Load determination method and device for wind turbine generator system, wind turbine generator system

Abstract

This disclosure provides a method and apparatus for determining the load of a wind turbine generator set, and the wind turbine generator set itself. The load determination method includes: acquiring an actual sensing signal sensed by a target load sensor; obtaining a reference sensing signal based on the current operating data of the wind turbine generator set using a preset virtual load sensor corresponding to the target load sensor; correcting the actual sensing signal based on the reference sensing signal to obtain a corrected sensing signal; and determining the load at the target location based on the corrected sensing signal. This disclosure solves the problem that inaccurate sensing results from hardware sensors may prevent the timely detection of safety hazards in the generator set. It allows for real-time load calibration of hardware sensors without affecting the power generation of the wind turbine generator set, more accurately determining the generator set load and ensuring the safety of the generator set.

Description

Technical Field

[0001] This disclosure relates to the field of wind power generation, and more specifically, to a method and apparatus for determining the load of a wind turbine generator set, and a wind turbine generator set. Background Technology

[0002] Wind turbine generators are typically designed for a service life of 20 years. However, during actual operation, wind resources, generator operating conditions, generator models, and control strategies are likely to differ from the initial design. Therefore, it is necessary to obtain data on the actual damage to the generator to ensure its safety.

[0003] For example, hardware sensors can be installed on different components or locations of the unit to obtain the actual load condition of the unit and then deduce the actual damage to the unit. However, due to environmental and other factors, the sensing results of hardware sensors may be affected, resulting in a certain deviation between the measured load value and the actual load value of the unit. This makes it impossible to accurately assess the load status of the unit and thus fail to detect potential safety hazards in the unit in a timely manner. Summary of the Invention

[0004] In view of the problem that inaccurate sensing results of hardware sensors may lead to the failure to detect potential safety hazards of the unit in a timely manner, this disclosure proposes a method and device for determining the load of a wind turbine generator set, and a wind turbine generator set, so as to solve or at least alleviate the above-mentioned problems.

[0005] The first aspect of this disclosure provides a method for determining the load of a wind turbine generator set. The method includes: acquiring an actual sensing signal sensed by a target load sensor, wherein the target load sensor is used to sense the load at a target location of the wind turbine generator set; obtaining a reference sensing signal based on the current operating data of the wind turbine generator set and using a preset virtual load sensor corresponding to the target load sensor; correcting the actual sensing signal based on the reference sensing signal to obtain a corrected sensing signal; and determining the load at the target location based on the corrected sensing signal.

[0006] Optionally, the load determination method further includes: determining a reference signal range including the reference sensing signal based on the reference sensing signal; comparing the actual sensing signal with the reference signal range; in response to the comparison result indicating that there is data in the actual sensing signal that exceeds the reference signal range, performing the step of correcting the actual sensing signal based on the reference sensing signal; and in response to the comparison result indicating that the actual sensing signal is within the reference signal range, determining the load at the target location based on the actual sensing signal.

[0007] Optionally, the load determination method further includes: in response to the comparison result indicating that the proportion of data in the actual sensed signal that exceeds the range of the reference signal exceeds a preset abnormal range, determining that the target load sensor is abnormal.

[0008] Optionally, the actual sensing signal includes multiple measured data arranged in time sequence, each measured data corresponding to a reference sensing signal. The step of correcting the actual sensing signal based on the reference sensing signal includes: determining a reference signal range corresponding to each measured data based on the reference sensing signal corresponding to each measured data; and obtaining the corrected sensing signal by correcting the measured data that exceed the corresponding reference signal range among the multiple measured data to the corresponding reference signal range.

[0009] Optionally, the corrected sensing signal is obtained by: while keeping the signal waveform of the current target measured data unchanged, correcting the current target measured data to obtain corrected measured data, wherein, in the initial correction, the target measured data is the measured data that exceeds the corresponding reference range among the plurality of measured data; using the measured data within the corresponding reference range in the corrected measured data as the corrected sensing signal; in response to the presence of measured data exceeding the corresponding reference range in the corrected measured data, using the measured data exceeding the corresponding reference range in the corrected measured data as the target measured data for the next correction, and returning to the step of correcting the current target measured data.

[0010] Optionally, the reference signal range is determined by: determining the reference signal range corresponding to the reference sensing signal based on the reference sensing signal and a preset confidence level, wherein the confidence level characterizes the error range of the virtual load sensor.

[0011] Optionally, the corrected measured data includes first corrected measured data and / or second corrected measured data. The first corrected measured data is obtained by: determining a first measured data point in the current target measured data, wherein the first measured data point is a target measured data point greater than the upper limit of the corresponding reference signal range; determining a first difference between the maximum value in the first measured data point and the upper limit of the corresponding reference signal range; subtracting the first difference from each first measured data point to obtain the first corrected measured data. The second corrected measured data is obtained by: determining a second measured data point in the current target measured data, wherein the second measured data point is a target measured data point less than the lower limit of the corresponding reference signal range; determining a second difference between the minimum value in the second measured data point and the lower limit of the corresponding reference signal range; adding the second difference to each second measured data point to obtain the second corrected measured data.

[0012] Optionally, the corrected sensing signal is obtained by: correcting the target measured data that exceeds the corresponding reference signal range in the plurality of measured data according to a window of preset duration, thereby obtaining the corrected sensing signal; wherein, in response to the proportion of the target measured data in the current window relative to all measured data in the current window being less than or equal to a preset proportion threshold, the target measured data in the current window is corrected to obtain the corrected sensing signal corresponding to the current window; wherein, in response to the proportion of the target measured data in the current window relative to all measured data in the current window being greater than the preset proportion threshold, the current window and the subsequent window are merged according to the proportion of the target measured data in the subsequent window after the current window relative to all measured data in the corresponding window, thereby obtaining a merged window; while keeping the signal waveform of the target measured data unchanged, the target measured data that exceeds the corresponding reference signal range in the measured data in the merged window is corrected to obtain the corrected sensing signal corresponding to the merged window.

[0013] Optionally, the virtual load sensor is obtained by: determining a first transfer function of the target load sensor based on the sensor type and sensing material of the target load sensor, wherein the first transfer function characterizes the conversion relationship between the sensing signal of the target load sensor and the load at the target location; and obtaining the virtual load sensor based on the first transfer function and a second transfer function, wherein the second transfer function characterizes the conversion relationship between the operating data of the wind turbine generator and the load at the target location.

[0014] Optionally, the input data of the virtual load sensor includes the configuration parameters and current operating data of the wind turbine generator set, and the output signal of the virtual load sensor has the same signal form as the sensing signal of the target load sensor.

[0015] A second aspect of this disclosure provides a load determination device for a wind turbine generator set. The load determination device includes: an acquisition unit configured to acquire an actual sensing signal sensed by a target load sensor, wherein the target load sensor is used to sense the load at a target location of the wind turbine generator set; a signal determination unit configured to obtain a reference sensing signal based on the current operating data of the wind turbine generator set and using a preset virtual load sensor corresponding to the target load sensor; a correction unit configured to correct the actual sensing signal based on the reference sensing signal to obtain a corrected sensing signal; and a load determination unit configured to determine the load at the target location based on the corrected sensing signal.

