Data analysis method and device for wind turbine generator, electronic equipment, storage medium and computer program product

Abstract

The invention relates to a data analysis method and device for a wind turbine generator, electronic equipment, a storage medium and a computer program product. The data analysis method comprises the steps of obtaining historical data of target parameters of the wind turbine generator; calculating a target probability density curve of the target parameter based on the historical data of the target parameter; and evaluating the operation performance of the wind turbine generator based on the target probability density curve. Therefore, the evaluation of the operation performance of the wind turbine generator does not depend on any input information, but calculates the probability density curve corresponding to the parameters of the wind turbine generator based on the real operation data of the wind turbine generator in the historical operation process, and then evaluates the operation performance of the wind turbine generator based on the calculated probability density curve. According to the method, the actual operation rule of the wind turbine generator is fully combined, namely the design principle and the control characteristics of the wind turbine generator are fully utilized, the control mode of the wind turbine generator can be truly restored, and the accuracy of evaluating the operation performance of the wind turbine generator is improved.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more specifically, to data analysis methods, apparatus, electronic devices, storage media, and computer program products for wind turbine generators. Background Technology

[0002] A wind turbine generator set is a power generation device that converts wind energy into mechanical energy and then into electrical energy. Its power generation capacity is affected by a combination of factors, including the turbine location environment, turbine configuration, electromechanical structure, and control strategy. Among these, the control strategy is a relatively important controllable element. The unit operates according to the instructions of the control strategy, and its operating data can include wind speed, engine speed, torque, power, pitch angle, etc.

[0003] In related technologies, the evaluation of wind turbine operating performance mainly relies on input information. For example, these technologies require users to input design values ​​for turbine control parameters, which are then used to classify the turbine's operating conditions. Because these evaluation methods heavily depend on input information and fail to fully incorporate the actual operating patterns of wind turbines, the accuracy of performance evaluations is often poor. Summary of the Invention

[0004] This disclosure provides data analysis methods, apparatus, electronic devices, storage media, and computer program products for wind turbine generators, to at least address the problem in the aforementioned related technologies that the evaluation methods are highly dependent on input information and fail to fully incorporate the actual operating patterns of wind turbine generators, resulting in poor accuracy in evaluating the operating performance of wind turbine generators.

[0005] According to a first aspect of the present disclosure, a data analysis method for wind turbine generators is provided, comprising: acquiring historical data of target parameters of the wind turbine generators; calculating a target probability density curve of the target parameters based on the historical data of the target parameters; and evaluating the operating performance of the wind turbine generators based on the target probability density curve.

[0006] Optionally, evaluating the operating performance of the wind turbine based on the target probability density curve includes: determining a target probability change rate curve corresponding to the target probability density curve; identifying the fluctuating portion of the target probability change rate curve; determining a target peak in the target probability density curve corresponding to the fluctuating portion; and evaluating the operating performance of the wind turbine based on the target peak.

[0007] Optionally, identifying the fluctuation portion of the target probability change rate curve includes: for each data point of the target probability change rate curve, calculating a statistical value for that data point based on the ordinate values ​​of all data points from the starting data point closest to that data point; for the preceding data point of each data point of the target probability change rate curve, calculating a statistical value for the preceding data point based on the ordinate values ​​of all data points from the starting data point to the preceding data point; calculating a first difference between the statistical value of each data point and the statistical value of the preceding data point; and responding to the first... If the difference is greater than a preset threshold, the corresponding data point is determined to be a fluctuating data point; if the first difference is less than or equal to the preset threshold, the corresponding data point is determined to be a non-fluctuating data point; one or more consecutive fluctuating data points between two non-fluctuating data points are determined as the fluctuating portion; wherein, the starting data point is the first data point of the target probability change rate curve, or a jump data point, wherein if any data point is a fluctuating data point and its preceding data point is a non-fluctuating data point, or if any data point is a non-fluctuating data point and its preceding data point is a fluctuating data point, then any data point is the jump data point.

[0008] Optionally, evaluating the operating performance of the wind turbine based on the target peak includes: calculating the absolute value of a second difference between the ordinate value of the target peak and the design value of the target parameter; and evaluating the degree of degradation of the target parameter based on the absolute value of the second difference, wherein the larger the absolute value of the second difference, the higher the degree of degradation.

[0009] Optionally, the target parameter includes a first parameter and a second parameter; calculating the target probability density curve of the target parameter based on historical data of the target parameter includes: constructing a first two-dimensional coordinate image based on historical data of the first parameter and the second parameter; performing compartment processing on the horizontal axis of the first two-dimensional coordinate image to obtain multiple compartments; for each of the multiple compartments, calculating the probability density curve of each compartment based on the data points on the first two-dimensional coordinate image contained in each compartment; wherein, evaluating the operating performance of the wind turbine based on the target probability density curve includes: evaluating the operating performance of the wind turbine based on the probability density curve of each of the multiple compartments.

[0010] Optionally, evaluating the operating performance of the wind turbine based on the probability density curve of each of the plurality of compartments includes: for each of the plurality of compartments, obtaining the maximum and minimum points contained in each compartment based on the probability density curve of each compartment; constructing an upper boundary curve based on the maximum points contained in each compartment; constructing a lower boundary curve based on the minimum points contained in each compartment; and identifying data points in the first two-dimensional coordinate image located outside the upper and lower boundary curves as outlier data points.

[0011] Optionally, the target parameter includes a third parameter and a fourth parameter; calculating the target probability density curve of the target parameter based on historical data of the target parameter includes: calculating the target probability density curve of the third parameter based on historical data of the third parameter; wherein, evaluating the operating performance of the wind turbine based on the target probability density curve includes: calculating the actual value of the third parameter based on the target probability density curve of the third parameter; and calculating the actual value of the fourth parameter based on the actual value of the third parameter, historical data of the third parameter and the fourth parameter.

[0012] Optionally, calculating the actual value of the fourth parameter based on the actual value of the third parameter and historical data of the third and fourth parameters includes: constructing a second two-dimensional coordinate image based on the third parameter and historical data of the fourth parameter; performing compartmentalization processing on the horizontal axis of the second two-dimensional coordinate image to obtain multiple compartments; determining the mean point of each compartment based on the data points on the second two-dimensional coordinate image contained in each compartment; constructing a mean curve based on the mean point of each compartment; and determining the actual value of the fourth parameter based on the actual value of the third parameter and the mean curve.

