Wind power processing method, device, equipment and storage medium
By acquiring wind condition data from wind turbines and using a pre-trained wind power processing model to handle abnormal power, the problem of inaccurate power data caused by equipment malfunctions and operating conditions of wind turbines was solved, thus improving the reliability of power data.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-10-25
- Publication Date
- 2026-05-15
AI Technical Summary
In existing technologies, the processing of abnormal operating power data caused by factors such as equipment malfunctions and abnormal operating conditions of wind turbine units is not effective, which affects the operation of wind farms and the development of the wind power industry.
By acquiring wind condition data of wind turbines within a target time period, abnormal power processing is performed using a pre-trained wind power processing model. Long short-term memory neural networks or other neural network models are used for multi-point to single-point mapping to mine abnormal power data within the optimal time period and restore and process the abnormal power data.
It improves the reliability of power data, resolves the impact of equipment malfunctions and abnormal operating conditions on power data, and ensures the effectiveness of abnormal power handling.
Smart Images
Figure CN115936919B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of power prediction technology, and in particular to a wind power processing method, apparatus, equipment and storage medium. Background Technology
[0002] Vigorously developing new energy sources is an inevitable choice to address the global climate / energy crisis and achieve dual-carbon goals. Wind power, as an important component of new energy, has the advantages of being clean, renewable, and having large reserves, and is an important part of building a new energy power system.
[0003] With the increasing installed capacity of wind power, the demand for technologies such as wind turbine performance evaluation, power prediction, and wind farm modeling based on grid connection stability is becoming increasingly prominent. The theoretical power of wind turbines serves as the cornerstone of these tasks, and their reliability directly impacts their effectiveness. However, due to factors such as equipment malfunctions and abnormal operating conditions of wind turbines, including sensor failures, power curtailment, and turbine malfunctions, a large amount of abnormal operating power can occur. Therefore, proposing a reliable wind power processing method for handling abnormal power is of great significance to wind farm operation and the development of the wind power industry. Summary of the Invention
[0004] To address the aforementioned technical problems, this disclosure provides a wind power processing method, apparatus, equipment, and storage medium.
[0005] In a first aspect, this disclosure provides a method for processing wind power, the method comprising:
[0006] Obtain wind condition data to be processed for wind turbines within a target time period, wherein the target time period is the optimal time period for wind power processing of the wind turbines, and the original power data corresponding to the wind condition data to be processed is abnormal power data.
[0007] An abnormal power processing model is used to process the wind condition data to be processed to obtain target power data, wherein the target power data is the normal operating power corresponding to the wind condition data to be processed.
[0008] Secondly, this disclosure provides a wind power processing device, which includes:
[0009] The wind condition data acquisition module is used to acquire wind condition data to be processed from the wind turbine within a target time period. The target time period is the optimal time period for processing wind power from the wind turbine. The original power data corresponding to the wind condition data to be processed is abnormal power data.
[0010] An abnormal power processing module is used to process the wind condition data to be processed using a pre-trained wind power processing model to obtain target power data, wherein the target power data is the normal operating power corresponding to the wind condition data to be processed.
[0011] Thirdly, embodiments of this disclosure also provide an electronic device, the device comprising:
[0012] One or more processors;
[0013] Storage device for storing one or more programs.
[0014] When one or more programs are executed by one or more processors, the one or more processors implement the methods provided in the first aspect.
[0015] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method provided in the first aspect.
[0016] The technical solution provided in this disclosure has the following advantages compared with the prior art:
[0017] This disclosure discloses a wind power processing method, apparatus, device, and storage medium. The method includes: acquiring wind condition data to be processed from a wind turbine within a target time period, wherein the target time period is the optimal time period for wind power processing of the wind turbine, and the original power data corresponding to the wind condition data to be processed is abnormal power data; and using a pre-trained wind power processing model to process the abnormal power data to be processed to obtain target power data, wherein the target power data is the normal operating power corresponding to the wind condition data to be processed. This allows for the identification of abnormal power data within the optimal time period, and, based on a multi-point to single-point mapping method, the abnormal power data can be restored and processed. This solves the problem of the impact of wind turbine equipment malfunctions and operating condition anomalies on power data, ensuring the effectiveness of abnormal power processing and ultimately improving the reliability of power data. Attached Figure Description
[0018] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.
