Wind power abnormal data identification method and device, equipment and storage medium
By constructing wind speed power regression model and mathematical morphological algorithms, abnormal data in wind farm operation data are identified, and the problem of low identification accuracy in the existing technology is solved, and higher accuracy and completeness are achieved, and data quality is improved.
Patent Information
- Application Number
- CN202510233855.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-28
- Publication Date
- 2025-07-11
AI Technical Summary
The prior art has low accuracy in identifying abnormal data in wind power operation data, resulting in misjudgment and misjudgment, affecting the accuracy and application effect of subsequent wind power operation data prediction.
Based on the actual power and wind speed output from the wind farm, a wind speed power regression model is constructed, and abnormal data is identified through mathematical morphological algorithms, and abnormal data is carefully identified through regression analysis and digital image processing technology.
It improves the accuracy and completeness of abnormal data in wind farm operation data, improves data quality, and provides accurate data support for the stable operation of the power system.
Smart Images
Figure CN120296693A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of data processing, and in particular, to a method, device, equipment, and storage medium for identifying abnormal wind power data. Background Art
[0002] Currently, the rapid development of wind power poses a severe challenge to the safe and stable operation of the power system. Therefore, it is necessary to predict the future operation data of wind power based on the actual operation data of wind power to improve the predictability of wind power operation data.
[0003] However, in the actual operation scenario, there is a large amount of abnormal operation data in the wind power operation data (including the output power of the wind farm and the corresponding wind speed) collected by the Supervisory Control And Data Acquisition (SCADA) system deployed in the wind farm. If the abnormal operation data is directly used for subsequent calculations, it will affect the accuracy of subsequent wind power operation data prediction and the effect of wind power data-driven applications. Therefore, it is crucial to accurately identify the abnormal operation data in the wind power operation data.
[0004] The existing methods have a low accuracy in identifying abnormal operation data in wind power operation data, and are prone to cause a large number of data omissions and misjudgments. Therefore, there is an urgent need for a method that can accurately and effectively identify abnormal operation data in wind power operation data to improve data quality. Summary of the Invention
[0005] To solve the above technical problems, the present disclosure provides a method, device, equipment, and storage medium for identifying abnormal wind power data.
[0006] The first aspect of the present disclosure provides a method for identifying abnormal wind power data, including:
[0007] Construct a wind speed-power regression model based on the actual power output by the wind farm and the actual wind speed corresponding to each actual power;
[0008] Input each actual wind speed into the wind speed-power regression model, and predict the power corresponding to each actual wind speed based on the wind speed-power regression model to obtain the predicted power corresponding to each actual wind speed;
[0009] Based on the difference between each predicted power and the actual power corresponding to the predicted power, identify the abnormal power in the actual power as the first abnormal power, determine the power other than the first abnormal power in the actual power as the power to be measured, and determine the wind speed corresponding to the power to be measured as the wind speed to be measured;
[0010] Construct a target data set based on the measured wind speed and the measured power. The target data set consists of data points, and each data point includes a measured wind speed and a measured power corresponding to the measured wind speed.
[0011] Construct a first digital image corresponding to the target data set and an envelope line corresponding to the first digital image based on the mathematical morphology algorithm. Each data point in the target data set corresponds to an element coordinate in the first digital image.
[0012] Identify the abnormal power in the target data set as the second abnormal power based on the positional relationship between each element coordinate in the first digital image and the envelope line.
[0013] Determine the first abnormal power and the actual wind speed corresponding to the first abnormal power, as well as the second abnormal power and the actual wind speed corresponding to the second abnormal power as the abnormal data in the actual operation data of the wind farm.
[0014] The second aspect of the present disclosure provides a device for identifying abnormal wind power data, including:
[0015] A first construction module for constructing a wind speed-power regression model based on the actual power output by the wind farm and the actual wind speed corresponding to each actual power.
[0016] A prediction module for inputting each actual wind speed into the wind speed-power regression model and predicting the power corresponding to each actual wind speed based on the wind speed-power regression model to obtain the predicted power corresponding to each actual wind speed.
[0017] A first identification module for identifying the abnormal power in the actual power as the first abnormal power based on the difference between each predicted power and the actual power corresponding to the predicted power, determining the power other than the first abnormal power in the actual power as the measured power, and determining the wind speed corresponding to the measured power as the measured wind speed.
[0018] A second construction module for constructing a target data set based on the measured wind speed and the measured power. The target data set consists of data points, and each data point includes a measured wind speed and a measured power corresponding to the measured wind speed.
[0019] A third construction module for constructing a first digital image corresponding to the target data set and an envelope line corresponding to the first digital image based on the mathematical morphology algorithm. Each data point in the target data set corresponds to an element coordinate in the first digital image.
[0020] A second identification module for identifying the abnormal power in the target data set as the second abnormal power based on the positional relationship between each element coordinate in the first digital image and the envelope line.