[0013] According to a second aspect of the present disclosure, a data analysis device for a wind turbine is provided, comprising: a historical data acquisition module configured to acquire historical data of a target parameter of the wind turbine; a probability density curve calculation module configured to calculate a target probability density curve of the target parameter based on the historical data of the target parameter; and a performance evaluation module configured to evaluate the operating performance of the wind turbine based on the target probability density curve.

[0014] Optionally, the performance evaluation module is configured to: determine a target probability change rate curve corresponding to the target probability density curve; identify the fluctuation portion of the target probability change rate curve; determine a target peak in the target probability density curve corresponding to the fluctuation portion; and evaluate the operating performance of the wind turbine based on the target peak.

[0015] Optionally, the performance evaluation module is configured to: for each data point of the target probability rate of change curve, calculate a statistical value for that data point based on the ordinate values ​​of all data points from the starting data point closest to that data point; for the preceding data point of each data point of the target probability rate of change curve, calculate a statistical value for the preceding data point based on the ordinate values ​​of all data points from the starting data point to the preceding data point; calculate a first difference between the statistical value of each data point and the statistical value of the preceding data point; and respond to a situation where the first difference is greater than a predetermined value. A threshold is set to determine the corresponding data point as a fluctuating data point; in response to the first difference being less than or equal to the preset threshold, the corresponding data point is determined as a non-fluctuating data point; one or more consecutive fluctuating data points between two non-fluctuating data points are determined as the fluctuating portion; wherein, the starting data point is the first data point of the target probability change rate curve, or a jump data point, wherein if any data point is a fluctuating data point and its preceding data point is a non-fluctuating data point, or if any data point is a non-fluctuating data point and its preceding data point is a fluctuating data point, then any data point is the jump data point.

[0016] Optionally, the performance evaluation module is configured to: calculate the absolute value of a second difference between the ordinate value of the target peak and the design value of the target parameter; and evaluate the degree of degradation of the target parameter based on the absolute value of the second difference, wherein the larger the absolute value of the second difference, the higher the degree of degradation.

[0017] Optionally, the target parameters include a first parameter and a second parameter; the probability density curve calculation module is configured to: construct a first two-dimensional coordinate image based on historical data of the first parameter and the second parameter; perform compartment processing on the horizontal axis of the first two-dimensional coordinate image to obtain multiple compartments; for each of the multiple compartments, calculate the probability density curve of each compartment based on the data points on the first two-dimensional coordinate image contained in each compartment; the performance evaluation module is configured to: evaluate the operating performance of the wind turbine based on the probability density curve of each of the multiple compartments.

[0018] Optionally, the performance evaluation module is configured to: for each of the plurality of sub-compartments, based on the probability density curve of each sub-compartment, obtain the maximum and minimum points contained in each sub-compartment; construct an upper boundary curve based on the maximum points contained in each sub-compartment; construct a lower boundary curve based on the minimum points contained in each sub-compartment; and determine the data points in the first two-dimensional coordinate image located outside the upper boundary curve and the lower boundary curve as outlier data points.

[0019] Optionally, the target parameter includes a third parameter and a fourth parameter; the probability density curve calculation module is configured to: calculate the target probability density curve of the third parameter based on historical data of the third parameter; the performance evaluation module is configured to: calculate the actual value of the third parameter based on the target probability density curve of the third parameter; and calculate the actual value of the fourth parameter based on the actual value of the third parameter, the historical data of the third parameter and the fourth parameter.

[0020] Optionally, the performance evaluation module is configured to: construct a second two-dimensional coordinate image based on historical data of the third parameter and the fourth parameter; perform compartmentalization processing on the horizontal axis of the second two-dimensional coordinate image to obtain multiple compartments; for each of the multiple compartments, determine the mean point of each compartment based on the data points on the second two-dimensional coordinate image contained in each compartment; construct a mean curve based on the mean point of each compartment; and determine the actual value of the fourth parameter based on the actual value of the third parameter and the mean curve.

[0021] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement a data analysis method for wind turbine generators according to the present disclosure.

[0022] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform a data analysis method for wind turbine generators according to the present disclosure.

[0023] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements a data analysis method for wind turbine generators according to the present disclosure.

[0024] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects:

[0025] In this invention, when evaluating the operating performance of a wind turbine, no input information is relied upon. Instead, the probability density curve corresponding to the parameters of the wind turbine is calculated based on the actual operating data of the wind turbine during its historical operation. The operating performance of the wind turbine is then evaluated based on the calculated probability density curve. This fully incorporates the actual operating rules of the wind turbine, that is, it makes full use of the design principles and control characteristics of the wind turbine. This ensures the accurate reproduction of the wind turbine's control mode and improves the accuracy of evaluating the operating performance of the wind turbine.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0027] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0028] Figure 1 This is a schematic diagram illustrating the mapping relationship between various parameters according to an exemplary embodiment of the present disclosure;

[0029] Figure 2 This is a schematic diagram illustrating the control features of each stage according to exemplary embodiments of the present disclosure;

[0030] Figure 3 This is a flowchart illustrating a data analysis method for wind turbine generators according to exemplary embodiments of the present disclosure;

[0031] Figure 4 This is a schematic diagram illustrating the probability density curve and probability rate of change curve of rotational speed according to an exemplary embodiment of the present disclosure;

[0032] Figure 5 This is a scatter plot showing the speed-power ratio according to an exemplary embodiment of the present disclosure;

[0033] Figure 6 This is a schematic diagram illustrating the detection of outlier data points using upper and lower boundary curves according to an exemplary embodiment of the present disclosure;

[0034] Figure 7 This is a schematic diagram illustrating the acquisition of rated wind speed based on rated power and mean curves according to an exemplary embodiment of the present disclosure;

[0035] Figure 8 This is a block diagram illustrating a data analysis apparatus for a wind turbine generator according to an exemplary embodiment of the present disclosure;

[0036] Figure 9 This is a block diagram illustrating an electronic device according to an exemplary embodiment of the present disclosure. Detailed Implementation

[0037] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0038] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following examples do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0039] It should be noted that the phrase "at least one of several items" in this disclosure refers to three parallel cases: "any one of the several items", "a combination of any number of the several items", and "all of the several items". For example, "including at least one of A and B" includes the following three parallel cases: (1) including A; (2) including B; (3) including A and B. As another example, "performing at least one of step one and step two" indicates the following three parallel cases: (1) performing step one; (2) performing step two; (3) performing both step one and step two.

