Controlling renewable energy power plant to detect control faults

Abstract

Aspects of the present invention relate to a method (200) of operating a renewable energy power plant (12) including one or more renewable energy generators (14). The method (200) comprises: receiving a measurement signal indicative of a measured power characteristic of the renewable energy power plant (12) at a connection point (20) with a power network (16); controlling the renewable energy power plant (12) according to a normal mode of operation to provide power by determining and scheduling a power setpoint determined by: obtaining samples of the measurement signal at a sampling rate; and determining the power setpoint based on a sampled measurement of the measured power characteristic and a target level; and monitoring the measurement signal to detect an undersampled oscillation of the measured power characteristic in the sampled measurements and thereby detect a control fault; and controlling the renewable energy power plant (12) according to a malfunctioning mode when the control malfunction is detected.

Description

Technical Field

[0001] The present invention relates to a method of controlling a renewable energy power plant, to a renewable energy power plant controller and more generally to a wind turbine power plant. Background Art

[0002] A wind power plant (WPP) typically includes a plurality of wind turbine generators (WTGs) and is also known as a wind park or wind farm. The regulation and general operation of the power plant is controlled by a power plant control system or controller (PPC), which implements the operating limits and requirements set by the transmission system operator (TSO) or in country-specific grid interconnection requirements or "grid codes". The TSO also communicates power delivery demand to the PPC.

[0003] Therefore, during normal operation of the WPP, the role of the PPC is to act as a command and control interface between the WPP and the grid operator or TSO, where the TSO is responsible for indicating the needs and requirements of the power network to the PPC. The PPC interprets the power delivery requirements required by the TSO and manages the WPP so as to meet these requirements.

[0004] To this end, the PPC may typically include one or more feedback control systems that control the WPP based on measurements of power characteristics at the connection to the power network, such as frequency, voltage level, active current or power level, and / or reactive current or power level. However, feedback control systems are susceptible to instability when the measured power characteristics fluctuate at relatively high frequencies.

[0005] An object of the present invention is to improve upon conventional solutions. Summary of the invention

[0006] According to one aspect of the present invention, a method for operating a renewable energy power plant comprising one or more renewable energy generators is provided. The method comprises: receiving a measurement signal indicating a measured power characteristic of the renewable energy power plant at a connection point to an electricity network; controlling the renewable energy power plant according to a normal operating mode to provide power by determining and scheduling a power set point, the power set point being determined by: obtaining samples of the measurement signal at a sampling rate; and determining the power set point based on a sampled measurement of the measured power characteristic and a target level; and monitoring the measurement signal to detect under-sampled oscillations of the measured power characteristic in the sampled measurement, and thereby detecting a control fault; and when the control fault is detected, controlling the renewable energy power plant according to a fault operating mode.

[0007] In this context, the undersampled oscillations are oscillations having a frequency greater than half the sampling rate. In this way, the invention monitors the measurement signal independently of the control of the power plant according to a normal operating mode, so that one or more undersampled oscillations of the measured power characteristic in the sampled measurement values ​​can be detected to identify a control fault. In this context, it is understood that the undersampled oscillations can be detected by performing a spectral analysis on the measurement signal, wherein the analysis is performed independently of a program, an algorithm or a "power controller" that determines the power set point. The power set point can be a reactive set point and / or an active set point.

[0008] In normal operation mode, the renewable energy power plant provides power to the power network in the form of active and / or reactive power by dispatching active and / or reactive power set points to the one or more renewable energy generators. The set points related to the active power set points may define the amount of active power to be generated, while the set points related to the reactive power set points may define the amount of reactive power to be generated or consumed.

[0009] The benefit of this is that control faults can be detected and the fault operating mode implemented to mitigate damage to the power plant and enable a quicker return to normal operation of the power plant.

[0010] For example, the target level may be received by a grid operator or determined based on a grid operator's requirements. Optionally, the fault operating mode includes generating an alarm for notifying an operator of the control fault.

[0011] Optionally, the fault operation mode comprises controlling the renewable energy power plant to provide electric power by determining and dispatching a power set point based on a reference reactive power level.

[0012] The fault operation mode may include controlling the renewable energy power plant by dispatching active and / or reactive set points to produce a certain amount of active power and / or to produce or consume a certain amount of reactive power.

[0013] In one example, the fault operating mode may further include determining a reference power level based on one or more power set points determined prior to the fault.

[0014] The power levels may be real and / or reactive power levels based on one or more real and / or reactive power set points, respectively.

[0015] In one example, determining and scheduling a power set point during the fault mode can include: calculating a rate of change for transitioning between a current power set point and the reference power level; and determining and scheduling the power set point based on the calculated rate of change.

[0016] Optionally, determining and scheduling a power set point during the fault mode further comprises determining and scheduling the power set point at the reference power level, once the reference power level is reached, for a remainder of the fault mode of operation.

[0017] In one example, the fault mode may be applied until one or more termination conditions are met. For example, the one or more termination conditions may include one or more selected from the following: a threshold time period has passed since the fault mode was applied; an override signal is received from an operator; and / or a non-oscillatory period of the measured power characteristic is detected.

[0018] In one instance, after the fault operation mode, the method further includes: implementing bumpless transfer operation; and controlling the renewable energy power plant according to the normal operation mode; wherein the bumpless transfer operation includes determining the target level of the power characteristic after the fault mode by: setting the target level of the power characteristic to a current sampled measurement level; and calculating a rate of change to transition between the current target level and a post-fault target level, wherein the post-fault target level is suitable for normal operation of the renewable energy power plant.

[0019] Optionally, the step of monitoring the measurement signal to detect under-sampled oscillations in the sampled measurements comprises performing a spectral analysis on successive periods of the measurement signal to detect oscillatory periods of the measured power characteristic having one or more under-sampled oscillations.

[0020] The one or more power set points determined during the non-oscillatory period may, for example, be stored in a memory.The reference power level may, for example, be determined based on the one or more set points stored in the memory.

[0021] Optionally, the memory stores one or more power set points determined during the last two non-oscillatory periods.