[0016] A third aspect of this disclosure provides an electronic device comprising: a processor; and a memory for storing processor-executable instructions, wherein, when executed by the processor, the processor causes the processor to perform a load determination method for a wind turbine generator according to this disclosure.

[0017] A fourth aspect of this disclosure provides a wind turbine generator set, the wind turbine generator set including electronic equipment according to this disclosure, or the control system of the wind turbine generator set being connected to electronic equipment according to this disclosure.

[0018] A fifth aspect of this disclosure provides a computer-readable storage medium that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform a load determination method for a wind turbine generator according to this disclosure.

[0019] A sixth aspect of this disclosure provides a computer program product including computer-executable instructions that, when executed by at least one processor, implement the load determination method for a wind turbine generator set according to this disclosure.

[0020] According to the load determination scheme for wind turbine generators disclosed herein, a reference sensing signal can be obtained using a preset virtual load sensor based on the current operating data of the wind turbine generator. Based on this reference sensing signal, the actual sensing signal of the target load sensor is corrected to obtain a corrected sensing signal, thereby determining the load at the target location. This allows for the calibration of the hardware sensor's sensing signal, correcting signal deviations and thus more accurately determining the generator load, ensuring generator safety. Furthermore, this scheme allows for real-time load calibration of the hardware sensor without affecting the wind turbine generator's power generation, eliminating the need for generator shutdown or setting specific pitch angles. While improving the accuracy of load assessment, it also avoids affecting the normal operation of the generator and does not introduce additional control actions, making it highly valuable in practical applications. Attached Figure Description

[0021] Figure 1 This is a schematic flowchart illustrating a method for determining the load of a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0022] Figure 2 This is a schematic block diagram illustrating the signal acquisition of hardware sensors in a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0023] Figure 3 This is a flowchart illustrating the step of obtaining a virtual load sensor in a load determination method for a wind turbine generator according to an exemplary embodiment of the present disclosure.

[0024] Figure 4 This is a schematic diagram illustrating the sensitivity analysis used to construct a virtual load sensor in a load determination method for a wind turbine generator according to an exemplary embodiment of the present disclosure.

[0025] Figure 5 This is a flowchart illustrating the step of correcting the actual sensed signal in a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0026] Figure 6A This is a schematic diagram illustrating the model error of a virtual load sensor in a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0027] Figure 6B This is a schematic diagram illustrating the frequency of occurrence of model errors of a virtual load sensor in a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0028] Figure 7 This is a waveform diagram illustrating the reference load signal and the true value of the load signal in a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0029] Figure 8 This is a flowchart illustrating the steps of obtaining the corrected sensing signal in a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0030] Figure 9 This is a waveform diagram illustrating the range of a reference signal in a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0031] Figure 10 This is a waveform diagram illustrating the actual load signal in a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0032] Figure 11 This is a flowchart illustrating an application example of a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0033] Figure 12 This is a schematic block diagram illustrating an example implementation of a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0034] Figure 13 This is a schematic block diagram illustrating a load determination device for a wind turbine generator set according to an exemplary embodiment of the present disclosure. Detailed Implementation

[0035] The following detailed embodiments are provided to aid the reader in gaining a comprehensive understanding of the methods, apparatus, and / or systems described herein. However, various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent upon understanding this disclosure. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but may be changed as will become clear upon understanding this disclosure, except for operations that must occur in a specific order. Furthermore, for clarity and conciseness, descriptions of features known in the art may be omitted.

[0036] The features described herein may be implemented in different forms and should not be construed as limited to the examples described herein. Rather, the examples described herein are provided only to illustrate some of the many feasible ways of implementing the methods, apparatus, and / or systems described herein, which will become clear upon understanding the disclosure of this application.

[0037] As used herein, the term “and / or” includes any one of the associated listed items and any combination of any two or more.

[0038] Although terms such as “first,” “second,” and “third” may be used herein to describe various components, assemblies, regions, layers, or parts, these components, assemblies, regions, layers, or parts should not be limited by these terms. Rather, these terms are used only to distinguish one component, assembly, region, layer, or part from another. Thus, without departing from the teaching of the examples described herein, the first component, first assembly, first region, first layer, or first part referred to as the first component, first assembly, first region, first layer, or first part may also be referred to as the second component, second assembly, second region, second layer, or second part.

[0039] In the specification, when an element (such as a layer, region, or substrate) is described as being "on" another element, "connected to," or "bonded to" another element, the element may be directly "on" another element, directly "connected to," or "bonded to" the other element, or one or more other elements may be present in between. Conversely, when an element is described as being "directly on" another element, "directly connected to," or "directly bonded to" another element, no other elements may be present in between.

[0040] The terminology used herein is for the purpose of describing various examples only and is not intended to limit disclosure. Unless the context clearly indicates otherwise, the singular form is intended to include the plural form as well. The terms “comprising,” “including,” and “having” indicate the presence of the described features, quantities, operations, components, elements, and / or combinations thereof, but do not preclude the presence or addition of one or more other features, quantities, operations, components, elements, and / or combinations thereof.

[0041] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains upon understanding this disclosure. Unless expressly defined herein, terms (such as those defined in a general dictionary) shall be interpreted as having a meaning consistent with their meaning in the context of the relevant field and in this disclosure, and shall not be interpreted in an idealized or overly formalistic manner.

[0042] Furthermore, in the description of the examples, detailed descriptions of well-known related structures or functions will be omitted when it is believed that such detailed descriptions would lead to a vague interpretation of this disclosure.

[0043] As mentioned earlier, the inaccurate sensing results of hardware sensors make it impossible to accurately assess the load status of the unit, resulting in the inability to detect potential safety hazards in the unit in a timely manner.

[0044] Specifically, both resistance strain gauge sensors and fiber optic sensors are affected by factors such as the environment in actual operation. Therefore, sensor calibration is required to obtain the correct load value of the unit, which is used to assess the unit's losses or to optimize and upgrade the control strategy.

[0045] In some cases, data can be acquired when the unit is shut down or at a set pitch angle to calibrate the hardware sensors. However, such calibration methods can cause power generation losses for the wind turbine.

[0046] Furthermore, the requirements for the accuracy of load measurement are becoming increasingly stringent, whether in the control of wind turbine generators or in the safety monitoring of wind turbine generators. However, during the operation of load sensors, there may be abnormalities such as sensor detachment or damage, which may lead to inaccurate measurements. However, in related technologies, there is a lack of early warning for the working status of the hardware sensors used for load measurement.

[0047] In view of these problems, this disclosure provides a method for determining the load of a wind turbine generator set, a device for determining the load of a wind turbine generator set, a wind turbine generator set, a computer-readable storage medium, and a computer program product, which can solve or at least alleviate the above-mentioned problems.

[0048] According to a first aspect of an exemplary embodiment of the present disclosure, a method for determining the load of a wind turbine generator is provided. This load determination method can be performed by an electronic device with computing capabilities. The electronic device may be, for example, a terminal device or a server, wherein the terminal device may be such as a tablet computer, a laptop computer, a digital assistant, etc.; the server may be a standalone server, a server cluster, a cloud computing platform, or a virtualization center.

[0049] Here, electronic devices can be installed at wind turbine generators or wind farms, and can be communicatively connected to measuring devices or data centers of wind turbine generators or wind farms, thereby acquiring the data required to perform the above methods, such as sensing signals from hardware sensors.