[0040] In the field of turbine generator set fault early warning and anomaly detection, it is often necessary to analyze and model the entire sample, or perform cluster analysis based solely on environmental data similarity, or roughly classify the operating status of the turbine based on the set values ​​of the turbine's control parameters. However, when analyzing and modeling the entire sample of wind turbine generator set operating data, the chosen algorithms often fail to characterize the features of the turbine's operating data constrained by control and measurement, resulting in unsatisfactory accuracy in outlier detection or fault identification models. Furthermore, classifying turbine operating conditions based on the design values ​​of the turbine's control parameters is highly dependent on information and fails to fully utilize the turbine's design principles and control characteristics, making it impossible to assess the degree to which the turbine executes the strategy.

[0041] To address the aforementioned problems in related technologies, the data analysis method, apparatus, electronic device, storage medium, and computer program product for wind turbine generators provided in this disclosure do not rely on any input information when evaluating the operating performance of wind turbine generators. Instead, they calculate the probability density curves corresponding to the parameters of the wind turbine generators based on real operating data from the historical operation of the wind turbine generators. Furthermore, they evaluate the operating performance of the wind turbine generators based on the calculated probability density curves. This fully incorporates the actual operating patterns of wind turbine generators, that is, it makes full use of the design principles and control characteristics of wind turbine generators. It can ensure the accurate reproduction of the control mode of the wind turbine generators and improve the accuracy of evaluating the operating performance of wind turbine generators.

[0042] As mentioned earlier, wind turbine generators are power generation devices that convert wind energy into mechanical energy and then into electrical energy output. Their power generation capacity is comprehensively affected by factors such as the turbine location environment, turbine configuration, electromechanical structure, and control strategy. Among these, the control strategy is a crucial controllable element. The generator operates according to the instructions of the control strategy, and its operating data can include wind speed, engine speed, torque, power, pitch angle, etc. Furthermore, the data distribution and variable relationships exhibit significant pattern characteristics, and these characteristic relationships can be characterized by estimated control parameters. These control parameters may include, but are not limited to, cut-in wind speed, cut-out wind speed, rated wind speed, grid-connected speed, rated speed, rated power, and optimal pitch angle. The Supervisory Control and Data Acquisition (SCADA) system records the generator's operating data. By constructing scatter plots of the operating data, including but not limited to: "wind speed-power," "wind speed-engine speed," "wind speed-pitch angle," "engine speed-torque," "engine speed-power," and "engine speed-pitch angle," it is helpful to visualize the interrelationships between the data.

[0043] According to the design principles and control laws of wind turbine generator sets, when the generator set is in grid-connected power generation and in a non-limited state, there is a specific mapping relationship between wind speed, power, speed, torque, and pitch angle. Figure 1 This is a schematic diagram illustrating the mapping relationships between various parameters according to exemplary embodiments of the present disclosure. (Refer to...) Figure 1 The diagram illustrates the mapping relationships between "wind speed-power", "rotation speed-torque", and "rotation speed-pitch angle". The operating conditions of the unit can be divided into four stages: startup stage, maximum wind energy tracking stage, rated speed stage, and rated power stage.

[0044] Figure 2 This is a schematic diagram illustrating the control features of each stage according to exemplary embodiments of the present disclosure. (Refer to...) Figure 2 The control characteristics for the "start-up phase" can be: "maintain grid-connected speed" and "maintain optimal pitch angle"; the control characteristics for the "maximum wind energy tracking phase" can be: "speed and torque are positively correlated" and "maintain optimal pitch angle"; the control characteristics for the "rated speed phase" can be: "maintain rated speed" and "maintain optimal pitch angle"; and the control characteristics for the "rated power phase" can be: "maintain rated power".

[0045] Therefore, by recording operational data through the SCADA monitoring system and quantifying the control parameters during the actual operation of the unit, including but not limited to: cut-in wind speed, grid-connected speed, optimal pitch angle, rated speed, rated wind speed, rated power, cut-out wind speed, etc., it helps to deepen the understanding of the unit's operating rules. This can build a two-way communication channel between the control mechanism and the data model for subsequent simulation and operational data comparison and analysis, unit operation effect evaluation, problem localization and fault diagnosis.

[0046] Figure 3 This is a flowchart illustrating a data analysis method for wind turbines according to an exemplary embodiment of the present disclosure.

[0047] Reference Figure 3 In step 301, historical data of the target parameters of the wind turbine can be obtained. For example, SCADA operating data of the turbine under test can be obtained, which may include, but is not limited to: operating status, power limitation status, wind speed, speed, torque, power, pitch angle, etc. Furthermore, based on the operating status and power limitation status, the operating data can be categorized into three types: off-grid generation status, grid-connected generation with power limitation status, and grid-connected generation without power limitation status.

[0048] In step 302, the target probability density curve (x,y) of the target parameter can be calculated based on the historical data of the target parameter.

[0049] In step 303, the operating performance of the wind turbine can be evaluated based on the target probability density curve (x,y).

[0050] According to an exemplary embodiment of this disclosure, a target probability change rate curve (x,y′) corresponding to a target probability density curve (x,y) can be determined. Then, the fluctuating portion of the target probability change rate curve (x,y′) can be identified. Next, a target peak in the target probability density curve (x,y) corresponding to the fluctuating portion can be determined. Then, the operating performance of the wind turbine can be evaluated based on the target peak.