[0022] Optionally, detecting the control fault comprises: using the spectrum analysis to detect at least two consecutive oscillatory periods with undersampled oscillations, each oscillation having a frequency greater than half the sampling rate. For example, the spectrum analysis may be performed using a Fast Fourier Transform.

[0023] According to another aspect of the present invention, a power plant controller for controlling the operation of a renewable energy power plant is provided, the renewable energy power plant comprising one or more renewable energy generators and connected to an electricity network. The power plant controller is configured to execute machine-readable instructions to: receive a measurement signal indicating a measured power characteristic of the renewable energy power plant at a connection point to the electricity network; control the renewable energy power plant according to a normal operating mode to provide power by determining and scheduling a power set point, the power set point being determined by: obtaining samples of the measurement level at a sampling rate; and determining the power set point based on the sampled measurement values ​​and target levels of the measured power characteristic; and monitoring the measurement signal to detect under-sampled oscillations of the measured power characteristic in the sampled measurement values, and thereby detecting a control fault; and when the control fault is detected, control the renewable energy power plant according to a fault operating mode.

[0024] According to one aspect of the present invention, there is provided a computer-readable storage medium containing instructions, which, when executed by a computer, cause the computer to perform the above method.

[0025] Within the scope of the present invention, it is expressly intended that the various aspects, embodiments, examples and alternatives described in the preceding paragraphs, in the claims and / or in the following description and drawings, in particular their respective features, may be adopted independently or in any combination. That is, all embodiments and / or features of any embodiment may be combined in any manner and / or combination unless such features are incompatible. Applicants reserve the right to change any originally filed claims accordingly or to file any new claims, including the right to amend any originally filed claim to be subordinate to any other claim and / or to incorporate any features of any other claim, even though protection was not originally claimed in that manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] One or more embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0027] Figure 1 is a schematic architecture of an electricity network including a wind farm and a main grid;

[0028] Figure 2 Is to show that due to Figure 1 Representation of an exemplary fault scenario caused by aliasing of sampled measurement signals used in the control of a wind power plant;

[0029] Figure 3 According to the embodiment of the present invention, the operation Figure 1 A flow chart of a method for a wind power plant;

[0030] Figure 4 According to the embodiment of the present invention Figure 3 A flowchart of exemplary sub-steps of a method of;

[0031] Figure 5 According to the embodiment of the present invention Figure 3 Another flow chart of exemplary sub-steps of the method;

[0032] Figure 6 is to show example failure scenarios and Figure 3 a representation of the response of the wind power plant according to the method; and

[0033] Figure 7 is to show another exemplary failure scenario and Figure 3 A representation of the response of a wind power plant using the method.

[0034] Figure 8 is a schematic architecture of a method and a power plant controller according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] Generally, the present invention relates to methods and systems for operating a renewable energy power plant, such as a wind power plant, comprising one or more renewable energy generators.

[0036] The method and system of the present invention are configured to receive a measurement signal indicative of a power characteristic of the power plant and to control the power plant according to a normal operating mode. In this context, the term "power characteristic" encompasses characteristics including frequency, active power level, active current level, reactive power level, reactive current level, power factor level and / or voltage level at the connection point of the power plant to the electricity network.

[0037] When controlling the power plant according to a normal operating mode, the amount of power provided by the power plant is controlled by determining and scheduling a power setpoint based on a target level of the power characteristic and samples of the measurement signal obtained at a sampling rate.

[0038] The sampling rate is usually optimized for the program, algorithm or "power controller" of the control system that determines the power set point. However, when controlling the renewable energy power plant according to the normal operating mode, relatively high frequency oscillations of the measured power characteristic can occur, making the power controller prone to under-sampling the measurement signal. As a result, the sampled measurement values ​​obtained by the power controller distort the measurement signal, resulting in erroneous operation of the renewable energy power plant. For example, when the measured power characteristic fluctuates at a frequency greater than the sampling rate, in particular when the measured power characteristic fluctuates at a frequency greater than the Nyquist frequency of the power controller, i.e., at a frequency greater than half the sampling rate, the sampled measurement value is under-sampled with respect to the measurement signal. Therefore, despite corrective control actions taken by the control system, the sampled measurement value may erroneously indicate that the measured power characteristic remains above / below a threshold value when the measured power characteristic is actually in an oscillatory state during this period. As a result, the normal operating mode will typically control the power plant in a manner that amplifies and / or exacerbates fluctuations in the measured power characteristic. The active and reactive power controller may comprise two separate controllers, one for reactive power control and another for active power control.

[0039] Advantageously, to counteract these problems, the method and system of the present invention is additionally configured to independently monitor the measurement signal to detect under-sampled oscillations of the measured power characteristic in the sampled measurement values ​​and thereby detect control faults. In this context, it should be understood that the under-sampled oscillations may be determined by performing a spectral analysis on a continuous period or fixed time window of the measurement signal, wherein such analysis is performed independently of the program, algorithm or "power controller" of the control system that determines the power set point.

[0040] When a control fault is detected, the renewable energy power plant can be controlled according to a fault operating mode, during which the renewable energy power plant can be controlled to generate an alarm for notifying an operator of the control fault, and / or to determine and dispatch active and / or reactive power set points, which can be specified for the fault mode or otherwise determined based on one or more of the active and / or reactive power set points determined before the fault.

[0041] This has the benefit of reducing oscillations in the measured power characteristics, mitigating damage to the power plant and enabling a quicker return to normal operation of the power plant.

[0042] Figure 1 A typical architecture is shown, where a wind power plant (WPP) is connected to a main transmission grid which is part of a wider electricity network. Figure 1 The schematic diagrams should be regarded as representative of the power network only. Alternative configurations of power networks and power plants are known and are contemplated in addition to those shown as Figure 1 In addition to the components shown and described in the drawings, other known components may be incorporated or used as Figure 1 In place of the components shown and described in the drawings, other known components may be incorporated. Such variations will be within the capabilities of those skilled in the art. For example, depending on the number of WTGs included in the plurality of WTGs, it may be contemplated to incorporate a substation or additional transformers into the WPP.