[0050] According to embodiments of this disclosure, such as Figure 1 As shown, the load determination method may include the following steps:

[0051] In step S110, the actual sensing signal sensed by the target load sensor can be obtained.

[0052] Here, the target load sensor can be used to sense the load at a target location on the wind turbine generator set. As an example, the target location may include, but is not limited to, the bottom of the wind turbine generator set, the top of the wind turbine generator set, the blade root, etc.

[0053] The target load sensor can be a hardware sensor used to measure the load, such as, but not limited to, a resistance strain gauge sensor or a fiber optic sensor such as a fiber optic grating. The target load sensor can emit a sensing signal, which can be sent, for example, to the unit's control system or monitoring platform, and this sensing signal can be converted into a load. As an example, the sensing signal of the target load sensor can be an electrical signal such as a voltage signal; however, it is not limited to this. Depending on the type of sensor and the sensing method, the sensing signal can also be other forms of signal such as an optical signal. As an example, other forms of signal can first be converted into electrical signals, and then the measured load data can be obtained based on the electrical signals.

[0054] Figure 2 An example procedure for acquiring the sensing signal from the target load sensor is shown. For example... Figure 2 As shown, the sensing signal of the target load sensor can be acquired. If the sensing signal of the target load sensor is a non-electrical signal, it can be converted into an electrical signal using existing signal conversion methods, for example, using... Figure 2 Taking a fiber optic sensor as an example, the optical signal generated by the fiber optic sensor can be connected to a fiber optic junction box through the tail fiber, and then connected to a fiber optic demodulator, thereby converting the optical signal into an electrical signal.

[0055] In step S120, a reference sensing signal can be obtained based on the current operating data of the wind turbine generator set by using a preset virtual load sensor corresponding to the target load sensor.

[0056] As an example, corresponding virtual load sensors can be set up for target load sensors used to sense loads at different target locations. These virtual load sensors can also be considered software sensors, which can determine the load at the corresponding target location through calculation or simulation. Here, the virtual load sensors can be pre-built, and the process of obtaining them will be described in detail below.

[0057] In embodiments of this disclosure, the input data of the virtual load sensor may include configuration parameters and current operating data of the wind turbine generator.

[0058] Here, the configuration parameters of the wind turbine generator set can include, for example, blade configuration parameters and tower configuration parameters. The operating data of the wind turbine generator set can include, for example, generator speed, power, pitch angle, and rotor azimuth angle. Depending on the target location, configuration parameters and operating data related to the target location can be used as input data for the virtual load sensor. Operating data can be, for example, real-time data such as SCADA signals. This designed virtual load sensor can establish the relationship between the current real-time state of the generator set and the load at the target location, thereby enabling real-time and dynamic load calculation, resulting in more accurate calculation results.

[0059] Furthermore, the output signal of the virtual load sensor can have the same signal form as the sensing signal of the target load sensor. Specifically, if the sensing signal of the target load sensor is an electrical signal, the output signal of the virtual load sensor can also be an electrical signal. The output signal of the virtual load sensor designed in this way can be directly used to correct the sensing signal of the hardware sensor. Compared with a virtual load sensor that outputs a calculated load value, it can provide a more accurate reference and avoid introducing errors in the process of converting the sensing signal to the load.

[0060] As an example, a virtual load sensor can be obtained through simulation calculation or testing. The virtual load sensor can be, for example, a machine learning model obtained through training or a statistical model constructed using statistical methods.

[0061] In one example, such as Figure 3 As shown, a virtual load sensor can be obtained in the following way:

[0062] In step S310, the first transfer function of the target load sensor can be determined based on the sensor type and sensing material of the target load sensor.

[0063] Here, the first transfer function can characterize the conversion relationship between the sensing signal of the target load sensor and the load at the target location.

[0064] Specifically, in the field of wind power, load measurement can typically be performed using methods such as resistance voltage strain gauges and fiber Bragg gratings. By applying a known load to different hardware sensors and processing the signal, the sensing signal (e.g., an electrical signal) corresponding to the known load can be obtained. From this, the relationship between the load and the sensing signal can be obtained, thereby constructing the first transfer function.

[0065] As an example, this process can be achieved through laboratory testing or in the routine calibration of wind turbine generators. Here, the transfer functions of load sensors of different types or with different sensing materials are different. For example, for a target load sensor, the conversion between the load and the sensed signal can be represented by the following equation (1):

[0066] M flap =W(i,j)*V (1)

[0067] Where W(i,j) represents the first transfer function, M flap V represents the load, V represents the sensing signal (e.g., a voltage signal), i represents the type of target load sensor, and j represents the sensing material of the target load sensor.

[0068] In step S320, a virtual load sensor can be obtained based on the first transfer function and the second transfer function.

[0069] Here, the second transfer function can characterize the conversion relationship between the operating data of the wind turbine generator and the load at the target location.

[0070] Specifically, the loads on different components or locations of the unit can be obtained through simulation calculations or tests. Furthermore, the relationship between the unit's operating data and the loads, i.e., the transfer function G, can be established using machine learning and other methods. Here, the unit's operating data can be the unit's current state, which may include, but is not limited to, speed, power, pitch angle, and impeller azimuth angle. In addition, depending on the unit's configuration parameters such as blades and tower, the second transfer function can also take different forms.

[0071] For example, the leaf root waving bending moment M flap For example, for the target location, the conversion between the load and the operating data of the wind turbine generator can be represented by the following equation (2):

[0072] M flap =G flap (r,p,b,a,m) (2)

[0073] Among them, M flap G represents the load. flap Let G represent the second transfer function, which can be a transfer function converted from rotational speed r, power p, pitch angle b, impeller azimuth angle a, and blade mass m into blade root flapping moment. Here, the second transfer function G varies depending on the target position. flap The independent variables can be different and can be determined based on existing physical analysis or simulation calculations. An example process for determining the operating data and / or configuration parameters associated with the second transfer function according to embodiments of this disclosure will be given later.

[0074] Based on equations (1) and (2) above, the load at the target location can be used as an intermediate quantity to establish the relationship between the first transfer function and the second transfer function, thereby constructing a virtual load sensor. Specifically, the following equation (3) can be obtained:

[0075] G flap (r,p,b,a,m)=W(i,j)*V (3)

[0076] Furthermore, the expression for the output signal of the virtual load sensor (e.g., the aforementioned reference sensing signal) can be obtained as follows:

[0077]

[0078] In this way, the load can be cleverly used as an intermediate quantity to establish the relationship between the real-time status and configuration of the unit and the load sensing signal, and to construct a virtual load sensor. The output signal of the virtual load sensor corresponds to the sensing signal of the target load sensor, so that it can be used to correct its actual sensing signal.

[0079] As described above, the second transfer function can vary depending on the target location. In one example, the operating data associated with the second transfer function can be determined based on existing physical analysis or simulation calculations. In another example, embodiments of this disclosure provide an example process for determining the operating data and / or configuration parameters associated with the second transfer function.

[0080] Sensitivity analysis can be performed on the data variables in the operating data of the wind turbine generator set and the load at the target location to determine the second transfer function.

