore also receive the Anti-Aliasing filter frequency, AAF_Frequency, and / or the sampling time, Ts, as respective inputs for mitigating the aliasing effects, as shown in Figure 2. Again, such inputs may be provided by an operator, such as the TSO 28 for example, or from another control input. The third signal, Frq_Sat+Rate, may therefore be filtered according to one or more anti-aliasing filtering methods based on the Anti-Aliasing filter frequency, AAF_Frequency, and the sampling time, Ts.

[0090] The output of the anti-aliasing filter block 106 is the frequency signal, f(feedback), which is output to the PPG 22 for frequency-dependent control of the WTGs 14 during the normal mode of operation. In this manner, the output frequency signal, f(feedback), is suitably filtered to remove / mitigate aliasing effects and conditioned to remove / mitigate the effects of noise on the indicated frequency level of the main grid 16.

[0091] It is anticipated that the combination of filtering and conditioning of the frequency signal, f(feedback), will substantially remove the measured noise and produce a more stable frequency signal for controlling the WTGs 14, ultimately improving grid stability.

[0092] In spite of such filtering and conditioning, it shall be appreciated that the determined frequency signal, f(feedback), may still be influenced by, and therefore show some variation due to, measured noise during a low voltage grid fault. Such noise can still dominate the frequency signal, f(feedback), and affect the reliability of the frequency signal for use in controlling the power production of the WTGs 14. For example, the frequency signal may vary due to the noise in an unexpected manner in response to the active power control in the normal mode of operation.

[0093] To mitigate this, the signal quality control module 32 is further configured to determine whether the frequency signal, f(feedback), is reliable or not. For example, the signal quality control module 32 may be configured to assess whether the signal is noise compromised such that it cannot be used for controlling the WPP 12. In particular, while the frequency signal, f(feedback), remains in an unreliable condition (due to noise), the PPC 22 is prevented from resuming the normal mode of operation. An exemplary control scheme for determined whether the frequency signal, f(feedback), is reliable or not shall now be discussed in more detail with reference to Figure 4.

[0094] Figure 4 schematically illustrates an exemplary control scheme 400 for determining the frequency signal condition in accordance with embodiments of the invention.

[0095] In this example, the signal quality control module 32 determines the condition of the frequency signal, f(feedback), output from the measurement evaluator 34. However, this example is not intended to be limiting on the scope of the invention and, in other examples, the signal quality control module 32 may determine the condition of the frequency signal, f(feedback), in parallel, for example based on the obtained series of measurements received from the PoM 24.

[0096] In step 402, the signal quality control module 32 compares the frequency signal, f(feedback), to the one or more reliability conditions. By way of example, the reliability conditions may include that:

[0097] (i) the frequency signal, f(feedback), is greater than or equal to an upper frequency threshold, which may be equal to Fsat_Upper used in the saturation block 102;

[0098] (ii) the frequency signal, f(feedback), is less than or equal to a lower frequency threshold, which may be equal to Fsat_Lower used in the saturation block 102; and / or

[0099] (iii) the rate of change of the frequency signal, f(feedback), is greater than or equal to a rate limit threshold, which may be equal to the upper rate limit used in the rate limiter block 104 - i.e. the change in frequency value between samples should not be greater the rate limit threshold. If the frequency signal, f(feedback), satisfies any one or more of these conditions, the signal quality control module 32 may determine that the frequency signal is unreliable, i.e. in an unreliable condition. Accordingly, the unreliable condition of the frequency signal is output as a signal, f( condition), to the LVRT control module 35, in step 404, to prevent resumption of the normal mode of operation.

[0100] Thereafter, in step 406, the signal quality control module 32 monitors for a change in the reliability of the frequency signal, f(frequency) by comparing the frequency signal, f(feedback), to the same thresholds, but the opposite conditions, checking whether:

[0101] (i) the frequency signal is less than the upper frequency threshold and greater than the lower frequency threshold; and

[0102] (ii) the rate of change of the frequency signal is less than the rate limit threshold.

[0103] If the frequency signal, f(feedback) satisfies all three of these conditions, or a similar set of conditions, the module 32 determines that the frequency signal, f(feedback), is in a reliable condition, in step 408.

