Method for detecting abnormality of wind turbine generator system and related device

Abstract

The application discloses an abnormality detection method of a wind turbine generator system and related devices, and the method comprises the following steps: acquiring operation data of each wind turbine generator in a wind farm in a historical time period; processing the operation data to obtain a plurality of operation data sets, each operation data set comprising operation data of each wind turbine generator in a continuous time period, and each wind turbine generator being in a yaw state in the continuous time period; screening a target operation data set in which each wind turbine generator is in a wind-aiming yaw state from the plurality of operation data sets; finding abnormal data deviating from a preset data range from the target operation data set; and determining a wind turbine generator corresponding to the abnormal data as a wind turbine generator with yaw vibration abnormality. The application solves the problem that yaw vibration abnormality is difficult to find.

Description

Technical Field

[0001] This application belongs to the field of wind power generation technology, and in particular relates to a method, device, equipment, computer-readable storage medium and computer program product for detecting anomalies in wind turbine generator sets. Background Technology

[0002] As heavy equipment operating for long periods, wind turbine generators work in complex environments. Vibration is a crucial indicator of a wind turbine's stability. Vibration within the permissible range is normal, but excessive vibration can damage components and even lead to major accidents. Currently, to prevent abnormal vibration, protection thresholds are typically set. When the vibration becomes too high and exceeds these thresholds, the turbine is automatically shut down to prevent excessive impact.

[0003] However, during the yaw process, the vibration is filtered out due to the settings of the sensor control and protection algorithm. As a result, some vibration impacts are filtered out during the yaw process, making it difficult to detect abnormal vibrations during the yaw process. Summary of the Invention

[0004] This application provides an anomaly detection method and related device for wind turbine generator sets, which can solve the problem that existing technologies have difficulty in detecting yaw vibration anomalies.

[0005] On one hand, embodiments of this application provide an anomaly detection method for wind turbine generator sets, the method comprising:

[0006] Obtain the operating data of each wind turbine generator in the wind farm within a historical time period;

[0007] The operational data is processed to obtain multiple operational datasets. Each operational dataset includes the operational data of each wind turbine generator set within a continuous time period, during which each wind turbine generator set is in a yaw state.

[0008] From multiple operational datasets, the target operational datasets in which each wind turbine is in a yaw state against the wind are selected;

[0009] Identify outlier data that deviates from the preset data range in the target dataset;

[0010] The wind turbines corresponding to the abnormal data were identified as wind turbines with abnormal yaw vibration.

[0011] Optionally, the operational data includes yaw flags, unmooring flags, and yaw residual pressure; from multiple operational datasets, a target operational dataset is selected to indicate that each wind turbine is in a yaw state against the wind, including:

[0012] Obtain the yaw flag, unmooring flag, and yaw residual pressure of each wind turbine in each running dataset;

[0013] Based on the yaw flag position of each wind turbine, determine the continuous yaw duration of each wind turbine within each continuous time period.

[0014] The target operating dataset is selected based on the following conditions: the continuous yaw duration meets the first preset condition, the unmooring flag meets the second preset condition, and the yaw residual pressure meets the third preset condition. The first preset condition is that within the continuous time period corresponding to the operating dataset, the continuous yaw duration of each wind turbine is greater than or equal to the preset duration. The second preset condition is that each unmooring flag in the operating dataset indicates that each wind turbine is in a non-unmoored state. The third preset condition is that each yaw residual pressure in the operating dataset is less than the preset residual pressure threshold.

[0015] Optionally, obtain the operating data of each wind turbine generator in the wind farm during a historical time period, including:

[0016] Obtain SCADA data for each wind turbine generator set within a historical time period;

[0017] SCADA data with the unit's operating status set to maintenance are removed to obtain the operating data.

[0018] Optionally, the operational data includes vibration acceleration; outlier data deviating from the preset data range is identified from the target operational dataset, including:

[0019] Calculate the statistical values ​​of vibration acceleration of each wind turbine in the target operating dataset over a continuous time period. The statistical values ​​include at least one of the maximum, minimum and mean values.

[0020] From the statistical values ​​corresponding to vibration acceleration, identify abnormal statistical values ​​whose values ​​are outside the preset data range. The operational data to which these abnormal statistical values ​​belong are considered abnormal data.