[0019] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0020] Figure 1 A schematic flowchart of a wind power processing method provided in an embodiment of this disclosure;
[0021] Figure 2 A schematic flowchart of another wind power processing method provided in this embodiment of the present disclosure;
[0022] Figure 3 A logical schematic diagram of a wind power processing method provided in an embodiment of this disclosure;
[0023] Figure 4 This is a schematic diagram of the structure of a wind power processing device provided in an embodiment of the present disclosure;
[0024] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation
[0025] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.
[0026] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.
[0027] To address abnormal power output, traditional theoretical power restoration methods are employed. These methods typically use historical data to establish a mapping model between wind conditions and theoretical power generation at individual time points, thereby obtaining the theoretical power generation corresponding to the wind conditions at each time point. However, due to variations in wind turbine rotational inertia and wind turbine operating efficiency, the relationship between wind conditions and theoretical power generation is not strictly cubic. This leads to significant errors in theoretical power restoration methods based on individual time points, resulting in poor reliability of the restored theoretical power output.
[0028] To address the aforementioned issues, this disclosure provides a wind power processing method, apparatus, device, and storage medium.
[0029] The following is combined with Figure 1 The wind power processing method provided in the embodiments of this disclosure will be described. In the embodiments of this disclosure, the wind power processing method can be executed by an electronic device, which can be a client device controlling the operation of a wind turbine.
[0030] Figure 1A schematic flowchart of a wind power processing method provided in an embodiment of this disclosure is shown.
[0031] like Figure 1 As shown, the wind power processing method may include the following steps.
[0032] S110. Obtain the wind condition data to be processed for the wind turbine within the target time length, wherein the target time length is the optimal time length for processing wind power for the wind turbine, and the original power data corresponding to the wind condition data to be processed is abnormal power data.
[0033] In practical applications, during wind turbine operation, the wind turbine generates power based on wind condition data at various times. Each moment of wind condition data corresponds to a specific power output. However, due to factors such as the rotational inertia of the wind turbine rotor and variations in the turbine's operating efficiency, abnormal power data may occur at certain times. To ensure effective processing of this abnormal power data, the electronic equipment needs to pre-determine a target time length that optimizes wind power processing and acquire the wind condition data to be processed within that target time length. Furthermore, the power output corresponding to this wind condition data must be abnormal power data.
[0034] The target time length is the optimal time length, which is the period during which wind power processing achieves the best results. Specifically, the target time length can be determined based on the wind power processing results over historical time periods.
[0035] The wind condition data to be processed refers to the wind condition data corresponding to the abnormal wind power. The wind condition data to be processed can be all the wind condition data within the target time period, or it can be a partial wind condition data within the target time period.
[0036] Optionally, the wind data to be processed may include one or more combinations of the following: wind speed data, wind direction data, and wind force data.
[0037] Abnormal power data refers to abnormal power caused by factors such as the rotational inertia of the wind turbine and changes in the operating efficiency of the wind turbine.
[0038] S120. Use a pre-trained wind power processing model to process the abnormal power of the wind condition data to be processed, and obtain the target power data, where the target power data is the normal operating power corresponding to the wind condition data to be processed.
[0039] In practical applications, electronic devices can use historical wind data and the corresponding normal operating power to train a wind power processing model. This allows the pre-trained wind power processing model to process abnormal power data and obtain the target power data. This enables the restoration and processing of abnormal power data based on a multi-point to single-point mapping method.
[0040] In this case, the wind power processing model can be a long short-term memory neural network, containing a multi-point to single-point mapping relationship between wind condition data and normal operating power. In other cases, the wind power processing model can also be other models such as convolutional neural networks or deep learning neural networks.
[0041] Among them, the normal operating power is the restored power corresponding to the abnormal power, that is, the operating power after eliminating the interference of factors such as the rotational inertia of the wind turbine and the operating efficiency of the wind turbine.