[0081] Specifically, the second transfer function can be determined as follows: acquire target operating data of the wind turbine generator set over a predetermined time period; remove data from the target operating data to obtain the data after removal; analyze the correlation between each data variable in the data after removal and the load at the target location to obtain the correlation analysis results; based on the correlation analysis results, determine the data variables in the data after removal that are related to the second transfer function, thereby determining the second transfer function. Here, data variables can refer to data types such as speed, power, and pitch angle mentioned above. Furthermore, the target operating data can be, for example, SCADA signal data.

[0082] As an example, removing data from target operational data may include: removing data from target operational data that is unrelated to the load at the target location; and / or removing non-general data from target operational data; and / or removing data from target operational data whose fluctuation differences within the aforementioned predetermined time period are less than a preset difference.

[0083] Specifically, data unrelated to the load at the target location may include environmental variables. For example, if the target location is the blade root, environmental variables may include measured wind speeds of non-flowing winds, such as wind speeds measured behind the rotor. Environmental variables may also include temperature. Non-universal data refers to data that differs across different wind turbine generator sets. This may include parameters of functional modules, such as those of radar, BeiDou satellite navigation systems, and general application platforms for field control. Different generator sets may be equipped with different radars, satellite positioning systems, or communication systems, therefore this part of the data may be disregarded.

[0084] Data fluctuations can be determined, for example, by analyzing the variance of statistical data. Specifically, the variance of all data for each feature (or data variable) is calculated. A larger variance indicates greater distinctiveness of the feature, while a smaller variance indicates less variation within the feature. When the variance is less than a preset variance (e.g., zero variance), it means the feature values ​​are identical, and features with zero variance can be removed. The preset variance can be set according to actual needs, and the statistical data can be calculated using methods other than variance, as long as it characterizes the degree of data fluctuation.

[0085] Correlation analysis results can characterize the correlation between each data variable and the load at the target location. Here, the correlation between each data variable and the load at multiple target locations can be analyzed.

[0086] As an example, correlation analysis results can be obtained by determining the maximum information coefficient. Specifically, the maximum information coefficient can be used to measure the linear or nonlinear relationship between two variables, and it has universality, fairness, and symmetry. The coefficients of each input variable (e.g., each data variable) with respect to each output variable (e.g., loads at multiple target locations) can be calculated separately. Figure 4 As shown, SCADA signals can include multiple data variables ( Figure 4 Each column in the table corresponds to a data variable, and the load variable can include loads at multiple target locations. Figure 4 Each row in the graph corresponds to a target location, and the intersection of the horizontal and vertical axes represents the correlation, such as the value of the maximum information coefficient.

[0087] In this example, based on the correlation analysis results, data variables with a correlation greater than a preset value, such as the maximum information coefficient, can be selected. The union of all selected data variables is then used as the data variables related to the second transfer function to determine the second transfer function.

[0088] By using the above methods, the data variables related to the second transfer function can be determined reasonably and accurately, thereby determining the second transfer function and enabling the construction of a virtual sensor that is more in line with the actual situation and provides more accurate load estimation.

[0089] Although the second transfer function has been described above as being determined based on operating data, the embodiments of this disclosure are not limited thereto. The second transfer function may also be related to unit configuration parameters. For example, the second transfer function may be constructed based on both the operating data and configuration parameters related to the second transfer function.

[0090] Here, configuration parameters related to each target location can be determined based on physical meaning. For example, when the target location is the top of the tower, the configuration parameters related to the second transfer function may include, but are not limited to, yaw angle, yaw rate, tower height, and nose mass; when the target location is the bottom of the tower, the configuration parameters related to the second transfer function may include, but are not limited to, tower weight and power; when the target location is the blade root, the configuration parameters related to the second transfer function may include, but are not limited to, blade weight, azimuth angle, pitch angle, rotational speed, and power.

[0091] Return to reference Figure 1 In step S130, the actual sensing signal can be corrected based on the reference sensing signal to obtain the corrected sensing signal.

[0092] As an example, the actual sensed signal may include multiple measured data points arranged in time sequence, each corresponding to a reference sensed signal. In this example, such as... Figure 5 As shown, the actual sensed signal can be corrected in the following ways:

[0093] In step S510, the range of the reference signal corresponding to each measured data can be determined based on the reference sensing signal corresponding to each measured data.

[0094] As an example, the reference signal range can be determined by: determining the reference signal range corresponding to the reference sensing signal based on the reference sensing signal and a preset confidence level.

[0095] Here, the confidence level can characterize the error range of the virtual load sensor.

[0096] Specifically, considering that the output signal of the virtual load sensor may have errors due to algorithm or testing errors during the construction of the virtual load sensor, sample data such as historical data can be used to compare the output signal of the virtual load sensor with the corresponding true signal value to obtain the model error of the virtual load sensor, and the frequency of error occurrence can be counted.

[0097] As an example, the model error can be represented as shown in equation (5):

[0098] Error = Measure - Output (5)

[0099] Where Error represents the model error, Output represents the output signal of the virtual load sensor, and Measure represents the corresponding signal truth value.

[0100] The model error can be calculated at multiple points within a time period, resulting in a schematic diagram of the model error, such as... Figure 6AAs shown, by statistically analyzing the various values ​​of the model error within this time period and the frequency corresponding to each value, we can also obtain the following: Figure 6B The diagram shows the distribution of model errors.

[0101] Based on, for example Figure 6B The error distribution shown allows for the selection of the confidence level for the virtual load sensor, thus determining its error range. Different confidence levels can be chosen according to actual needs; for example, a confidence level of 90% can be used. Therefore, a 90% confidence interval for the output signal of the virtual load sensor can be determined, meaning the probability that the true signal value falls within this interval is considered to be 90%. For example, as... Figure 7 As shown, the upper and lower limits of the reference signal range corresponding to each reference sensing signal can be determined based on the confidence level, and the measured data can fall within the reference signal range at the corresponding time. Figure 7 And what will be described below Figure 9 and Figure 10 The term "in" is intended to reflect the relationship between the reference signal range and the actual sensed signal, or the overall waveform of the signal. Therefore, the specific value and unit of the sensed signal are not shown.

[0102] Here, as an example, the confidence level can be used as a configurable parameter of the virtual load sensor. After obtaining the reference sensing signal, the virtual load sensor can further determine the reference signal range based on the confidence level. In this example, the virtual load sensor can directly output the reference signal range. For example, the virtual load sensor can output the upper and lower limits of the reference signal range.

[0103] Using the above method, the range of reference signals into which the true value of the signal may fall can be determined based on the reference sensing signal. In subsequent steps, the actual sensing signal can be compared with this range to determine whether the actual sensing signal needs to be corrected. The actual sensing signal can be corrected based on this range so that the corrected sensing signal is more likely to be close to the true value.

[0104] Although the above description illustrates an example of determining the reference signal range using a confidence level, the embodiments of this disclosure are not limited thereto. Other methods can be used to determine the reference signal range, such as defining a range with a preset signal width centered on a reference sense signal. Furthermore, although the above description illustrates an example of determining the confidence level, the embodiments of this disclosure are not limited thereto. The error range of the virtual load sensor can also be determined using statistical methods to predetermine the confidence level.

[0105] In step S520, the corrected sensing signal can be obtained by correcting the measured data that exceeds the corresponding reference signal range from multiple measured data to the corresponding reference signal range.