[0021] Optionally, from the statistical values ​​corresponding to vibration acceleration, abnormal statistical values ​​whose values ​​are outside the preset data range are identified, including:

[0022] In a coordinate system with time as the first coordinate axis and vibration acceleration magnitude as the second coordinate axis, the statistical values ​​corresponding to the vibration acceleration of the wind turbine generator in each consecutive time period are used as input data to form a scatter plot;

[0023] Capture discrete points in a scatter plot that are outside the preset data range; the statistical values ​​corresponding to these discrete points are outlier statistical values.

[0024] Optionally, after identifying the wind turbine corresponding to the abnormal data as a wind turbine with abnormal yaw vibration, the method further includes:

[0025] Obtain the generator unit identification of wind turbine generators with abnormal yaw vibration;

[0026] The system sends alerts, including abnormal data and unit identification information, to the operation and maintenance terminal.

[0027] On the other hand, embodiments of this application provide an anomaly detection device for wind turbine generator sets, the device comprising:

[0028] The acquisition module is used to acquire the operating data of each wind turbine generator in the wind farm within a historical time period;

[0029] The processing module is used to process the operating data to obtain multiple operating datasets. Each operating dataset includes the operating data of each wind turbine generator set within a continuous time period, during which each wind turbine generator set is in a yaw state.

[0030] The filtering module is used to filter out the target operational datasets from multiple operational datasets, which are in the wind-yawing state of each wind turbine.

[0031] The search module is used to find abnormal data that deviates from the preset data range from the target running dataset;

[0032] The determination module is used to identify wind turbine generators with abnormal data as having abnormal yaw vibration.

[0033] Furthermore, embodiments of this application provide an anomaly detection device for wind turbine generator sets, the device comprising:

[0034] Processor and memory storing computer program instructions;

[0035] The processor implements the above-described anomaly detection method for wind turbine generators when executing computer program instructions.

[0036] In another aspect, embodiments of this application provide a computer storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, they implement the anomaly detection method for wind turbine generator sets as described above.

[0037] In another aspect, embodiments of this application provide a computer program product that stores computer program instructions, which, when executed by a processor, implement the above-described method for detecting anomalies in wind turbine generator sets.

[0038] The anomaly detection method and related apparatus for wind turbine generators in this application can acquire the operating data of each wind turbine generator in a wind farm within a historical time period; process the operating data to obtain multiple operating datasets, and then filter out a target operating dataset from these datasets; identify abnormal data that deviates from a preset data range from the target operating dataset; and designate the wind turbine generators corresponding to the abnormal data as those with yaw vibration anomalies. Because the filtering and processing ultimately yields the operating data of each wind turbine generator when it is in a yaw state against the wind within each continuous time period, it is possible to analyze and identify wind turbine generators with abnormal data during yaw under the same time conditions. Therefore, by detecting yaw vibration anomalies early based on historical data, the problem of difficulty in detecting yaw vibration anomalies in existing technologies is solved. Attached Figure Description

[0039] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating an anomaly detection method for a wind turbine generator set provided in one embodiment of this application;

[0041] Figure 2 This is a detailed flowchart of step S103 in an embodiment of the anomaly detection method for a wind turbine generator provided in this application.

[0042] Figure 3 This is a detailed flowchart of S104 in an anomaly detection method for a wind turbine generator provided in one embodiment of this application;

[0043] Figure 4 This is an optional schematic diagram of a scatter plot generated in an anomaly detection method for a wind turbine generator provided in one embodiment of this application.

[0044] Figure 5 This is a timing diagram of the yaw position drawn in the anomaly detection method for a wind turbine generator provided in one embodiment of this application.

[0045] Figure 6 This is a timing diagram of nacelle acceleration drawn in an anomaly detection method for a wind turbine generator provided in one embodiment of this application.