[0104] However, this may be a transient state and so, in examples, the signal quality control module 32 may continue to output the signal, f( condition), indicative of the unreliable condition (preventing resumption of the normal mode of operation) and start a timer, in step 410. The timer is used to monitor whether the reliable condition is maintained for a threshold period and so the signal quality control module 32 continues to monitor the frequency signal, f(feedback).

[0105] If the signal quality control module 32 subsequently determines that the frequency signal, f(feedback), is unreliable the timer is reset, in step 412, and the process returns to step 406. For example, this may occur if the signal quality control module 32 determines that:

[0106] (i) the frequency signal, f(feedback), is greater than or equal to the upper frequency threshold;

[0107] (ii) the frequency signal, f(feedback), is less than or equal to the lower frequency threshold; or (iii) the rate of change of the frequency signal, f(feedback), is greater than or equal to the rate limit threshold.

[0108] However, if the signal quality control module 32 determines that a threshold time period has elapsed since starting the timer, the signal quality control module 32 outputs a signal, f(condition), indicating the reliable condition of the frequency signal to the LVRT control module 35, in step 414. The PPC 22 is therefore able to resume the normal mode of operation providing the fault has cleared, i.e. if the LVRT control module 35 further receives the LVRT FLAG OFF.

[0109] In overview, a method of operating the WPP 12 therefore involves obtaining a grid frequency signal, controlling the WTGs 14 in a normal mode of operation or a fault ride through mode of operation in dependence on detecting a grid fault, and assessing the reliability of the frequency signal in order to generate pause-release commands once the grid fault has been resolved.

[0110] An exemplary method 500 of operating the WPP 12 in accordance with an embodiment of the invention shall now be described in more detail with additional reference to Figure 5.

[0111] In step 502, the PPC 22 obtains the frequency signal, f(feedback), indicative of the frequency level of the power network 16. The frequency signal, f(feedback), may be derived based on measurements of the current and / or voltage determined by a power meter at the PoM 24, for example.

[0112] In examples, the frequency signal, f(feedback) may be suitably processed by the signal quality control module 32 of the PPC 22, substantially as described in relation to Figure 3, to remove / mitigate noise and / or aliasing effects in the measurements. For example, the signal, FrqOut, may be derived from the series of measurements obtained from the PoM 24, and provided as an input to the signal quality control module 32. In turn, the signal quality control module 32 may pass the signal, FrqOut, through the saturation block 102, the rate limiter block 104, and / or the anti-aliasing filter block 106, to determine the frequency signal, f(feedback) for use in controlling the power output of the WTGs 14. In step 504, the PPC 22 therefore controls the WTGs 14 in a normal mode of operation based on the frequency level of the power network 16, as indicated by the frequency signal, f(feedback).

[0113] The Q controller 36 and the P controller 37 may therefore receive the respective input signals (Q(feedback), V(feedback), f(feedback), P(feedback)) from the measurement evaluator 34 and determine and dispatch corresponding active and reactive power set points, Pset, Qset, to the WTG controllers 15 for controlling the respective WTGs 14.

[0114] In turn, the WTG controllers 15 control the individual WTGs 14 to produce corresponding amounts of active and / or reactive power, providing frequency and / or voltage support to the main grid 16.

[0115] In step 506, the PPC 22 detects a grid fault, such as a low voltage fault. For example, the LVRT detector 38 of the PPC 22 may detect a drop of the voltage measurements below a fault threshold (e.g. which persists for a minimum-time interval), and the LVRT detector 38 may produce an LVRT flag 39 indicating that a fault has happened, i.e. activating the LVRT flag “On”.

[0116] In step 508, the normal-operation software of the PPC 22 and the WTG controllers 15 is interrupted, and the controllers enter a fault ride through mode of operation in response to the grid fault, such as a low-voltage ride through mode of operation. During the fault ride through mode of operation, the centralised PPC control is frozen or paused, and the WTGs 14 are operated under local control by the WTG controllers 15.

[0117] For example, in sub-step 510, a pause command may be generated by the LVRT control module 35, in response to the LVRT flag “On”. The pause command may be output to each of the P controller 37 and the Q controller 36 and held.