[0042] This disclosure discloses a wind power processing method, which includes: acquiring wind condition data of a wind turbine generator within a target time period, wherein the target time period is the optimal time for wind power processing of the wind turbine generator, and the original power data corresponding to the wind condition data to be processed is abnormal power data; and using a pre-trained wind power processing model to process the abnormal power data of the wind condition data to be processed to obtain target power data, wherein the target power data is the normal operating power corresponding to the wind condition data to be processed. This method can identify abnormal power data within the optimal time period and, based on a multi-point to single-point mapping method, restore and process the abnormal power data, thus solving the problem of the impact of wind turbine generator equipment malfunctions and operating condition anomalies on power data, ensuring the effectiveness of abnormal power processing, and ultimately improving the reliability of power data.
[0043] In another embodiment of this disclosure, the power range can be determined based on the original power data, and the wind condition data corresponding to the abnormal power data outside the power range can be obtained from the original wind condition data based on the power range, as the wind condition data to be processed.
[0044] Optionally, in this embodiment of the disclosure, S110 may specifically include the following steps:
[0045] S1101. Obtain the original wind condition data of the wind turbine within the target time period and the original power data corresponding to the original wind condition data.
[0046] S1102. Divide the raw wind data into multiple raw data intervals, and determine the power of two quantiles from the raw power data corresponding to each raw data interval.
[0047] S1103. Based on the power of the two quantiles corresponding to each original data interval, obtain the wind condition data to be processed from the original wind condition data.
[0048] The raw wind data consists of all wind data within the target time period, and the raw power data consists of all power generation within the target time period.
[0049] Here, the raw data interval refers to a continuous data interval of raw wind condition data. Specifically, the raw wind condition data can be divided into multiple raw data intervals based on the sampling frequency or a custom frequency. Each raw data interval contains the same raw wind condition data, and each raw wind condition data corresponds to one raw power data. One raw power data corresponds to one or more raw wind condition data, such that the number of raw power data corresponding to each raw data interval can be equal to or unequal to the number of raw wind condition data.
[0050] The quantile power is determined by sorting the raw power data corresponding to each raw data interval according to a preset quantile method. Optionally, the sorting order of the raw power data corresponding to each raw data interval can be ascending or descending.
[0051] In some embodiments, if the preset quantile method is the quartile method, then the power of the two quantile points is the power of the first quantile point and the power of the third quantile point, or the power of the two quantile points is the power of the second quantile point and the power of the fourth quantile point.
[0052] In other embodiments, if the preset quantile method is the quartile method, then the power of the two quantile points is the power of the third quantile point and the power of the fifth quantile point, or the power of the two quantile points is the power of the first quantile point and the power of the sixth quantile point.
[0053] For S1102, the original data interval is a quartile interval, and the power of the two quartiles includes the power of the first quartile and the power of the third quartile. Accordingly, S1103 may specifically include the following steps:
[0054] S11031. Calculate the power range of each original data interval based on the power of the first quantile and the power of the third quantile.
[0055] S11032. Take the original wind condition data corresponding to the original power data that is outside the power range of each original data interval as the wind condition data to be processed.
[0056] Specifically, for S11031, the difference between the power at the first quantile and the power at the third quantile is calculated; based on this difference, the power at the first quantile, and a preset ratio, the lower limit of the power corresponding to each quartile interval is calculated; based on this difference, the power at the third quantile, and a preset ratio, the upper limit of the power corresponding to each quartile interval is calculated; the range formed by the lower limit of the power and the upper limit of the power is taken as the power range of each original data interval.
[0057] For example, the power at the first quantile and the power at the third quantile are represented by Q1 and Q3, respectively, the preset ratio is represented by k, and the difference between the power at the first quantile and the power at the third quantile in each quartile interval is represented by I.QR =Q3-Q1, then the lower power limit corresponding to each quartile interval is expressed as F1=Q1-k*I QR The upper limit of power corresponding to each quartile is represented by F. u =Q3+k*I QR The power range of each original data interval is then expressed as [Q1-k*I]. QR Q3+k*I QR ]
[0058] Specifically, for S11032, firstly, abnormal power data outside the power range is obtained from the original power data corresponding to each quartile interval, and then the original wind condition data corresponding to the abnormal power data is used as the wind condition data to be processed.