[0046] Figure 7 This is a schematic diagram of the structure of an anomaly detection device for a wind turbine generator set provided in another embodiment of this application;

[0047] Figure 8This is a structural schematic diagram of an anomaly detection device for a wind turbine generator set provided in another embodiment of this application. Detailed Implementation

[0048] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0049] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0050] A wind turbine's yaw system is a system that controls the nacelle's rotation to face the wind based on measured wind direction. It is generally divided into active yaw systems and passive yaw systems. The yaw system ensures that the wind turbine's nacelle continuously changes direction to follow changes in wind direction, always remaining in a windward position.

[0051] The yaw system can also work in conjunction with the control system to keep the wind turbine rotor always in the windward position, thereby improving power generation efficiency. In addition, the yaw system can also provide locking torque to ensure the safe operation of the wind turbine generator set. Therefore, the yaw system is one of the essential systems for wind turbine generator sets.

[0052] In the field of wind power generation, the vibration of wind turbine generators is a crucial indicator of unit stability. Specifically, during normal operation, components such as the generator and tower within the wind turbine generator will vibrate, and vibration within permissible ranges is normal. However, if the vibration of the wind turbine generator intensifies, it may lead to component damage or even a major accident. Therefore, relevant technologies typically set corresponding protection values ​​for wind turbine generators. When the vibration of the wind turbine generator exceeds the protection value, a protective shutdown is initiated to prevent the wind turbine generator from suffering excessive impact.

[0053] In the vibration testing of wind turbine generators, there is a special scenario: abnormal vibration detection during yaw. In fact, when the acceleration of the wind turbine generator exceeds the protection value during yaw vibration, it will not only increase the failure rate of the wind turbine generator but also shorten the service life of the yaw braking components.

[0054] In addition, wind turbine generators are equipped with control protection algorithms to prevent shutdowns caused by sensor malfunctions. These algorithms filter vibrations, which means that some vibrations and impacts during yaw are filtered out and difficult to detect.

[0055] Therefore, existing solutions for abnormal vibration during yaw are too late in protecting wind turbine generators. By the time the abnormality is detected, the generator has already lost power, and the lifespan of the yaw components of the wind turbine generator has been affected, with an increased failure rate, thus impacting the normal operation of the wind turbine generator.

[0056] To address the aforementioned technical problems, this application provides a method, apparatus, device, computer-readable storage medium, and computer program product for detecting anomalies in wind turbine generator sets. The method for detecting anomalies in wind turbine generator sets provided in this application will be described first.

[0057] Figure 1 A flowchart illustrating an anomaly detection method for a wind turbine generator set according to an embodiment of this application is shown. Figure 1 As shown, the method includes:

[0058] S101: Obtain the operating data of each wind turbine generator in the wind farm during the historical time period.

[0059] S102, process the operating data to obtain multiple operating datasets. Each operating dataset includes the operating data of each wind turbine generator set within a continuous time period, during which each wind turbine generator set is in a yaw state.

[0060] S103, from multiple operational datasets, select the target operational datasets in which each wind turbine is in a yaw state against the wind.

[0061] S104: Find out abnormal data that deviates from the preset data range from the target running dataset.

[0062] S105, identify the wind turbine generator corresponding to the abnormal data as the wind turbine generator with abnormal yaw vibration.

[0063] This application's embodiments can acquire operational data of each wind turbine generator in a wind farm within a historical time period; process the operational data to obtain multiple operational datasets; then filter out a target operational dataset from these datasets; identify abnormal data deviating from a preset data range within the target operational dataset; and designate the wind turbine generators corresponding to the abnormal data as those exhibiting yaw vibration anomalies. Because the filtering and processing ultimately yields operational data for each wind turbine generator when it is yawing against the wind within each continuous time period, it is possible to analyze and identify wind turbine generators with abnormal data during yawing against the wind under the same time conditions. Therefore, by using historical data, abnormal yaw vibration can be detected early, thus solving the problem of difficulty in detecting abnormal yaw vibration in existing technologies.

[0064] In the embodiments of this application, yaw vibration anomaly detection is performed based on the historical operating data of the wind turbine generator set. It should be noted that the operating data of the wind turbine generator set records data under different wind conditions and operating states for all time periods. This operating data can be obtained from the SCADA (Supervisory Control and Data Acquisition) system or directly from the wind turbine generator set and / or sensors.

[0065] The historical time period can be all previous time periods, or it can be a certain period of time since the extraction time, such as the running data within 1 year since the extraction time.

