Wind turbine yaw control method and system

Abstract

Provided are a wind turbine group clearance control method and system, the method comprising: in response to a clearance device of a target wind turbine group failing, obtaining predicted monitoring data of the clearance device through a clearance device model of the target wind turbine group; determining a predicted clearance value of the target wind turbine group based on the predicted monitoring data of the clearance device; and determining a clearance control strategy for the target wind turbine group based on the predicted clearance value, wherein the clearance device model is used to predict monitoring data of the clearance device of the target wind turbine group based on group-related data of the target wind turbine group and monitoring data of the clearance device of an associated wind turbine group associated with the target wind turbine group.

Description

Technical Field

[0001] This application relates to the field of wind power, and more specifically, to a method and system for controlling the airspace clearance of wind turbine generators. Background Technology

[0002] In the wind power sector, complex terrain at wind farm sites or extreme weather conditions often lead to extreme operating conditions, causing significant deformation of wind turbine blades, insufficient clearance, and even tower sweep hazards. To more accurately monitor the clearance status of wind turbines and implement protective controls, clearance devices are typically installed on the turbines. Although the types of clearance devices and testing methods vary, most are hardware-based and therefore subject to significant failure risks. When clearance devices fail, the wind turbines face substantial risks.

[0003] When air clearance equipment fails, existing solutions typically revert to a primitive, coarser software control mode. However, this control method suffers from poor performance and cannot quickly and accurately identify tower sweep risks due to the lack of air clearance detection information input. Alternatively, existing solutions may employ a conservative control strategy to ensure unit safety as much as possible. Such a strategy cannot perform refined control based on air clearance monitoring results, which is detrimental to the power generation performance of the wind turbine.

[0004] Therefore, there is a need for a solution that can quickly and accurately obtain the airspace clearance results of wind turbines after the airspace clearance equipment fails, and to implement appropriate airspace control strategies for wind turbines. Summary of the Invention

[0005] In order to at least solve the above-mentioned problems in the prior art, this application provides a wind turbine headroom control method and system.

[0006] According to one aspect of this disclosure, a wind turbine airspace control method is provided, the method comprising: in response to the failure of airspace equipment of a target wind turbine, acquiring predictive monitoring data of airspace equipment through an airspace equipment model of the target wind turbine; determining a predicted airspace value of the target wind turbine based on the predicted monitoring data of the airspace equipment; and determining an airspace control strategy for the target wind turbine based on the predicted airspace value; wherein the airspace equipment model is used to predict the monitoring data of the airspace equipment of the target wind turbine based on turbine-related data of the target wind turbine and monitoring data of the airspace equipment of associated wind turbines related to the target wind turbine.

[0007] Optionally, the method further includes generating the air clearance equipment model, wherein the step of generating the air clearance equipment model includes: acquiring training samples, wherein each training sample includes sample historical feature data and sample historical labels, the sample historical feature data includes the unit-related data of the target wind turbine at the corresponding historical sampling time and the monitoring data of the air clearance equipment of the associated wind turbine at the corresponding historical sampling time, and the sample labels include the actual monitoring data of the air clearance equipment of the target wind turbine at the corresponding historical sampling time; training a benchmark model based on the acquired training samples, and using the trained benchmark model as the air clearance equipment model.

[0008] Optionally, the training samples may include a first training sample and a second training sample, wherein the first training sample includes historical feature data and historical labels of samples collected when the target wind turbine does not trigger the airspace control strategy, and the second training sample includes historical feature data and historical labels of samples collected when the target wind turbine triggers the airspace control strategy.

[0009] Optionally, the method may further include: periodically updating the air clearance equipment model.

[0010] Optionally, in response to the failure of the air clearance equipment of the target wind turbine, the step of obtaining predictive monitoring data of the air clearance equipment through the air clearance equipment model of the target wind turbine may include: in response to the failure of the air clearance equipment of the target wind turbine, obtaining the turbine-related data of the target wind turbine at the current time and the monitoring data of the air clearance equipment of the associated wind turbine at the current time, and inputting the obtained turbine-related data of the target wind turbine at the current time and the monitoring data of the air clearance equipment of the associated wind turbine at the current time into the air clearance equipment model to obtain the predictive monitoring data of the target wind turbine.

[0011] Optionally, determining the airspace control strategy for the target wind turbine based on the predicted airspace value may include: determining the lower boundary value of the airspace based on the predicted airspace value and the airspace value deviation confidence interval, wherein the airspace value deviation confidence interval indicates the fluctuation range of the predicted airspace value under a preset confidence level; and determining the airspace control strategy for the target wind turbine based on the predicted airspace value, the lower boundary value of the airspace, and a preset airspace level.

[0012] Optionally, the preset airspace levels may include multiple levels, and different airspace levels may correspond to different airspace control strategies. Determining the airspace control strategy for the target wind turbine based on the predicted airspace value, the lower airspace boundary value, and the preset airspace levels may include: determining whether the lower airspace boundary value is at a second airspace level based on the predicted airspace value being at a first airspace level, wherein the second airspace level indicates that the probability of a tower sweeping accident occurring at the airspace value corresponding to the second airspace level is equal to or higher than a preset threshold; in response to the lower airspace boundary value being at the second airspace level, determining the airspace control strategy corresponding to the second airspace level as the airspace control strategy for the target wind turbine; and in response to the lower airspace boundary value not being at the second airspace level, determining the airspace control strategy corresponding to the first airspace level as the airspace control strategy for the target wind turbine.

[0013] Optionally, the method may further include: testing the generated air clearance equipment model and determining the air clearance value deviation distribution based on the test results; determining the air clearance value deviation confidence interval based on the air clearance value deviation distribution; wherein, testing the generated air clearance equipment model and determining the air clearance value deviation distribution based on the test results includes: acquiring multiple test samples, wherein each test sample includes historical feature data of the test sample and historical labels of the test sample, wherein the historical feature data of the test sample includes the unit operation data of the target wind turbine at the corresponding historical sampling time and the air clearance equipment of the associated wind turbine at the corresponding historical sampling time. The monitoring data includes the actual monitoring data of the air clearance equipment of the target wind turbine at the corresponding historical sampling time. The test samples are input into the air clearance equipment model to obtain the predicted monitoring data corresponding to each test sample. Based on the actual monitoring data and predicted monitoring data corresponding to each test sample, the actual air clearance value and predicted air clearance value corresponding to each test sample are calculated respectively. The air clearance value deviation of each test sample is determined based on the ratio of the actual air clearance value to the predicted air clearance value. Statistical analysis is performed on the air clearance value deviations of the multiple test samples to determine the air clearance value deviation distribution.

