Fatigue load in wind turbines and use of operational metadata

Abstract

The present disclosure relates to methods and systems for determining fatigue loads in wind turbines. Such methods can include obtaining operational metadata representative of an operating condition, and determining whether the operational metadata corresponds to operational metadata from a plurality of previously defined operating conditions for which fatigue loads are known, the plurality of previously defined operating conditions being stored in an operating condition database. If the operational metadata representative of the operating condition substantially corresponds to operational metadata from a stored previous operating condition, the fatigue load for the stored previous defined operating condition is aggregated to a historical cumulative fatigue load to determine a total cumulative fatigue load. Methods for registering operating conditions in a wind turbine are also provided.

Description

Technical Field

[0001] This disclosure relates to wind turbines, and more particularly to methods and systems for registering and determining fatigue loads in wind turbines. This disclosure further relates to the use and generation of operational metadata in wind turbines. Background Technology

[0002] Modern wind turbines are used to supply electricity to the grid. A wind turbine typically consists of a tower upon which a nacelle is supported. The wind turbine rotor, including a hub and multiple turbine blades, can be rotatably mounted to the nacelle.

[0003] Wind turbine blades are kept in motion by wind. The hub of the wind turbine can be operatively coupled to the rotor of a generator. As the hub and blades rotate, the kinetic energy of the wind is converted into the mechanical kinetic energy of the wind turbine rotor, and ultimately into electrical energy or power in the generator. The generator is typically housed inside the nacelle.

[0004] In a so-called direct-drive wind turbine, the wind turbine rotor can be directly coupled to the generator rotor. Alternatively, the wind turbine rotor may include a main rotor shaft (the so-called "low-speed shaft") leading to a gearbox. The high-speed shaft of the gearbox then drives the generator. Regardless of the wind turbine topology, the generator's electrical power output can be fed into the power grid. The connection between the generator and the grid can include, for example, converters, transformers, medium-voltage lines, etc.

[0005] Wind turbines can be grouped together in so-called wind farms or wind farms, which may have a common connection point to the power grid. The connection point defines the power plant from the perspective of the distribution network.

[0006] A control system or “controller” can be provided for the wind turbine, which may be physically located within the wind turbine (e.g., in the nacelle) or remotely located. Typically, the wind turbine control system may be partially integrated into the wind turbine itself and partially located outside the wind turbine, i.e., at the wind farm level and / or remotely.

[0007] Control systems for wind turbines may include so-called SCADA (“Monitoring and Data Acquisition”) systems. SCADA is a control system architecture that includes a computer and data communication capabilities for advanced process supervision and management. SCADA may include programmable logic controllers (PLCs), a type of industrial digital computer.

[0008] Control via a SCADA system typically comprises different control levels. Level 0 of such a system may include sensors (temperature sensors, flow sensors, position sensors, etc.) and actuators. Level 1 may include industrial input / output (I / O) modules and their associated distributed electronic processors, and may include one or more PLCs. Level 2 may be a supervisory computer, which may be locally located within the wind turbine. Level 2 represents the computer and software responsible for communicating with remote terminal units (RTUs, which are connected to the sensors and actuators) and the PLC. The PLC is connected to the sensors and actuators during the process and networked to the supervisory system. Level 3 may be a wind farm controller, and does not need to be included in or form part of a separate wind turbine, but is typically provided physically within the wind farm. Level 4 may be a remote control center capable of operating remotely from the wind farm and effectively monitoring and collecting data from multiple wind farms.

[0009] Wind turbine controllers can be configured to determine the appropriate actuator setpoints for the wind turbine based on the prevailing circumstance. Actuator setpoints for modern variable-speed wind turbines include, for example, generator torque and blade pitch angle. By controlling the pitch angle of one or more blades and the generator torque, the rotor speed, electrical power output, aerodynamic thrust, and other mechanical loads can be controlled. The goal of the control system is typically to maximize electrical power output while keeping the loads in the wind turbine at acceptable levels. Additional actuators in the wind turbine are conceivable in this regard for controlling the aerodynamic torque of the wind turbine rotor, the electromagnetic counter-torque on the generator, and the movement of the nacelle and tower. Such actuators may include aerodynamic flaps, ailerons, spoilers, thrusters, deformable blade surfaces, and so on.

[0010] Normal operation of a wind turbine typically follows a predefined power curve, which dictates the turbine's operation based on the dominant wind speed. Normal operation encompasses different operating ranges. In the lower wind speed range, the objective is usually to maximize electrical power output. In the higher wind speed range (especially above the nominal wind speed), turbine operation focuses on controlling the load while maintaining electrical power output at a predetermined level.