[0059] For example, obtain the data located in [Q1-k*I] from the raw power data corresponding to each quartile interval. QR Q3+k*I QR Abnormal power data outside of [Q1-k*I], and will be located in [Q1-k*I] QR Q3+k*I QR The original wind condition data corresponding to the abnormal power data outside the area is used as the wind condition data to be processed.
[0060] For S1102, the original data interval is a quartile interval, and the two quartile powers include the third quartile power and the fifth quartile power of the current quartile interval. Accordingly, S1103 may specifically include the following steps:
[0061] S11033. Calculate the power range of each original data interval based on the power of the third quantile and the power of the fifth quantile.
[0062] S11034. Take the original wind condition data corresponding to the original power data that is outside the power range of each original data interval as the wind condition data to be processed.
[0063] It should be noted that the specific implementation of S11033 to S11034 is similar to that of S11031 to S11032, and will not be elaborated here.
[0064] Therefore, the power range is determined based on the original power data, and abnormal power data is obtained from the original power data based on the power range. The original wind condition data corresponding to the abnormal power data is used as the wind condition data to be processed. Thus, the wind condition data to be processed that needs to be restored based on the actual power is determined, which improves the reliability of the method for determining the wind condition data to be processed.
[0065] In another embodiment of this disclosure, a wind power processing model is trained using sample pairs and the model is tested. The target time length is determined based on the model test results.
[0066] Figure 2 A schematic flowchart of another wind power processing method provided in an embodiment of this disclosure is shown.
[0067] like Figure 2 As shown, the wind power processing method may include the following steps.
[0068] S210: Obtain historical wind data and historical power data for multiple time periods.
[0069] In practical applications, when it is necessary to train the wind power processing model, wind condition data and power data are collected in historical time periods such as the past four months or the past six months. The historical time period can be divided into multiple time lengths according to the preset sampling frequency or a custom frequency, so as to obtain historical wind condition data and historical power data within multiple time lengths.
[0070] Optionally, historical wind data may include one or more combinations of the following: wind speed data, wind direction data, and wind force data.
[0071] S220: Obtain training sample pairs and test sample pairs from historical wind data and historical power data.
[0072] In practical applications, historical wind data and historical power data can be cleaned to obtain normal data pairs, and then the normal data pairs can be divided into training sample pairs and test sample pairs.
[0073] Specifically, S220 may include the following steps:
[0074] S2201. Divide the historical wind data into multiple historical data intervals, and determine the power of two quantiles from the historical power data corresponding to each historical data interval.
[0075] S2202. Based on the power of the two quantiles corresponding to each historical data interval, historical normal wind data is obtained from historical wind data, and historical normal power data is obtained from historical power data. The historical normal wind data and historical normal power data constitute a sample pair for the wind power processing model.
[0076] S2203. According to the preset ratio, the sample pairs are divided into training sample pairs and test sample pairs.
[0077] In some embodiments, if the historical data interval is a quartile interval, then the two quantile powers are the first quantile power and the third quantile power, or the two quantile powers are the second quantile power and the fourth quantile power.
[0078] In other embodiments, if the historical data interval is a quartile interval, then the power of the two quartiles is the power of the third quartile and the power of the fifth quartile, or the power of the two quartiles is the power of the first quartile and the power of the sixth quartile.
[0079] For S2201, the historical data interval is a quartile interval, and the two quartile powers include the first quartile power and the third quartile power of the quartile interval. Accordingly, S2202 may specifically include the following steps:
[0080] S22021. Calculate the power range of each historical data interval based on the power of the first quantile and the power of the third quantile.
[0081] S22022. Historical power data within the power range of each historical data interval shall be used as historical normal power data, and historical wind data corresponding to historical normal power data shall be used as historical normal wind data.