[0014] Optionally, the method may further include: dividing the target wind turbine into multiple sectors; and identifying wind turbines within a preset distance from the target wind turbine in each sector as associated wind turbines of the target wind turbine.

[0015] Optionally, the relevant data of the turbine unit may include at least one of the turbine unit operation data and wind parameter data of the target wind turbine unit, wherein the wind parameter data may include at least one of wind speed, wind direction, wind shear, and inflow angle; the turbine unit operation data may include at least one of yaw angle, pitch angle, speed, and torque; and the monitoring data of the air clearance equipment may include monitoring values ​​obtained by measuring using the air clearance equipment.

[0016] According to another aspect of this disclosure, a wind turbine air clearance control system is provided, the system comprising: a monitoring data acquisition unit configured to acquire predictive monitoring data of the air clearance equipment of the target wind turbine through an air clearance equipment model of the target wind turbine in response to the air clearance equipment failure of the target wind turbine; and a control strategy determination unit configured to determine a predicted air clearance value of the target wind turbine based on the predicted monitoring data of the air clearance equipment, and to determine an air clearance control strategy for the target wind turbine based on the predicted air clearance value, wherein the air clearance equipment model is used to predict the monitoring data of the air clearance equipment of the target wind turbine based on the turbine-related data of the target wind turbine and the monitoring data of the air clearance equipment of associated wind turbines associated with the target wind turbine.

[0017] Optionally, the monitoring data acquisition unit may include: an airspace equipment model generation unit, configured to generate the airspace equipment model, wherein the airspace equipment model generation unit may be configured to: acquire training samples, wherein each training sample includes sample historical feature data and sample historical labels, the sample historical feature data includes the unit-related data of the target wind turbine at the corresponding historical sampling time and the monitoring data of the airspace equipment of the associated wind turbine at the corresponding historical sampling time, and the sample labels include the actual monitoring data of the airspace equipment of the target wind turbine at the corresponding historical sampling time; train a benchmark model based on the acquired training samples, and use the trained benchmark model as the airspace equipment model.

[0018] Optionally, the training samples may include a first training sample and a second training sample, wherein the first training sample may include historical feature data and historical tags of samples collected when the target wind turbine does not trigger the airspace control strategy, and the second training sample may include historical feature data and historical tags of samples collected when the target wind turbine triggers the airspace control strategy.

[0019] Optionally, the air clearance equipment model generation unit can also be configured to periodically update the air clearance equipment model.

[0020] Optionally, the monitoring data acquisition unit can be configured to: in response to the failure of the air clearance equipment of the target wind turbine, acquire the turbine-related data of the target wind turbine at the current time and the monitoring data of the air clearance equipment of the associated wind turbine at the current time, input the acquired turbine-related data of the target wind turbine at the current time and the monitoring data of the air clearance equipment of the associated wind turbine at the current time into the air clearance equipment model to obtain the predicted monitoring data of the target wind turbine.

[0021] Optionally, the control strategy determination unit can be configured to: determine the lower boundary value of the airspace based on the predicted airspace value and the airspace value deviation confidence interval, wherein the airspace value deviation confidence interval indicates the fluctuation range of the predicted airspace value under a preset confidence level; and determine the airspace control strategy for the target wind turbine based on the predicted airspace value, the lower boundary value of the airspace, and the preset airspace level.

[0022] Optionally, the preset airspace levels may include multiple levels, and different airspace levels may correspond to different airspace control strategies. The control strategy determination unit may be configured to determine the airspace control strategy for the target wind turbine based on the following operations: based on the predicted airspace value being at the first airspace level, determine whether the lower boundary value of the airspace is at the second airspace level, wherein the second airspace level indicates that the probability of a tower sweeping accident occurring at the airspace value corresponding to the second airspace level is equal to or higher than a preset threshold; in response to the lower boundary value of the airspace being at the second airspace level, determine the airspace control strategy corresponding to the second airspace level as the airspace control strategy for the target wind turbine; in response to the lower boundary value of the airspace not being at the second airspace level, determine the airspace control strategy corresponding to the first airspace level as the airspace control strategy for the target wind turbine.

[0023] Optionally, the monitoring data acquisition unit may further include: a testing unit configured to test the generated airspace equipment model and determine the airspace value deviation distribution based on the test results; and to determine the airspace value deviation confidence interval based on the airspace value deviation distribution; wherein the testing unit is configured to determine the airspace value deviation distribution by: acquiring multiple test samples, wherein each test sample includes historical feature data of the test sample and historical labels of the test sample, the historical feature data of the test sample including the unit operation data of the target wind turbine at the corresponding historical sampling time and the airspace equipment of the associated wind turbine at the corresponding historical sampling time. The monitoring data at the sampling time includes the actual monitoring data of the air clearance equipment of the target wind turbine at the corresponding historical sampling time. The test samples are input into the air clearance equipment model to obtain the predicted monitoring data corresponding to each test sample. Based on the actual monitoring data and predicted monitoring data corresponding to each test sample, the actual air clearance value and predicted air clearance value corresponding to each test sample are calculated respectively. The air clearance value deviation of each test sample is determined based on the ratio of the actual air clearance value to the predicted air clearance value. Statistical analysis is performed on the air clearance value deviations of the multiple test samples to determine the air clearance value deviation distribution.

[0024] Optionally, the monitoring data acquisition unit can also be configured to: divide multiple sectors with the target wind turbine as the center; and determine the wind turbines within a preset distance from the target wind turbine in each sector as the associated wind turbines of the target wind turbine.