[0011] As mentioned earlier, the actuator setpoints for torque and pitch (and other actuators, such as yaw) can be varied depending on the circumstances. These circumstances can include, for example, average wind speed, turbulence, wind shear, air density, and other meteorological conditions, as well as internal conditions such as vibration, mechanical loading, or component temperature. They can also include specific external requirements for noise reduction, operational interruptions for maintenance, grid-based conditions or grid events such as reduced active power requirements (e.g., low-voltage events, zero-voltage events, increases in grid frequency, etc.).

[0012] Wind turbine controllers can be programmed to send signals to various systems (such as generators, pitch systems, and yaw systems) based on a set of measured variables received from multiple sensors to influence the operation of the wind turbine. Sensors may include rotor speed sensors, load sensors (strain gauges or accelerometers), anemometers, wind vanes, etc.

[0013] To ensure that wind turbines or their components do not fail prematurely, numerical models for load estimation are known to be used to estimate the loads that wind turbines will experience throughout their lifetime. For example, fatigue loads can be estimated based on average wind speed and turbulence within a short time window. These potential fatigue loads can be estimated based on wind speed measurements before wind turbine commissioning. Actual fatigue loads can also be recorded during operation. Fatigue loads occurring during the life of a wind turbine can be determined, and if the fatigue limits of the wind turbine components are known, the remaining lifetime of the components can be estimated. Operational changes or planned maintenance can be made based on such estimates of the remaining lifetime.

[0014] However, it has been found that grid conditions defined by grid requirements are often not considered in the estimation of fatigue loads on wind turbines, even though these grid conditions can have a significant impact on the loads and fatigue life of wind turbine (components). During certification calculations, loads caused by transients and extended operation are typically not considered in response to grid conditions, leading to increased uncertainty in fatigue load and lifespan estimates.

[0015] Grid events can cause operational changes that deviate from what is considered normal operation for a wind turbine. Electrical power output can typically decrease in these situations (or fall below the optimal operating performance of the wind turbine at that time). Based on the lower power output during such transients, lower loads can be assumed, but the loads can actually remain high or even increase. Grid events can, for example, reduce stall margins, cause oscillations, require auxiliary systems, and / or bring operation closer to critical frequencies.

[0016] This disclosure provides examples of systems and methods that at least partially address some of the aforementioned drawbacks.

[0017] definition

[0018] Throughout this disclosure, the following definitions and descriptions shall apply unless otherwise indicated.

[0019] The operational lifespan of a wind turbine can be viewed as a series of continuous slices or time periods. Time slices can have variable lengths, and during such time slices, the operation of the wind turbine can be constant (steady-state) or change rapidly (transient).

[0020] Each of these time slices is considered an operating condition herein, within which the operation of the wind turbine can be described using a given set of operating data. Therefore, an operating condition can involve steady-state operation or transient conditions. As will be explained below, each operating condition can be described by operating metadata, which can be used to determine fatigue loads using a numerical model of load estimation. Different operating conditions can have different time lengths.

[0021] Steady-state operation can be considered as normal operation of a wind turbine in the absence of significant changes in operating metadata.

[0022] As used throughout this disclosure, an operation event can be viewed as a change in operation metadata, indicating a shift between two operation states. The operation event itself is also an operation state in the aforementioned sense.

[0023] Grid events are a specific example of operational events. Grid events can be detected by changes in one or more setpoints, which can be seen as a response to grid sensors (frequency, line voltage) or to wind farm commands (reactive and / or active power setpoints). Wind farm controllers may receive such setpoints centrally from the grid operator and assign them to individual wind turbines, or individual turbine controllers may be able to receive setpoints directly from the grid operator.

[0024] Operating data for a wind turbine is data available within the wind turbine controller and can include all signals or settings of systems controlled by or connected to the wind turbine controller. These systems include, for example, generators (signals or settings may include, for example, electrical power output, generator rotor speed) and pitch systems (signals or settings may include pitch angle), and the operating data may specifically include signals within one or more PLCs of the wind turbine controller. In this sense, the controller itself is one of the sensors or sensor systems describing the operating conditions of the wind turbine. Therefore, the calculations performed by the wind turbine controller and the setpoints processed and set by the controller can form part of the operating data that can be considered as characteristics describing specific operating conditions.

[0025] Operational metadata is derived from operational data by processing and / or compressing it based on statistical parameters. This operational metadata provides, for example, a description of a "fingerprint" of a specific operational condition. Each operational condition can have a different time length; that is, the operational data can remain substantially the same, or can be considered to represent a 5, 10, or 30-minute period. In other cases, much shorter time periods can be used, such as 30-second or 1-minute time periods. Operational metadata may specifically include statistical metadata, such as the average, mean, and standard deviation of the operational data over the time period of the operational condition. Operational metadata may also include the first or second (or additional) time derivatives of the operational data, as well as data composed of different variables, i.e., the ratio between two variables (e.g., active power output and mechanical load) or the product of two variables (e.g., pitch speed and pitch activation time). Summary of the Invention

[0026] In a first aspect, a method for determining cumulative fatigue loads in a wind turbine is provided. The method includes obtaining operational metadata representing an operating condition, and determining whether the operational metadata corresponds to operational metadata of one of a plurality of previously defined operating conditions, for which fatigue loads are known, and the plurality of previously defined operating conditions are stored in an operating condition database. If the operational metadata representing an operating condition substantially corresponds to operational metadata from one of the previously defined operating conditions, the fatigue loads of the previously defined operating conditions corresponding to the operational metadata are summed to historical cumulative fatigue loads to determine a total cumulative fatigue load.