[0106] By determining the range of the reference signal, the portion of the actual sensed signal that falls outside the range can be corrected back into the range, thereby making the portion of the signal that may deviate from the true value closer to the real situation.

[0107] In this step, data correction, or signal waveform correction, can be achieved using existing common methods, such as, but not limited to, translation and filtering.

[0108] As an example, the corrected sensing signal is obtained by: correcting the target measured data that exceeds the corresponding reference signal range in multiple measured data according to a window of preset duration, and obtaining the corrected sensing signal.

[0109] Here, in response to the fact that the proportion of the target measured data in the current window relative to all measured data in the current window is less than or equal to a preset proportion threshold, the target measured data in the current window is corrected to obtain the corrected sensing signal corresponding to the current window.

[0110] In response to a situation where the proportion of the target measured data in the current window relative to all measured data in the current window exceeds a preset proportion threshold, the current window and subsequent windows are merged based on the proportion of the target measured data in subsequent windows relative to all measured data in their respective windows, resulting in a merged window. While maintaining the signal waveform of the target measured data unchanged, the target measured data in the merged window that exceeds the corresponding reference signal range is corrected to obtain a corrected sensing signal corresponding to the merged window.

[0111] As an example, the preset duration and preset percentage threshold can be set according to actual needs. The preset duration can be, for example, but not limited to, 10 minutes, and the preset percentage threshold can be, for example, but not limited to, 50%.

[0112] Furthermore, the current window and subsequent windows can be merged based on the proportion of the target measured data in a preset number of subsequent windows (n) relative to all measured data in the corresponding windows. Here, the preset number n can be set according to actual needs, for example, it can be 2. That is, if the proportion of the target measured data in n+1 consecutive windows (including the current window) is greater than a preset proportion threshold, these windows can be merged, and the target measured data in the merged window can be uniformly corrected.

[0113] Taking translation as an example, any point in time can be used as the starting point of the window. Continuous windows are set according to a preset duration. It is determined whether there are waveforms or measured data exceeding the reference signal range in the current window. If so, the proportion of scattered points exceeding the corresponding reference signal range in the current window to the total data volume of the window is calculated to see if it exceeds a preset proportion threshold (e.g., 50%). If the proportion in the current window exceeds the preset proportion threshold, it is determined whether the proportion of the target measured data in a preset number of subsequent windows all exceeds this preset proportion threshold. If the proportion of the target measured data in multiple consecutive windows, including the current window, exceeds this preset proportion threshold, these windows can be merged into a single window, and the measured data in the merged window can be corrected using the same correction method. For example, in the correction process described below, corrections can be performed using the same step size. If the current window and subsequent windows do not continuously exceed the preset proportion threshold, the current window is corrected separately.

[0114] By using the above method, the densely occurring continuous time periods of measured data that exceed the range of the reference signal can be merged and uniformly corrected, thereby simplifying the correction process. Furthermore, the signal waveform can remain unchanged during the correction process to preserve the trend of the original signal, ensuring that the corrected signal still reflects the characteristics of the original signal.

[0115] Although the above describes the correction of the target measured data according to a preset window, the embodiments of this disclosure are not limited thereto, and any number of measured data can be corrected without dividing the window.

[0116] The following describes a specific example of correcting the measured data of a target that exceeds the range of the reference signal. The correction example described here can be used to correct the measured data of a target in a single window or a merged window, or it can be used to correct any number of measured data without dividing the window.

[0117] Specifically, in the embodiments of this disclosure, such as Figure 8 As shown, the corrected sensing signal can be obtained in the following way:

[0118] In step S810, the current target measured data can be corrected while keeping the signal waveform of the current target measured data unchanged, so as to obtain the corrected measured data.

[0119] For example, a uniform correction can be performed on the data in the current target measured data that exceeds the corresponding reference signal range, such as by shifting by the same amplitude, to obtain the corrected measured data.

[0120] Here, during the initial correction, the target measured data can be any measured data point that exceeds the corresponding reference range from among multiple measured data points. As an example, in an example where the target measured data is corrected according to a window of preset duration, during the initial correction, the target measured data can be any measured data point in the current window or a merged window that exceeds the corresponding reference range.

[0121] In step S820, the measured data within the corresponding reference range in the corrected measured data can be used as the corrected sensing signal.

[0122] In step S830, in response to the presence of measured data exceeding the corresponding reference range in the corrected measured data, the measured data exceeding the corresponding reference range in the corrected measured data can be used as the target measured data for the next correction, and the step of correcting the current target measured data can be returned.

[0123] For the corrected measured data, the measured data that are already within the corresponding reference signal range can be used as part of the corrected sensing signal. However, the measured data that are still outside the corresponding reference signal range after correction needs to be further corrected. For example, it can be used as target measured data and returned to step S810. The above steps can be executed cyclically until all measured data are corrected to the reference signal range, and the final corrected sensing signal can be obtained.

[0124] By using the above method, while keeping the signal waveform unchanged, the measured data of the actual sensed signal can be corrected to the corresponding reference range through multiple corrections. This way, the trend of the original signal can be preserved while correcting, and sudden changes in the signal waveform can be avoided.

[0125] As an example, in each of the above correction processes, in step S810, the corrected measured data may include first corrected measured data and / or second corrected measured data. The first corrected measured data may represent data obtained by correcting measured data that is greater than the upper limit of the corresponding reference signal range, and the second corrected measured data may represent data obtained by correcting measured data that is less than the lower limit of the corresponding reference signal range.

[0126] For example, the first corrected measured data can be obtained by: determining the first measured data in the current target measured data, wherein the first measured data is the target measured data that is greater than the upper limit of the corresponding reference signal range; determining the first difference between the maximum value in the first measured data and the upper limit of the corresponding reference signal range; and subtracting the first difference from each first measured data to obtain the first corrected measured data.

[0127] For example, the second corrected measured data can be obtained by: determining the second measured data in the current target measured data, wherein the second measured data is the target measured data that is less than the lower limit of the corresponding reference signal range; determining the second difference between the minimum value in the second measured data and the lower limit of the corresponding reference signal range; and adding the second difference to each second measured data to obtain the second corrected measured data.

[0128] Specifically, the first and second measured data in the target measured data to be corrected can be determined. Taking the correction of the target measured data in a single window or a merged window as an example, it can be determined whether the actual sensing signal in the window is out of range and is positive or negative.

[0129] This can include three scenarios: all target measured data are the first measured data, all target measured data are the second measured data, or the target measured data include both the first and second measured data. Correspondingly, the corrected measured data can all be the first corrected measured data, all be the second corrected measured data, or the corrected measured data include both the first and second corrected measured data.

[0130] Specifically, in response to the fact that all target measured data are the first measured data, that is, all target measured data are greater than the upper limit of the corresponding reference signal range, the maximum value of the target measured data that exceeds the range can be obtained, the difference between the maximum value and the upper limit of the corresponding reference signal range can be calculated, and the difference can be subtracted from all target measured data to obtain the corrected measured data.

[0131] Since all target measured data are second measured data, that is, all target measured data are less than the lower limit of the corresponding reference signal range, the minimum value of the target measured data that exceeds the range can be obtained, the difference between the minimum value and the lower limit of the corresponding reference signal range can be calculated, and the difference can be added to all target measured data to obtain the corrected measured data.