[0025] Optionally, the relevant data of the turbine unit may include at least one of the turbine unit operation data and wind parameter data of the target wind turbine unit, wherein the wind parameter data may include at least one of wind speed, wind direction, wind shear, and inflow angle; the turbine unit operation data may include at least one of yaw angle, pitch angle, speed, and torque; and the monitoring data of the air clearance equipment may include monitoring values ​​obtained by measuring using the air clearance equipment.

[0026] According to another aspect of this disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored, which, when executed by a processor, implement the aforementioned wind turbine headroom control method.

[0027] According to another aspect of this disclosure, a computer device is provided, including a readable medium storing computer program instructions, characterized in that the computer program instructions include instructions for performing the aforementioned wind turbine headroom control method.

[0028] Beneficial effects

[0029] By applying the wind turbine clearance control method and system according to exemplary embodiments of this disclosure, when the clearance equipment of a wind turbine in a wind farm fails, the wind turbine can quickly obtain the wind turbine's clearance equipment prediction monitoring data by establishing a connection between the wind turbine and surrounding wind turbines. The clearance value can be calculated according to the algorithm used when the clearance equipment is not failed, and the wind turbine can be controlled using the control strategy used when the clearance equipment is not failed. This maximizes the preservation of the advantages of the clearance control scheme based on the current clearance equipment monitoring results, and allows for continued clearance monitoring and control after the clearance equipment fails. It quickly and accurately obtains the clearance results of the wind turbine and adopts appropriate clearance control strategies for the wind turbine. This not only avoids the risk of tower sweeping accidents caused by clearance equipment failure, but also ensures the safety of the wind turbine while taking into account its economic efficiency with a more accurate control strategy. Attached Figure Description

[0030] These and / or other aspects and advantages of this application will become clearer and more readily understood from the following detailed description of embodiments of this application taken in conjunction with the accompanying drawings, wherein:

[0031] Figure 1 A flowchart illustrating a wind turbine headroom control method according to an exemplary embodiment of the present disclosure is shown.

[0032] Figure 2 A schematic diagram illustrating the selection of associated wind turbine units according to an exemplary embodiment of the present disclosure is shown.

[0033] Figure 3 This is a schematic flowchart illustrating a method for generating a headroom device model according to an exemplary embodiment of the invention.

[0034] Figure 4 This is a schematic flowchart illustrating a method for determining the net clearance deviation distribution according to an exemplary embodiment of the invention.

[0035] Figure 5 This is a flowchart illustrating a method for determining a wind turbine's airspace control strategy according to an exemplary embodiment of the invention.

[0036] Figure 6 This is a block diagram illustrating a wind turbine headroom control system according to an exemplary embodiment of the present disclosure.

[0037] Figure 7 This is a block diagram illustrating an electronic device according to an exemplary embodiment of the present disclosure.

[0038] The present application will be described in detail below with reference to the accompanying drawings, throughout which the same or similar elements will be indicated by the same or similar reference numerals. Detailed Implementation

[0039] The following description, taken with reference to the accompanying drawings, is provided to aid in a full understanding of exemplary embodiments of the present disclosure as defined by the claims and their equivalents. The description includes various specific details to aid understanding, but these details are considered exemplary only. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Furthermore, descriptions of known functions and constructions may be omitted for clarity and brevity.

[0040] Figure 1 A flowchart illustrating a wind turbine clearance control method 100 according to an exemplary embodiment of the present disclosure is shown. In an exemplary embodiment of the present invention, the wind turbine clearance control method 100 may be executed by the turbine master controller of the wind turbine. In some other exemplary embodiments, the wind turbine clearance control method 100 may also be executed by a field-level controller.

[0041] Reference Figure 1 In step S101, in response to the failure of the air defense equipment of the target wind turbine, predictive monitoring data of the air defense equipment can be obtained through the air defense equipment model of the target wind turbine.

[0042] Specifically, when the air clearance equipment of the target wind turbine is not faulty, various monitoring data used to formulate or select the air clearance control strategy for the target wind turbine can be directly measured through the air clearance equipment. However, when the air clearance equipment of the target wind turbine fails, since the air clearance equipment cannot provide accurate measured monitoring data, an air clearance equipment model can be used to provide (e.g., predict) monitoring data for the air clearance control strategy of the target wind turbine. For example, when the air clearance equipment no longer returns data or the returned data is abnormal, the main controller of the wind turbine can determine that the air clearance equipment has failed. In an exemplary embodiment of this disclosure, the target wind turbine can be any wind turbine in the wind farm. A separate air clearance equipment model can be set for each wind turbine in the wind farm for use after air clearance equipment failure.

[0043] In an exemplary embodiment of this disclosure, the air clearance equipment model used may be monitoring data of the air clearance equipment of the target wind turbine obtained through training. For example, the air clearance equipment model may predict the monitoring data of the air clearance equipment of the target wind turbine based on the turbine-related data of the target wind turbine and the monitoring data of the air clearance equipment of associated wind turbines related to the target wind turbine.

[0044] In an exemplary embodiment of this disclosure, the associated wind turbine unit may be another turbine unit located in the same wind farm as the target wind turbine unit. After the locations of all turbine sites in the wind farm are determined, corresponding associated wind turbine units can be set for each turbine unit in the wind farm. That is, in an exemplary embodiment of this disclosure, before obtaining the monitoring data of the airspace clearance equipment of the target wind turbine unit in step S101, the wind turbine airspace clearance control method according to an exemplary embodiment of this disclosure first needs to set the associated wind turbine units of the target wind turbine unit.

[0045] Figure 2 A schematic diagram illustrating the selection of associated wind turbine units according to an exemplary embodiment of the present disclosure is shown.

[0046] Reference Figure 2 As an example only, multiple sectors can be divided around the target wind turbine. Each sector corresponds to a different direction on a circular area centered on the target wind turbine. Wind turbines within a preset distance from the target wind turbine in each sector can be selected as associated wind turbines of the target wind turbine. There can be one or more associated wind turbines.