[0082] Specifically, for S22021, the difference between the power at the first quantile and the power at the third quantile is calculated; based on this difference, the power at the first quantile, and a preset ratio, the lower limit of power corresponding to each quartile interval is calculated; based on this difference, the power at the third quantile, and a preset ratio, the upper limit of power corresponding to each quartile interval is calculated; and the range formed by the lower limit of power to the upper limit of power is taken as the power range of each historical data interval.
[0083] Specifically, for S22022, firstly, historical normal power data within the power range is obtained from the historical power data corresponding to each quartile interval, and then the historical wind condition data corresponding to the historical normal power data is used as the historical normal wind condition data.
[0084] For S2201, the historical data interval is the quartile interval, and the two quantile powers include the third and fifth quantile powers of the current quantile interval. Accordingly, S2202 may specifically include the following steps:
[0085] S22023. Calculate the power range of each historical data interval based on the power of the third quantile and the power of the fifth quantile.
[0086] S22024. Historical power data within the power range of each historical data interval shall be used as historical normal power data, and historical wind data corresponding to historical normal power data shall be used as historical normal wind data.
[0087] It should be noted that the specific implementation of S22023 to S22024 is similar to that of S22021 to S22022, and will not be elaborated here.
[0088] Specifically, for S2203, the preset ratio can be determined based on the model's accuracy or it can be a fixed ratio predetermined manually. Therefore, sample pairs with the first preset ratio can be used as training sample pairs, and sample pairs with the second preset ratio can be used as test sample pairs.
[0089] Optionally, the preset ratio can be 70% or 80%.
[0090] S230. Use the normal wind condition data and normal power data in the training sample pair to train the model and obtain the wind power processing model.
[0091] Specifically, normal wind data from the training sample pairs can be input into a preset neural network to output estimated power data. The preset neural network can then be iteratively adjusted using the estimated power data and normal power data until the number of iterations, learning rate, and time step of the preset neural network reach preset values, thus obtaining a wind power processing model.
[0092] S240. Calculate the model evaluation index of the wind power processing model using the normal wind condition data and normal power data in the test sample pair, and take the time length during which the model evaluation index is the optimal evaluation index as the target time length.
[0093] Specifically, normal wind data for multiple time periods from the test sample pair are input into the wind power processing model, which outputs test power data. Based on the test power data and normal power data, model evaluation indicators corresponding to multiple time periods are calculated, and the time period for which the model evaluation indicator is the optimal evaluation indicator is taken as the target time period.
[0094] The model evaluation index refers to the assessment data used to select the time period that best reflects the wind power treatment effect. Optionally, the model evaluation index can be an index such as the root mean square error (RMSE).
[0095] Therefore, the model is trained based on training sample pairs, and after the model training is completed, the model evaluation index of the wind power processing model is calculated using test sample pairs. The target time length is then determined based on the model evaluation index, which improves the reliability of the target time length determination method.
[0096] S250. Obtain the wind condition data to be processed for the wind turbine within the target time length, where the target time length is the optimal time length for processing the wind power of the wind turbine, and the original power data corresponding to the wind condition data to be processed is abnormal power data.
[0097] S260. Use a pre-trained wind power processing model to process the abnormal power of the wind condition data to be processed, and obtain the target power data, where the target power data is the normal operating power corresponding to the wind condition data to be processed.
[0098] S250 to S260 are similar to S110 to S120, and will not be described in detail here.
[0099] In another embodiment of this disclosure, the wind power processing logic is described in its entirety.
[0100] Figure 3 A logical schematic diagram of a wind power processing method provided in an embodiment of this disclosure is shown.
[0101] like Figure 3 As shown, the wind power processing method may include the following steps.
[0102] S310 collects wind condition data and power data.
[0103] S320. Perform data cleaning on wind condition data and power data to determine normal datasets and abnormal datasets. The normal dataset includes normal wind condition data and normal power data, while the abnormal dataset includes abnormal wind condition data and abnormal power data.
[0104] S330. Use the training samples in the normal dataset to train the model and obtain the wind power processing model. Use the test samples in the normal dataset to determine the optimal time length as the target time length.