[0027] In this regard, a method for determining fatigue loads is provided, which uses operating data available in a wind turbine control system to determine the fatigue loads. This method does not require additional load sensors to provide reliable estimates of fatigue loads under different operating conditions, including, for example, grid conditions. Therefore, this method can be used on standard wind turbines without the need for specific dedicated sensors.

[0028] It has been found that fatigue loads can be reliably determined by relying on operational metadata. Therefore, even if meteorological or power grid data may differ, fatigue damage can be assumed to be the same for known operating conditions if the operational metadata (including, for example, actuator setpoints) is identical to that for previously defined operating conditions.

[0029] The known loads for known operating conditions may have been determined previously from numerical simulations.

[0030] In another aspect, a system for determining fatigue loads in a wind turbine is provided. The system includes: a local wind turbine module for determining operational metadata representing the operating conditions of the wind turbine; a load estimation module configured to determine fatigue loads of the operating conditions characterized by the operational metadata; and a module for determining cumulative fatigue loads, configured to determine the cumulative damage equivalent load of the wind turbine by summing the damage equivalent loads of the operating conditions that have occurred during the lifespan of the wind turbine.

[0031] In this regard, the local wind turbine module may be or may include software or a combination of hardware and software connected to the wind turbine controller.

[0032] In another aspect, a method for registering operating conditions in a wind turbine is provided. The method includes: determining operating data, said operating data including signals and / or set points of one or more wind turbine components under the control of a wind turbine controller; and determining changes in the operating data indicating the end of an operating condition. The method further includes determining operating metadata of the wind turbine representing the operating condition, wherein the operating metadata is derived from operating data during the operating condition.

[0033] In this regard, it enables efficient communication and high-frequency data transmission for wind turbine controllers. Operational data can be effectively "encapsulated" and compressed by deriving relevant operational metadata and transmitting only the metadata. This reduces the data storage requirements per turbine. Metadata can be used to capture essential points of specific operating conditions.

[0034] The present invention provides a set of technical solutions, as follows.

[0035] Technical Solution 1. A method for determining the cumulative fatigue load in a wind turbine, the method comprising:

[0036] Obtain operation metadata representing the operation status;

[0037] Determine whether the operation metadata corresponds to operation metadata from one of a plurality of previously defined operation conditions, for which fatigue loads are known, and which are stored in an operation condition database; and

[0038] If the operation metadata representing the operation state substantially corresponds to the operation metadata of one of the plurality of previously defined operation states, then

[0039] The fatigue loads corresponding to the previously defined operating conditions of the metadata are summed up to the historical cumulative fatigue loads of the wind turbine to determine the total cumulative fatigue load.

[0040] Technical Solution 2. The method as described in Technical Solution 1, further comprising sending the operation metadata of the operation state to the load estimation module when the operation metadata during the operation state does not correspond to the operation metadata of any of the plurality of previously defined operation states.

[0041] Technical Solution 3. The method as described in Technical Solution 2, wherein the load estimation module creates a new operating condition based on the operating metadata and the estimated fatigue load.

[0042] Technical Solution 4. The method as described in Technical Solution 3, wherein the new operating status is stored in the operating status database.

[0043] Technical Solution 5. The method of any one of Technical Solutions 1-4 further includes determining the damage equivalent load for the operating conditions based on rainflow counting cycles for known operating conditions.

[0044] Technical Solution 6. The method of any one of Technical Solutions 1-5, further comprising estimating the remaining service life of one or more components of the wind turbine based on the total cumulative fatigue load.

[0045] Technical Solution 7. The method as described in any one of Technical Solutions 1-6, wherein the operating condition is a power grid event.

[0046] Technical Solution 8. The method of any one of technical solutions 1-7, wherein the operation metadata includes one or more of the average value, mean, and standard deviation of operation data during the time period of the operation status.

[0047] Technical Solution 9. The method as described in Technical Solution 8, wherein the operating data includes one or more of the following: one or more pitch angles of one or more blades; the rotor speed of the generator or wind turbine rotor; and the electrical power output from the generator.

[0048] Technical Solution 10. The method of any one of technical solutions 1-9, wherein the local wind turbine module determines operational data including signals and / or set points of one or more wind turbine components under the control of the wind turbine controller;

[0049] Determine changes in the operation data that indicate the end of the operation status;

[0050] Determine the operational metadata of the wind turbine representing the operational condition, wherein the operational metadata is derived from the operational data during the operational condition.