[0047] like Figure 2 As shown, multiple associated wind turbines can be selected in each sector according to their distance from the target wind turbine, from closest to furthest, but the furthest distance cannot exceed a preset distance X. lim If the distance exceeds a certain threshold, the state of the wind turbine can be considered irrelevant to the target wind turbine. In an exemplary embodiment of this disclosure, the preset distance X...lim These can be empirical values ​​or experimental values ​​obtained through multiple experiments. Furthermore, in exemplary embodiments of this disclosure, the number of sectors can be as large as possible (e.g., more than 8) to ensure that there are associated wind turbines providing relevant data in each incoming flow direction of the target wind turbine, thereby improving the accuracy of the airspace equipment model.

[0048] return Figure 1 In exemplary embodiments of this disclosure, the turbine-related data may include at least one of turbine operating data and wind parameter data detected by sensors of the target wind turbine. Here, wind parameter data may include, but is not limited to, at least one of wind speed, wind direction, wind shear, and inflow angle; turbine operating data may include, but is not limited to, at least one of yaw angle, pitch angle, speed, and torque. Furthermore, the monitoring data of the air clearance device may include monitoring values ​​obtained by measuring using the air clearance device, for example, directly detected values. As an example only, in the case of using a multi-line laser air clearance device, its monitoring value may be the distance [d1,d2,d3,d4,…] from where the laser is emitted to where it is reflected. In this case, the relationship between the air clearance value D and the monitoring value may be D = func([d1,d2,d3,d4,…]), where func may represent a preset calculation formula. However, it should be understood that this application is not limited to this; if other types of air clearance devices are used, the monitoring value may be other data, and the method of calculating the air clearance value using the monitoring value may also be other. Furthermore, it should be understood that the input data for the air clearance equipment model is not limited to the various data listed above, but can be set to other data as needed, which will not be detailed here.

[0049] The following will combine Figure 3 A detailed description is provided of the headroom equipment model used in the exemplary embodiments of this disclosure. Figure 3 This is a schematic flowchart illustrating a method for generating a clearance device model according to an exemplary embodiment of the invention. In an exemplary embodiment of this disclosure, the clearance device model is generated by training a baseline model using training samples, which will be described in detail below.

[0050] Reference Figure 3 In step S301, training samples can be obtained first.

[0051] Each training sample may include historical feature data and historical labels. The historical feature data includes the relevant data of the target wind turbine at the corresponding historical sampling time and the monitoring data of the air clearance equipment of the associated wind turbine at the corresponding historical sampling time. The sample labels include the actual monitoring data of the air clearance equipment of the target wind turbine at the corresponding historical sampling time.

[0052] In an exemplary embodiment of this disclosure, when the wind turbine is operating and the air clearance equipment is functioning properly, various turbine-related data (including turbine operating data and wind parameter data) can be measured or detected using various sensors installed on the wind turbine. Air clearance monitoring data can also be detected using the wind turbine's air clearance equipment. This data can be stored as historical data according to its sampling time. Therefore, when training the air clearance equipment model, training samples can be obtained from this stored data. For example, the turbine-related data collected by the target wind turbine at time t and the monitoring data collected by the air clearance equipment of its associated wind turbine at time t can be obtained as the sample feature data of the t-th training sample. Furthermore, the actual monitoring data collected by the air clearance equipment of the target wind turbine at time t can be obtained as the sample label of the t-th training sample, and so on. In exemplary embodiments of this disclosure, the aforementioned historical data may be stored in the relevant memory of each wind turbine and periodically transmitted to a networked server for storage. Alternatively, the historical data may be directly transmitted back to the server and stored thereafter after being acquired. When a wind turbine needs to use this historical data (e.g., for training, updating, and / or testing the airspace equipment model, as described later), this historical data may be transmitted to the wind turbine's main controller. Alternatively, this historical data may not be transmitted to the wind turbine's main controller but may be used by the server to perform training, updating, and / or testing of the airspace equipment model, as explained later. In exemplary embodiments of this disclosure, training samples may be obtained from data collected over a recent period (e.g., the past year or several years). Furthermore, to ensure the most comprehensive coverage of operating conditions, the training samples may also include samples triggered by airspace control strategies. In other words, the training samples according to the exemplary embodiments of this disclosure may include a first training sample and a second training sample. The first training sample may include sample feature data and sample labels collected when the target wind turbine does not trigger the airspace control strategy, and the second training sample may include sample feature data and sample labels collected when the target wind turbine triggers the airspace control strategy. In the exemplary embodiments of this disclosure, the second training sample may be obtained from more historical data than the historical data used to obtain the first training sample. As an example only, the first training sample may be obtained from historical data from the past year, and the second training sample may be obtained from historical data from the past three years. This is because the amount of data that triggers the airspace control strategy within a year may be much less than the amount of data that does not trigger it. If the proportion of the second training sample is too small, the airspace equipment model may have a large calculation deviation for situations where the airspace control strategy needs to be triggered. Conversely, when the amount of data that triggers the airspace control strategy within a year is much greater than the amount of data that does not trigger it, the proportion of the first training sample may be too small. In this case, the first training sample may be obtained from more historical data than the historical data used to obtain the second training sample.In this way, it can be ensured that the number of the two types of training samples is basically the same, and the training samples can cover more application scenarios. In an exemplary embodiment of this disclosure, the amount of data used for training can be determined according to the storage capacity and computing power of the wind turbine.

[0053] In step S302, the baseline model can be trained based on the acquired training samples, and the trained baseline model can be used as the clearance equipment model. Given that the training samples are acquired and the training objective of the training task is known, the method of training the model using the training samples is known to those skilled in the art and therefore will not be explained in detail here.

[0054] The methods for generating the airspace clearance model have been described above. However, it should be understood that after the airspace clearance model is generated, it can be updated periodically (e.g., periodically) using the latest data to avoid the airspace clearance model trained using past data failing to make accurate predictions in the current scenario due to changes in the wind farm or other factors. In exemplary embodiments of this disclosure, the airspace clearance model can be updated in the same way as the method for generating the airspace clearance model; that is, the airspace clearance model can be retrained using the latest stored data or data from a recent period, which will not be described in more detail here.