[0105] S340. Input the wind condition data in the abnormal dataset into the wind power processing model to obtain the normal power data corresponding to the abnormal power data in the abnormal dataset.
[0106] This disclosure also provides a wind power processing device for implementing the above-described wind power processing method, which is described below in conjunction with... Figure 4 The following explanation is provided. In this embodiment of the disclosure, the wind power processing device can be configured at the client corresponding to the target wind and solar power station among multiple wind and solar power stations in the area to be predicted.
[0107] Figure 4 A schematic diagram of the structure of a wind power processing device provided in an embodiment of this disclosure is shown.
[0108] like Figure 4 As shown, the wind power processing device 400 may include:
[0109] The wind condition data acquisition module 401 is used to acquire wind condition data to be processed of the wind turbine within a target time length, wherein the target time length is the optimal time length for wind power processing of the wind turbine, and the original power data corresponding to the wind condition data to be processed is abnormal power data.
[0110] The abnormal power processing module 402 is used to process the wind condition data to be processed using a pre-trained wind power processing model to obtain target power data, wherein the target power data is the normal operating power corresponding to the wind condition data to be processed.
[0111] This disclosure discloses a wind power processing device that acquires wind condition data of a wind turbine generator within a target time period. The target time period is the optimal time for wind power processing of the wind turbine generator. The original power data corresponding to the wind condition data is abnormal power data. A pre-trained wind power processing model is used to process the abnormal power data to obtain target power data, which is the normal operating power corresponding to the wind condition data. This allows for the identification of abnormal power data within the optimal time period. Based on a multi-point to single-point mapping method, the abnormal power data is restored and processed. This addresses the impact of equipment malfunctions and abnormal operating conditions of the wind turbine generator on the power data, ensuring the effectiveness of abnormal power processing and ultimately improving the reliability of the power data.
[0112] In one optional embodiment of this disclosure, the wind condition data acquisition module 401 includes:
[0113] The data acquisition unit is used to acquire the original wind condition data of the wind turbine within the target time period and the original power data corresponding to the original wind condition data;
[0114] An interval division unit is used to divide the original wind condition data into multiple original data intervals and determine two quantile power from the original power data corresponding to each original data interval.
[0115] The wind condition data acquisition unit acquires the wind condition data to be processed from the original wind condition data based on the power of the two quantiles corresponding to each of the original data intervals.
[0116] In one optional embodiment of this disclosure, the original data interval is a quartile interval, and the two quartile power points include the first quartile power point and the third quartile power point of the quartile interval. Accordingly, the wind condition data acquisition unit to be processed is specifically used to calculate the power range of each of the original data intervals based on the first quartile power point and the third quartile power point; and to take the original wind condition data corresponding to the original power data located outside the power range of each of the original data intervals as the wind condition data to be processed.
[0117] In one optional embodiment of this disclosure, the device further includes:
[0118] The historical data acquisition module is used to acquire historical wind condition data and historical power data over multiple time periods;
[0119] The sample pair partitioning module is used to obtain training sample pairs and test sample pairs from the historical wind data and the historical power data.
[0120] The model training module is used to train the model using the normal wind condition data and normal power data in the training sample pair to obtain the wind power processing model.
[0121] The model testing module is used to calculate the model evaluation index of the wind power processing model using normal wind condition data and normal power data in the test sample pair, and to take the time length during which the model evaluation index is the optimal evaluation index as the target time length.
[0122] In one optional embodiment of this disclosure, the sample pair partitioning module includes:
[0123] The quantile power determination unit is used to divide the historical wind data into multiple historical data intervals and determine two quantile powers from the historical power data corresponding to each historical data interval.
[0124] The normal data acquisition unit is used to acquire historical normal wind condition data from the historical wind condition data and historical normal power data from the historical power data based on the two quantile power corresponding to each of the historical data intervals, wherein the historical normal wind condition data and the historical normal power data constitute a sample pair of the wind power processing model.
[0125] The sample pair partitioning unit is used to partition the sample pair into the training sample pair and the test sample pair according to a preset ratio.