[0043] As a skilled reader will appreciate, a WPP includes at least one wind turbine generator (WTG), more typically referred to as a "wind turbine". A WPP is also known as a wind park or wind farm. The examples shown are representative only, and a skilled reader will appreciate that other specific architectures related to wind power plants, power plants for other renewable energy sources, wind turbine generators, and other renewable energy generation sources are possible. Therefore, the present disclosure generally relates to renewable energy power plants and renewable energy generators, rather than being specific to wind power plants and generators as shown in the accompanying drawings. In addition, a skilled reader will appreciate that the methods, systems, and techniques also described below are applicable to many different configurations of power networks. In addition, the components of wind power plants and power networks are conventional and therefore familiar to skilled readers. It is contemplated that other known components may be incorporated in addition to the components shown and described in the drawings, or other known components may be incorporated as alternatives to the components shown and described in the drawings. Such changes are within the capabilities of those skilled in the art. In particular, it will be understood that a WPP may also be considered to include STATCOM equipment, substation equipment, cables, plant controllers, and other components required to connect the WPP to the main grid.

[0044] Furthermore, readers skilled in the art will understand that Figure 1 is a schematic diagram and therefore does not explicitly describe the manner in which the control commands are transmitted. The interconnection may be a direct or "point-to-point" connection, or may be part of a local area network (LAN) operating under an appropriate protocol (such as CAN bus or Ethernet). In addition, it should be understood that the control commands may be transmitted wirelessly, rather than using cables, via an appropriate wireless network, such as WiFi. TM Standard or ZigBee TMWireless networks that operate under standards (IEEE802.11 and 802.15.4, respectively).

[0045] Figure 1 There is shown an electricity network 10 incorporating a WPP 12. The WPP 12 comprises a plurality of WTGs 14. Each of the plurality of WTGs 14 converts wind energy into electrical energy which is transmitted from the WTG 14 to a main electricity network or grid 16 as active current for distribution.

[0046] The WTG 14 generates both active and reactive power. The main grid 16 typically has specific active and reactive power level requirements that the WPP 12 is required to comply with, and the output of the WTG 14 may be varied in real time to match these requirements. Other grid requirements may be specified for reactive and active current levels and voltage levels at specific points within the power network.

[0047] Each of the WTGs 14 is associated with a respective WTG controller 15. In some instances, a group of WTGs may share a semi-centralized WTG controller, such that the number of WTG controllers is less than the number of WTGs. As will be appreciated by those skilled in the art, the WTG controller 15 may be viewed as a computer system capable of operating the WTG 14 in the manner specified herein, and may include a plurality of modules that control the various components of the WTG, or may include only one controller. The computer system of the WTG controller 15 may operate according to software downloaded via a communication network or software programmed thereon by a computer-readable storage medium.

[0048] During normal operation of the WPP 12, the WTG controller 15 operates to implement the active and reactive current and / or power requirements received from the PPC 22 to provide frequency and voltage support to the main grid 16. During special conditions, the WTG controller 15 operates to meet predetermined network requirements and also serves to protect the WTG 14 from any potentially harmful conditions.

[0049] exist Figure 1 , the WPP 12 is shown to include a connection network 17 for connecting the WPP 12 to the main grid 16. The connection network is known in the art and includes a combination of transmission lines, buses and / or transformers to couple the WPP 12 to the main grid 16. Other components such as circuit breakers, reclosers, and other systems known in the art may also be incorporated into the connection network. In this example, the WPP 12 and the main grid 16 are connected at a point of interconnection (PoI) 20, which is an interface between the WPP 12 and the main grid 16. The PoI 20 may also be referred to as a point of common coupling, which may be abbreviated as "PCC" or "PoCC".

[0050] A power plant controller (PPC) 22 is connected to the main grid 16 at a point of measurement (PoM) 24 and to the WTG controller 15. For example, the PPC 22 may be configured to receive one or more measurement signals from the PoM 24, the measurement signals including measurements of characteristics of power supply from the WPP 12 to the main grid 16. Such power characteristics may include frequency levels, power factors, voltage levels, active current levels, active power levels, reactive current levels, and / or reactive power levels exchanged between the WPP 12 and the main grid 16. Since the PoM 24 is not at the PoI 20 in this example, the parameters measured at the PoM 24 are only considered as representative values, because the losses in the line between the PoM 24 and the PoI 20 and the losses in the line between the PoM 24 and the PPC 22 may have an impact on the measurements. Therefore, the measurements may be appropriately compensated to account for the losses and ensure that the measurements used by the PPC 22 are accurate.

[0051] PPC 22 functions as a command and control interface between WPP 12 and grid 16, and more specifically, a grid operator 26, such as a transmission system operator (TSO) or a distribution system operator (DSO).

[0052] In its role as a command and control interface, the PPC 22 interprets the power delivery requirements requested of it by the grid operator 26 and monitors the measured power characteristics to manage the WTGs 14 of the WPP 12 so as to meet the power requirements, as will be described in detail below. The PPC 22 is therefore a suitable computer system for executing control and command as described herein, and thus may incorporate a processing module 28, a connection module 30, a memory module 32, and a sensing module 34, as described herein. Figure 1 Shown in.

[0053] The WTGs 14 are capable of varying their current or power output in response to commands received from the PPC 22. Specifically, as part of its operation, the PPC 22 generates a dispatch signal and sends the signal to the WTG controller 15, and the WTG controller 15 controls the WTGs 14 according to the set points contained within the dispatch signal.

[0054] During the normal operating mode, the WTG 14 is controlled substantially continuously according to control signals dispatched by the PPC 22. When controlling the power plant according to the normal operating mode, the PPC 22 is configured to receive measurement signals, such as measurement signals from the PoM 24, indicating a power characteristic of the WPP 12, and is configured to control the amount of power provided by the WPP 12 so as to meet a target level of the power characteristic. In this context, the target level may correspond to a demand required of the WPP 12 by the grid operator 26, for example.