[0021] A determination module, configured to determine the first abnormal power and the actual wind speed corresponding to the first abnormal power, and the second abnormal power and the actual wind speed corresponding to the second abnormal power as abnormal data in the actual operation data of the wind farm.
[0022] A third aspect of the present disclosure provides a computer device, including a memory and a processor. Wherein, a computer program is stored in the memory, and when the computer program is executed by the processor, the wind power abnormal data identification method of the first aspect can be implemented.
[0023] A fourth aspect of the present disclosure provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by the processor, the wind power abnormal data identification method of the first aspect can be implemented.
[0024] The technical solution provided by the present disclosure has the following advantages compared with the prior art:
[0025] The present disclosure constructs a wind speed-power regression model based on the actual power output by the wind farm and the actual wind speed corresponding to each actual power; inputs each actual wind speed into the wind speed-power regression model, predicts the power corresponding to each actual wind speed based on the wind speed-power regression model, and obtains the predicted power corresponding to each actual wind speed; based on the difference between each predicted power and the actual power corresponding to the predicted power, identifies the abnormal power in the actual power as the first abnormal power, determines the power other than the first abnormal power in the actual power as the power to be measured, and determines the wind speed corresponding to the power to be measured as the wind speed to be measured; constructs a target data set based on the wind speed to be measured and the power to be measured, the target data set is composed of data points, and each data point includes a wind speed to be measured and a power to be measured corresponding to the wind speed to be measured; constructs a first digital image corresponding to the target data set and an envelope line corresponding to the first digital image based on the mathematical morphology algorithm, and each data point in the target data set corresponds to an element coordinate in the first digital image; based on the positional relationship between each element coordinate in the first digital image and the envelope line, identifies the abnormal power in the target data set as the second abnormal power; determines the first abnormal power and the actual wind speed corresponding to the first abnormal power, and the second abnormal power and the actual wind speed corresponding to the second abnormal power as abnormal data in the actual operation data of the wind farm. The present disclosure can initially identify abnormal data in the actual operation data of the wind farm based on the regression model, and then further accurately identify other abnormal data based on the image processing technology of mathematical morphology. By comprehensively using regression analysis and mathematical morphology, it can comprehensively and accurately identify abnormal data in the actual operation data of the wind farm, effectively improve the accuracy and integrity of abnormal data identification, improve the data quality of the wind farm operation data, and provide accurate data support for the subsequent operation of the power system. Description of the Drawings
[0026] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with the present disclosure and, together with the specification, are used to explain the principles of the present disclosure.
[0027] To more clearly illustrate the technical solutions in the embodiments of the present disclosure or in the prior art, the following will briefly introduce the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.
[0028] Figure 1 is a flowchart of a method for identifying abnormal wind power data provided by an embodiment of the present disclosure;
[0029] Figure 2 is a flowchart of a method for generating an envelope provided by an embodiment of the present disclosure;
[0030] Figure 3 is a schematic structural diagram of a device for identifying abnormal wind power data provided by an embodiment of the present disclosure;
[0031] Figure 4 is a schematic structural diagram of a computer device provided by an embodiment of the present disclosure. Detailed implementation manners
[0032] In order to be able to more clearly understand the above-mentioned objects, features and advantages of the present disclosure, the following will further describe the solutions of the present disclosure. It should be noted that, without conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other.
[0033] Many specific details are set forth in the following description in order to fully understand the present disclosure, but the present disclosure can also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only a part of the embodiments of the present disclosure, rather than all the embodiments.
[0034] It should be understood that the various steps recorded in the method embodiments of the present disclosure can be executed in different orders and / or executed in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.
[0035] It should be noted that in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the said element.
[0036] It should be noted that the modifications of "one" and "multiple" mentioned in this disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless expressly stated otherwise in the context, it should be construed as "one or more".
[0037] The wind power anomaly data identification method provided by the embodiments of the present disclosure can be executed by a computer device, which can be understood as any device with processing and computing capabilities. Such a device may include, but is not limited to, mobile terminals such as smart phones, laptop computers, tablet computers (PADs), wearable devices, etc., and fixed electronic devices such as digital TVs, desktop computers, etc.
[0038] To better understand the inventive concept of the embodiments of the present disclosure, the technical solutions of the embodiments of the present disclosure will be described below in conjunction with exemplary embodiments.
[0039] Figure 1 is a flowchart of a wind power anomaly data identification method provided by the embodiments of the present disclosure, which can be executed by a computer device, as Figure 1 shown, the wind power anomaly data identification method provided in this embodiment includes the following steps:
[0040] Step 110: Construct a wind speed-power regression model based on the actual power output by the wind farm and the actual wind speed corresponding to each actual power.