[0055] In the exemplary embodiments of this disclosure, after generating the airspace clearance equipment model as described above, the model can be used to predict airspace clearance equipment monitoring data in the event of airspace clearance equipment failure. Specifically, the relevant data of the target wind turbine and the monitoring data of the airspace clearance equipment of associated wind turbines at the current moment can be obtained first. Then, the obtained relevant data of the target wind turbine and the monitoring data of the airspace clearance equipment of associated wind turbines at the current moment are input into the airspace clearance equipment model. The output of the airspace clearance equipment model can be the predicted monitoring data of the airspace clearance equipment of the target wind turbine.

[0056] In the exemplary embodiments of this disclosure, the training process of the air clearance equipment model described above can be performed in the main control unit of the corresponding wind turbine or in a server connected to the main control unit. When the training of the air clearance equipment model is performed by the main control unit of the wind turbine, the main control unit can obtain historical data from the server to generate training samples for training. When the training of the air clearance equipment model is performed by the server, the server can generate training samples based on the stored historical data and train the air clearance equipment model, and then transmit the trained air clearance equipment model to the main control unit of the wind turbine. Alternatively, the server may not transmit the trained air clearance equipment model to the main control unit of the wind turbine, but instead complete the function of the air clearance equipment model (i.e., predict the monitoring data of the air clearance equipment) on the server, and then directly provide the output result of the air clearance equipment model (i.e., the predicted monitoring data) to the main control unit of the wind turbine for subsequent use.

[0057] It should be understood that monitoring data predicted through the airspace equipment model may still contain errors because it is not actual monitoring data. Such errors can lead to inaccuracies in the airspace values ​​when using this monitoring data to determine airspace clearance values, thus affecting the selection of airspace control strategies. To address this, in an exemplary embodiment of this disclosure, the trained airspace control model can be tested, and the airspace value deviation distribution related to the airspace equipment model can be determined based on the test results. Then, the airspace value deviation confidence interval can be determined based on the airspace value deviation distribution. The following will combine... Figure 4 This will be explained in detail.

[0058] Figure 4 This is a schematic flowchart illustrating a method for determining the net clearance deviation distribution according to an exemplary embodiment of the invention.

[0059] Reference Figure 4 In step S401, multiple test samples can be acquired. Each test sample includes historical feature data of the test sample and historical tags of the test sample. The historical feature data of the test sample may include the unit operation data of the target wind turbine at the corresponding historical sampling time and the monitoring data of the air clearance equipment of the associated wind turbine at the corresponding historical sampling time. The historical tags of the test sample include the actual monitoring data of the air clearance equipment of the target wind turbine at the corresponding historical sampling time. That is, test samples can be acquired in the same way as training samples. In an exemplary embodiment of this disclosure, the test samples may be a subset of samples selected from the training samples, or they may be newly constructed samples from stored historical data.

[0060] Then, in step S402, test samples can be input into the airspace equipment model to obtain predictive monitoring data corresponding to each test sample.

[0061] In step S403, based on the actual monitoring data and predicted monitoring data corresponding to each test sample, the actual net air value and predicted net air value corresponding to each test sample can be calculated respectively, and the net air value deviation of each test sample can be determined based on the ratio of the actual net air value to the predicted net air value. For example, the actual net air value D can be calculated based on the actual monitoring data B of test sample t. real Based on the predictive monitoring data B' of the test sample t, the predicted net gap value D' is calculated, thereby determining the net gap value deviation of the test sample t as D. real / D'.

[0062] In step S404, statistical analysis can be performed on the net void deviation of multiple test samples to determine the net void deviation distribution. Calculating the distribution of a large number of sample values ​​using statistical methods is known to those skilled in the art, and therefore will not be detailed here for the sake of brevity.

[0063] The airspace deviation distribution can be used to determine the airspace control strategy to ensure the safety of the wind turbine, which will be described later.

[0064] Furthermore, it should be understood that the training process for models similar to those used for clearing equipment... Figure 4 The method shown can be performed on the main control unit of the corresponding wind turbine, or on a server connected to the main control unit's network. When both the training and testing of the air clearance equipment model are performed by the main control unit of the wind turbine, the main control unit can obtain historical data from the server to generate test samples to complete the process. Figure 4 The method shown, where both training and testing of the air defense device model are performed by the server, allows the server to generate test samples based on stored historical data to complete the process. Figure 4 The method shown is then used to provide the results of the method (i.e., the net clearance deviation distribution) to the main controller of the wind turbine unit.

[0065] Return to reference Figure 1 In step S102, the predicted air clearance value of the target wind turbine can be determined based on the predicted monitoring data of the air clearance equipment. The method of determining the predicted air clearance value of the target wind turbine based on the monitoring data of the air clearance equipment is known to those skilled in the art. For example, the method of calculating the air clearance value using actual monitoring data when the air clearance equipment is not in failure, as described above, can be used for calculation, and will not be described in more detail here.

[0066] Subsequently, in step S103, a clearance control strategy for the target wind turbine can be determined based on the predicted clearance value determined in step S102. In an exemplary embodiment of this disclosure, the clearance control strategy adopted in the event of clearance equipment failure can be the same as when the clearance equipment is not failure. In this case, the corresponding clearance control strategy can be determined in the same manner as when the clearance equipment is not failure, based on the determined clearance value (e.g., the clearance level at which the clearance value is located).

[0067] Furthermore, in exemplary embodiments of this disclosure, when the airspace deviation distribution is also determined as described above, the airspace deviation distribution may also be incorporated when determining the airspace control strategy.

[0068] In an exemplary embodiment of the present invention, a lower air clearance boundary value can be determined based on the predicted air clearance value and the air clearance value deviation confidence interval, and an air clearance control strategy for the target wind turbine can be determined based on the predicted air clearance value, the lower air clearance boundary value, and a preset air clearance level. The air clearance value deviation confidence interval can indicate the fluctuation range of the predicted air clearance value at a preset confidence level, and can be selected from the air clearance value deviation distribution based on the preset confidence level.