[0055] In this manner, the PPC 22 controls the WPP 12 by determining and scheduling active and / or reactive power set points based on the measured power characteristic and the target level of the power characteristic. For example, the measured power characteristic may indicate or correspond to a power factor, an active power level, an active current level, a reactive power level, a reactive current level, and / or a voltage level at a PoI 20 between the WPP 12 and the main grid 16. The target level of the power characteristic may be received at the PPC 22 or otherwise determined by the PPC 22 based on a demand of a grid operator 26.

[0056] To determine the power set point, according to an embodiment of the present invention, the PPC 22 may include a program, algorithm or "active and reactive power controller" 100, which forms part of the processing module 28 of the PPC 22. The active and reactive power controller 100 may include, for example, one or more feedback loops and obtain samples of the measurement signal at a sampling rate to determine a successive active and / or reactive power set point. The sampling rate is typically optimized for the program, algorithm or "active and reactive power controller" 100 and may typically be in the range of, for example, 10 to 20 samples per second.

[0057] Various methods and control system architectures are known in the art for determining a power set point based on a comparison between a sampled measurement of a power characteristic and a target level of the power characteristic, which are not described in detail herein to avoid obscuring the present invention. Nevertheless, it should be understood that the active and reactive power controller 100 may include one or more feedback loops, schemes, rules, algorithms and / or lookup tables for this purpose.

[0058] As will be further understood by those skilled in the art, during normal operation mode, the WPP 12 may operate according to one of a voltage control mode, a power factor control mode, a frequency control mode, an active power control mode, and / or a reactive power control mode, and the PPC 22 will determine active and / or reactive power set points according to the corresponding mode. For example, a WPP operating in a voltage control mode is controlled to maintain a voltage level, and thus the active and reactive power controller 100 will determine a reactive power set point by comparing a sampled measurement of the voltage at the PoI 20 with a target voltage. A WPP operating in a power factor control mode is controlled to maintain a power factor, and thus the active and reactive power controller 100 will determine a reactive power set point by comparing a sampled measurement of the power factor with a target power factor. A WPP operating in a frequency control mode is controlled to maintain a frequency level, and thus the active and reactive power controller 100 will determine an active power set point by comparing a sampled measurement of the frequency at the PoI 20 with a target frequency. A WPP operating in an active power control mode is controlled to provide an active power level, and thus the active and reactive power controller 100 will determine an active power set point by comparing sampled measurements of active power at the PoI 20 with a target active power level. Finally, a WPP operating in a reactive power control mode is controlled to consume or supply a reactive power level, and thus the active and reactive power controller 100 will determine a reactive power set point by comparing sampled measurements of reactive power at the PoI 20 with a target reactive power level. Thus, the PPC 22 may receive one or more measurement signals and one or more indications of an operating mode, and based on these inputs determine a reactive power set point to be dispatched to the WPP 12. The reactive power set point may be signal conditioned prior to dispatching to the WPP 12, wherein the WTG 14 is controlled so that the reactive power set point is met.

[0059] Thus, during the normal operating mode, the PPC 22 is able to operate the WPP 12 to meet the requirements of the grid operator 26 .

[0060] However, when the renewable energy power plant is controlled according to a normal operating mode, relatively high frequency oscillations of the measured power characteristic may occur, for example due to a grid fault. In this case, the active and reactive power controller 100 is prone to under-sampling the measurement signal, so that the sampled measurement values ​​obtained by the active and reactive power controller 100 distort the measurement signal in a manner known as aliasing, resulting in erroneous operation of the WPP 12.

[0061] As an example, provide Figure 2 To illustrate control faults that may be caused by aliasing of measurement signals in the active and reactive power controller 100.

[0062] For purposes of this example, the measured power characteristic is the reactive power level at the PoI 20. Thus, the graph depicts a measurement signal representing the measured reactive power level, a target signal representing target levels of active and reactive power, a sampled signal depicting sampled measurements obtained by the active and reactive power controller 100 at a sampling rate, a setpoint signal depicting a reactive power setpoint determined by the reactive power controller 100, and a WTG signal depicting the reactive power level of the WTG 14.

[0063] like Figure 2 , initially, the measurement signal is substantially equal to the target signal, and the set point signal remains substantially constant until time T1. After time T1, the measurement signal begins to oscillate, for example due to a fault or disturbance of the main grid 16. The oscillation of the measurement signal is too high for the sampling rate of the active and reactive power controller 100, so the sampled measurement value distorts the measurement signal. In particular, aliasing occurs and the sampled signal indicates that the measured reactive power level remains below the target signal, for example until time T2. As a result, the active and reactive power controller 100 determines and schedules a reactive power set point that increases the reactive power demand to the reactive power upper limit at time T3, as shown by the set point signal. As a result, the measured reactive power is still oscillating, and the fault persists as the average reactive power increases above the target signal. This results in a control fault in which the PPC 22 controls the WTG 14 to exacerbate or amplify the oscillation, and the active and reactive power controller 100 cannot correct the fault. Therefore, after time T2, the sampled measurement signal is still mostly below the target signal, but fluctuates significantly due to distorted oscillations, and the setpoint signal is still substantially at the upper reactive power limit, and the fault cannot be corrected. This situation results in unsafe operation and / or damage to the WPP 12.

[0064] To address this problem, the PPC 22 of the present invention is additionally configured to independently monitor the measurement signal to detect under-sampled oscillations of the power characteristic, and thereby detect control faults. That is, the PPC 22 monitors the measurement signal independently of a program, algorithm, or active and reactive power controller 100 that obtains samples of the measurement signal at a sampling rate and thereby determines a power set point, so as to detect under-sampled oscillations of the measured power characteristic. In this context, it should be understood that an under-sampled oscillation is any oscillation having a frequency greater than the Nyquist frequency of the active and reactive power controller 100 (i.e., a frequency greater than half the sampling rate).