[0041] In the embodiments of the present disclosure, the computer device can obtain the actual power output by the wind farm and the actual wind speed corresponding to each actual power from a Supervisory Control And Data Acquisition (SCADA) system deployed in the wind farm, and then construct a wind speed-power regression model based on the actual power output by the wind farm and the actual wind speed corresponding to each actual power.
[0042] Specifically, based on the actual power output of the wind farm and the actual wind speed corresponding to each actual power, a wind speed-power regression model can be constructed, which may include steps 1101-1105:
[0043] Step 1101: Obtain the actual operation dataset of the wind farm. The actual operation dataset includes n samples, and each sample includes an actual power output by the wind farm and the actual wind speed corresponding to this actual power. n is a positive integer.
[0044] Specifically, the actual operation dataset U of the wind farm can be expressed as:
[0045] U = {(v1, p1), …, (v i , p i ), (v n , p n )};
[0046] Among them, n is the number of samples in the actual operation dataset U, and n is a positive integer; v i is the actual wind speed of the i-th sample; p i is the actual power of the i-th sample.
[0047] Step 1102: Divide the samples in the actual operation dataset into k different sample layers according to the magnitude of the actual power in the actual operation dataset. k is a positive integer.
[0048] Step 1103: Based on the proportion of the number of samples included in each sample layer in the actual operation dataset, randomly sample the samples in each sample layer to obtain a training dataset.
[0049] Specifically, the training dataset U t can be expressed as:
[0050] U t = {(v'1, p'1), …, (v' i , p' i ), (v' m , p' m )};
[0051] Among them, m is the number of samples in the training dataset U t , and m is a positive integer; v i ’ is the actual wind speed of the i-th sample; p i ’ is the actual power of the i-th sample.
[0052] Step 1104: Perform a cubic polynomial expansion on each actual wind speed in the training dataset to obtain a target wind speed expansion matrix.
[0053] Specifically, the wind speed dataset V in the training dataset t=[v'1,…,v' i ,…,v' m perform a cubic polynomial expansion to expand each actual wind speed in the wind speed dataset V t into x i = [1, v i ', (v i ') 2 , (v i ') 3 . Thus, the wind speed dataset V t is converted into the target wind speed expansion matrix X:
[0054]
[0055] By performing a cubic polynomial expansion on the data, the drastic fluctuations of the data can be avoided, the stability of the data can be improved, and thus the accuracy of subsequent model training can be improved.
[0056] Step 1105: Use the target wind speed expansion matrix as the input of the regression model, and use the actual power in the training dataset as the output of the regression model to train the regression model to obtain a trained wind speed-power regression model.
[0057] For example, the computer device can initialize a linear regression model based on the Random Sample Consensus (RANSAC) algorithm. Use the obtained target wind speed expansion matrix X as the input of the linear regression model, and use the power dataset P t in the obtained training dataset U t = [p'1,…,p' i ,…,p' m as the output of the linear regression model to train the linear regression model to obtain a trained wind speed-power regression model.
[0058] The RANSAC algorithm in this embodiment can be implemented using the Sklearn toolbox in the Python environment.
[0059] Step 120: Input each actual wind speed into the wind speed-power regression model, and predict the power corresponding to each actual wind speed based on the wind speed-power regression model to obtain the predicted power corresponding to each actual wind speed.
[0060] For example, the computer device can perform the actual wind speed dataset V = [v1,…,v i ,…,v nPerform a cubic polynomial expansion on each actual wind speed in [], obtain the actual wind speed expansion matrix, input the actual wind speed expansion matrix into the wind speed-power regression model, and predict the power corresponding to each actual wind speed based on the wind speed-power regression model to obtain the predicted power corresponding to each actual wind speed.
[0061] Step 130: Based on the difference between each predicted power and the actual power corresponding to the predicted power, identify the abnormal power in the actual power as the first abnormal power, determine the power other than the first abnormal power in the actual power as the power to be measured, and determine the wind speed corresponding to the power to be measured as the wind speed to be measured.
[0062] Specifically, the computer device can calculate, for each predicted power, the difference between the predicted power and the actual power corresponding to the predicted power; when the difference between the predicted power and the actual power corresponding to the predicted power is within the preset error range, determine the actual power corresponding to the difference as the first abnormal power; when the difference between the predicted power and the actual power corresponding to the predicted power is outside the preset error range, determine the actual power corresponding to the difference as the power to be measured, and determine the wind speed corresponding to the power to be measured as the wind speed to be measured.
[0063] Among them, the preset error range can be set as needed and is not limited here. For example, by analyzing the error distribution of regression learning, a suitable preset error range t = {t0, t1} can be set, where t0 represents the lower boundary threshold and t1 represents the upper boundary threshold.
[0064] For example, the computer device can calculate the predicted power and the difference with the actual power dataset P = [p1,..., p i ,..., p n in the actual operation dataset U Then compare each difference e i in the difference E with the preset error range, determine the actual power corresponding to the difference within the preset error range as the first abnormal power, determine the first abnormal power and the actual wind speed corresponding to the first abnormal power as abnormal operation data, and denote the set composed of the abnormal operation data as U ab :
[0065] U ab = {(v i , p i ) | (v i , p i ) ∈ U, e i ≤ to or e i ≥ t1}.