[0069] Specifically, a pre-set confidence interval for the net air value deviation can be selected from the net air value deviation distribution. Based on the selected confidence interval for the net air value deviation, the upper and lower boundary values ​​of the predicted net air value determined in step S102 are calculated. Then, based on the predicted net air value and the lower boundary value, the corresponding level of net air control strategy is selected as the net air control strategy for the target wind turbine.

[0070] As an example only, assuming the predicted net air value calculated from the predictive monitoring data obtained in step S101 is D, and a net air value deviation confidence interval [a%, c%] at, for example, a 95% confidence level is selected to calculate the boundary values, then the upper and lower boundary values ​​of the predicted net air value D can be calculated as D0. low =D×a% and D high =D×c%.

[0071] In an exemplary embodiment of this disclosure, the predicted net clearance value D and its corresponding lower net clearance boundary value D can be used as a basis. low To select different levels of airspace control strategies. As an example only, consider predicting the airspace value D and its corresponding lower airspace boundary value D0. low The airspace clearance level determines the corresponding airspace control strategy. In the following explanation, it is assumed that the airspace values ​​corresponding to airspace clearance levels L1 to Ln-1 decrease, and the probability of a tower sweeping accident occurring at each corresponding airspace value is lower than a preset threshold. Airspace clearance level Ln indicates that the probability of a tower sweeping accident occurring at the corresponding airspace value is equal to or higher than the preset threshold. The following will combine... Figure 5 This needs to be explained.

[0072] Figure 5 This is a flowchart illustrating a method for determining a wind turbine's airspace control strategy according to an exemplary embodiment of the invention.

[0073] Reference Figure 5 After obtaining the predictive monitoring data from the airspace equipment model, in steps S501 and S502, the predicted airspace value D and its corresponding lower airspace boundary value D can be calculated respectively according to the methods described above. low Then, in step S503, based on the preset control logic for the air clearance equipment in the wind turbine, it can be determined that the predicted air clearance value D is at control level Lx. When the predicted air clearance value D is at air clearance level Lx, in step S504, the lower boundary value D of the air clearance can be determined. low Is it at clearance level Ln? If the lower clearance boundary value D low If the airspace is not at clearance level Ln, it means that even with the most conservative airspace result, the airspace protection boundary will not be reached. Therefore, to ensure the power generation of the wind turbine, in step S505, a control strategy x corresponding to the airspace level Lx can be selected. If the lower boundary value of the airspace is D... low If the airspace level is Ln, it means that even with the most conservative airspace prediction, the airspace might reach the airspace protection boundary. In this case, there is a significant risk of tower sweeping if the predicted airspace result fluctuates within a certain range. Therefore, in step S506, the most conservative control strategy can be selected, namely control strategy n corresponding to airspace level Ln. This method of selecting an airspace control strategy effectively balances the safety and economy of the wind turbine.

[0074] In summary, this invention, by establishing a clearance equipment model, can quickly and accurately obtain the predicted clearance value of a wind turbine after the actual clearance equipment fails, and based on this, perform clearance control on the wind turbine, avoiding the risk of tower sweeping accidents caused by clearance equipment failure. Compared with general software solutions based on control logic, it has higher control accuracy and can ensure the safety of the wind turbine while also considering its economic efficiency. Furthermore, this invention can continue clearance monitoring and control after clearance equipment failure, and because it also trains the clearance equipment model by associating it with the wind turbine, it can improve the accuracy of the clearance equipment model and further enhance control precision. On the other hand, this invention also introduces the concept of confidence intervals in the selection of control logic for the clearance control strategy, fully considering the uncertainty of the clearance equipment model, thus maximizing the balance between the safety and economy of the turbine even after clearance equipment failure.

[0075] Figure 6 This is a block diagram illustrating a wind turbine headroom control system 600 according to an exemplary embodiment of the present disclosure.

[0076] Reference Figure 6The wind turbine airspace control device 600 according to an exemplary embodiment of the present disclosure may include a monitoring data acquisition unit 610 and a control strategy determination unit 620.

[0077] In an exemplary embodiment of this disclosure, the monitoring data acquisition unit 610 can acquire predictive monitoring data of the air clearance equipment of the target wind turbine through the air clearance equipment model of the target wind turbine in response to the air clearance equipment failure of the target wind turbine. The control strategy determination unit 620 can determine the predicted air clearance value of the target wind turbine based on the predicted monitoring data of the air clearance equipment, and determine the air clearance control strategy for the target wind turbine based on the determined predicted air clearance value. Here, the air clearance equipment model can predict the monitoring data of the air clearance equipment of the target wind turbine based on the turbine-related data of the target wind turbine and the monitoring data of the air clearance equipment of associated wind turbines related to the target wind turbine.

[0078] In an exemplary embodiment of this disclosure, the monitoring data acquisition unit 610 may include a clearance equipment model generation unit (not shown), which can generate a clearance equipment model. Specifically, the clearance equipment model generation unit (not shown) can generate a clearance equipment model through the following operations: acquiring training samples, wherein each training sample includes sample historical feature data and sample historical labels, the sample historical feature data including the unit-related data of the target wind turbine at the corresponding historical sampling time and the monitoring data of the clearance equipment of the associated wind turbine at the corresponding historical sampling time, and the sample labels including the actual monitoring data of the clearance equipment of the target wind turbine at the corresponding historical sampling time; training a benchmark model based on the acquired training samples, and using the trained benchmark model as the clearance equipment model.

[0079] In an exemplary embodiment of this disclosure, the monitoring data acquisition unit 610 can acquire the predictive monitoring data of the air clearance equipment through the following operations: in response to the failure of the air clearance equipment of the target wind turbine, it acquires the turbine-related data of the target wind turbine at the current time and the monitoring data of the air clearance equipment of the associated wind turbine at the current time, and inputs the acquired turbine-related data of the target wind turbine at the current time and the monitoring data of the air clearance equipment of the associated wind turbine at the current time into the air clearance equipment model to obtain the predictive monitoring data of the target wind turbine.