[0051] Specifically, taking a wind turbine in a grid-connected, unrestricted state as an example: First, operating data for this state can be filtered out, including but not limited to: speed, power, torque, pitch angle, etc. Then, for each variable, kernel density estimation can be performed to obtain the probability density curve corresponding to each variable, thereby determining the probability rate of change curve (x,y′) corresponding to the probability density curve (x,y). Next, sequence segmentation algorithms, such as the mean-variance test, can be used to identify the constant and fluctuating portions of the probability rate of change curve. Based on the identified constant and fluctuating portions, sequence segmentation points can be determined. Then, based on the sequence segmentation points, the local maxima of the portion of the probability density curve corresponding to the fluctuating portion of the probability rate of change curve can be calculated, and the operating performance of the wind turbine can be evaluated based on these local maxima.

[0052] Below, we will take rotor speed as an example to explain in detail how to evaluate the performance of wind turbines based on historical rotor speed data.

[0053] During the startup phase of the unit, constrained by the control strategy, the actual speed of the unit fluctuates around the grid-connected speed. After reaching the rated speed, the actual speed of the unit fluctuates around the rated speed. Therefore, it can be inferred that the speed values ​​should exhibit a clustering phenomenon near the grid-connected speed and the rated speed. The corresponding probability density curves should form wave packets resembling a normal distribution in the range of grid-connected speed and rated speed, with the peaks corresponding to the grid-connected speed and the rated speed, respectively.

[0054] Assume (x1, x2, ..., x N The obtained historical rotational speed data is denoted as , where N is the data sample size. The probability density function for kernel density estimation can be:

[0055]

[0056] Where x is the random variable rotational speed, h is the smoothing bandwidth, and K is the kernel density function.

[0057] According to formula (1), the rotational speed vector x = (x1) can be calculated. <x2<…<x n |x1=inf(A),x n =sup(A), and the probability density value y = (y1, y2, ..., y) corresponding to the rotational speed vector x. n ), where A is the speed range (x1, x2, ..., x N The closure of ). Where sup() is the upper bound function and inf() is the lower bound function.

[0058]

[0059] Then (x, y) are the x and y coordinates of the probability density curve, respectively. Figure 4 This is a schematic diagram illustrating the probability density curve and probability rate of change curve (density slope) of rotational speed according to an exemplary embodiment of the present disclosure. Wherein, Figure 4 (a) in the figure shows the probability density curve of the rotational speed. Figure 4 As shown in (a), the two peaks of the probability density curve are the actual grid-connected speed and the rated speed, respectively.

[0060] It should be noted that identifying the peak of the probability density curve can be achieved through algorithms such as sequence segmentation. For example, to improve the accuracy of sequence segmentation, the vector y can be subjected to first-order difference or fitting, thereby obtaining the slope y′ of the probability density curve. Then (x, y′) are the x and y coordinates of the probability rate of change curve, respectively.

[0061] According to an exemplary embodiment of this disclosure, for each data point of the target probability rate of change curve (x, y′), a statistical value for that data point can be calculated based on the ordinate values ​​of all data points from the starting data point closest to that data point. For example, the statistical value can be the mean and / or variance. Furthermore, the aforementioned "starting data point" can be the first data point of the target probability rate of change curve (x, y′), i.e., the first data point, or it can be a jump data point. Wherein, if any data point is a fluctuating data point and its preceding data point is a non-fluctuating data point, or if any data point is a non-fluctuating data point and its preceding data point is a fluctuating data point, then that data point is a jump data point.

[0062] Then, for each data point preceding the target probability rate of change curve (x,y′), the statistical value of the preceding data point can be calculated based on the ordinate values ​​of all data points from the starting data point to the preceding data point.

[0063] Next, a first difference can be calculated between the statistical value of each data point and the statistical value of the preceding data point. If the first difference is greater than a preset threshold, the corresponding data point can be identified as a fluctuating data point; if the first difference is less than or equal to the preset threshold, the corresponding data point can be identified as a non-fluctuating data point. Then, one or more consecutive fluctuating data points between two non-fluctuating data points can be identified as the aforementioned fluctuating portion.

[0064] Specifically, the mean-variance test method based on sequence segmentation can be used to analyze the probability change rate y′=(y1′,y2′,…,y n Perform hypothesis testing on the probability distribution, where:

[0065] If the difference between the statistical value of the current data point and the statistical value of the previous data point is less than or equal to a preset threshold, the current data point can be determined as a non-fluctuating data point, and in this case, it can be included in set H0. If the difference between the statistical value of the current data point and the statistical value of the previous data point is greater than a preset threshold, the current data point can be determined as a fluctuating data point, and in this case, it can be included in set H0. α Within the set. Specifically, set H0 and set H... α The data points contained in the set satisfy:

[0066] H0: μ1=μ2=…=μ n

[0067] σ1=σ2=…=σ n

[0068]

[0069] Where, μ i Let σ be the mean of the random variable. i Let k be the variance of the random variable. j This is the sequence number corresponding to the segmentation point of the sequence.

[0070] Furthermore, suitable algorithms can be selected, such as Maximum Likelihood Estimation (MLE) and Successive Interference Cancellation (SIC), to solve for k. j Return to reference Figure 4 , Figure 4 (b) shows the probability rate of change curve of the rotational speed. Figure 4 As shown in (b) in the figure, k j j = 1, 2, 3, 4 are the sequence numbers of the split points. Figure 4 As shown in (b), the probability rate of change curve has a constant part that is easy to distinguish, namely: 1~k1, k2+1~k3, k4+1~n; and a fluctuating part, namely: k1~k2, k3~k4.

[0071] Actual grid-connected speed of the unit Rated speed It can be obtained from formulas (5) and (6):

[0072]

[0073] Figure 5 This is a scatter plot showing the speed-power ratio according to an exemplary embodiment of the present disclosure. (Refer to...) Figure 5The horizontal axis represents rotational speed, and the vertical axis represents power. Figure 5 The data also shows the grid speed and the rated speed.

[0074] Similarly, the actual values ​​of control parameters such as rated power, rated torque, and optimal pitch angle can be obtained when the unit is in grid-connected and non-limited operation. Furthermore, the data analysis method for wind turbines disclosed herein can be applied to off-grid operation, grid-connected operation with limits, etc., in which case the actual values ​​of relevant control parameters when the unit is in off-grid operation or grid-connected operation with limits can be obtained.