[0066] Step 140: Construct a target data set based on the measured wind speed and the measured power. The target data set consists of data points, and each data point includes a measured wind speed and the measured power corresponding to the measured wind speed.
[0067] Specifically, the target data set can be constructed based on the measured wind speed and the measured power obtained above. The target data set consists of data points, and each data point includes a measured wind speed and the measured power corresponding to the measured wind speed. For example, the target data set U n can be expressed as:
[0068] U n = {(v i , p i ) | (v i , p i ) ∈ U, t0 < t i < t1}.
[0069] Step 150: Construct a first digital image corresponding to the target data set and an envelope line corresponding to the first digital image based on the mathematical morphology algorithm. Each data point in the target data set corresponds to an element coordinate in the first digital image.
[0070] Mathematical morphology is an image analysis discipline based on lattice theory and topology, and it is the basic theory of mathematical morphology image processing. It can convert data into images for analysis.
[0071] In the embodiments of the present disclosure, the computer device can construct a first digital image corresponding to the target data set and an envelope line corresponding to the first digital image based on the mathematical morphology algorithm. Each data point in the target data set corresponds to an element coordinate in the first digital image, that is, an element coordinate includes a measured wind speed and the measured power corresponding to the measured wind speed.
[0072] The elements in the digital image can be understood as the pixels of the digital image, and the element coordinates can represent the position of the element in the digital image.
[0073] Step 160: Identify the abnormal power in the target data set as the second abnormal power based on the positional relationship between each element coordinate in the first digital image and the envelope line.
[0074] In the embodiments of the present disclosure, the computer device can identify the abnormal power in the target data set as the second abnormal power based on the positional relationship between each element coordinate in the first digital image and the envelope line.
[0075] Specifically, for each element coordinate in the first digital image, the computer device can determine whether the element coordinate is within the envelope line corresponding to the first digital image;
[0076] If the coordinates of the element are within the envelope, the data points corresponding to the coordinates of the element can be determined as normal data points, and the power to be measured in the normal data points can be determined as the normal power.
[0077] If the coordinates of the element are not within the envelope, the data points corresponding to the coordinates of the element can be determined as abnormal operation data points, and the power to be measured in the abnormal operation data points can be determined as the second abnormal power.
[0078] Step 170: Determine the first abnormal power and the actual wind speed corresponding to the first abnormal power, and the second abnormal power and the actual wind speed corresponding to the second abnormal power as the abnormal data in the actual operation data of the wind farm.
[0079] Thus, the present disclosure can initially identify the abnormal data in the actual operation data of the wind farm based on the regression model, and then further finely identify other abnormal data based on the image processing technology of mathematical morphology. By comprehensively performing regression analysis and mathematical morphology, the abnormal data in the actual operation data of the wind farm can be accurately and comprehensively identified, the accuracy and integrity of the abnormal data identification can be effectively improved, the data quality of the operation data of the wind farm can be improved, and accurate data support can be provided for the subsequent operation of the power system.
[0080] In some embodiments of the present disclosure, for constructing the first digital image corresponding to the target data set and the envelope corresponding to the first digital image based on the mathematical morphology algorithm, the computer device may execute Figure 2 The flowchart of a method for generating an envelope provided, as Figure 2 shown, may include the following steps:
[0081] Step 210: Perform standardization processing on each data point in the target data set to obtain the standardized data set corresponding to the target data set.
[0082] Standardization processing refers to converting the data through a certain mathematical transformation into a form with zero mean (average value is 0) and unit variance (standard deviation is 1), or making the data fall into a specific interval (such as 0 to 1 or -1 to 1) to eliminate the differences in characteristic attributes such as the nature, dimension, and order of magnitude between different variables.
[0083] For example, the computer device may perform standardization processing on each data point in the target data set U Figure 1 obtained above n to obtain the standardized data set corresponding to the target data set
[0084] In some embodiments, the standard wind speed of the i-th sample in the standardized data set It can be expressed as:
[0085]
[0086] Among them, v min is the minimum wind speed in the target data set; v max is the maximum wind speed in the target data set;
[0087] Standardized dataset The standard power of the i-th sample in It can be expressed as:
[0088]
[0089] Among them, p min is the minimum power in the target dataset; p max is the maximum power in the target dataset.
[0090] Step 220: Perform image conversion on the standardized data set to obtain a first digital image.
[0091] In the embodiment of the present disclosure, after obtaining the standardized data set corresponding to the target data set, the computer device may perform image conversion on the standardized data set to obtain a first digital image.