[0065] It should be appreciated that PPC 22 may detect under-sampling oscillations according to one or more mathematical and / or graphical methods known in the art for detecting signal oscillations, which methods are not described in detail herein to avoid obscuring the present invention.

[0066] To this end, the PPC 22 may additionally include an algorithm, process or "fault detection module" 102, which forms part of the processing module 28 of the PPC 22, for monitoring the measurement signal. To give an example, the fault detection module 102 may be configured to monitor the measurement signal by performing a spectral analysis on successive periods of the measurement signal to detect so-called oscillatory periods having one or more or a threshold amount of under-sampled oscillations. For example, the fault detection module 102 may be configured to perform the spectral analysis using a Fast Fourier Transform to determine the frequency characteristics of the measurement signal.

[0067] Upon detection of a control fault, the PPC 22 of the present invention is additionally configured to control the WPP 12 according to a fault operating mode rather than a normal operating mode. During the fault operating mode, the PPC 22 may take active or passive measures to address the control fault.

[0068] For example, according to the fault operating mode, the PPC 22 may generate an alarm to notify the operator of the control fault so that the operator can control or instruct corrective action. At the same time, the PPC 22 may continue to determine and schedule power set points as described in the normal operating mode until the operator controls or triggers corrective action.

[0069] In another example, the PPC 22 can additionally or alternatively be configured to take active control and be configured to determine and schedule active and / or reactive power set points based on reference active and / or reactive power levels, respectively, for the fault operating mode. The reference active and reactive power levels can be pre-programmed for the fault operating mode or otherwise determined, for example, based on one or more of the active and reactive power set points determined prior to the fault.

[0070] However, there may be a significant difference between the power level of the set point determined during the control fault and the reference power level, and a sudden change in the power level is undesirable because it may cause instability in the power grid. Therefore, during the fault operating mode, the PPC 22 may determine a power set point that is ramped from the level when the control fault was detected to the reference power level. Thereafter, when the fault operating mode continues, the power set point may be frozen at the reference power level until one or more termination conditions are met for returning to the normal operating mode. In this way, the PPC 22 can counteract the control fault and stabilize the control of the WPP 12, suppressing any oscillatory behavior in the measured power characteristics.

[0071] Now we will refer to Figures 3 to 7 The operation of the PPC 22 will be described.

[0072] Figure 3 An exemplary method 200 of controlling a WPP 12 to generate or consume power according to an embodiment of the present invention is shown.

[0073] In step 202 , the PPC 22 receives a measurement signal indicative of at least one of the power characteristics of the WPP 12 at the PoI 20 between the WPP 12 and the main grid 16 .

[0074] For example, PPC 22 may receive measurement signals determined at PoM 24 that indicate voltage levels, power factor levels, active current levels, active power levels, reactive current levels, and / or reactive power levels at PoI 20 .

[0075] In step 204 , according to the normal operating mode, the PPC 22 is configured to control the WPP 12 to provide power by determining and dispatching power set points to meet the requirements of the grid operator 26 .

[0076] To this end, the active and reactive power controller 100 is configured to obtain samples of the measurement signal at a sampling rate in sub-step 206 and to determine a continuous power set point based on the sampled measurement values ​​of the power characteristic and a target level in sub-step 208. In this regard, the target level may correspond to a demand for power supply required by the grid operator 26, such as the target level may be received or otherwise determined at the active and reactive power controller 100.

[0077] As previously described, various methods are known in the art for determining a power set point based on sampled measurements of power characteristics and a target level, which methods may involve the use of one or more control loops, schemes, rules, algorithms, or lookup tables for determining the power set point. However, these methods are not described in detail here to avoid obscuring the present invention.

[0078] The determined power set point is dispatched to the WPP 12, and the WTG controller 15 operates the WTG 14 to produce or consume a corresponding amount of reactive power. The power set point may be an active and / or reactive power set point.

[0079] In parallel with the normal operating mode, in step 210, PPC 22 independently monitors the received measurement signal to detect one or more under-sampled oscillations of the measured power characteristic.

[0080] In particular, the fault detection module 102 of the PPC 22 monitors the measurement signal independently of the active and reactive power controller 100 to detect any under-sampled oscillations of the measured power characteristic in the sampled measurement values. To this end, it should be understood that the algorithm, process, or fault detection module 102 of the PPC 22 may process the measurement signal without sampling, or otherwise process the measurement signal at a significantly higher sampling rate than the active and reactive power controller 100 to facilitate detection of such under-sampled oscillations.

[0081] As mentioned above, various mathematical and graphical methods are known in the art for detecting oscillations in a signal and identifying undersampled oscillations, which methods are not described in detail herein to avoid obscuring the present invention.

[0082] However, to give an example, according to an embodiment of the present invention, Figure 4 Exemplary sub-steps 212 to 216 are shown for monitoring a measurement signal to detect one or more under-sampled oscillations by performing a spectral analysis on consecutive time periods of the measurement signal.

[0083] In sub-step 212, the PPC 22 performs spectral analysis on successive time periods of the measurement signal. For example, the fault detection module 102 may segment the measurement signal into successive fixed time windows and perform spectral analysis on each time period, such as using an FFT, to determine the frequency components of the measurement signal and thereby detect any under-sampled oscillations of the measured power characteristic during the time period.

[0084] If the analyzed time period is non-oscillatory, i.e., it does not contain any under-sampling oscillations or it contains under-sampling oscillations less than a threshold amount (i.e., oscillations having a frequency greater than half the sampling rate of the active and reactive power controller 100), then in sub-step 214, the fault detection module 102 may store information about the non-oscillatory time period in the memory module 32 of the PPC22.

[0085] For example, the fault control module 102 may store one or more of the active and reactive power set points determined during the non-oscillation period in a memory, which may be used as a reference level for determining the active power and / or reactive power set points in the event of a subsequent control fault. In an example, the memory module 32 of the PPC 22 may be configured to store information about a specified number (e.g., two) of the most recent non-oscillation periods. Thus, when a third consecutive non-oscillation period is determined, the information stored in the memory module 32 may be updated accordingly to retain information about the two most recent non-oscillation periods.