[0066] To obtain abnormal data during the yaw process, it is necessary to find operational data matching the yaw process from all historical operational data within a given time period. Currently, the yaw process of wind turbine generators is automatically completed by the unit's control logic, and is affected by wind conditions and the operating conditions of the wind turbine generator, resulting in variations in parameter settings, number of yaws, and duration of each yaw. Therefore, how to sift through massive amounts of operational data to obtain abnormal data during the yaw process of wind turbine generators is one of the technical problems that this application's embodiments aim to solve in order to achieve abnormal yaw vibration detection.

[0067] Therefore, in some embodiments, the operating data of wind turbine generators in a non-maintenance state (i.e., in standby, power generation, etc.) can be screened from all data first. Taking the acquisition of operating data from the SCADA system as an example, in some optional examples, the above S101 process may include: acquiring SCADA data of each wind turbine generator within a historical time period; removing SCADA data of generators in a maintenance state to obtain the operating data. This achieves a first-stage screening of operating data, facilitating the subsequent rapid acquisition of abnormal data screening for wind yaw.

[0068] It should be noted that the wind farm SCADA system, namely the data acquisition and monitoring control system, is a computer-based DCS (Distributed Control System) and power automation monitoring system that can store data recorded by various sensors of the wind turbine generator set as well as various operating information of the unit.

[0069] In this embodiment, the SCADA system can store SCADA data for all wind turbine generators in the entire site. This SCADA data may include the generator number, operating status word, yaw flag, etc., of each wind turbine generator. To obtain the non-maintenance operating data of each wind turbine generator, SCADA data with the operating status word indicating maintenance can be filtered out.

[0070] In other examples, after removing SCADA data with the status word "maintenance," keywords can be set to retrieve the corresponding data from the SCADA data as key operational information. This key operational information is then used for subsequent processing and diagnosis, reducing the amount of data processing. This SCADA data should at least include the extraction time (or data generation time), the wind turbine generator number, the operational status word, the yaw flag, the unmooring flag, the vibration acceleration, and the yaw residual pressure.

[0071] In some embodiments, after obtaining the non-maintenance state operating data, the non-maintenance state operating data can be arranged in chronological order of extraction and deduplicated, and then the data within a continuous time period can be grouped together. By using the yaw flag, the non-yaw state data in each group can be removed, and the data in the yaw state can be retained. This allows for the acquisition of multiple operating datasets of each wind turbine in the yaw state within a continuous time period, so as to reconstruct the operating process of each wind turbine.

[0072] Understandably, due to various factors, wind turbines do not always yaw against the wind. Therefore, in order to find abnormal vibration data during the yaw process based on historical operating data, it is necessary to filter out some non-wind-yawing operating data and find the target operating dataset for wind-yawing.

[0073] In some embodiments, the yaw flag, unmooring flag, and yaw residual pressure in the operational data can be used to filter out the target operational dataset of each wind turbine in the wind yaw state from multiple operational datasets, so as to reconstruct the wind yaw operation process of each wind turbine.

[0074] For example, please see Figure 2 The process of selecting the target operational dataset for each wind turbine's yaw state may include:

[0075] S201, obtain the yaw flag, unmooring flag, and yaw residual pressure of each wind turbine in each running dataset.

[0076] S202, based on the yaw flag position of each wind turbine, determine the continuous yaw duration of each wind turbine in each continuous time period.

[0077] S203, select the operating dataset that meets the first preset condition for continuous yaw duration, the second preset condition for unmooring flag, and the third preset condition for yaw residual pressure as the target operating dataset. The first preset condition is that within the continuous time period corresponding to the operating dataset, the continuous yaw duration of each wind turbine is greater than or equal to the preset duration. The second preset condition is that each unmooring flag in the operating dataset indicates that each wind turbine is in a non-unmooring state. The third preset condition is that each yaw residual pressure in the operating dataset is less than the preset residual pressure threshold.

[0078] The system can simultaneously filter the data in each running dataset to obtain the target running dataset for the yaw state, or it can select running datasets one by one in a preset order for filtering and judgment. After filtering and judging all running datasets, the one that meets the requirements is the target running data for the yaw state.