[0075] According to an exemplary embodiment of this disclosure, the absolute value of a second difference between the ordinate value of the target peak and the design value of the target parameter can be calculated, and the degree of degradation of the target parameter can be evaluated based on the absolute value of the second difference, wherein the larger the absolute value of the second difference, the higher the degree of degradation of the target parameter.

[0076] Specifically, assuming ω is the setpoint of the control parameter and ω′ is the actual value of the control parameter calculated based on the method of this disclosure, the aforementioned second difference can be expressed as: In this way, based on historical data, the difference between the actual and design values ​​of control parameters in each time period can be obtained, i.e., the degree of deviation. Furthermore, based on the degree of deviation between the actual and design values ​​of control parameters in each time period, the actual operating performance of the unit and the degree of degradation of characteristic parameters can be evaluated, ensuring the objectivity and accuracy of the evaluation results.

[0077] According to an exemplary embodiment of this disclosure, the target parameters may further include a first parameter and a second parameter. A first two-dimensional coordinate image can be constructed based on historical data of the first and second parameters. Then, the first two-dimensional coordinate image can be segmented along its horizontal axis to obtain multiple segments. For each of the multiple segments, a probability density curve (x, y) can be calculated based on the data points on the first two-dimensional coordinate image contained within each segment. Next, the operating performance of the wind turbine can be evaluated based on the probability density curve (x, y) of each of the multiple segments.

[0078] Specifically, the graphical relationships between unit operation data variables can reveal characteristics of the control mode. Taking feature extraction from a two-dimensional coordinate image (x, y) as an example, the variable x represented by the horizontal axis can be divided into sections, and the statistical parameters of the vector y represented by the vertical axis within each section can be calculated. For example, the statistical parameters of vector y may include, but are not limited to: mean, variance, quantiles, coefficient of variation, etc.

[0079] Furthermore, to improve the accuracy of the analysis, the probability distribution of variable y within each compartment can be fitted, relevant statistical parameters can be calculated, and the characteristic curves (b, s) of the corresponding compartments can be obtained. For example, these characteristic curves (b, s) can be, but are not limited to, mean curves, boundary curves, dispersion curves, etc. Next, appropriate clustering, fitting, or sequence segmentation methods can be combined to obtain the true values ​​of the unit's control parameters, the boundary points of operating phases, and to identify the unit's special control modes, etc.

[0080] According to an exemplary embodiment of this disclosure, for each of a plurality of sub-compartments, the maximum and minimum points contained in each sub-compartment can be obtained based on the probability density curve (x, y) of each sub-compartment. For example, the 5% and 95% quantiles of the probability density curve (x, y) of each sub-compartment can be calculated, and the data point at the 5% quantile of the probability density curve (x, y) can be determined as the aforementioned minimum point; and the data point at the 95% quantile of the probability density curve (x, y) can be determined as the aforementioned maximum point.

[0081] Then, an upper boundary curve can be constructed based on the maximum value point contained in each compartment; and a lower boundary curve can be constructed based on the minimum value point contained in each compartment. Next, data points in the first two-dimensional coordinate image that are outside the upper and lower boundary curves can be identified as outlier data points.

[0082] Figure 6 This is a schematic diagram illustrating the detection of outlier data points using upper and lower boundary curves according to an exemplary embodiment of the present disclosure. (Refer to...) Figure 6 The diagram shows the upper boundary curve C1 and the lower boundary curve C2. The data points contained in the rectangular region above the upper boundary curve C1 are outlier data points; similarly, the data points contained in the elliptical region below the lower boundary curve C2 are also outlier data points. Specifically, in... Figure 6 The illustrated two-dimensional "Rotor Speed-Power" coordinate image allows for the acquisition of upper and lower boundary feature curves for power at speeds between rated speed and grid speed. Outlier data points within rectangular regions can then be identified based on the upper boundary curve, and outlier data points within elliptical regions can be identified based on the lower boundary curve. Furthermore, based on strategy understanding, speed jump control identification can be performed.

[0083] According to an exemplary embodiment of this disclosure, the target parameter may further include a third parameter and a fourth parameter. A target probability density curve (x, y) for the third parameter can be calculated based on historical data of the third parameter. Then, the actual value of the third parameter can be calculated based on the target probability density curve (x, y) of the third parameter. The process of calculating the actual value of the parameter based on its probability density curve (x, y) has been described in detail above and will not be repeated here. Next, the actual value of the fourth parameter can be calculated based on the actual value of the third parameter and the historical data of the third and fourth parameters.

[0084] According to an exemplary embodiment of this disclosure, a second two-dimensional coordinate image can be constructed based on historical data of the third and fourth parameters. Then, the second two-dimensional coordinate image can be segmented along its horizontal axis to obtain multiple segments. Next, for each of the multiple segments, the mean point of each segment can be determined based on the data points on the second two-dimensional coordinate image contained within that segment. Then, a mean curve can be constructed based on the mean points of each segment. Finally, the actual value of the fourth parameter can be determined based on the actual value of the third parameter and the mean curve.

[0085] For example, suppose (x) i ,y i ) i＝1,…,N To obtain historical data on wind speed and power, where x i Represents wind speed, y i Represents power, N is the data sample size, (b i ,s i ) i＝1,…,n The characteristic curve obtained by the sub-marketing method is n, where n is the number of sub-markets, (b i ,s i Let be the average power of the data points contained in warehouse i. It should be noted that the rated wind speed is the wind speed corresponding to the unit reaching its rated power. After calculating the rated power based on the aforementioned probability density curve of utilized power, the power can be calculated using this rated power and the aforementioned average curve (b). i ,s i The rated wind speed is obtained. Figure 7 This is a schematic diagram illustrating the acquisition of rated wind speed based on rated power and mean curves according to an exemplary embodiment of the present disclosure. (Refer to...) Figure 7 The horizontal axis represents wind speed, and the vertical axis represents power. The curve in the figure is the mean curve A, which is the mean curve obtained by connecting the mean points in each compartment. Figure 7 The vertical axis corresponding to the horizontal dashed line represents the rated power (Rated), and the horizontal axis of the intersection of the horizontal dashed line and the mean curve is the rated wind speed.