[0092] In some embodiments, performing image conversion on the standardized data set to obtain the first digital image may include steps 2201-2205:
[0093] Step 2201: Map each standard data point in the standardized data set to a preset positive integer interval to obtain a first data set.
[0094] Specifically, the computer device can standardize the data set Each standard data point in is mapped to a preset positive integer interval to obtain the first data set The preset positive integer interval can be set as needed and is not limited here.
[0095] For example, the preset positive integer interval may be 1-q, where q is a positive integer greater than 1. The first wind speed of the i-th sample in It can be expressed as:
[0096]
[0097] Represents the first data set The standard wind speed of the i-th sample in;
[0098] First Dataset The first power of the i-th sample in It can be expressed as:
[0099]
[0100] Denote the standard power of the i-th sample in the first data set in the first data set.
[0101] Where, denote f(x) as the floor function, and the expression of f(x) can be:
[0102] [x] is the largest integer not greater than x.
[0103] Step 2202: Initialize a two-dimensional zero matrix based on a preset positive integer interval.
[0104] For example, the computer device can initialize a two-dimensional zero matrix M of size (q + 1) × (q + 1).
[0105] Step 2203: Map each data point in the first data set into the two-dimensional zero matrix to determine the target position of each data point in the first data set in the two-dimensional zero matrix.
[0106] For example, the computer device can each data point in the first data set
[0107] map into the two-dimensional zero matrix M to determine the target position of each data point in the first data set in the two-dimensional zero matrix.
[0108] Step 2204: Mark the value at each target position in the two-dimensional zero matrix as 1.
[0108] Step 2205: Use the first color pixel to represent the value 1 in the two-dimensional zero matrix, and use the second color pixel to represent the value 0 in the two-dimensional zero matrix, and perform pixel conversion on the two-dimensional zero matrix to obtain a first digital image, where the first color pixel and the second color pixel are different.
[0109] Specifically, the first color pixel and the second color pixel are different, and the first color pixel and the second color pixel can be set as needed, which is not limited here. For example, the first color pixel can be a black pixel, and the second color pixel can be a white pixel.
[0110] In the embodiments of the present disclosure, the computer device can use the first color pixel to represent the value 1 in the two-dimensional zero matrix, and use the second color pixel to represent the value 0 in the two-dimensional zero matrix, and perform pixel conversion on the two-dimensional zero matrix to obtain a first digital image A.
[0111] Step 230: Perform an erosion operation on the first digital image based on a preset structural element to obtain a second digital image.
[0112] Specifically, the preset structural element can be set as needed, which is not limited here. For example, the preset structural element can be a circular structural element R.
[0113] For example, the computer device may select a circular structure element R based on the mathematical morphology theory to perform an erosion operation on the first digital image A to obtain the second digital image B.
[0114] Step 240: Perform a dilation operation on the second digital image to obtain a third digital image.
[0115] Specifically, the computer device may perform a dilation operation on the second digital image B to obtain the third digital image C.
[0116] Step 250: Identify boundary pixels among non-zero pixels in the third digital image, and extract data points corresponding to each boundary pixel as boundary coordinates of the third digital image.
[0117] Specifically, the computer device can traverse each column of the third digital image C, identify boundary pixels among non-zero pixels in the third digital image C, where boundary pixels can be understood as pixels located at the boundary, and then extract the data point corresponding to each boundary pixel as the boundary coordinates of the third digital image.
[0118] Step 260: interpolate and smoothly connect the boundary coordinates of the third digital image to generate an envelope corresponding to the first digital image.
[0119] Specifically, the computer device may perform interpolation processing on the boundary coordinates of the third digital image and smoothly connect them to generate an envelope corresponding to the first digital image.
[0120] Therefore, the remaining data other than the initially identified abnormal data can be standardized and mapped into a two-dimensional matrix according to the rules. The matrix is further represented by a digital image, and the image is eroded and expanded based on mathematical morphology operations, and the boundary coordinates are extracted and the envelope is generated. In order to subsequently identify the abnormal data by judging whether the data point is within the envelope, the abnormal data in the actual operation data of the wind farm can be identified as a whole, which can effectively improve the accuracy and completeness of abnormal data identification, improve the data quality of wind farm operation data, and provide accurate data support for the subsequent operation of the power system.
[0121] In some embodiments of the present disclosure, after determining the abnormal data in the actual operation data of the wind farm, the computer device can remove the abnormal data in the actual operation data of the wind farm to obtain the target operation data of the wind farm. This can improve the data quality of the wind farm operation data and provide accurate data support for the subsequent operation of the power system.