[0132] Since the target measured data includes both the first and second measured data (i.e., data exceeding the upper limit of the corresponding reference signal range and data falling below the lower limit of the corresponding reference signal range), the minimum and maximum values ​​of the out-of-range target measured data can be calculated separately. The difference between the maximum value and the upper limit of the corresponding reference signal range is calculated, and this difference is subtracted from all measured data exceeding the upper limit of the corresponding reference signal range to obtain the first part of the corrected measured data. Similarly, the difference between the minimum value and the lower limit of the corresponding reference signal range is calculated, and this difference is added to all measured data falling below the lower limit of the corresponding reference signal range to obtain the second part of the corrected measured data. Thus, both parts of data can be used together as the corrected measured data.

[0133] This method facilitates the rapid correction of signal portions that fall outside the reference signal range back to the corresponding reference signal range, minimizing the number of correction cycles.

[0134] In step S140, the load at the target location can be determined based on the corrected sensing signal.

[0135] In this step, the corrected sensing signal can be converted into the load at the target position based on the conversion relationship between the sensing signal of the target load sensor and the load (e.g., the first transfer function W mentioned above), so as to perform relevant evaluation or control based on the load.

[0136] Here, by correcting the actual sensing signals, the hardware sensors can be calibrated in real time without affecting the power generation of the wind turbine. This provides reliable load data for applications requiring high-precision loads, such as load-optimized control strategies, thereby improving the safety and control accuracy of the unit while ensuring power generation.

[0137] The process of correcting the actual sensed signal based on the reference signal range has been described above. In addition, in the embodiments of this disclosure, it can also be determined whether to correct the actual sensed signal based on the reference signal range.

[0138] Specifically, as described above, a reference signal range containing the reference sensing signal can be determined based on the reference sensing signal. Based on this reference signal range, the actual sensing signal can be compared with the reference signal range; in response to the comparison result indicating that there is data in the actual sensing signal that exceeds the reference signal range, a step of correcting the actual sensing signal based on the reference sensing signal can be performed; in response to the comparison result indicating that the actual sensing signal is within the reference signal range, the load at the target location can be determined based on the actual sensing signal.

[0139] Figure 9 and Figure 10 The waveform diagrams for the reference signal range and the actual load signal are shown respectively. Given the confidence level of the virtual load sensor, the upper and lower limits of the reference signal range can be output (e.g., ...). Figure 9 As shown), in this case, the actual sensing signal of the target load sensor (such as...) can be used. Figure 10 The system performs statistical analysis to determine whether the actual sensed signal is within the reference signal range. It then corrects the waveform of any time-series data points that exceed the range until each measured data point is within the corresponding reference signal range. Finally, it outputs the calibrated sensed signal or the corrected sensed signal.

[0140] By setting a reference signal range, a certain degree of fluctuation in the actual sensed signal can be allowed, avoiding unnecessary correction actions.

[0141] It should be noted that although the determination of whether the actual sensing signal needs to be corrected based on the above-mentioned reference signal range is described herein, the embodiments of this disclosure are not limited to this. For example, the deviation between the actual sensing signal and the virtual reference signal can also be calculated, and correction can be made when the deviation exceeds a preset value.

[0142] Furthermore, as an example, the load determination method according to an embodiment of the present disclosure may further include: in response to a comparison result indicating that the proportion of data in the actual sensed signal that exceeds the range of the reference signal exceeds a preset abnormal range, determining that the target load sensor is abnormal.

[0143] Specifically, the proportion (or statistical probability) of time series data points that exceed the corresponding reference signal range to the total number of collected time series (e.g., the total number of measured data) can be counted. If the proportion of measured data that exceeds the reference signal range exceeds the preset abnormal range, it can be considered that the hardware sensor has a measurement abnormality, and there may be problems such as detachment or damage.

[0144] Here, the preset range can be, for example, more than 60%, but it is not limited to this. The preset range can be set according to the actual application scenario.

[0145] In this way, during the operation of the load sensor, abnormalities such as sensor detachment or damage can be detected, thereby providing an alarm for abnormal operating status of the hardware sensor and enabling timely maintenance.

[0146] Furthermore, when the aforementioned sensor malfunction is detected, the step of correcting the actual sensing signal can continue to be performed in order to obtain a continuous sensing signal.

[0147] Furthermore, as an example, a correction range can be preset. In response to the comparison result indicating that the proportion of data in the actual sensing signal that exceeds the reference signal range exceeds the preset correction range but does not exceed the aforementioned preset abnormal range, a step of correcting the actual sensing signal can be performed. In response to the comparison result indicating that the proportion of data in the actual sensing signal that exceeds the reference signal range does not exceed the preset correction range, the load at the target location can be determined based on the actual sensing signal.

[0148] Figure 11 An application example of a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure is shown. For example... Figure 11 As shown, in step S1101, the current operating data of the wind turbine generator can be received. The virtual load sensor can obtain a reference load signal based on this operating data. In step S1102, the upper and lower limits of the reference signal range can be obtained based on the reference load signal and a preset confidence level. In step S1103, the actual sensing signal of the target load sensor can be received. In step S1104, it can be determined whether the actual sensing signal is within the reference signal range. In response to the fact that the actual sensing signals are all within the reference signal range, in step S1105, the actual sensing signals can be used as the final sensing signals.

[0149] In response to the presence of data outside the reference signal range in the actual sensed signal, in step S1106, the proportion of data outside the reference signal range can be counted, and in step S1107, it can be determined whether the proportion of data outside the reference signal range in the total data exceeds a preset range or a preset acceptance level (e.g., 60%). In response to not exceeding the preset range, step S1108 can be executed to correct the actual sensed signal; in response to exceeding the preset range, in step S1109, it can be determined that the target load sensor is faulty, and step S1108 can be executed.

[0150] After correcting the actual sensed signal, in step S1110, it can be determined whether the corrected signal is within the reference signal range. If the corrected signal is within the reference signal range, the final sensed signal can be obtained in step S1105. If there is still data outside the reference signal range in the corrected signal, the process can return to step S1108 to correct the data currently outside the reference signal range again. Steps S1110 and S1108 can be executed repeatedly once or multiple times until the corrected signal is within the reference signal range.

[0151] In step S1111, the load at the target position can be obtained based on the final sensing signal (e.g., the corrected sensing signal) and the conversion relationship between the load of the target load sensor and the sensing signal (e.g., the first transfer function mentioned above).

[0152] Figure 12 An example implementation of a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure is shown. Figure 12 As shown, the configuration parameters of the wind turbine generator set and real-time operating data such as SCADA signals can be input to the virtual load sensor 1210. The virtual load sensor 1210 can output a reference signal range, such as the upper and lower limits of the reference signal range. The actual sensing signal of the target load sensor can be input to the signal acquisition unit 1220. The signal acquisition unit 1220 can directly output the actual sensing signal, or it can perform signal processing on the actual sensing signal (e.g., convert the optical signal into an electrical signal) and output the processed actual sensing signal.