[0086] If the fault detection module 102 determines that one or more consecutive oscillatory periods contain one or more or a threshold amount of undersampled oscillations, the fault detection module 102 may detect a control fault in sub-step 216 and control the PPC 22 to enter a fault operating mode.

[0087] For example, if the spectrum analysis performed in sub-step 212 indicates that two consecutive periods are oscillatory (with one or more under-sampled oscillations), the fault detection module 102 can be configured to detect a control fault. In this regard, a threshold of two consecutive periods can be applied to confirm that the oscillation is not transient or otherwise has not been resolved by the operation of the WPP 12.

[0088] Back to Figure 3 Once a control fault is detected, the PPC 22 continues to control the WPP to provide power according to the fault operation mode in step 218 .

[0089] During the fault operating mode, the PPC 22 may take active or passive measures to address the control fault, where passive measures may involve alerting a plant or grid operator and waiting for the operator to dictate or control corrective action, while active measures may involve one or more rules, algorithms, or processes for adjusting reactive power set points.

[0090] To give an example, according to an embodiment of the present invention, Figure 5 Exemplary sub-steps 220 through 226 are shown for controlling the WPP 12 according to the fault operating mode.

[0091] In sub-step 220, PPC 22 generates an alarm for notifying an operator of WPP 12 of the control fault. For example, PPC 22 may generate an alarm to notify a power plant or grid operator via connection module 30.

[0092] In sub-step 222 , the PPC 22 controls the WPP 12 to provide power by determining and scheduling a power setpoint based on the reference power level.

[0093] The reference power level may be a specified or preprogrammed level of active and / or reactive power suitable for the fault operating mode, or the reference power level may be determined based on one or more active and / or reactive power set points determined prior to controlling the fault.

[0094] In one example, sub-step 222 may further include sub-step 224 and sub-step 226. Figure 5 , for determining a reference power level based on one or more power set points determined before a control fault.

[0095] Thus, in one example, in sub-step 224 , the PPC 22 may recall information stored in the memory module 32 , specifically information relating to the two most recent non-oscillatory periods of the measurement signal.

[0096] In sub-step 226, the PPC 22 may determine a reference power level based on active and / or reactive power set points determined during those pre-fault periods, for example, as an average of those power set points determined before the control fault, or as an average of the power set points determined during the earliest period stored in the memory module 32.

[0097] Once the PPC 22 has determined the reference power level, the PPC 22 may proceed to determine active and / or reactive power set points for the fault operating mode accordingly.

[0098] However, to mitigate any sudden changes in power levels, PPC 22 may be configured to determine a power set point that ramps up, down, or otherwise transitions to a reference power level.

[0099] In one example, sub-step 222 may therefore further include sub-steps 228 and 232 for determining a power set point during the fault operating mode, such as Figure 5 as shown in .

[0100] In sub-step 228, the PPC 22 may calculate a rate of change for transitioning between the current power set point and the reference power level. Specifically, the PPC 22 may determine a ramp up or ramp down rate for regulating power based on a comparison between the current power set point and the reference power level. In this context, the ramp rate may be determined according to one or more rules, algorithms, or lookup tables that take into account corresponding ramp rate limits that may be dictated by grid requirements, such as to maintain a stable frequency level and / or to maintain a stable voltage level at the grid 16. For example, the grid code may specify a specific time period during which the transition must be completed, thereby providing a minimum ramp rate for the transition period.

[0101] In sub-step 230, the PPC 22 may determine a power set point based on the calculated rate of change and thereby transition to a reference power level.

[0102] In sub-step 232, once the current power set point has reached the reference power level, the PPC 22 may continue to determine a power set point corresponding to the reference power level for the remainder of the fault operating mode. In this way, the PPC 22 may effectively freeze the power of the WPP 22 at the reference level.

[0103] Back to Figure 3 In step 234 of method 200 , the PPC 22 may check whether one or more termination conditions have been met for terminating the fault operating mode and returning to the normal operating mode.

[0104] For example, the one or more termination conditions may include one or more selected from: a threshold time period having passed since the fault mode was applied; receiving an override signal from an operator; and / or detecting a steady state of the measured power characteristic, ie, a return to a normal level.

[0105] If the one or more fault conditions are not met, the PPC 22 may continue to control the WPP 12 according to the fault mode of operation, as described in step 218 .

[0106] However, if the one or more fault conditions are met, the PPC 22 may return to normal operating mode in step 204 .

[0107] First, in some embodiments, such as Figure 3 As shown in FIG. 2 , in step 236 , upon detecting that the one or more termination conditions are satisfied, the PPC 22 may be configured to implement a bumpless switching operation.

[0108] Specifically, upon return to normal operation mode, PPC 22 determines and schedules active and / or reactive power set points based on sampled measurements of measured power characteristics and target levels. However, when returning to the normal operation mode, changes in operation may result in sudden changes in power levels.

[0109] To mitigate this situation, the PPC 22 may implement a bumpless handover operation in step 236. Bumpless handover operation is known in the art and is not described in detail herein to avoid obscuring the present invention.

[0110] In short, in step 236, the PPC 22 may determine a target level of the power characteristic based on the bumpless switching operation by: (i) setting the target level of the power characteristic to a current sampled measurement value; and (ii) calculating a rate of change to transition between the target level and a post-fault target level suitable for normal operation of the renewable energy power plant. The post-fault target level may be received at the PPC 22 or otherwise determined by the PPC 22, for example, based on a power delivery demand of a grid operator before or after a controlled fault.

[0111] Using this approach, the target level received at or determined by the PPC 22 is changed according to a predetermined rate of change or "ramp rate". The predetermined ramp rate is designed to comply with the grid code requirements of the main grid. For example, the grid code may specify a specific time period during which the transition must be completed, thereby providing a minimum ramp rate for the transition period. The PPC 22 implements the ramp rate using a series of step changes to the reactive power reference level, which it provides to the WTG 14. Using a predetermined ramp rate advantageously allows a controlled transition between a fault operating mode and a normal operating mode, and reduces the possibility of instability in the grid. The ramp rate of active power may be different from the ramp rate of reactive power. Those skilled in the art will appreciate that the ramp rate may be implemented in any conventional manner.