[0080] In an exemplary embodiment of this disclosure, the control strategy determination unit 620 can determine a lower boundary value of the airspace based on the predicted airspace value and the confidence interval of the deviation from the predicted airspace value, and determine an airspace control strategy for the target wind turbine based on the predicted airspace value, the lower boundary value of the airspace, and a preset airspace level. Here, the confidence interval of the deviation from the airspace value can indicate the fluctuation range of the predicted airspace value under a preset confidence level.

[0081] In an exemplary embodiment of this disclosure, the preset airspace levels may include multiple levels, and different airspace levels may correspond to different airspace control strategies. Furthermore, the control strategy determination unit 620 may determine the airspace control strategy for the target wind turbine based on the following operations: based on a predicted airspace value at a first airspace level, determining whether the lower airspace boundary value is at a second airspace level, wherein the second airspace level indicates that the probability of a tower sweeping accident occurring at the airspace value corresponding to the second airspace level is equal to or higher than a preset threshold; in response to the lower airspace boundary value being at the second airspace level, determining the airspace control strategy corresponding to the second airspace level as the airspace control strategy for the target wind turbine; in response to the lower airspace boundary value not being at the second airspace level, determining the airspace control strategy corresponding to the first airspace level as the airspace control strategy for the target wind turbine.

[0082] In an exemplary embodiment of this disclosure, the monitoring data acquisition unit 610 may further include a testing unit (not shown). The testing unit (not shown) may test the generated airspace equipment model, determine the airspace value deviation distribution based on the test results, and may also determine the airspace value deviation confidence interval based on the airspace value deviation distribution. Specifically, the test unit (not shown) can determine the airspace deviation distribution through the following operations: acquiring multiple test samples, wherein each test sample includes historical feature data and historical labels, the historical feature data including the operating data of the target wind turbine at the corresponding historical sampling time and the monitoring data of the airspace equipment of the associated wind turbine at the corresponding historical sampling time, and the test sample labels including the actual monitoring data of the airspace equipment of the target wind turbine at the corresponding historical sampling time; inputting the test samples into the airspace equipment model to obtain the predicted monitoring data corresponding to each test sample; calculating the actual airspace value and the predicted airspace value corresponding to each test sample based on the actual monitoring data and the predicted monitoring data corresponding to each test sample; determining the airspace deviation of each test sample based on the ratio of the actual airspace value to the predicted airspace value corresponding to each test sample; and performing statistical analysis on the airspace deviations of the multiple test samples to determine the airspace deviation distribution. In an exemplary embodiment of the present invention, the airspace deviation confidence interval can be determined from the airspace deviation distribution based on a preset confidence level.

[0083] In an exemplary embodiment of this disclosure, the monitoring data acquisition unit 610 may divide the target wind turbine into multiple sectors, and select the wind turbines within a preset distance from the target wind turbine in each sector as associated wind turbines of the target wind turbine.

[0084] In an exemplary embodiment of this disclosure, the air clearance device model generation unit (not shown) may also periodically update the air clearance device model.

[0085] The above has been referred to Figures 1 to 5 The specific operation of each component of the airspace control system 600 has been described in detail, so for the sake of brevity, it will not be repeated here.

[0086] Figure 7 This is a block diagram illustrating an electronic device 700 according to an exemplary embodiment of the present disclosure.

[0087] According to embodiments of the present disclosure, an electronic device may be provided. The electronic device includes at least one memory 701 and at least one processor 702, wherein the at least one memory stores a set of computer-executable instructions, which, when executed by the at least one processor, perform a wind turbine headroom control method according to embodiments of the present disclosure.

[0088] As an example, electronic device 700 may be a PC, tablet, personal digital assistant, smartphone, or other device capable of executing the aforementioned set of instructions. Here, electronic device 700 is not necessarily a single electronic device, but may be a collection of any devices or circuits capable of executing the aforementioned instructions (or instruction sets) individually or in combination. Electronic device 700 may also be part of an integrated control system or system manager, or may be configured to interconnect with a portable electronic device locally or remotely (e.g., via wireless transmission) through an interface.

[0089] In electronic device 700, processor 702 may 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, processor 702 may also include analog processors, digital processors, microprocessors, multi-core processors, processor arrays, network processors, etc.

[0090] The processor 702 can execute instructions or code stored in memory, wherein memory 701 can also store data. Instructions and data can also be sent and received via a network through a network interface device, wherein the network interface device can employ any known transmission protocol.

[0091] The memory 701 can be integrated with the processor 702, for example, by placing RAM or flash memory within an integrated circuit microprocessor. Alternatively, the memory 701 can include a separate device, such as an external disk drive, a storage array, or other storage device usable by any database system. The memory 701 and the processor 702 can be operatively coupled, or can communicate with each other, for example, via I / O ports, network connections, etc., enabling the processor 702 to read files stored in the memory 701.

[0092] In addition, the electronic device 700 may also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, mouse, touch input device, etc.). All components of the electronic device can be interconnected via a bus and / or network.

[0093] According to embodiments of this disclosure, a computer-readable storage medium may also be provided, wherein when 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 wind turbine headroom control method of the present 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. The computer program in the aforementioned computer-readable storage medium can run in an environment deployed in computer devices such as clients, hosts, agent devices, servers, etc. Furthermore, 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.

[0094] According to an embodiment of this disclosure, a computer program product is provided, including computer instructions, which, when executed by a processor, implement the wind turbine headroom control method of this disclosure.

[0095] 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 following claims.

[0096] 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.

Claims

1. A method for controlling the airspace clearance of a wind turbine generator, the method comprising: In response to the failure of the air defense equipment of the target wind turbine, predictive monitoring data of the air defense equipment is obtained through the air defense equipment model of the target wind turbine. The predicted air clearance value of the target wind turbine is determined based on the predictive monitoring data of the air clearance equipment. Based on the predicted air clearance value, determine the air clearance control strategy for the target wind turbine; The air clearance equipment model is used to predict the monitoring data of the air clearance equipment of the target wind turbine based on the turbine-related data of the target wind turbine and the monitoring data of the air clearance equipment of the associated wind turbines. Among them, determining the airspace control strategy for the target wind turbine based on the predicted airspace value includes: The lower boundary value of the net airspace is determined based on the confidence interval between the predicted net airspace value and the net airspace value deviation, wherein the confidence interval between the net airspace value deviation indicates the fluctuation range of the predicted net airspace value under a preset confidence level. Based on the predicted airspace value, the lower airspace boundary value, and the preset airspace level, an airspace control strategy is determined for the target wind turbine.