[0051] Technical Solution 11. A system for determining fatigue loads in a wind turbine, comprising:

[0052] A local wind turbine module, used to determine operational metadata of the wind turbine representing its operational status;

[0053] A load estimation module configured to determine the fatigue load of the operating condition characterized by the operating metadata; and

[0054] A module for determining cumulative fatigue load is configured to determine the cumulative damage equivalent load for the wind turbine by summing the damage equivalent load for operating conditions that have occurred during the service life of the wind turbine.

[0055] Technical Solution 12. The system as described in Technical Solution 11, wherein the local wind turbine module is configured to send the operating metadata representing the operating condition to the module used to determine the cumulative fatigue load.

[0056] Technical Solution 13. The system as described in Technical Solution 11 or 12, wherein the module for determining the cumulative fatigue load is configured to determine whether the operating metadata of the operating condition substantially corresponds to the operating metadata of one of a plurality of previously defined operating conditions, and

[0057] If the operation metadata corresponds to one of the stored operation states...

[0058] The fatigue load for the aforementioned operating condition is then determined to be equal to the fatigue load for the corresponding stored operating condition.

[0059] Technical Solution 14. The system of Technical Solution 13, wherein the module for determining the cumulative fatigue load is configured to send the operation metadata of the operation condition to the load estimation module when the operation metadata of the operation condition does not correspond to any operation condition of the stored operation condition.

[0060] Technical Solution 15. The system of any one of technical solutions 11-14, wherein the module for determining the cumulative fatigue load is configured to determine the remaining service life of the wind turbine or one or more components of the wind turbine based on the total damage equivalent load. Attached Figure Description

[0061] Non-limiting examples of this disclosure will now be described with reference to the accompanying drawings, in which:

[0062] Figure 1 The illustration is a perspective view of an example wind turbine;

[0063] Figure 2The illustration shows a detailed interior view of the nacelle of an example wind turbine;

[0064] Figure 3 The schematic diagram illustrates an example of a system used to determine fatigue loads in a wind turbine under grid conditions;

[0065] Figure 4 Schematic diagrams illustrate examples of methods for determining cumulative fatigue loads in wind turbines; and

[0066] Figure 5 The diagram illustrates an example of a method for registering the operating conditions of a wind turbine. Detailed Implementation

[0067] In these figures, the same reference numerals are used to denote matching elements.

[0068] Figure 1 The illustration shows a perspective view of an example wind turbine 160. As shown, the wind turbine 160 includes: a tower 170 extending from a support surface 150; a nacelle 161 mounted on the tower 170; and a rotor 115 coupled to the nacelle 161. The rotor 115 includes: a rotatable hub 110; and at least one rotor blade 120 coupled to and extending outward from the hub 110. For example, in the illustrated embodiment, the rotor 115 includes three rotor blades 120. However, in alternative embodiments, the rotor 115 may include more or fewer than three rotor blades 120. Each rotor blade 120 may be spaced around the hub 110 to facilitate the rotation of the rotor 115 so that kinetic energy can be converted from wind into usable mechanical energy and subsequently into electrical energy. For example, the hub 110 may be rotatably coupled to a generator 162. Figure 2 The generator 162 is positioned within the engine room 161 to allow electrical energy to be generated.

[0069] Figure 2 Illustration Figure 1 A simplified internal view of an example of the nacelle 161 of a wind turbine 160. As shown, a generator 162 may be disposed within the nacelle 161. Typically, the generator 162 may be coupled to the rotor 115 of the wind turbine 160 to generate electrical power from the rotational energy generated by the rotor 115. For example, the rotor 115 may include a main rotor shaft 163 coupled to a hub 110 for rotation therewith. The generator 162 may then be coupled to the rotor shaft 163 such that rotation of the rotor shaft 163 drives the generator 162. For example, in the illustrated embodiment, the generator 162 includes a generator shaft 166 rotatably coupled to the rotor shaft 163 via a gearbox 164.

[0070] It should be understood that the rotor shaft 163, gearbox 164 and generator 162 are typically supported within the nacelle 161 by a support frame or platform positioned on top of the wind turbine tower 170.

[0071] The nacelle 161 is rotatably coupled to the tower 170 via a yaw system 20 in such a way that the nacelle 161 can rotate about a yaw axis YA. The yaw system 20 includes a yaw bearing having two bearing assemblies configured to rotate relative to the other. The tower 170 is coupled to one of the bearing assemblies, and the platform or support frame 165 of the nacelle 161 is coupled to the other bearing assembly. The yaw system 20 includes a ring gear 21 and a plurality of yaw actuators 22, each having a motor 23, a gearbox 24, and a pinion 25 for meshing with the ring gear 21 to rotate one of the bearing assemblies relative to the other.