[0126] In one optional embodiment of this disclosure, the historical data interval is a quartile interval, and the two quartile power values include the first quartile power and the third quartile power of the quartile interval. Correspondingly, the normal data acquisition unit is specifically used to calculate the power range of each of the historical data intervals based on the first quartile power and the third quartile power; take the historical power data located within the power range of each of the historical data intervals as the historical normal power data, and take the historical wind condition data corresponding to the historical normal power data as the historical normal wind condition data.
[0127] In one optional embodiment of this disclosure, the wind condition data to be processed includes one or more of the following combinations:
[0128] Wind speed data, wind direction data, wind force data.
[0129] It should be noted that, Figure 4 The wind power processing device 400 shown can perform Figures 1-3The various steps in the method embodiment shown are implemented. Figures 1-3 The processes and effects in the method embodiments shown are not described in detail here.
[0130] This disclosure also provides a wind power processing device for implementing the above-described wind power processing method, which is described below in conjunction with... Figure 5 The following explanation is provided. In this embodiment of the disclosure, the wind power processing device can be configured in an electronic device.
[0131] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure is shown.
[0132] like Figure 5 As shown, the electronic device may include a processor 501 and a memory 502 storing computer program instructions.
[0133] Specifically, the processor 501 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.
[0134] Memory 502 may include a large-capacity storage for information or instructions. For example, and not limitingly, memory 502 may include a hard disk drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 502 may include removable or non-removable (or fixed) media. Where appropriate, memory 502 may be internal or external to the integrated gateway device. In a particular embodiment, memory 502 is a non-volatile solid-state memory. In a particular embodiment, memory 502 includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (Electrically Programmable ROM, EPROM), an electrically erasable programmable PROM (EEPROM), an electrically alterable ROM (EAROM), or flash memory, or a combination of two or more of these.
[0135] The processor 501 reads and executes computer program instructions stored in the memory 502 to perform the steps of the wind power processing method provided in the embodiments of this disclosure.
[0136] In one example, the electronic device may also include a transceiver 503 and a bus 504. Wherein, as... Figure 5 As shown, the processor 501, memory 502 and transceiver 503 are connected via bus 504 and communicate with each other.
[0137] Bus 504 may include hardware, software, or both. For example, and not limitingly, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial 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 Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 504 may include one or more buses. Although specific buses are described and illustrated in the embodiments of this application, this application considers any suitable bus or interconnection.
[0138] The following are embodiments of a computer-readable storage medium provided in this disclosure. This computer-readable storage medium belongs to the same inventive concept as the wind power processing methods in the above embodiments. For details not described in detail in the embodiments of the computer-readable storage medium, please refer to the embodiments of the wind power processing methods described above.
[0139] This embodiment provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a wind power processing method.
[0140] Of course, the computer-executable instructions provided in the embodiments of this disclosure are not limited to the above-described method operations, but can also perform related operations in the wind power processing method provided in any embodiment of this disclosure.
[0141] Based on the above description of the implementation methods, those skilled in the art can clearly understand that this disclosure can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer cloud platform (which may be a personal computer, server, or network cloud platform, etc.) to execute the wind power processing methods provided in the various embodiments of this disclosure.
[0142] Note that the above description is merely a preferred embodiment and the technical principles employed in this disclosure. Those skilled in the art will understand that this disclosure is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this disclosure. Therefore, although this disclosure has been described in detail through the above embodiments, it is not limited to the above embodiments. Many other equivalent embodiments may be included without departing from the concept of this disclosure, and the scope of this disclosure is determined by the scope of the appended claims.
Claims
1. A method for processing wind power, characterized in that, include: The wind condition data to be processed of the wind turbine is obtained within a target time period. The target time period is the optimal time period for processing the wind power of the wind turbine. The original power data corresponding to the wind condition data to be processed is abnormal power data. Abnormal power data refers to abnormal power caused by changes in the rotational inertia of the wind turbine and the operating efficiency of the wind turbine. An abnormal power processing method is used to process the wind condition data to be processed using a pre-trained wind power processing model to obtain target power data. The target power data is the normal operating power corresponding to the wind condition data to be processed. The wind power processing model contains a multi-point to single-point mapping relationship between wind condition data and normal operating power. The normal operating power is the restored power corresponding to the abnormal power after excluding the interference of wind turbine rotational inertia and wind turbine operating efficiency factors.