[0153] The sensor fusion unit 1230 can receive the reference signal range from the virtual load sensor 1210 and the actual sensed signal from the signal acquisition unit 1220, and determine the calibrated transfer function. It then uses the calibrated transfer function to obtain the calibrated load from the actual sensed signal. Furthermore, when the proportion of the actual sensed signal outside the reference signal range exceeds a preset abnormal range, the sensor fusion unit 1230 can also issue an alarm for any abnormality in the target load sensor.

[0154] According to embodiments of this disclosure, a load measurement scheme for wind turbine generators based on multi-mode sensor fusion technology can be provided for the wind power field. This scheme can perform load calibration of hardware sensors without adding additional testing equipment or causing loss of generator power generation. It combines real-time operating data of the generator, generator configuration parameters and actual sensing signals, and combines virtual load sensors with hardware load sensors to improve the accuracy of load measurement.

[0155] Furthermore, in the embodiments of this disclosure, a virtual load sensor, which serves as a load estimator, can be obtained through a data-driven approach based on the simulated load, thereby eliminating or reducing the deviation between the measured load and the load estimator, and calibrating the hardware sensors of the wind turbine generator set in real time.

[0156] According to a second aspect of an exemplary embodiment of the present disclosure, a load determination device for a wind turbine generator set is provided, such as... Figure 13 As shown, the load determination device may include an acquisition unit 1310, a signal determination unit 1320, a correction unit 1330, and a load determination unit 1340.

[0157] The acquisition unit 1310 is configured to acquire the actual sensing signal actually sensed by the target load sensor, wherein the target load sensor is used to sense the load at the target location of the wind turbine generator set.

[0158] The signal determination unit 1320 is configured to obtain a reference sensing signal based on the current operating data of the wind turbine generator set, using a preset virtual load sensor corresponding to the target load sensor.

[0159] The correction unit 1330 is configured to correct the actual sensing signal based on the reference sensing signal to obtain the corrected sensing signal.

[0160] The load determination unit 1340 is configured to determine the load at the target location based on the corrected sensing signal.

[0161] As an example, the load determination device may further include a comparison unit, which may be configured to: determine a reference signal range containing the reference sensing signal based on the reference sensing signal; compare the actual sensing signal with the reference signal range; in response to the comparison result indicating that there is data in the actual sensing signal that exceeds the reference signal range, cause the correction unit 1330 to perform a step of correcting the actual sensing signal based on the reference sensing signal; and in response to the comparison result indicating that the actual sensing signal is within the reference signal range, cause the load determination unit 1340 to determine the load at the target location based on the actual sensing signal.

[0162] As an example, the comparison unit can also be configured to: determine that the target load sensor is abnormal in response to the comparison result indicating that the proportion of data in the actual sensed signal that exceeds the range of the reference signal exceeds a preset abnormal range.

[0163] As an example, the actual sensing signal includes multiple measured data arranged in time sequence, each measured data corresponding to a reference sensing signal. The correction unit 1330 is further configured to: determine a reference signal range corresponding to each measured data based on the reference sensing signal corresponding to each measured data; and obtain a corrected sensing signal by correcting the measured data that exceeds the corresponding reference signal range to the corresponding reference signal range.

[0164] As an example, the correction unit 1330 is further configured to obtain the corrected sensing signal by: correcting the current target measured data while keeping the signal waveform of the current target measured data unchanged, to obtain the corrected measured data, wherein, in the initial correction, the target measured data is the measured data that exceeds the corresponding reference range among multiple measured data; the measured data within the corresponding reference range in the corrected measured data is used as the corrected sensing signal; in response to the presence of measured data exceeding the corresponding reference range in the corrected measured data, the measured data exceeding the corresponding reference range in the corrected measured data is used as the target measured data for the next correction, and the step of correcting the current target measured data is returned.

[0165] As an example, the comparison unit can also be configured to determine the reference signal range by determining the reference signal range corresponding to the reference sensing signal based on the reference sensing signal and a preset confidence level, wherein the confidence level characterizes the error range of the virtual load sensor.

[0166] As an example, a virtual load sensor can be obtained by: determining a first transfer function of the target load sensor based on the sensor type and sensing material of the target load sensor, wherein the first transfer function characterizes the conversion relationship between the sensing signal of the target load sensor and the load at the target location; and obtaining a virtual load sensor based on the first transfer function and a second transfer function, wherein the second transfer function characterizes the conversion relationship between the operating data of the wind turbine generator and the load at the target location.

[0167] As an example, the input data of the virtual load sensor includes the configuration parameters and current operating data of the wind turbine generator set, and the output signal of the virtual load sensor has the same signal form as the sensing signal of the target load sensor.

[0168] Regarding the apparatus in the above embodiments, the specific manner in which each unit performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0169] According to a third aspect of an exemplary embodiment of the present disclosure, an electronic device is provided, the electronic device comprising: a processor; and a memory for storing processor-executable instructions, wherein the processor-executable instructions, when executed by the processor, cause the processor to perform a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0170] As an example, an electronic device can be a PC, tablet, personal digital assistant, smartphone, or other device capable of executing the aforementioned set of instructions. Here, the electronic device is not necessarily a single device; it can be any collection of devices or circuits capable of executing the aforementioned instructions (or instruction sets) individually or in combination. The electronic device can also be part of an integrated control system or system manager, or can be configured to interconnect locally or remotely (e.g., via wireless transmission) through an interface.

[0171] In electronic devices, a processor can include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller, or a microprocessor. By way of example and not limitation, a processor can also include analog processors, digital processors, microprocessors, multi-core processors, processor arrays, network processors, etc.

[0172] The processor can execute instructions or code stored in memory, which can also store data. Instructions and data can also be sent and received over a network via a network interface device, which can employ any known transport protocol.

[0173] Memory can be integrated with the processor; for example, RAM or flash memory can be housed within an integrated circuit microprocessor. Alternatively, memory can comprise a separate device, such as an external disk drive, storage array, or other storage device that can be used by any database system. Memory and processor can be operatively coupled, or can communicate with each other, for example, via I / O ports, network connections, etc., enabling the processor to read files stored in the memory.

[0174] In addition, electronic devices may include video displays (such as liquid crystal displays) and user interaction interfaces (such as keyboards, mice, touch input devices, etc.). All components of the electronic device may be interconnected via buses and / or networks.

[0175] According to a fourth aspect of exemplary embodiments of the present disclosure, a wind turbine generator set is provided, which may include electronic equipment described in exemplary embodiments of the present disclosure, or the control system of the wind turbine generator set may be connected to electronic equipment described in exemplary embodiments of the present disclosure.

[0176] According to a fifth aspect of an exemplary embodiment of the present disclosure, a computer-readable storage medium is provided such that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform a load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0177] Specifically, the load determination method for a wind turbine generator set according to embodiments of this disclosure can be programmed into a computer program and stored on a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by at least one processor, the at least one processor causes the processor to perform the pitch control method for the wind turbine generator set according to exemplary embodiments of this disclosure. Examples of computer-readable storage media include: read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disc storage, hard disk drive (HDD), solid-state drive (SSD), card storage (such as multimedia cards, secure digital (SD) cards, or ultra-fast digital (XD) cards), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid-state drive, and any other device configured to store a computer program and any associated data, data files, and data structures in a non-transitory manner and to provide the computer program and any associated data, data files, and data structures to a processor or computer so that the processor or computer can execute the computer program. In one example, the computer program and any associated data, data files, and data structures are distributed across a networked computer system, such that the computer program and any associated data, data files, and data structures are stored, accessed, and executed in a distributed manner through one or more processors or computers.