[0072] Blade 120 is coupled to hub 110, with pitch bearing 100 positioned between blade 120 and hub 110. Pitch bearing 100 includes an inner ring and an outer ring. Wind turbine blades may be attached to either the inner or outer bearing ring, while the hub is connected to another bearing ring. When pitch system 107 is actuated, blade 120 can perform relative rotational movement with respect to hub 110. Therefore, the inner bearing ring can perform rotational movement with respect to the outer bearing ring. Figure 2 The pitch system 107 includes a pinion 108 that meshes with an annular gear 109 provided on an inner bearing ring to cause the wind turbine blades to rotate about the pitch axis PA.

[0073] The energy generated by the generator can be transferred to a converter, which adapts the generator's output power to the requirements of the power grid. The generator may include electrical phases, such as three phases. The converter may be located inside the nacelle, inside the tower, or externally.

[0074] Figure 3 The schematic diagram illustrates an example of a system 220 used to determine fatigue loads in a wind turbine.

[0075] according to Figure 3 The system 220 for determining fatigue loads in a wind turbine includes: a local wind turbine module 400 for determining operational metadata of the wind turbine representing operational conditions; a load estimation module 300 configured to determine fatigue loads corresponding to operational conditions characterized (i.e., described by) the operational metadata; and a module for determining cumulative fatigue loads configured to determine the cumulative damage equivalent load of the wind turbine by summing the damage equivalent loads of operational conditions that have occurred during the lifespan of the wind turbine.

[0076] The local wind turbine module 400 can be provided locally, i.e., inside the wind turbine (e.g., in the nacelle) or adjacent to the wind turbine. The module 400 may have access to the internal signals of the controller.

[0077] In the example, the local wind turbine module 400 is configured to send operational metadata representing the operating condition to the module 500 used to determine the cumulative fatigue load.

[0078] The local wind turbine module 400 may include direct control of wind turbine operation, i.e., a wind turbine controller represented by a network of PLC signal 405, buffer 430, and separately linked PLC 420.

[0079] The buffer 430 can continuously receive and store signals 405 from the direct-control wind turbine controller. The PLC 420 can be configured to send operational metadata to the module 500 used to determine cumulative fatigue loads.

[0080] PLC 420 can also be configured to determine the end of an operating state and the beginning of a new operating state. The transition from one operating state to the next is marked by changes in the operating data and operating metadata. Although the metadata record in box 420 is depicted herein as a separate PLC, it should be understood that any combination of hardware, firmware, and / or software suitable for the functions described herein can be used. In one example, box 420 may reside on the same PLC as a wind turbine controller.

[0081] In this disclosure, a method for registering operating conditions in a wind turbine is provided (examples thereof are shown in...). Figure 5 (Illustrated in the diagram). The method includes: determining operational data, said operational data including signal and / or setpoints of one or more wind turbine components under the control of a wind turbine controller; and determining changes in the operational data indicating the end of an operational condition. The method further includes determining operational metadata of the wind turbine representing the operational condition, wherein the operational metadata is derived from operational data during the operational condition.

[0082] As used in this article, the change can be considered an operational event, that is, a significant change in operational data that exceeds the normal continuous changes belonging to the same operational condition.

[0083] This method can be executed by a local wind turbine module 400. Throughout the operation of the wind turbine, the controller can read operational and internal signals from one or more PLCs. Operational signals may include, for example, actuator setpoints, such as generator settings (speed, output), pitch system settings (e.g., pitch angles of different blades, power of the pitch motor), yaw system settings (e.g., yaw angle, power output of one or more yaw motors), and sensor measurements including wind direction, temperature, etc. Internal signals may include, for example, PID gain.

[0084] Operational and / or internal signals can be stored in buffer 430. When the end of an operational condition is detected (i.e., a new operational condition is beginning) or the transmission of an operational condition ends, a signal corresponding to the behavior of the wind turbine during the operational condition can be sent to module 500. Such operational metadata can be viewed as a fingerprint of a specific operational condition.

[0085] In some examples, operational metadata may include one or more of the following: the average, mean, and standard deviation of operational data over a time period of operational conditions. Operational data may include one or more pitch angles of one or more blades, the rotor speed of the generator or wind turbine rotor, and the electrical power output from the generator. Additional operational data may include internal signals calculated for monitoring or signals that trigger control decisions. The operational data from which operational metadata is derived may also vary over time and according to different operational conditions.

[0086] In addition to statistical metadata, other metadata, such as that derived from AI-based classification techniques, can be used. Diagnostic metadata related to the failure of one or more systems controlled by the wind turbine controller can also be used in the example.

[0087] In some examples, the operating condition could be a power grid event.