[0079] The following example illustrates the process of filtering and judging each of the M groups of operational datasets to obtain the target operational dataset. Test operational datasets can be selected sequentially from the M groups of operational datasets according to the extraction time. Upon selecting a test operational dataset, the yaw flag, unmooring flag, and yaw residual pressure of each wind turbine generator in that dataset are extracted.

[0080] Specifically, based on the extracted yaw flag, the continuous yaw duration of each wind turbine generator within the corresponding continuous time period of the test run dataset can be determined. Then, it can be determined whether the test run dataset meets the first, second, and third preset conditions.

[0081] It should be noted that the first, second, and third preset conditions can be evaluated simultaneously or sequentially. The final target dataset that meets all three preset conditions is the running dataset. The following example demonstrates how to first filter using the second preset condition, then using the first preset condition, and finally using the third preset condition for further illustration.

[0082] It can be determined whether the untethering flags extracted from the test run dataset all indicate that the wind turbine generators are in a non-untethered state, that is, whether each wind turbine generator was in a non-untethered state when the operation data in the test run dataset was generated.

[0083] If not all untether flags in the current test dataset are in the untethered state, then the next test dataset is selected. Otherwise, the first preset condition can be used to further filter and judge the test dataset.

[0084] When using the first preset condition for judgment, you can check whether the continuous yaw duration of the wind turbine generator is greater than or equal to a preset duration within the continuous time period corresponding to the test run dataset. For example, the preset duration can be 20 seconds. If the continuous yaw duration of the test run dataset is greater than or equal to the preset duration, the test run data can be considered to meet the first preset condition.

[0085] Whether the test run dataset that meets the first preset condition is the target run dataset can be determined based on whether the yaw residual pressure meets the third preset condition. The preset residual pressure threshold set in the third preset condition can be set according to actual conditions, for example, it can be 40 bar.

[0086] Once the yaw residual pressure meets the third preset condition, i.e., the test run dataset is the target run dataset, the next run dataset can be selected as the test run dataset until all M test run datasets are selected. Then, all target run datasets are output, and all target run datasets are the target run datasets that meet the wind yaw state.

[0087] If the test dataset does not meet the first or third preset conditions, the next test dataset can be selected.

[0088] In the above scheme, by searching for non-maintenance operating data from the operating data of wind turbine generators, a yaw state operating dataset is formed. First, second, and third preset conditions are set, and combined with yaw flag, unmooring flag, and yaw residual pressure, the target dataset of each wind turbine generator's yaw state is obtained from massive data under the complex wind and operating conditions of the wind farm. This reconstructs the yaw process of each wind turbine generator and provides a technical basis for the anomaly detection of wind turbine generators.

[0089] In some embodiments, please refer to Figure 3 In S104, identifying abnormal data that deviates from the preset data range from the target running dataset can include:

[0090] S301, Calculate the statistical values ​​of the vibration acceleration of each wind turbine generator in the target operating dataset over a continuous time period. The statistical values ​​include at least one of the maximum, minimum and mean values.

[0091] S302, from the statistical values ​​corresponding to the vibration acceleration, find the abnormal statistical values ​​whose values ​​are not within the preset data range. The operating data to which the abnormal statistical values ​​belong is abnormal data.

[0092] In the above embodiments, based on the target operational dataset of the yaw state, vibration acceleration parameters in the operational data are used to diagnose yaw anomalies, thereby solving the problem of difficulty in detecting yaw vibration anomalies. Specifically, anomaly diagnosis is performed by calculating statistical values ​​related to vibration acceleration. These statistical values ​​include not only the maximum, minimum, and mean values, but also other statistical reference quantities such as median values. Thus, the mean, maximum, and minimum vibration acceleration values ​​corresponding to each set of target operational datasets can be output.

[0093] The vibration acceleration fluctuation range can be set and determined based on the vibration threshold to obtain the preset data range corresponding to the vibration acceleration. When the statistical value corresponding to the vibration acceleration of each wind turbine generator set is not within the preset data range, the statistical value outside the preset data range is considered an abnormal statistical value. The operating data to which the abnormal statistical value belongs is abnormal data, and the wind turbine generator set corresponding to the abnormal data is a wind turbine generator set with abnormal yaw vibration.