2. The method according to claim 1, characterized in that, The acquisition of wind condition data to be processed for wind turbines within a target time period includes: Acquire the original wind condition data of the wind turbine within the target time period and the original power data corresponding to the original wind condition data; The original wind data is divided into multiple original data intervals, and two quantile power points are determined from the original power data corresponding to each original data interval. Based on the power of the two quantiles corresponding to each of the original data intervals, the wind condition data to be processed is obtained from the original wind condition data.
3. The method according to claim 2, characterized in that, The original data interval is a quartile interval, and the two quantile powers include the first quantile power and the third quantile power of the quartile interval. The step of obtaining the wind condition data to be processed from the original wind condition data based on the two quantile powers corresponding to each of the original data intervals includes: Calculate the power range of each of the original data intervals based on the power of the first quantile and the power of the third quantile; The original wind condition data corresponding to the original power data located outside the power range of each of the original data intervals is taken as the wind condition data to be processed.
4. The method according to claim 1, characterized in that, The method further includes: Acquire historical wind and power data over multiple time periods; Training sample pairs and test sample pairs are obtained from the historical wind data and the historical power data; The wind power processing model is obtained by training the model using the normal wind condition data and normal power data in the training sample pair. The model evaluation index of the wind power processing model is calculated using the normal wind condition data and normal power data in the test sample pair, and the length of time during which the model evaluation index is the optimal evaluation index is taken as the target time length.
5. The method according to claim 4, characterized in that, The step of obtaining training sample pairs and test sample pairs from the historical wind data and the historical power data includes: The historical wind data is divided into multiple historical data intervals, and two quantile power points are determined from the historical power data corresponding to each historical data interval. Based on the power of two quantiles corresponding to each of the historical data intervals, historical normal wind data is obtained from the historical wind data, and historical normal power data is obtained from the historical power data, wherein the historical normal wind data and the historical normal power data constitute a sample pair of the wind power processing model. According to a preset ratio, the sample pairs are divided into training sample pairs and test sample pairs.
6. The method according to claim 5, characterized in that, The historical data interval is a quartile interval, and the two quantile power points include the first quantile power and the third quantile power of the quartile interval. The step of obtaining historical normal wind condition data from the historical wind condition data based on the two quantile power points corresponding to each historical data interval, and obtaining historical normal power data from the historical power data, includes: Calculate the power range of each of the historical data intervals based on the power of the first quantile and the power of the third quantile; Historical power data within the power range of each of the historical data intervals are taken as historical normal power data, and historical wind data corresponding to the historical normal power data are taken as historical normal wind data.
7. The method according to any one of claims 1 to 6, characterized in that, The wind condition data to be processed includes one or more combinations of the following: Wind speed data, wind direction data, wind force data.
8. A wind power processing device, characterized in that, include: The wind condition data acquisition module is used to acquire wind condition data to be processed for the wind turbine within a target time period. The target time period is the optimal time period for processing wind power for the wind turbine. The original power data corresponding to the wind condition data to be processed is abnormal power data. Abnormal power data refers to abnormal power caused by changes in the rotational inertia of the wind rotor and the operating efficiency of the wind turbine. An abnormal power processing module is used to process the wind condition data to be processed using a pre-trained wind power processing model to obtain target power data. The target power data is the normal operating power corresponding to the wind condition data to be processed. The wind power processing model contains a multi-point to single-point mapping relationship between wind condition data and normal operating power. The normal operating power is the restored power corresponding to the abnormal power after excluding the interference of wind turbine rotational inertia and wind turbine operating efficiency factors.
9. An electronic device, characterized in that, include: processor; Memory, used to store executable instructions; The processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the method of any one of claims 1-7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The storage medium stores a computer program that, when executed by a processor, causes the processor to implement the method described in any one of claims 1-7.