[0088] In some examples, as will be illustrated herein, operational metadata can be used to determine (cumulative) fatigue loads. In other examples, operational metadata can be used for remote diagnostics or troubleshooting.

[0089] Module 500 for determining cumulative fatigue loads can be configured at block 510 to determine whether the operational metadata of an operating condition substantially corresponds to the operational metadata of one of a plurality of previously defined operating conditions. If the operational metadata corresponds to one of the stored operating conditions, block 520, then the rainflow cycle for the operating condition is determined to be equal to the rainflow cycle of the corresponding stored operating condition. The rainflow cycle can be converted into a damage equivalent load, as will be explained herein.

[0090] The module 520 for determining cumulative fatigue load may be further configured to send 524 operational metadata of the operational condition to the load estimation module 300 when the operational metadata of the operational condition does not correspond to any operational condition stored in the operational condition.

[0091] The module 500 used to determine the cumulative fatigue load can also be configured to determine the remaining lifetime of a wind turbine or one or more components 560 of a wind turbine based on the total damage equivalent load.

[0092] In the example, module 500 for determining cumulative fatigue loads can be stored on a server at a location away from the wind turbine. Similarly, load estimation module 300 can also be located away from the wind turbine.

[0093] Therefore, the system can be configured to create new operating conditions and store the operating metadata for the new operating conditions. The load calculation module 300 is configured to determine the fatigue load 330 for the new operating condition. The fatigue load for the new operating system can be represented as rainflow counts. The new operating condition contains the corresponding operating metadata in the database 515.

[0094] The load estimation module 300 can be, for example, a high-performance computing load calculation cluster in a remote data center. The load estimation module can be a digital or virtual "twin" of the wind turbine, as described herein. The load estimation module 300 can simulate the output of a numerical model given a set of operational metadata. The numerical module can be the same as the numerical module used for certification of the wind turbine. Because the numerical model can be the same as the numerical model used for certification, fidelity and accuracy can be high.

[0095] Load estimation module 300 can be configured to create new operating conditions based on operating metadata and estimated fatigue loads. Module 310 can interpret the operating metadata received from module 500 as input settings for numerical load estimation model 320. Module 310 can generate multiple variants capable of parallel computation.

[0096] At 330°, the Rainflow Count (RFC) algorithm can be used to determine fatigue loads. The Rainflow Count algorithm is used in the analysis of fatigue data to reduce the varying stress spectrum to an equivalent set of simple stress reversals.

[0097] The load estimation module 300 can calculate fatigue loads based on the RFC cycle for all relevant wind turbine components.

[0098] Now that the main components of system 220 and the functions of the local wind turbine module 400 have been discussed, other aspects of this disclosure will be explained in more detail.

[0099] In this disclosure, a method for determining cumulative fatigue loads in a wind turbine is provided. This method can be performed at least in part by a module 500 for determining fatigue loads. An example of such a method can be seen with reference to the figures illustrating it. Figure 4 .

[0100] The method includes obtaining operation metadata representing an operation condition at 410, and determining at 510 and 520 whether the operation metadata corresponds to the operation metadata of one of a plurality of previously defined operation conditions, for which fatigue loads are known, and the plurality of previously defined operation conditions are stored in an operation condition database 505.

[0101] The previously defined operating conditions stored in databases 505 and 515 may include actual operating conditions that have occurred during the lifespan of the wind turbine, as well as simulated operating conditions that may have been performed before or during the lifespan of the wind turbine. Some previously defined operating conditions may be based on design load conditions defined in standards such as IEC 61400.

[0102] The method then further includes summing the fatigue loads of the previously defined operating conditions corresponding to the metadata to the historical cumulative fatigue loads of the wind turbine to determine the total cumulative fatigue load.

[0103] Based on the total cumulative fatigue load, wind turbines can perform operational changes, such as downrating. 、 Increase power operation (uprating) or pitching.

[0104] If the operational metadata of an operating condition matches (i.e., the operational metadata substantially corresponds to) the operational metadata of a previously defined operating condition, then the fatigue load can be assumed to be the same as the fatigue load of a known condition. Matching can be understood as the operational metadata of the current operating condition not significantly deviating from the operational metadata of a previously defined operating condition, such that the corresponding fatigue load still sufficiently corresponds to the load of the known condition. Insignificant deviation can be defined in several ways. One example is that all or a selection of the variables constituting the operational metadata deviate from the stored metadata from the previously defined operating condition by less than a corresponding predefined threshold. Another example is that the deviation of the metadata from the known condition is weighted. Other more complex algorithms involving artificial neural networks or other machine learning techniques (e.g., ensemble learning) can be used to determine when the correspondence of the metadata is sufficient, i.e., when the operational metadata sufficiently corresponds to the known condition. In further examples, different matching algorithms can be used for different operating ranges and conditions. The matching algorithm can be different for, for example, high wind speeds, low wind speeds, high turbulence, power grid events, etc.