[0094] To facilitate the identification of abnormal data and provide clearer results for wind turbine generators exhibiting abnormal yaw vibration, a visual interface can be created to display the statistical results of each wind turbine generator over time.

[0095] In a coordinate system with time as the first axis and vibration acceleration magnitude as the second axis, the statistical values ​​corresponding to the vibration acceleration of the wind turbine generator within each consecutive time period are used as input data to form a scatter plot. Then, discrete points in the scatter plot that are outside the preset data range are captured, and the statistical values ​​corresponding to the discrete points are outlier statistical values.

[0096] It should be noted that the above visualization process can be implemented using software such as Tableau. Tableau is a business intelligence tool for desktop systems. For an example, please refer to... Figure 4 The graph shows a scatter plot constructed based on the maximum vibration acceleration during wind yaw from January to September of XXXX. Figure 4 In the data, the preset data range for vibration acceleration is [-0.13, +0.13]. The data points at positions ① and ② are the abnormal data that are outside the preset range. The wind turbine corresponding to this abnormal data is the wind turbine with the final acquired yaw vibration anomaly.

[0097] It can also obtain the unit identification of wind turbine generators with abnormal yaw vibration, and then send early warning information including abnormal data and unit identification back to the operation and maintenance terminal.

[0098] Still with Figure 4 Taking the abnormal data points at locations ① and ② as examples, these abnormal data points all correspond to wind turbine generator sets #29. "29#" is the generator set identifier, and wind turbine generator set #29 is the wind turbine generator set with abnormal yaw vibration. By sending early warning information to the operation and maintenance terminal, operation and maintenance personnel can be promptly notified to conduct abnormality inspections and maintenance, and abnormalities of the generator sets can be detected as early as possible.

[0099] In the above embodiments, millions of operational data points from the entire field over a historical period were extracted and their dimensionality reduced. The statistical values ​​of each wind turbine during the yaw process were displayed, which can quickly assess the operation of the wind turbines and identify abnormal data and corresponding abnormal wind turbines.

[0100] Further, please see Figure 5 and Figure 6 You can also select Figure 4 The file on the xy-direction over-limit fault during the yaw of the No. 29 wind turbine generator set, along with the time sequence diagram of the yaw position and nacelle acceleration, can further confirm that the No. 29 wind turbine generator set did indeed have an over-limit acceleration during the yaw, thus helping to further clarify the cause of the abnormal yaw vibration.

[0101] In the above solution, abnormal data is identified through a visual interface, and further analysis reveals the cause of the anomaly. This allows for timely feedback to maintenance personnel for correction. This significantly reduces the workload of maintenance staff and avoids load overruns and system safety risks caused by abnormal wind-induced yaw vibrations. This ensures operational safety and reliability.

[0102] Please refer to Figure 7 , Figure 7 A schematic diagram of the anomaly detection device for the wind turbine generator set of this application is shown. Figure 7 The device includes:

[0103] The acquisition module 710 is used to acquire the operating data of each wind turbine generator in the wind farm during a historical time period.

[0104] The processing module 720 is used to process the operating data to obtain multiple operating datasets. Each operating dataset includes the operating data of each wind turbine generator set within a continuous time period, during which each wind turbine generator set is in a yaw state.

[0105] The filtering module 730 is used to filter out the target operating dataset from multiple operating datasets, which shows each wind turbine in a yaw state.

[0106] The search module 740 is used to find abnormal data that deviates from the preset data range from the target running dataset;

[0107] The determination module 750 is used to identify the wind turbine generators corresponding to abnormal data as wind turbine generators with abnormal yaw vibration.

[0108] In one optional embodiment, the operational data includes yaw marker, unmooring marker, and yaw residual pressure; the filtering module includes:

[0109] The first acquisition unit is used to acquire the yaw flag, unmooring flag, and yaw residual pressure of each wind turbine in each running dataset;

[0110] The determination unit is used to determine the continuous yaw duration of each wind turbine in each continuous time period based on the yaw flag position of each wind turbine.