[0122] Figure 3 This is a schematic structural diagram of a wind power anomaly data recognition device provided by an embodiment of the present disclosure. This device can be understood as the above-mentioned computer device or a partial functional module in the above-mentioned computer device. As Figure 3 shown, the wind power anomaly data recognition device 300 includes:
[0123] A first construction module 310, configured to construct a wind speed-power regression model based on the actual power output by the wind farm and the actual wind speed corresponding to each actual power;
[0124] A prediction module 320, configured to input each actual wind speed into the wind speed-power regression model, and predict the power corresponding to each actual wind speed based on the wind speed-power regression model to obtain the predicted power corresponding to each actual wind speed;
[0125] A first recognition module 330, configured to identify the abnormal power in the actual power as the first abnormal power based on the difference between each predicted power and the actual power corresponding to the predicted power, determine the power other than the first abnormal power in the actual power as the power to be measured, and determine the wind speed corresponding to the power to be measured as the wind speed to be measured;
[0126] A second construction module 340, configured to construct a target data set based on the wind speed to be measured and the power to be measured. The target data set is composed of data points, and each data point includes a wind speed to be measured and the power to be measured corresponding to the wind speed to be measured;
[0127] A third construction module 350, configured to construct a first digital image corresponding to the target data set and an envelope line corresponding to the first digital image based on a mathematical morphology algorithm. Each data point in the target data set corresponds to an element coordinate in the first digital image;
[0128] A second recognition module 360, configured to identify the abnormal power in the target data set as the second abnormal power based on the positional relationship between each element coordinate in the first digital image and the envelope line;
[0129] A determination module 370, configured to determine the first abnormal power and the actual wind speed corresponding to the first abnormal power, and the second abnormal power and the actual wind speed corresponding to the second abnormal power as abnormal data in the actual operation data of the wind farm.
[0130] Optionally, the above-mentioned first construction module includes:
[0131] An acquisition sub-module, configured to acquire the actual operation data set of the wind farm. The actual operation data set includes n samples, and each sample includes an actual power output by the wind farm and the actual wind speed corresponding to the actual power. n is a positive integer;
[0132] A sub-module for dividing samples in the actual operation dataset into k different sample layers according to the magnitude of the actual power in the actual operation dataset, where k is a positive integer;
[0133] A sampling sub-module for randomly sampling the samples in each sample layer based on the proportion of the number of samples included in each sample layer in the actual operation dataset to obtain a training dataset;
[0134] An expansion sub-module for performing a cubic polynomial expansion on each actual wind speed in the training dataset to obtain a target wind speed expansion matrix;
[0135] A training sub-module for training a regression model with the target wind speed expansion matrix as the input of the regression model and the actual power in the training dataset as the output of the regression model to obtain a trained wind speed-power regression model.
[0136] Optionally, the above first recognition module includes:
[0137] A calculation sub-module for calculating the difference between each predicted power and the actual power corresponding to the predicted power;
[0138] A first determination sub-module for determining the actual power corresponding to the difference as the first abnormal power when the difference is within a preset error range;
[0139] A second determination sub-module for determining the actual power corresponding to the difference as the power to be measured when the difference is outside the preset error range.
[0140] Optionally, the above second construction module includes:
[0141] A normalization sub-module for performing normalization processing on each data point in the target dataset to obtain a normalized dataset corresponding to the target dataset;
[0142] A conversion sub-module for performing image conversion on the normalized dataset to obtain a first digital image;
[0143] An erosion sub-module for performing an erosion operation on the first digital image based on a preset structural element to obtain a second digital image;
[0144] A dilation sub-module for performing a dilation operation on the second digital image to obtain a third digital image;
[0145] An extraction sub-module for identifying boundary pixels among non-zero pixels in the third digital image and extracting the data points corresponding to each boundary pixel as the boundary coordinates of the third digital image;
[0146] A generating sub-module, configured to perform interpolation processing on the boundary coordinates of the third digital image and smoothly connect them to generate an envelope line corresponding to the first digital image.
[0147] Optionally, the above conversion sub-module includes:
[0148] A mapping unit, configured to map each standard data point in the standardized data set to a preset positive integer interval to obtain a first data set;
[0149] An initialization unit, configured to initialize a two-dimensional zero matrix based on the preset positive integer interval;
[0150] A determination unit, configured to map each data point in the first data set to the two-dimensional zero matrix and determine the target position of each data point in the first data set in the two-dimensional zero matrix;
[0151] A marking unit, configured to mark the value of each target position in the two-dimensional zero matrix as 1;
[0152] A conversion unit, configured to perform pixel conversion on the two-dimensional zero matrix by representing the value of 1 in the two-dimensional zero matrix with a first color pixel and representing the value of 0 in the two-dimensional zero matrix with a second color pixel, to obtain a first digital image, where the first color pixel and the second color pixel are different.
[0153] Optionally, the above second recognition module includes:
[0154] A judgment sub-module, configured to judge whether each element coordinate in the first digital image is within the envelope line;
[0155] A third determination sub-module, configured to, if the element coordinate is within the envelope line, determine the data point corresponding to the element coordinate as a normal data point and determine the measured power in the normal data point as the normal power;
[0156] A fourth determination sub-module, configured to, if the element coordinate is not within the envelope line, determine the data point corresponding to the element coordinate as an abnormal operation data point and determine the measured power in the abnormal operation data point as the second abnormal power.