[0105] In some examples, the operating condition can be a grid event. To illustrate the above principle: if the PLC's operating signal is the same as a previously defined operating condition stored in memory, the fatigue load can be determined to be the same as the known load for a known operating condition. If a grid event occurs (e.g., a low voltage event or a decrease in the power setpoint), the wind turbine will react with certain control settings for rotor speed, electrical power output, and pitch angle. If an operation with similar rotor speed, electrical power output, pitch angle (even if this involves different operating conditions) and other external conditions can be found in the database, the fatigue load is assumed to be the same as the fatigue load for a known condition.

[0106] In some examples, the method may further include sending the operation metadata of operation condition 524 to the load estimation module 300 if the operation metadata during the operation condition does not correspond to the operation metadata of any operation condition from multiple previously defined operation conditions.

[0107] In some examples, new operating conditions (operating metadata and fatigue loads) can be stored in operating condition databases 505 and 515. By defining new operating conditions, their corresponding operating metadata, and corresponding fatigue loads, the database of known operating conditions can be continuously enriched. Therefore, new operating conditions become known or previously defined operating conditions, and they can be matched with future operating conditions.

[0108] As previously indicated, the method in the example may further include determining the damage equivalent load 540 for the operating condition based on the rainflow count cycle 505 for the known operating condition. The rainflow count can be converted into the damage equivalent load (DEL) at block 540 via a Markov matrix and accumulation.

[0109] When operational conditions occur during the operation and lifespan of a wind turbine, the fatigue loads corresponding to these operational conditions can be registered, and each received metadata encapsulation can be used to determine and update the fatigue loads that the wind turbine has endured throughout its lifespan.

[0110] In some examples, the method may further include estimating the remaining lifetime of one or more components of a wind turbine based on the total cumulative fatigue load 540. For one or more components of the wind turbine, a threshold or limit regarding fatigue load may be defined in a database 550. By comparing the applied load 540 with the limit 550, the remaining operational lifetime 565 can be estimated for a selection of the wind turbine or its components.

[0111] Based on all recorded conditions, a condition report (e.g., a list) can be generated at point 575.

[0112] The examples disclosed herein can provide one or more of the following advantages: damage corresponding to transient power grid events can be quantified. Such quantification can be used in the determination of countermeasures and can help determine the causes of damage to wind turbine components.

[0113] It can provide insights into the conditions that wind turbines actually encounter during their lifespan, which can help define future guidelines and requirements.

[0114] The use of wind turbines can be optimized. By taking grid events into account, the remaining lifetime can be determined more accurately. If the remaining lifetime is determined to be longer than expected, operational changes can be made to maximize power output. Preventative maintenance can be planned when necessary.

[0115] In addition, new design load conditions that may be incorporated into the standard can be generated based on grid conditions or other new conditions (reductions caused by noise or temperature, wind misalignment, unusual turbulence, etc.) and their loads.

[0116] Examples of the methods disclosed herein may be implemented using hardware, software, firmware, and / or combinations thereof. Examples of the methods disclosed herein may employ one or more of virtual machines, cloud computing, and edge computing.

[0117] Those skilled in the art will also appreciate that the various illustrative logic blocks, modules, circuits, and algorithmic steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps are generally described above in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art can implement the functionality in variations for each specific application.

[0118] The various illustrative logic blocks, modules, and circuits described herein can be designed to perform the functions described herein using one or more general-purpose processors, digital signal processors (DSPs), cloud computing architectures, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination thereof. The general-purpose processor may be a microprocessor, but alternatively, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration.

[0119] This disclosure also relates to computational systems suitable for performing the methods disclosed herein.

[0120] This disclosure also relates to a computer program or computer program product containing instructions (code) that, when executed, perform any of the methods disclosed herein.

[0121] Computer programs may take the form of source code, object code, intermediate source code, and object code (e.g., in a partially compiled form), or any other form intended for use in the implementation of a process. The carrier may be any entity or device capable of carrying a computer program.

[0122] If implemented via software / hardware, the functionality can be stored on a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, encompassing any medium that facilitates the transfer of a computer program from one location to another. Storage media can be any available medium accessible by a general-purpose or special-purpose computer. By way of example and not limitation, such computer-readable media can include RAM, ROM, EEPROM, CD / DVD or other optical disc storage devices, magnetic disk storage devices or other magnetic storage devices, or any other medium capable of carrying or storing intended program code components in the form of instructions or data structures and accessible by a general-purpose or special-purpose computer or a general-purpose or special-purpose processor. Furthermore, any connection is appropriately referred to as a computer-readable medium. For example, if software / firmware is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technology (e.g., infrared, radio, and microwave), then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technology (e.g., infrared, radio, and microwave) is included in the definition of medium. As used herein, disks and optical discs include compact optical discs (CDs), laser discs, optical discs, digital versatile optical discs (DVDs), floppy disks, and Blu-ray discs. Disks typically reproduce data magnetically, while optical discs use lasers to reproduce data optically. Combinations described above should also be included within the scope of computer-readable media.