[0112] Once the target level of the power characteristic has reached the post-fault target, the bumpless transfer operation is completed and normal operation of the WPP 12 is resumed.

[0113] It is therefore expected that the present invention will indicate and counteract oscillations in measured power characteristics, mitigating damage to the power plant and enabling a quicker return to normal operation of the power plant.

[0114] As an example, according to method 200, providing Figure 6 and Figure 7 To illustrate the response of the PPC 22 to a control fault. Figure 6shows an exemplary scenario in which a persistent oscillation of the measured power characteristic is a result of aliasing of the measurement signal, while Figure 7 Another exemplary scenario is shown, in which aliasing of the measurement signal promotes and exacerbates an oscillatory behavior due to an external fault, such as a grid fault.

[0115] To be consistent with the previous example, Figure 2 As shown in , the measured power characteristics are plotted in each graph as a reactive power level at PoI 20, and each graph includes a measurement signal representing the measured reactive power level, a target signal representing the target level of reactive power, a sampling signal representing sampled measurement values ​​obtained by the active and reactive power controller 100 at a sampling rate, a set point signal representing a reactive power set point determined by the active and reactive power controller 100, and a WTG signal representing the reactive power level of WTG 14.

[0116] like Figure 6 As shown in FIG. 2 , initially, the PPC 22 receives a measurement signal in step 202 and controls the WPP 12 to provide, thereby generate or consume reactive power according to a normal operating mode. For a period not exceeding time T1, the measurement signal is substantially equal to the target signal, so that the set point signal remains substantially constant. During this period, the PPC 22 independently monitors the measurement signal in step 210 and determines consecutive non-oscillation periods in sub-step 212 and stores information of the two most recent non-oscillation periods in the memory module 32, such as the determined reactive power set point. In particular, as shown in FIG. 2 ... example, as shown in FIG. 2 , the PPC 22 receives a measurement signal in step 202 and controls the WPP 12 to provide, thereby generating or consume reactive power according to a normal operating mode. For example, as shown in FIG. 2 , the PPC 22 receives a measurement signal in step 202 and controls the WPP 12 to provide, thereby generating or consume reactive power according to a normal operating mode. For example, as shown in FIG. 2 , the PPC 22 receives a measurement signal in step 202 and controls the WPP 12 to provide, thereby generating or consume reactive power according to a normal operating mode. Figure 6As shown in , a first non-oscillatory period is detected between time T2 and time T3, and a second non-oscillatory period is detected between time T3 and time T4. However, after time T1, the measurement signal begins to oscillate, for example due to a fault or disturbance of the WPP 12 and / or the main grid 16. The oscillation of the measurement signal is too high for the sampling rate of the active and reactive power controller 100, so the sampled measurement value distorts the measurement signal. In particular, aliasing occurs and the sampled signal indicates that the measured reactive power level remains below the target signal. As a result, the active and reactive power controller 100 determines and schedules a reactive power set point that increases the reactive power demand, as shown in the set point signal. This has the effect of continuing and / or exacerbating the oscillatory behavior, which may have been initially caused by, for example, a grid fault. In this example, although the PPC 22 can detect that under-sampled oscillations have occurred between time T1 and time T4, the PPC 22 can still classify the period T3 to T4 as non-oscillatory because there are less than a threshold number of under-sampled oscillations. However, between time T4 and time T5, the PPC 22 determines in sub-step 212 a pair of consecutive oscillatory periods (T4 to T5 and T5 to T6), and therefore the PPC 22 detects a control fault at time T6 in sub-step 216. Therefore, in step 218, the PPC 22 enters a fault operating mode, and in sub-step 220, generates an alarm for notifying the operator of the WPP 12 of the control fault. During the fault operating mode, the PPC 22 determines a reactive power set point based on the reactive power set point determined before the fault, particularly during the earliest non-oscillatory period stored in the memory module 32 (i.e., between time T2 and time T3). Therefore, the PPC 22 calls the set point in sub-step 224, and determines the reference reactive power level as the average of those set points in sub-step 226. On this basis, according to sub-steps 228 to 230, the PPC 22 decreases the reactive power set point to the reference level between time T6 and time T7. In this example, the original fault that caused the oscillatory behavior may have been resolved shortly after T1, however, the continued oscillation may be the result of aliasing of the measured reactive power level. Therefore, when the PPC 22 decreases the reactive power set point, the oscillation of the measured signal can be substantially eliminated, and the measured power level can be reduced accordingly. After time T7, the PPC 22 freezes the reactive power set point at the reference reactive power level for the remainder of the fault operating mode, i.e., until time T8. Therefore, the measured signal remains constant. At time T8, the PPC 22 detects in step 234 that a termination condition has been met, such as the prescribed time period for the fault operating mode has passed.Therefore, in step 236 , the PPC 22 implements a bumpless switching operation to regulate the target level between time T8 and time T9 , and returns to controlling the WPP 12 to generate or consume reactive power according to the normal operating mode in step 204 .

[0117] In this manner, oscillations in the measured power characteristic are counteracted and the WPP 12 returns to a normal operating mode to meet the power demands of the grid operator 26 .

[0118] exist Figure 7 , the PPC continues to control the WPP 12 to provide power in substantially the same manner. However, in this example, although the set point is determined and scheduled according to the fault operating mode between times T6 and T8, the oscillation of the measured signal continues because the source of the fault is external. In this scenario, the present invention is therefore used to notify the operator of a fault having an external source, while the fault operating mode mitigates further amplification, maintenance or exacerbation of the oscillations caused by aliasing. Of course, the fault mode can be applied until one or more termination conditions are met, where the termination condition in the example shown is that a threshold time period has passed. In other examples, for example, the termination condition may include an override signal received from the operator and / or detection of a non-oscillatory period of the measured power characteristic. As shown in FIG. Figure 7 As shown in , a bumpless switching operation is then implemented at time T8 to adjust the target level, and the PPC 22 returns to the normal operating mode. However, in this scenario, it can be seen that the determined set point begins to rise again due to aliasing of the oscillatory measurement signal, and it should be appreciated that the method 200 can be repeated, substantially as described above, until the fault is resolved.