2. The method as described in claim 1, characterized in that, The method also includes generating the clearance equipment model. The steps for generating the air clearance equipment model include: Acquire training samples, wherein each training sample includes historical feature data and historical labels. The historical feature data includes the relevant data of the target wind turbine at the corresponding historical sampling time and the monitoring data of the air clearance equipment of the associated wind turbine at the corresponding historical sampling time. The historical labels include the actual monitoring data of the air clearance equipment of the target wind turbine at the corresponding historical sampling time. The baseline model is trained based on the acquired training samples, and the trained baseline model is used as the air clearance equipment model.

3. The method as described in claim 2, characterized in that, The training samples include a first training sample and a second training sample, wherein, The first training sample includes historical feature data and historical labels of samples collected when the target wind turbine did not trigger the airspace control strategy. The second training sample includes historical feature data and historical labels of samples collected when the target wind turbine triggers the airspace control strategy.

4. The method as described in claim 2, characterized in that, The method further includes periodically updating the air clearance equipment model.

5. The method as described in claim 1, characterized in that, In response to the failure of the air defense equipment of the target wind turbine, the steps for obtaining predictive monitoring data of the air defense equipment through the air defense equipment model of the target wind turbine include: In response to a failure of the air clearance equipment of the target wind turbine, the system acquires the relevant data of the target wind turbine at the current moment, as well as the monitoring data of the air clearance equipment of the associated wind turbine at the current moment. The relevant data of the target wind turbine at the current moment and the monitoring data of the air clearance equipment of the associated wind turbine at the current moment are input into the air clearance equipment model to obtain the predicted monitoring data of the target wind turbine.

6. The method as described in claim 1, characterized in that, There are multiple preset airspace levels, and different airspace levels correspond to different airspace control strategies. Based on the predicted airspace clearance value, the lower airspace boundary value, and the preset airspace clearance level, an airspace control strategy is determined for the target wind turbine, including: Based on the predicted air clearance value being at the first air clearance level, it is determined whether the lower boundary value of the air clearance is at the second air clearance level, wherein the second air clearance level indicates that the probability of a tower sweeping accident occurring at the air clearance value corresponding to the second air clearance level is equal to or higher than a preset threshold. In response to the lower boundary value of the airspace being at the second airspace level, the airspace control strategy corresponding to the second airspace level is determined as the airspace control strategy for the target wind turbine. In response to the lower boundary value of the airspace not being at the second airspace level, the airspace control strategy corresponding to the first airspace level is determined as the airspace control strategy for the target wind turbine.

7. The method as described in claim 1, characterized in that, The method further includes: testing the generated airspace equipment model and determining the airspace value deviation distribution based on the test results; and determining the airspace value deviation confidence interval based on the airspace value deviation distribution. The generated airspace equipment model is tested, and the airspace deviation distribution is determined based on the test results, including: Multiple test samples are obtained, wherein each test sample includes test sample historical feature data and test sample historical tags. The test sample historical feature data includes the unit operation data of the target wind turbine at the corresponding historical sampling time and the monitoring data of the air clearance equipment of the associated wind turbine at the corresponding historical sampling time. The test sample historical tags include the actual monitoring data of the air clearance equipment of the target wind turbine at the corresponding historical sampling time. The test samples are input into the airspace equipment model to obtain predictive monitoring data corresponding to each test sample; Based on the actual monitoring data and predicted monitoring data corresponding to each test sample, the actual net air value and predicted net air value corresponding to each test sample are calculated respectively. The net miss deviation for each test sample is determined based on the ratio of the actual net miss value to the predicted net miss value for each test sample; and The net void deviation distribution is determined by statistical analysis of the net void deviation of the multiple test samples.

8. The method according to any one of claims 1 to 7, characterized in that, The method further includes: Divide the target wind turbine into multiple sectors; Wind turbines within a preset distance from the target wind turbine in each sector are identified as associated wind turbines of the target wind turbine.

9. The method according to any one of claims 1 to 7, characterized in that, The relevant data for the wind turbine unit includes at least one of the following: the unit operation data and wind parameter data of the target wind turbine unit. in, Wind parameter data includes at least one of wind speed, wind direction, wind shear, and inflow angle; The unit's operating data includes at least one of yaw angle, pitch angle, speed, and torque; The monitoring data of the air clearance equipment includes the monitoring values ​​obtained by measuring using the air clearance equipment.

10. A wind turbine air clearance control system, the system comprising: The monitoring data acquisition unit is configured to acquire predictive monitoring data of the air clearance equipment through the air clearance equipment model of the target wind turbine in response to the air clearance equipment failure of the target wind turbine. The control strategy determination unit is configured to determine the predicted air clearance value of the target wind turbine based on the predicted monitoring data of the air clearance equipment, and to determine the air clearance control strategy for the target wind turbine based on the predicted air clearance value. The air clearance equipment model is used to predict the monitoring data of the air clearance equipment of the target wind turbine based on the turbine-related data of the target wind turbine and the monitoring data of the air clearance equipment of the associated wind turbines. The control strategy determination unit is configured to: determine the lower boundary value of the airspace based on the predicted airspace value and the airspace value deviation confidence interval, wherein the airspace value deviation confidence interval indicates the fluctuation range of the predicted airspace value under a preset confidence level; and determine the airspace control strategy for the target wind turbine based on the predicted airspace value, the lower boundary value of the airspace, and the preset airspace level.

11. A system comprising at least one computing device and at least one storage device for storing instructions, wherein, When the instruction is executed by the at least one computing device, it causes the at least one computing device to perform the method as described in any one of claims 1 to 9.

12. A computer-readable storage medium for storing instructions, wherein, When the instruction is executed by at least one computing device, it causes the at least one computing device to perform the method as described in any one of claims 1 to 9.

Images