[0118] In sub-step 512, a state of each of the P controller 37 and the Q controller 37 may be frozen in response to the LVRT flag “On”, stopping the determination and / or dispatch of new active / reactive power reference values I set points, and frozen values may be sent instead. At the same time, in sub-step 514, one or more control variables of the P controller 37 and / or the Q controller 36 may be stored in a memory of the PPC 22, as a pre-fault value. The pre-fault values may, for example, include current values or values of one or more previous values (pre-fault) and relate to variables used in the active / reactive power loop for determining the active / reactive power set points.

[0119] The local WTG controllers 15 stop receiving active / reactive power set points and execute stored instructions, in sub-step 516, to control the WTGs 14 to ride through the fault (under local control). For example, the local control may be implemented by the WTG controllers 15 based on local power characteristic measurements at the respective WTGs 14.

[0120] During the fault mode of operation, the PPC 22 continues to monitor the grid frequency signal, f(feedback), to determine whether the fault has been cleared and the normal mode of operation can be resumed.

[0121] In step 518, the PPC 22 may therefore determine that the fault has cleared. For example, the LVRT detector 38 of the PPC 22 may detect a voltage rise above a return threshold (e.g. which persists for a minimum-time interval), and the LVRT detector 38 may produce an LVRT flag 39 indicating that the fault has been cleared, i.e. activating the LVRT flag “Off”. A grid fault may typically be cleared within about 100ms to 1000ms, for example.

[0122] However, in order to avoid unexpected events when the normal mode of operation is resumed, the PPC 22 prevents resumption of the normal mode of operation unless it is further confirmed that the grid frequency signal, f(feedback), is in a reliable condition (i.e. the signal is not compromised by noise).

[0123] Accordingly, in step 520, it is further determined whether the grid frequency signal, f(feedback), is in a reliable condition by comparison to the reliability condition(s). For example, the signal quality control module 32 of the PPC 22 may compare the frequency signal, f(feedback), to the reliability condition(s), substantially as described in Figure 4, and output a signal, f( condition), that is indicative of the frequency signal condition to the LVRT control module 35. Here, it shall be appreciated that the condition of the frequency signal, f(feedback), may be continuously monitored, or selectively monitored, for example in response to receiving the LVRT flag “Off”. While the signal, f( condition), indicates that the frequency signal is unreliable, the fault ride through mode of operation is maintained and the pause command is held or frozen. However, when the signal, f(condition), indicates that the frequency signal, f(feedback), is in a reliable condition, the normal-operation software of the PPC 22 and the WTG controllers 15 is resumed, in step 522.

[0124] For example, in sub-step 524, a pause-release command may be generated by the LVRT control module 35, in response to the LVRT flag “Off”, and output to each of the P controller s? and the Q controller 36. In sub-step 526, a state of each of the P controller 37 and the Q controller 37 may therefore be unfrozen, and the PPC 22 may resume determining and dispatching new active and / or reactive power reference values I set points. For example, the P controller s? may resume determining and dispatching active power set points, Pset, based on the P(feedback) and f(feedback) signals, and the prefault values (stored in the memory of the PPC 22) may be used for this purpose.

[0125] In this manner, it is expected that embodiments of the invention will provide fault ride through of the renewable energy power plant and reduce or avoid unexpected shutdowns or other events once the fault has been cleared, ultimately contributing to greater power generating capabilities.

[0126] It will be appreciated that various changes and modifications can be made to the examples described above without departing from the scope of the present invention.

Claims

CLAIMS1 . A method of operating a renewable energy power plant connected to a power network, the method comprising: obtaining a frequency signal at a power plant controller, the frequency signal being indicative of a frequency level of the power network; controlling at least one renewable energy generator of the renewable energy power plant in a normal mode of operation based on the frequency level of the power network, the normal mode of operation being executed by the power plant controller via one or more local controllers of the at least one renewable energy generator; and in the event of a power network fault: controlling the at least one renewable energy generator according to a fault ride through mode of operation using the one or more local controllers; determining if the frequency signal is reliable or unreliable based on a comparison to one or more reliability conditions; preventing resumption of the normal mode of operation while the frequency signal is in an unreliable condition; and resuming the normal mode of operation once the fault has cleared and the frequency signal is in a reliable condition.