[0157] Optionally, the above wind power abnormal data recognition device includes:
[0158] An elimination module, configured to eliminate abnormal data in the actual operation data of the wind farm to obtain the target operation data of the wind farm.
[0159] The wind power abnormal data recognition device provided by the embodiments of the present disclosure can implement the methods of any of the above embodiments, and its execution manners and beneficial effects are similar, which will not be elaborated here.
[0160] An embodiment of the present disclosure also provides a computer device, which includes a processor and a memory. Among them, a computer program is stored in the memory. When the computer program is executed by the processor, the methods of any of the above embodiments can be implemented. Their execution manners and beneficial effects are similar and will not be elaborated here.
[0161] The computer device provided by the embodiment of the present disclosure may include, but is not limited to, mobile terminals such as smart phones, laptop computers, tablet computers (PADs), wearable devices, etc., and fixed electronic devices such as digital TVs, desktop computers, etc.
[0162] Figure 4 FIG. is a schematic structural diagram of a computer device provided by an embodiment of the present disclosure. As Figure 4 shown, the computer device 400 may include a processor 410 and a memory 420. Among them, a computer program 421 is stored in the memory 420. When the computer program 421 is executed by the processor 410, the methods provided by any of the above embodiments can be implemented. Their execution manners and beneficial effects are similar and will not be elaborated here.
[0163] Of course, for simplicity, Figure 4 only some of the components related to the present invention in the computer device 400 are shown in, and components such as buses, input / output interfaces, input devices, and output devices are omitted. In addition, according to specific application scenarios, the computer device 400 may further include any other appropriate components.
[0164] An embodiment of the present disclosure provides a computer-readable storage medium. A computer program is stored in the storage medium. When the computer program is executed by a processor, the methods of any of the above embodiments can be implemented. Their execution manners and beneficial effects are similar and will not be elaborated here.
[0165] The above computer-readable storage medium may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, include, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the readable storage medium (non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0166] The above computer program can be written in any combination of one or more programming languages to write program code for performing the operations of the embodiments of the present disclosure. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer device, partially on the user's device, executed as an independent software package, partially on the user's computer device and partially on a remote computer device, or entirely on a remote computer device or server.
[0167] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the present disclosure.
[0168] In addition, although the operations are depicted in a specific order, this should not be construed as requiring that the operations be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments can also be implemented combinatorially in a single embodiment. Conversely, the various features described in the context of a single embodiment can also be implemented separately or in any suitable sub-combination in multiple embodiments.
[0169] The above is only the specific implementation manner of the present disclosure, enabling those skilled in the art to understand or implement the present disclosure. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to these embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for identifying abnormal wind power data, characterized in that, Including: Construct a wind speed-power regression model based on the actual power output of the wind farm and the actual wind speed corresponding to each actual power. Input each of the actual wind speeds into the wind speed-power regression model, and predict the power corresponding to each of the actual wind speeds based on the wind speed-power regression model to obtain the predicted power corresponding to each of the actual wind speeds. Based on the difference between each predicted power and the actual power corresponding to the predicted power, identify the abnormal power in the actual power as the first abnormal power, determine the power other than the first abnormal power in the actual power as the power to be measured, and determine the wind speed corresponding to the power to be measured as the wind speed to be measured. Construct a target data set based on the wind speed to be measured and the power to be measured. The target data set is composed of data points, and each data point includes a wind speed to be measured and the power to be measured corresponding to the wind speed to be measured. Construct a first digital image corresponding to the target data set and an envelope line corresponding to the first digital image based on the mathematical morphology algorithm. Each data point in the target data set corresponds to an element coordinate in the first digital image. Identify the abnormal power in the target data set as the second abnormal power based on the positional relationship between each element coordinate in the first digital image and the envelope line. Determine the first abnormal power and the actual wind speed corresponding to the first abnormal power, and the second abnormal power and the actual wind speed corresponding to the second abnormal power as the abnormal data in the actual operation data of the wind farm.
2. The method according to claim 1, wherein The constructing a wind speed-power regression model based on the actual power output of the wind farm and the actual wind speed corresponding to each actual power includes: Obtain the actual operation data set of the wind farm. The actual operation data set includes n samples, and each sample includes an actual power output by the wind farm and the actual wind speed corresponding to the actual power. n is a positive integer. Divide the samples in the actual operation data set into k different sample layers according to the magnitude of the actual power in the actual operation data set. k is a positive integer. Randomly sample the samples in each sample layer based on the proportion of the number of samples included in each sample layer in the actual operation data set to obtain a training data set. Perform a third-degree polynomial expansion on each actual wind speed in the training data set to obtain a target wind speed expansion matrix. Use the target wind speed expansion matrix as the input of the regression model and the actual power in the training data set as the output of the regression model to train the regression model to obtain a trained wind speed-power regression model.