[0123] This written description uses examples, including preferred embodiments, to disclose the invention and also enables those skilled in the art to practice the invention, including making and using any apparatus or system, and performing any combination method. The patentable scope of the invention is defined by the claims and may include other examples that may occur to those skilled in the art. Such other examples are intended to fall within the scope of the claims if they have structural elements that are exactly the same as the literal language of the claims, or if they include equivalent structural elements that have a non-substantially different literal language from the claims. Aspects from the various embodiments described, and other known equivalents of each such aspect, can be mixed and matched by those skilled in the art to constitute additional embodiments and techniques according to the principles of this application. If reference numerals associated with the drawings are enclosed in parentheses in the claims, they are used only to attempt to increase the comprehensibility of the claims and should not be construed as limiting the scope of the claims.

Claims

1. A method for determining cumulative fatigue loads in a wind turbine, the method comprising: Obtain operational metadata representing the operational status, wherein the operational metadata provides a description of a specific operational status and is derived from the operational data by processing and / or compressing the operational data based on statistical parameters, wherein the operational metadata is a fingerprint of the specific operational status; Determine whether the operation metadata corresponds to the operation metadata of one of a plurality of previously defined operation conditions, for which the fatigue load is known and the plurality of previously defined operation conditions are stored in an operation condition database; as well as If the operation metadata representing the operation state substantially corresponds to the operation metadata of one of the plurality of previously defined operation states, then The fatigue loads corresponding to the previously defined operating conditions of the metadata are summed to the historical cumulative fatigue loads of the wind turbine to determine the total cumulative fatigue load, and wherein... The local wind turbine module determines operational data including signals and / or setpoints of one or more wind turbine components under the control of a wind turbine controller, and determines changes in the operational data indicating the end of the operational state, and determines the operational metadata of the wind turbine representing the operational state. If the operation metadata during the operation state does not correspond to the operation metadata of any of the plurality of previously defined operation states, the operation metadata of the operation state is sent to the load estimation module.

2. The method as described in claim 1, wherein, The load estimation module creates new operating conditions based on the operating metadata and estimated fatigue load.

3. The method as described in claim 2, wherein, The new operational status is stored in the operational status database.

4. The method of any one of claims 1-3, further comprising determining the damage equivalent load for the operating condition based on rainflow counting cycles for a known operating condition.

5. The method of any one of claims 1-3, further comprising estimating the remaining service life of one or more components of the wind turbine based on the total cumulative fatigue load.

6. The method according to any one of claims 1-3, wherein, The operational status mentioned is a power grid event.

7. The method according to any one of claims 1-6, wherein, The operational metadata includes one or more of the average, median, and standard deviation of operational data over the time period of the operational status.

8. The method of claim 3, wherein, The operating data includes one or more of the following: one or more pitch angles of one or more blades; the rotor speed of the generator or wind turbine rotor; and the electrical power output from the generator.

9. A system for determining fatigue loads in a wind turbine, comprising: A local wind turbine module is used to determine the operating metadata of the wind turbine representing the operating condition, wherein the operating metadata provides a description of a specific operating condition and is derived from the operating data by processing and / or compressing the operating data based on statistical parameters, wherein the operating metadata is a fingerprint of the specific operating condition. A load estimation module configured to determine the fatigue load of the operating condition characterized by the operating metadata; as well as A module for determining cumulative fatigue load is configured to determine the cumulative damage equivalent load for the wind turbine by summing the damage equivalent load for operational conditions that have occurred during the service life of the wind turbine, wherein the local wind turbine module is configured to: Determine the operational data of signals and / or setpoints for one or more wind turbine components under the control of the wind turbine controller; Determine changes in the operation data that indicate the end of the operation status; And configured as Determine the operational metadata of the wind turbine representing the operational status. The module for determining the cumulative fatigue load is configured to determine whether the operation metadata of the operating condition substantially corresponds to the operation metadata of one of a plurality of previously defined operating conditions, and If the operation metadata corresponds to one of the stored operation states... The fatigue load for the aforementioned operating condition is then determined to be equal to the fatigue load for the corresponding stored operating condition. The module for determining the cumulative fatigue load is configured to send the operation metadata of the operation condition to the load estimation module when the operation metadata of the operation condition does not correspond to any operation condition of the stored operation condition.

10. The system of claim 9, wherein, The local wind turbine module is configured to send operational metadata representing the operational status to the module used to determine the cumulative fatigue load.

11. The system according to any one of claims 9-10, wherein, The module used to determine the cumulative fatigue load is configured to determine the remaining service life of the wind turbine or one or more components of the wind turbine based on the total damage equivalent load.

Images