2. A method according to claim 1 , wherein each renewable energy generator of the renewable energy power plant is controlled in a normal mode of operation by: determining an active power reference value based, at least in part, on the frequency level of the power network; and dispatching the active power reference value from the power plant controller to the one or more local controllers.

3. A method according to claim 1 or claim 2, wherein the fault ride through mode of operation comprises: freezing, upon entering the fault ride through mode, a state of the power plant controller so as to stop determining and / or dispatching the active power reference value; storing a pre-fault-operation value of at least one control variable for determining the active power reference value, and controlling the at least one renewable energy generator to ride through the fault using the one or more local controllers; and wherein the control is resumed in the normal mode of operation using the at-least- one stored control-variable value from pre-fault operation.

4. A method according to any preceding claim, wherein the one or more reliability conditions include a first frequency threshold, and wherein determining if the frequency signal is reliable or unreliable comprises: determining whether the frequency signal is greater than or equal to the first frequency threshold.

5. A method according to any preceding claim, wherein the one or more reliability conditions include a second frequency threshold, and wherein determining if the frequency signal is reliable or unreliable comprises: determining whether the frequency signal is less than or equal to the second frequency threshold.

6. A method according to any preceding claim, wherein the one or more reliability conditions include a frequency rate of change threshold, and wherein determining if the frequency signal is reliable or unreliable comprises: determining a rate of change of the frequency signal; and determining whether the rate of change of the frequency signal is greater than or equal to the frequency rate of change threshold.

7. A method according to claims 4 to 6, wherein the unreliable condition is determined if: the frequency signal is greater than or equal to the first frequency threshold; the frequency signal is less than or equal to the second frequency threshold; or the rate of change of the frequency signal is greater than or equal to the frequency rate of change threshold.

8. A method according to claims 4 to 6, or claim 7, wherein the reliable condition is determined if: the frequency signal is less than the first frequency threshold; the frequency signal is greater than the second frequency threshold; and the rate of change of the frequency signal is less than the frequency rate of change threshold.

9. A method according to any preceding claim, wherein resumption of the normal mode of operation is prevented until a minimum hold period has elapsed since determining that the frequency signal has changed from the unreliable condition to the reliable condition.

10. A method according to any preceding claim, wherein the frequency signal is determined based on a series of measurements, obtained by the power plant controller, indicative of a frequency level of the power network, the frequency signal being determined by: limiting an indicated frequency level of the obtained series of measurements based on one or more frequency limits; and limiting a rate of change of the obtained series of measurements based on one or more frequency rate of change limits.

11. A method according to claim 10, when dependent on claims 4 and 5, wherein the one or more frequency limits comprise: a frequency limit corresponding to the first frequency threshold; and a frequency limit corresponding to the second frequency threshold.

12. A method according to claim 10 or claim 11 , wherein determining the frequency signal further comprises filtering the obtained series of measurements using an antialiasing filter.

13. A power plant controller for a renewable energy power plant connected to a power network, the power plant controller being configured to execute machine readable instructions to:obtain a frequency signal indicative of a frequency level of the power network; control at least one renewable energy generator of the renewable energy power plant in a normal mode of operation based on the frequency level of the power network, the normal mode of operation being executed via one or more local controllers of the at least one renewable energy generator; and in the event of a power network fault: enter a fault ride through mode of operation, during which the at least one renewable energy generator is controlled by the one or more local controllers; determine if the frequency signal is reliable or unreliable based on a comparison to one or more reliability conditions; prevent resumption of the normal mode of operation while the frequency signal is in an unreliable condition; and resume the normal mode of operation once the fault has cleared and the frequency signal is in a reliable condition.

14. A method of operating a renewable energy power plant connected to a power network, the renewable energy power plant comprising one or more renewable energy generators, the method comprising: obtaining a series of measurements indicative of a frequency level of the power network; determining a frequency signal based on the obtained series of measurements, the frequency signal being determined by: limiting a frequency level of the obtained series of measurements based on one or more frequency limits; and limiting a rate of change of the obtained series of measurements based on one or more frequency rate of change limits; andcontrolling the one or more renewable energy generators based on the determined frequency signal.

15. A method according to claim 14, wherein the determining the frequency signal further comprises filtering the obtained series of frequency measurements using an antialiasing filter.

Images