[0111] The selection unit is used to select an operating dataset that meets the first preset condition for continuous yaw duration, the second preset condition for untethering flag, and the third preset condition for yaw residual pressure as the target operating dataset. The first preset condition is that within the continuous time period corresponding to the operating dataset, the continuous yaw duration of each wind turbine is greater than or equal to the preset duration. The second preset condition is that each untethering flag in the operating dataset indicates that each wind turbine is in a non-untethered state. The third preset condition is that each yaw residual pressure in the operating dataset is less than the preset residual pressure threshold.

[0112] In another alternative embodiment, the acquisition module includes:

[0113] The second acquisition unit is used to acquire SCADA data of each wind turbine generator set within a historical time period.

[0114] The rejection unit is used to reject SCADA data whose operating status is maintenance, and obtain the operating data.

[0115] In another alternative embodiment, the operating data includes vibration acceleration and yaw residual pressure; the lookup module may include:

[0116] The calculation unit is used to calculate the statistical values ​​of the vibration acceleration of each wind turbine in the target operating dataset over a continuous time period. The statistical values ​​include at least one of the maximum value, minimum value, and mean value.

[0117] The search unit is used to find abnormal statistical values ​​from the statistical values ​​corresponding to vibration acceleration that are outside the preset data range. The operating data to which the abnormal statistical values ​​belong are abnormal data.

[0118] In yet another alternative embodiment, the search unit may include:

[0119] The first setting sub-unit, in a coordinate system with time as the first coordinate axis and vibration acceleration magnitude as the second coordinate axis, uses the statistical values ​​corresponding to the vibration acceleration of the wind turbine generator in each consecutive time period as input data to form a scatter plot;

[0120] The capture sub-unit is used to capture discrete points in the scatter plot that are outside the preset data range. The statistical values ​​corresponding to the discrete points are abnormal statistical values.

[0121] In yet another alternative embodiment, the device may further include:

[0122] The identifier acquisition module is used to acquire the unit identifier of wind turbine generator sets with abnormal yaw vibration.

[0123] The sending module is used to send early warning information, including abnormal data and unit identification, to the operation and maintenance terminal.

[0124] Figure 8A schematic diagram of the hardware structure of the anomaly detection device for wind turbine generator sets provided in an embodiment of this application is shown.

[0125] The anomaly detection device for the wind turbine generator set may include a processor 801 and a memory 802 storing computer program instructions.

[0126] Specifically, the processor 801 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0127] Memory 802 may include mass storage for data or instructions. For example, and not limitingly, memory 802 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 802 may include removable or non-removable (or fixed) media. Where appropriate, memory 802 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 802 is non-volatile solid-state memory.

[0128] In certain embodiments, the memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described in the above-disclosed method for detecting anomalies in wind turbine generators.

[0129] The processor 801 reads and executes computer program instructions stored in the memory 802 to implement any of the abnormal detection methods for wind turbine generator sets in the above embodiments.

[0130] In one example, the anomaly detection device for the wind turbine generator may also include a communication interface 803 and a bus 810. Wherein, as... Figure 8 As shown, the processor 801, memory 802, and communication interface 803 are connected through bus 810 and complete communication with each other.

[0131] The communication interface 803 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0132] Bus 810 includes hardware, software, or both, that couples components of an anomaly detection device for a wind turbine generator set together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 810 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0133] The anomaly detection device for this wind turbine generator set can execute the anomaly detection method for the wind turbine generator set in the embodiments of this application, thereby achieving a combination of Figures 1 to 7 The method and apparatus for detecting anomalies in wind turbine generator sets are described.

[0134] Furthermore, in conjunction with the anomaly detection method for wind turbine generator sets in the above embodiments, this application embodiment can provide a computer storage medium for implementation. This computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the anomaly detection methods for wind turbine generator sets in the above embodiments.

[0135] In addition, this application also provides a computer program product, including a computer program, which, when executed by a processor, can implement the steps and corresponding content of the aforementioned method embodiments.