[0086] Similarly, the actual values ​​of control parameters such as wind speed when the unit reaches its rated speed can be obtained. In addition, the upper and lower boundary curves of the "speed-torque" curve can be used to identify whether the unit has speed jump control; the lower boundary curve of the "speed-pitch angle" curve can be used to identify whether the unit has pitch angle optimization control, and so on.

[0087] Thus, the data analysis method for wind turbines provided in this disclosure does not rely on information input and can realize the true values ​​of the control parameters of the automatic computer unit. In this way, it can make full use of the control and mechanism characteristics of the unit, ensure the accurate reproduction of the unit's control mode, and build a two-way communication channel between the control mechanism and the data model for simulation and operation data comparison and analysis, unit operation effect evaluation, problem location and fault diagnosis, which can improve the efficiency and accuracy of data outlier detection and model development.

[0088] Figure 8 This is a block diagram illustrating a data analysis apparatus for a wind turbine generator according to an exemplary embodiment of the present disclosure.

[0089] Reference Figure 8 The data analysis device 800 for wind turbine units may include a historical data acquisition module 801, a probability density curve calculation module 802, and a performance evaluation module 803.

[0090] The historical data acquisition module 801 can acquire historical data of the target parameters of the wind turbine. For example, it can acquire SCADA operating data of the turbine under test, which may include, but is not limited to: operating status, power limitation status, wind speed, speed, torque, power, pitch angle, etc. Furthermore, based on the operating status and power limitation status, the operating data can be categorized into three types: off-grid generation, grid-connected generation with power limitation status, and grid-connected generation without power limitation status.

[0091] The probability density curve calculation module 802 can calculate the target probability density curve (x,y) of the target parameter based on the historical data of the target parameter.

[0092] The performance evaluation module 803 can evaluate the operating performance of wind turbine units based on the target probability density curve (x,y).

[0093] According to an exemplary embodiment of this disclosure, the performance evaluation module 803 can determine a target probability change rate curve (x,y′) corresponding to a target probability density curve (x,y). Then, the fluctuating portion of the target probability change rate curve (x,y′) can be identified. Next, a target peak in the target probability density curve (x,y) corresponding to the fluctuating portion can be determined. Then, the operating performance of the wind turbine can be evaluated based on the target peak.

[0094] Specifically, taking a wind turbine in a grid-connected, unrestricted state as an example: First, operating data for this state can be filtered out, including but not limited to: speed, power, torque, pitch angle, etc. Then, for each variable, kernel density estimation can be performed to obtain the probability density curve corresponding to each variable, thereby determining the probability rate of change curve (x,y′) corresponding to the probability density curve (x,y). Next, sequence segmentation algorithms, such as the mean-variance test, can be used to identify the constant and fluctuating portions of the probability rate of change curve. Based on the identified constant and fluctuating portions, sequence segmentation points can be determined. Then, based on the sequence segmentation points, the local maxima of the portion of the probability density curve corresponding to the fluctuating portion of the probability rate of change curve can be calculated, and the operating performance of the wind turbine can be evaluated based on these local maxima.

[0095] According to an exemplary embodiment of this disclosure, the performance evaluation module 803 can calculate a statistical value for each data point on the target probability rate of change curve (x, y′), based on the ordinate values ​​of all data points from the starting data point closest to the data point before that data point to the data point itself. For example, the statistical value can be the mean and / or variance. Furthermore, the aforementioned "starting data point" can be the first data point on the target probability rate of change curve (x, y′), i.e., the first data point, or it can be a jump data point. Wherein, if any data point is a fluctuating data point and its preceding data point is a non-fluctuating data point, or if any data point is a non-fluctuating data point and its preceding data point is a fluctuating data point, then that data point is a jump data point.

[0096] Then, for each data point preceding the target probability change rate curve (x,y′), the performance evaluation module 803 can calculate the statistical value of the preceding data point based on the ordinate values ​​of all data points from the starting data point to the preceding data point.

[0097] Next, the performance evaluation module 803 can calculate a first difference between the statistical value of each data point and the statistical value of the preceding data point. If the first difference is greater than a preset threshold, the performance evaluation module 803 can determine the corresponding data point as a fluctuating data point; if the first difference is less than or equal to the preset threshold, the performance evaluation module 803 can determine the corresponding data point as a non-fluctuating data point. Then, the performance evaluation module 803 can determine one or more consecutive fluctuating data points between two non-fluctuating data points as the aforementioned fluctuating portion.

[0098] According to an exemplary embodiment of this disclosure, the performance evaluation module 803 can calculate the absolute value of a second difference between the ordinate value of the target peak and the design value of the target parameter. Furthermore, the performance evaluation module 803 can assess the degree of degradation of the target parameter based on the absolute value of this second difference, wherein the larger the absolute value of the second difference, the higher the degree of degradation of the target parameter.

[0099] Specifically, assuming ω is the setpoint of the control parameter and ω′ is the actual value of the control parameter calculated based on the method of this disclosure, the aforementioned second difference can be expressed as: In this way, based on historical data, the difference between the actual and design values ​​of control parameters in each time period can be obtained, i.e., the degree of deviation. Furthermore, based on the degree of deviation between the actual and design values ​​of control parameters in each time period, the actual operating performance of the unit and the degree of degradation of characteristic parameters can be evaluated, ensuring the objectivity and accuracy of the evaluation results.

[0100] According to an exemplary embodiment of this disclosure, the target parameters may further include a first parameter and a second parameter. The probability density curve calculation module 802 can construct a first two-dimensional coordinate image based on historical data of the first and second parameters. Then, the probability density curve calculation module 802 can perform compartmentalization processing on the horizontal axis of the first two-dimensional coordinate image to obtain multiple compartments. For each of the multiple compartments, the probability density curve calculation module 802 can calculate the probability density curve (x, y) of each compartment based on the data points on the first two-dimensional coordinate image contained within each compartment. Next, the performance evaluation module 803 can evaluate the operating performance of the wind turbine based on the probability density curve (x, y) of each of the multiple compartments.