[0178] According to a sixth aspect of an exemplary embodiment of the present disclosure, a computer program product is provided, including computer-executable instructions that, when executed by at least one processor, implement the load determination method for a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0179] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0180] Furthermore, it should be noted that although several examples of each step have been described above with reference to the specific accompanying drawings, it should be understood that the embodiments of this disclosure are not limited to the combinations given in the examples. The steps appearing in different drawings can be combined, and the execution order of each step can be changed, which will not be exhaustive here.

[0181] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

[0182] The specific embodiments of this disclosure have been described in detail above. Although some embodiments have been shown and described, those skilled in the art should understand that modifications and variations can be made to these embodiments without departing from the principles and spirit of this disclosure, which are defined by the claims and their equivalents. Such modifications and variations should also be within the protection scope of the claims of this disclosure.

Claims

1. A method of load determination for a wind turbine generator system, characterized by, The load determination method includes: Acquire the actual sensing signal sensed by the target load sensor, wherein the target load sensor is used to sense the load at the target location of the wind turbine generator set; Based on the current operating data of the wind turbine generator set, a reference sensing signal is obtained using a preset virtual load sensor corresponding to the target load sensor. Based on the reference sensing signal, the actual sensing signal is corrected to obtain the corrected sensing signal; Based on the corrected sensing signal, the load at the target location is determined.

2. The load determination method according to claim 1, characterized by, The load determination method further includes: Based on the reference sensing signal, a reference signal range including the reference sensing signal is determined; The actual sensed signal is compared with the range of the reference signal; In response to the comparison result indicating that there is data in the actual sensed signal that exceeds the range of the reference signal, the step of correcting the actual sensed signal based on the reference sensed signal is performed; In response to the result of the comparison indicating that the actual sensed signal is within the range of the reference signal, the load at the target location is determined based on the actual sensed signal.

3. The load determination method according to claim 2, characterized by, The load determination method further includes: If the comparison result indicates that the proportion of data in the actual sensed signal that exceeds the range of the reference signal exceeds a preset abnormal range, it is determined that the target load sensor is abnormal.

4. The load determination method according to claim 1, characterized by, The actual sensed signal includes multiple measured data arranged in time sequence, each measured data corresponding to a reference sensed signal, wherein the step of correcting the actual sensed signal based on the reference sensed signal includes: Based on the reference sensing signal corresponding to each measured data point, the range of the reference signal corresponding to each measured data point is determined; The corrected sensing signal is obtained by correcting the measured data that are outside the corresponding reference signal range to the corresponding reference signal range.

5. The load determination method according to claim 4, characterized by, The corrected sensing signal is also obtained in the following way: While keeping the signal waveform of the current target measured data unchanged, the current target measured data is corrected to obtain corrected measured data. In the initial correction, the target measured data is the measured data that exceeds the corresponding reference range among the plurality of measured data. The measured data within the corresponding reference range in the corrected measured data are used as the corrected sensing signal; In response to the presence of measured data exceeding the corresponding reference range in the corrected measured data, the measured data exceeding the corresponding reference range in the corrected measured data is taken as the target measured data for the next correction, and the process returns to the step of correcting the current target measured data.

6. The load determination method according to claim 5, characterized by, The corrected measured data includes first corrected measured data and / or second corrected measured data, wherein the first corrected measured data is obtained in the following manner: Determine the first measured data in the current target measured data, wherein the first measured data is the target measured data that is greater than the upper limit of the corresponding reference signal range; Determine the first difference between the maximum value in the first measured data and the upper limit of the corresponding reference signal range; Subtract the first difference from each of the first measured data to obtain the first corrected measured data; The second corrected measured data was obtained in the following manner: Determine the second measured data in the current target measured data, wherein the second measured data is the target measured data that is less than the lower limit of the corresponding reference signal range; Determine the second difference between the minimum value in the second measured data and the lower limit of the corresponding reference signal range; The second difference is added to each of the second measured data to obtain the second corrected measured data.

7. The load determination method according to claim 4, characterized in that, The corrected sensing signal is obtained in the following manner: According to a preset time window, the target measured data that exceeds the corresponding reference signal range among the multiple measured data are corrected to obtain the corrected sensing signal. Wherein, in response to the fact that the proportion of the target measured data in the current window relative to all measured data in the current window is less than or equal to a preset proportion threshold, the target measured data in the current window is corrected to obtain a corrected sensing signal corresponding to the current window; Specifically, if the proportion of the target measured data in the current window relative to all measured data in the current window is greater than a preset proportion threshold, the current window and the subsequent windows are merged based on the proportion of the target measured data in the subsequent windows relative to all measured data in the corresponding windows, resulting in a merged window. While keeping the signal waveform of the target measured data unchanged, the target measured data in the merged window that exceeds the corresponding reference signal range is corrected to obtain the corrected sensing signal corresponding to the merged window.

8. The load determination method according to any one of claims 2 to 7, characterized in that, The range of the reference signal is determined in the following manner: Based on the reference sensing signal and a preset confidence level, a reference signal range corresponding to the reference sensing signal is determined. The confidence level characterizes the error range of the virtual load sensor.

9. The load determination method according to claim 1, characterized in that, The virtual load sensor is obtained in the following manner: Based on the sensor type and sensing material of the target load sensor, a first transfer function of the target load sensor is determined, wherein the first transfer function characterizes the conversion relationship between the sensing signal of the target load sensor and the load at the target location; Based on the first transfer function and the second transfer function, the virtual load sensor is obtained. The second transfer function characterizes the conversion relationship between the operating data of the wind turbine generator set and the load at the target location.

10. The load determination method according to claim 1, characterized in that, The input data of the virtual load sensor includes the configuration parameters and current operating data of the wind turbine generator set, and the output signal of the virtual load sensor has the same signal form as the sensing signal of the target load sensor.

11. A load determination device for a wind turbine generator set, characterized in that, The load determination device includes: The acquisition unit is configured to acquire the actual sensing signal actually sensed by the target load sensor, wherein the target load sensor is used to sense the load at the target location of the wind turbine generator set; The signal determination unit is configured to obtain a reference sensing signal based on the current operating data of the wind turbine generator set using a preset virtual load sensor corresponding to the target load sensor. The correction unit is configured to correct the actual sensing signal based on the reference sensing signal to obtain the corrected sensing signal; The load determination unit is configured to determine the load at the target location based on the corrected sensing signal.

12. An electronic device, characterized in that, The electronic device includes: processor; Memory used to store processor-executable instructions. Wherein, when the processor executes the executable instructions, it causes the processor to perform the load determination method for a wind turbine generator set according to any one of claims 1 to 10.

13. A wind turbine generator set, characterized in that, The wind turbine generator set includes the electronic equipment according to claim 12, or the control system of the wind turbine generator set is connected to the electronic equipment according to claim 12.

14. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the load determination method for a wind turbine generator set according to any one of claims 1 to 10.

15. A computer program product comprising computer-executable instructions, characterized in that, When the computer-executable instructions are executed by at least one processor, they implement the load determination method for a wind turbine generator set according to any one of claims 1 to 10.

Images