[0136] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0137] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0138] The above are merely specific embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for detecting anomalies in a wind turbine generator set, characterized in that, include: Obtain the operating data of each wind turbine generator in the wind farm within a historical time period; The operational data is processed to obtain multiple operational datasets. Each operational dataset includes the operational data of each wind turbine generator set within a continuous time period, during which each wind turbine generator set is in a yaw state. From the multiple operational datasets, the target operational datasets in which each wind turbine is in a yaw state are selected; Identify abnormal data that deviates from the preset data range from the target running dataset; The wind turbine generator corresponding to the abnormal data is identified as a wind turbine generator generator with abnormal yaw vibration.

2. The method according to claim 1, characterized in that, The operational data includes yaw flag, unmooring flag, and yaw residual pressure; the step of selecting the target operational dataset from multiple operational datasets to determine the wind turbine's yaw state includes: Obtain the yaw flag, unmooring flag, and yaw residual pressure of each wind turbine in each of the said running datasets; Based on the yaw flag position of each wind turbine generator set, determine the continuous yaw duration of each wind turbine generator set within each continuous time period. The target operating dataset is selected based on the following conditions: the continuous yaw duration meets the first preset condition, the unmooring flag meets the second preset condition, and the yaw residual pressure meets the third preset condition. The first preset condition is that within the continuous time period corresponding to the operating dataset, the continuous yaw duration of each wind turbine is greater than or equal to a preset duration. The second preset condition is that each unmooring flag in the operating dataset indicates that each wind turbine is in a non-unmoored state. The third preset condition is that each yaw residual pressure in the operating dataset is less than a preset residual pressure threshold.

3. The method according to claim 1, characterized in that, The acquisition of operational data for each wind turbine generator unit in the wind farm during a historical time period includes: Obtain SCADA data for each wind turbine generator set within a historical time period; The SCADA data with the unit operating status set to maintenance is removed to obtain the operating data.

4. The method according to claim 1, characterized in that, The operational data includes vibration acceleration; the step of identifying abnormal data deviating from the preset data range from the target operational dataset includes: Calculate the statistical value of the vibration acceleration of each wind turbine generator in the target operating dataset during the continuous time period, wherein the statistical value includes at least one of the maximum value, minimum value and mean value; From the statistical values ​​corresponding to the vibration acceleration, identify abnormal statistical values ​​whose values ​​are outside the preset data range. The operational data to which these abnormal statistical values ​​belong are considered abnormal data.

5. The method according to claim 4, characterized in that, The step of identifying abnormal statistical values ​​from the statistical values ​​corresponding to the vibration acceleration that are outside the preset data range includes: In a coordinate system with time as the first coordinate axis and vibration acceleration magnitude as the second coordinate axis, the statistical values ​​corresponding to the vibration acceleration of the wind turbine generator in each consecutive time period are used as input data to form a scatter plot; The system captures discrete points in the scatter plot that are outside the preset data range, and the statistical values ​​corresponding to these discrete points are abnormal statistical values.

6. The method according to any one of claims 1 to 5, characterized in that, After identifying the wind turbine corresponding to the abnormal data as a wind turbine with abnormal yaw vibration, the method further includes: Obtain the unit identifier of the wind turbine generator set with abnormal yaw vibration; The system sends a warning message, including the abnormal data and the unit identifier, to the operation and maintenance terminal.

7. A device for detecting abnormal yaw vibration of a wind turbine generator set, characterized in that, The device includes: The acquisition module is used to acquire the operating data of each wind turbine generator in the wind farm within a historical time period; The processing module is used to process the operating data to obtain multiple operating datasets. Each operating dataset includes the operating data of each wind turbine generator set within a continuous time period, during which each wind turbine generator set is in a yaw state. The filtering module is used to filter out the target operating datasets from the multiple operating datasets, where each of the wind turbine generators is in a yaw state. The search module is used to find abnormal data that deviates from the preset data range from the target running dataset; The determination module is used to identify the wind turbine generator corresponding to the abnormal data as a wind turbine generator with abnormal yaw vibration.

8. A device for detecting abnormal yaw vibration of a wind turbine generator set, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the yaw vibration anomaly detection method for wind turbine generator sets as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the yaw vibration anomaly detection method for wind turbine generator sets as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product stores computer program instructions, which, when executed by a processor, implement the yaw vibration anomaly detection method for wind turbine generator sets as described in any one of claims 1 to 6.

Images