3. The method according to claim 1, wherein The identifying the abnormal power in the actual power as the first abnormal power based on the difference between each predicted power and the actual power corresponding to the predicted power, and determining the power other than the first abnormal power in the actual power as the power to be measured includes: For each predicted power, calculate the difference between the predicted power and the actual power corresponding to the predicted power. When the difference is within the preset error range, determine the actual power corresponding to the difference as the first abnormal power. When the difference is outside the preset error range, determine the actual power corresponding to the difference as the power to be measured.
4. The method according to claim 1, wherein The constructing the first digital image corresponding to the target data set and the envelope line corresponding to the first digital image based on the mathematical morphology algorithm includes: Perform standardization processing on each data point in the target data set to obtain a standardized data set corresponding to the target data set; Perform image conversion on the standardized data set to obtain a first digital image; Perform an erosion operation on the first digital image based on a preset structural element to obtain a second digital image; Perform a dilation operation on the second digital image to obtain a third digital image; Identify the boundary pixels among the non-zero pixels in the third digital image, and extract the data points corresponding to each boundary pixel as the boundary coordinates of the third digital image; Perform interpolation processing and smooth connection on the boundary coordinates of the third digital image to generate the envelope line corresponding to the first digital image.
5. The method according to claim 4, wherein The performing image conversion on the standardized data set to obtain a first digital image includes: Map each standard data point in the standardized data set to a preset positive integer interval to obtain a first data set; Initialize a two-dimensional zero matrix based on the preset positive integer interval; Map each data point in the first data set to the two-dimensional zero matrix, and determine the target position of each data point in the first data set in the two-dimensional zero matrix; Mark the value at each target position in the two-dimensional zero matrix as 1; Perform pixel conversion on the two-dimensional zero matrix, representing the value 1 in the two-dimensional zero matrix with a first color pixel and the value 0 in the two-dimensional zero matrix with a second color pixel, to obtain a first digital image, where the first color pixel and the second color pixel are different.
6. The method according to claim 1, wherein The identifying the abnormal power in the target data set as the second abnormal power based on the positional relationship between each element coordinate in the first digital image and the envelope line includes: For each element coordinate in the first digital image, determine whether the element coordinate is within the envelope line; If the element coordinate is within the envelope line, determine the data point corresponding to the element coordinate as a normal data point, and determine the power to be measured in the normal data point as the normal power; If the element coordinate is not within the envelope line, determine the data point corresponding to the element coordinate as an abnormal operation data point, and determine the power to be measured in the abnormal operation data point as the second abnormal power.
7. The method according to claim 1, characterized in that, After determining the first abnormal power and the actual wind speed corresponding to the first abnormal power, and the second abnormal power and the actual wind speed corresponding to the second abnormal power as the abnormal data in the actual operation data of the wind farm, the method further includes: Eliminate the abnormal data in the actual operation data of the wind farm to obtain the target operation data of the wind farm.
8. An abnormal wind power data recognition device, characterized in that including: A first construction module, configured to construct a wind speed-power regression model based on the actual power output by the wind farm and the actual wind speed corresponding to each actual power; A prediction module, configured to input each of the actual wind speeds into the wind speed-power regression model, and predict the power corresponding to each of the actual wind speeds based on the wind speed-power regression model, so as to obtain the predicted power corresponding to each of the actual wind speeds; A first identification module, configured to identify the abnormal power in the actual power as the first abnormal power based on the difference between each predicted power and the actual power corresponding to the predicted power, determine the power other than the first abnormal power in the actual power as the power to be measured, and determine the wind speed corresponding to the power to be measured as the wind speed to be measured; A second construction module, configured to construct a target data set based on the wind speed to be measured and the power to be measured, where the target data set is composed of data points, and each data point includes a wind speed to be measured and the power to be measured corresponding to the wind speed to be measured; A third construction module, configured to construct a first digital image corresponding to the target data set and an envelope line corresponding to the first digital image based on a mathematical morphology algorithm, and each data point in the target data set corresponds to an element coordinate in the first digital image; A second identification module, configured to identify the abnormal power in the target data set as the second abnormal power based on the positional relationship between each element coordinate in the first digital image and the envelope line; A determination module, configured to determine the first abnormal power and the actual wind speed corresponding to the first abnormal power, and the second abnormal power and the actual wind speed corresponding to the second abnormal power as the abnormal data in the actual operation data of the wind farm.
9. A computer device, characterized in that, Comprising: A memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method for identifying abnormal wind power data according to any one of claims 1-7 is implemented.
10. A computer-readable storage medium, characterized in that, A computer program is stored in the storage medium, and when the computer program is executed by a processor, the method for identifying abnormal wind power data according to any one of claims 1-7 is implemented.