[0101] Specifically, the graphical relationships between unit operation data variables can reveal characteristics of the control mode. Taking feature extraction from a two-dimensional coordinate image (x, y) as an example, the variable x represented by the horizontal axis can be divided into sections, and the statistical parameters of the vector y represented by the vertical axis within each section can be calculated. For example, the statistical parameters of vector y may include, but are not limited to: mean, variance, quantiles, coefficient of variation, etc.

[0102] Furthermore, to improve the accuracy of the analysis, the probability distribution of variable y within each compartment can be fitted, relevant statistical parameters can be calculated, and the characteristic curves (b, s) of the corresponding compartments can be obtained. For example, these characteristic curves (b, s) can be, but are not limited to, mean curves, boundary curves, dispersion curves, etc. Next, appropriate clustering, fitting, or sequence segmentation methods can be combined to obtain the true values ​​of the unit's control parameters, the boundary points of operating phases, and to identify the unit's special control modes, etc.

[0103] According to an exemplary embodiment of this disclosure, for each of a plurality of sub-compartments, the performance evaluation module 803 can obtain the maximum and minimum points contained in each sub-compartment based on the probability density curve (x,y) of each sub-compartment. For example, the 5% and 95% quantiles of the probability density curve (x,y) of each sub-compartment can be calculated, and the data point at the 5% quantile of the probability density curve (x,y) can be determined as the aforementioned minimum point; and the data point at the 95% quantile of the probability density curve (x,y) can be determined as the aforementioned maximum point.

[0104] Then, the performance evaluation module 803 can construct an upper boundary curve based on the maximum value point contained in each sub-compartment; and can construct a lower boundary curve based on the minimum value point contained in each sub-compartment. Next, the performance evaluation module 803 can identify data points in the first two-dimensional coordinate image that are outside the upper and lower boundary curves as outlier data points.

[0105] According to an exemplary embodiment of this disclosure, the target parameter may further include a third parameter and a fourth parameter. The probability density curve calculation module 802 can calculate the target probability density curve (x, y) of the third parameter based on historical data of the third parameter. Then, the performance evaluation module 803 can calculate the actual value of the third parameter based on the target probability density curve (x, y) of the third parameter. The process of calculating the actual value of the parameter based on the probability density curve (x, y) of the parameter has been described in detail above and will not be repeated here. Next, the performance evaluation module 803 can calculate the actual value of the fourth parameter based on the actual value of the third parameter and historical data of the third and fourth parameters.

[0106] According to an exemplary embodiment of this disclosure, the performance evaluation module 803 can construct a second two-dimensional coordinate image based on historical data of the third and fourth parameters. Then, the performance evaluation module 803 can perform compartmentalization processing on the horizontal axis of the second two-dimensional coordinate image to obtain multiple compartments. Next, for each of the multiple compartments, the performance evaluation module 803 can determine the mean point of each compartment based on the data points on the second two-dimensional coordinate image contained within each compartment. Then, the performance evaluation module 803 can construct a mean curve based on the mean point of each compartment. Next, the performance evaluation module 803 can determine the actual value of the fourth parameter based on the actual value of the third parameter and the mean curve.

[0107] Figure 9 This is a block diagram illustrating an electronic device 900 according to an exemplary embodiment of the present disclosure.

[0108] Reference Figure 9The electronic device 900 includes at least one memory 901 and at least one processor 902. The at least one memory 901 stores instructions that, when executed by the at least one processor 902, perform a data analysis method for a wind turbine according to an exemplary embodiment of the present disclosure.

[0109] As an example, electronic device 900 can be a PC, tablet, personal digital assistant, smartphone, or other device capable of executing the aforementioned instructions. Here, electronic device 900 is not necessarily a single electronic device, but can be any collection of devices or circuits capable of executing the aforementioned instructions (or instruction sets) individually or in combination. Electronic device 900 can also be part of an integrated control system or system manager, or can be configured to interconnect with a portable electronic device locally or remotely (e.g., via wireless transmission) through an interface.

[0110] In electronic device 900, processor 902 may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller, or a microprocessor. By way of example and not limitation, processor may also include analog processors, digital processors, microprocessors, multi-core processors, processor arrays, network processors, etc.

[0111] The processor 902 can execute instructions or code stored in the memory 901, which can also store data. Instructions and data can also be sent and received via a network through a network interface device, which can employ any known transmission protocol.

[0112] The memory 901 may be integrated with the processor 902, for example, by placing RAM or flash memory within an integrated circuit microprocessor. Alternatively, the memory 901 may include a separate device, such as an external disk drive, a storage array, or other storage device usable by any database system. The memory 901 and the processor 902 may be operatively coupled, or may communicate with each other, for example, via I / O ports, network connections, etc., enabling the processor 902 to read files stored in the memory.

[0113] In addition, the electronic device 900 may also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, mouse, touch input device, etc.). All components of the electronic device 900 can be interconnected via a bus and / or network.

[0114] According to exemplary embodiments of this disclosure, a computer-readable storage medium may also be provided, which, when executed by a processor of an electronic device, enables the electronic device to perform the aforementioned data analysis method for wind turbine generators. Examples of computer-readable storage media include: read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disc storage, hard disk drive (HDD), solid-state drive (SSD), card storage (such as multimedia cards, secure digital (SD) cards, or ultra-fast digital (XD) cards), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid-state drive, and any other device configured to store a computer program and any associated data, data files, and data structures in a non-transitory manner and to provide the computer program and any associated data, data files, and data structures to a processor or computer so that the processor or computer can execute the computer program. The computer program in the aforementioned computer-readable storage medium can run in an environment deployed in computer devices such as clients, hosts, agent devices, servers, etc. Furthermore, in one example, the computer program and any associated data, data files, and data structures are distributed across a networked computer system, such that the computer program and any associated data, data files, and data structures are stored, accessed, and executed in a distributed manner through one or more processors or computers.

[0115] According to exemplary embodiments of the present disclosure, a computer program product may also be provided, including a computer program that, when executed by a processor, implements the data analysis method for wind turbine generators according to the present disclosure.