[0119] Figure 8 A schematic diagram of an embodiment of the present invention is shown. Using a power meter, a power characteristic is measured at a measuring point 24 in a cable (shown as a three-phase cable) connecting the WPP 12 to the main grid 16. The measurement signal is forwarded to a normal controller 27 operating according to a normal operating mode and to an oscillation detector 29, which monitors the measurement signal to detect under-sampled oscillations of the measured power characteristic. The normal controller 27 determines a power set point based on the sampled measurement values ​​and the target power level. A fault mode controller 31 is also shown, which determines the power set point according to a fault operating mode. The oscillation detector controls a (symbolic) switch 33 so that if no oscillation is detected, the switch is positioned as shown in the figure, where the power set point generated from the normal controller 27 is dispatched to one or more renewable energy generators 14; however, if an oscillation is detected, the power set point determined by the fault mode controller 31 is dispatched to the one or more renewable energy generators 14.

[0120] Many modifications may be made to the examples described above without departing from the scope of the present invention as defined in the accompanying claims.

[0121] For example, it should be appreciated that in other instances, the corrective power control described in sub-step 222 of method 200 may be automatically triggered upon application of a fault operating mode, or following a command, for example, from an operator of WPP 12 .

[0122] Furthermore, in some embodiments, alternative renewable energy generators or hybrid power plants may be operated according to the methods described herein.

Claims

1. A method (200) of operating a renewable energy power plant (12) comprising one or more renewable energy generators (14), the method (200) comprising: receiving a measurement signal indicative of a measured power characteristic of the renewable energy power plant (12) at a connection point (20) to an electricity network (16); The renewable energy power plant (12) is controlled according to a normal operating mode to provide electrical power by determining and dispatching a power set point, the power set point being determined by: obtaining samples of the measurement signal at a sampling rate; and determining the power set point based on sampled measurements of the sensed power characteristic and a target level; and monitoring said measurement signal to detect under-sampled oscillations of said measured power characteristic in said sampled measurement values ​​and thereby detect a control fault; and When the control fault is detected, the renewable energy power plant (12) is controlled according to a fault operation mode.

2. The method (200) of claim 1, wherein the fault operating mode includes generating an alarm for notifying an operator of the control fault.

3. The method (200) according to claim 1 or claim 2, wherein the fault operation mode comprises: The renewable energy power plant (12) is controlled to provide electrical power by determining and scheduling a power set point based on a reference power level.

4. The method (200) of claim 3, wherein the fault operation mode further comprises: The reference power level is determined based on one or more power set points determined prior to the fault.

5. The method (200) of claim 3 or claim 4, wherein determining and scheduling a power set point during the fault mode comprises: calculating a rate of change for transitioning between a current power set point and the reference power level; and The power set point is determined and scheduled based on the calculated rate of change.

6. The method (200) of claim 5, wherein determining and scheduling a power set point during the fault mode further comprises: The power set point is determined and scheduled at the reference power level for a remainder of the fault operating mode once the reference power level is reached.

7. The method (200) of any one of the preceding claims, wherein the failure mode is applied until one or more termination conditions are met.

8. The method (200) of claim 7, wherein the one or more termination conditions include one or more selected from the group consisting of: a threshold period of time has elapsed since the failure mode was applied; Receiving an override signal from the operator; and / or A non-oscillatory period of the measured power characteristic is detected.

9. The method (200) according to claim 7 or claim 8, wherein: After the fault operating mode, the method further comprises: Implement bumpless transfer operations; and According to the normal operating mode, the renewable energy power plant (12) is controlled; wherein the bumpless switching operation comprises determining the target level of the power characteristic after the fault mode by: setting the target level of the power characteristic to a currently sampled measurement level; and A rate of change is calculated to transition between a present target level and a post-fault target level, the post-fault target level being suitable for normal operation of the renewable energy power plant (12).

10. The method (200) according to any one of the preceding claims, wherein the step of monitoring the measurement signal to detect under-sampled oscillations in the sampled measurement values ​​comprises: A spectral analysis is performed on successive periods of the measurement signal to detect oscillatory periods of the measured power characteristic having one or more undersampled oscillations.

11. The method (200) of claim 10, when appended to claim 3, wherein the one or more power set points determined during the non-oscillatory period are stored in a memory (32); and The reference power level is determined based on the one or more set points stored in the memory (32).

12. The method (200) of claim 11, wherein the memory (32) stores one or more power set points determined during the last two non-oscillatory periods.

13. The method (200) according to any one of claims 10 to 12, wherein detecting the control fault comprises: Using the spectrum analysis, at least two consecutive oscillatory periods with undersampled oscillations are detected, each oscillation having a frequency greater than half the sampling rate.

14. The method (200) according to any one of claims 10 to 13, wherein: The spectral analysis is performed using Fast Fourier Transform.

15. A power plant controller (22) for controlling the operation of a renewable energy power plant (12), the power plant comprising one or more renewable energy generators (14) and connected to an electricity network (16), the power plant controller (22) being configured to execute machine-readable instructions to: receiving a measurement signal indicative of a measured power characteristic of the renewable energy power plant (12) at a connection point (12) to an electricity network (16); According to a normal operating mode, the renewable energy power plant (12) is controlled to provide electricity by determining and dispatching a power set point, the power set point being determined by: obtaining samples of said measurement level at a sampling rate; and determining the power set point based on the sampled measurements of the sensed power characteristic and a target level; and monitoring said measurement signal to detect under-sampled oscillations of said measured power characteristic in said sampled measurement values ​​and thereby detect a control fault; and When the control fault is detected, the renewable energy power plant (12) is controlled according to a fault operation mode.

Images