[0116] According to the data analysis method, apparatus, electronic equipment, storage medium, and computer program product for wind turbines disclosed herein, when evaluating the operating performance of wind turbines, no input information is required. Instead, the probability density curves corresponding to the parameters of the wind turbines are calculated based on the actual operating data of the wind turbines during historical operation. The operating performance of the wind turbines is then evaluated based on the calculated probability density curves. This fully incorporates the actual operating patterns of wind turbines, that is, it makes full use of the design principles and control characteristics of wind turbines. This ensures the accurate reproduction of the control mode of the wind turbines and improves the accuracy of evaluating the operating performance of wind turbines.

[0117] According to exemplary embodiments of this disclosure, the difference between the actual and design values ​​of control parameters in each time period, i.e., the degree of deviation, can be obtained based on historical data. Furthermore, the actual operating performance of the unit and the degree of degradation of characteristic parameters can be evaluated based on the degree of deviation between the actual and design values ​​of control parameters in each time period, ensuring the objectivity and accuracy of the evaluation results.

[0118] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0119] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A data analysis method for wind turbine generators, characterized in that, include: Obtain historical data of the target parameters of the wind turbine; Based on the historical data of the target parameters, calculate the target probability density curve of the target parameters; The operating performance of the wind turbine is evaluated based on the target probability density curve.

2. The data analysis method as described in claim 1, characterized in that, The evaluation of the wind turbine's operating performance based on the target probability density curve includes: Determine the target probability change rate curve corresponding to the target probability density curve; Identify the fluctuating portion of the target probability change rate curve; Determine the target peak in the target probability density curve that corresponds to the fluctuating portion; The operating performance of the wind turbine is evaluated based on the target peak.

3. The data analysis method as described in claim 2, characterized in that, The identification of the fluctuation portion of the target probability change rate curve includes: For each data point on the target probability rate of change curve, the statistical value of the data point is calculated based on the ordinate values ​​of all data points from the starting data point that precedes and is closest to the data point to the data point. For each data point preceding the target probability rate of change curve, the statistical value of the preceding data point is calculated based on the ordinate values ​​of all data points from the starting data point to the preceding data point. Calculate the first difference between the statistical value of each data point and the statistical value of the previous data point for each data point; In response to the first difference being greater than a preset threshold, the corresponding data point is determined to be a fluctuating data point; In response to the first difference being less than or equal to the preset threshold, the corresponding data point is determined to be a non-fluctuating data point; One or more consecutive fluctuating data points between two non-fluctuating data points are defined as the fluctuating portion; The starting data point is either the first data point of the target probability change rate curve or a jump data point. If any data point is a fluctuating data point and its preceding data point is a non-fluctuating data point, or if any data point is a non-fluctuating data point and its preceding data point is a fluctuating data point, then any data point is the jump data point.

4. The data analysis method as described in claim 2, characterized in that, The evaluation of the wind turbine's operating performance based on the target peak includes: Calculate the absolute value of the second difference between the ordinate value of the target peak and the design value of the target parameter; The degree of degradation of the target parameter is evaluated based on the absolute value of the second difference, wherein the larger the absolute value of the second difference, the higher the degree of degradation.

5. The data analysis method as described in claim 1, characterized in that, The target parameters include a first parameter and a second parameter; The calculation of the target probability density curve of the target parameter based on historical data of the target parameter includes: Based on historical data of the first parameter and the second parameter, a first two-dimensional coordinate image is constructed; The first two-dimensional coordinate image is divided into compartments based on its horizontal axis to obtain multiple compartments; For each of the multiple sub-compartments, the probability density curve of each sub-compartment is calculated based on the data points on the first two-dimensional coordinate image contained in each sub-compartment; The step of evaluating the operating performance of the wind turbine based on the target probability density curve includes: The operating performance of the wind turbine is evaluated based on the probability density curve of each of the multiple sub-compartments.

6. The data analysis method as described in claim 5, characterized in that, The evaluation of the wind turbine's operating performance based on the probability density curve of each of the multiple sub-compartments includes: For each of the multiple sub-compartments, based on the probability density curve of each sub-compartment, obtain the maximum and minimum points contained in each sub-compartment; Construct an upper boundary curve based on the maximum value point contained in each sub-compartment; Construct a lower boundary curve based on the minimum value point contained in each sub-compartment; Data points located outside the upper and lower boundary curves in the first two-dimensional coordinate image are identified as outlier data points.

7. The data analysis method as described in claim 1, characterized in that, The target parameters include a third parameter and a fourth parameter; The calculation of the target probability density curve of the target parameter based on historical data of the target parameter includes: Based on the historical data of the third parameter, calculate the target probability density curve of the third parameter; The step of evaluating the operating performance of the wind turbine based on the target probability density curve includes: Calculate the actual value of the third parameter based on the target probability density curve of the third parameter; The actual value of the fourth parameter is calculated based on the actual value of the third parameter and the historical data of the third and fourth parameters.

8. The data analysis method as described in claim 7, characterized in that, The calculation of the actual value of the fourth parameter based on the actual value of the third parameter, the historical data of the third parameter and the fourth parameter includes: Based on the historical data of the third and fourth parameters, a second two-dimensional coordinate image is constructed; The second two-dimensional coordinate image is divided into compartments based on its horizontal axis to obtain multiple compartments; For each of the multiple sub-compartments, the mean point of each sub-compartment is determined based on the data points on the second two-dimensional coordinate image contained in each sub-compartment. Construct a mean curve based on the mean point of each sub-warehouse; The actual value of the fourth parameter is determined based on the actual value of the third parameter and the mean curve.

9. A data analysis device for wind turbine generators, characterized in that, include: The historical data acquisition module is configured to acquire historical data of the target parameters of the wind turbine. The probability density curve calculation module is configured to calculate the target probability density curve of the target parameter based on historical data of the target parameter. The performance evaluation module is configured to evaluate the operating performance of the wind turbine based on the target probability density curve.

10. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the data analysis method for wind turbine generators as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the data analysis method for wind turbine generators as described in any one of claims 1 to 8.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the data analysis method for wind turbine generators as described in any one of claims 1 to 8.

Images