Method and device for determining power characteristics of a wind turbine

Abstract

The present disclosure provides a method and device for determining power characteristics of a wind turbine. The method comprises: obtaining first operating data of the wind turbine before applying a predetermined control strategy and second operating data of the wind turbine after applying the predetermined control strategy, the first operating data comprising first environmental data and first output power data, and the second operating data comprising second environmental data and second output power data; determining an environmental space intersection of the first operating data and the second operating data by performing environmental space matching on the first operating data and the second operating data according to the first environmental data and the second environmental data; predicting a first predicted power of the wind turbine before applying the predetermined control strategy and a second predicted power of the wind turbine after applying the predetermined control strategy based on the environmental space intersection; and determining a power characteristic difference of the wind turbine before and after applying the predetermined control strategy according to the first predicted power and the second predicted power.

Description

Technical Field

[0001] This disclosure relates to the field of wind power generation, specifically to a method and apparatus for determining the power characteristics of a wind turbine generator set. Background Technology

[0002] Against the backdrop of new energy power market reform, wind power investors and operators are increasingly sensitive to changes in production capacity. The power characteristics of wind turbine generators are the core of wind power capacity; analyzing changes in power characteristics allows for more accurate predictions of wind power capacity. In particular, while various power-increasing and efficiency-enhancing technologies have emerged in recent years, limitations have led to short verification cycles and difficulties in meeting the power characteristic testing standards of the IEC (International Electrotechnical Commission), making it challenging to obtain clear assessments of their effectiveness.

[0003] Although wind power operators currently have the ability to record and analyze operational data, they lack practical and effective assessment methods for the differences in the power characteristics of wind turbine generators. This makes it difficult to effectively analyze changes in power characteristics and thus makes it difficult to accurately predict the production capacity of wind turbine generators. Summary of the Invention

[0004] The purpose of the embodiments disclosed herein is to provide a method and apparatus for determining the power characteristics of a wind turbine generator set, providing an effective and practically valuable assessment measure for the relative differences in the power characteristics of wind turbine generator sets.

[0005] According to embodiments of this disclosure, a method for determining the power characteristics of a wind turbine generator set is provided. The method includes: acquiring first operating data of the wind turbine generator set before applying a predetermined control strategy and second operating data of the wind turbine generator set after applying the predetermined control strategy, wherein the first operating data includes first environmental data and first output power data of the wind turbine generator set, and the second operating data includes second environmental data and second output power data of the wind turbine generator set; determining the environmental spatial intersection of the first operating data and the second operating data by performing environmental spatial matching on the first operating data and the second operating data based on the first environmental data and the second environmental data; predicting a first predicted power of the wind turbine generator set before applying the predetermined control strategy and a second predicted power of the wind turbine generator set after applying the predetermined control strategy based on the first predicted power and the second predicted power; and determining the power characteristic difference of the wind turbine generator set before and after applying the predetermined control strategy based on the first predicted power and the second predicted power.

[0006] According to embodiments of this disclosure, a power characteristic determination device for a wind turbine generator set is provided. The device includes: an acquisition unit configured to acquire first operating data of the wind turbine generator set before the application of a predetermined control strategy and second operating data of the wind turbine generator set after the application of the predetermined control strategy, wherein the first operating data includes first environmental data and first output power data of the wind turbine generator set, and the second operating data includes second environmental data and second output power data of the wind turbine generator set; an environmental spatial intersection determination unit configured to determine the environmental spatial intersection of the first operating data and the second operating data by performing environmental spatial matching between the first environmental data and the second environmental data; a power prediction unit configured to predict a first predicted power of the wind turbine generator set before the application of the predetermined control strategy and a second predicted power of the wind turbine generator set after the application of the predetermined control strategy based on the environmental spatial intersection; and a power characteristic difference determination unit configured to determine the power characteristic difference of the wind turbine generator set before and after the application of the predetermined control strategy based on the first predicted power and the second predicted power.

[0007] According to embodiments of the present disclosure, a computer-readable storage medium storing a computer program is provided, which, when executed by a processor, implements the power characteristic determination method as described above.

[0008] According to an embodiment of this disclosure, a computing device is provided, the computing device comprising: a processor; and a memory storing a computer program, wherein when the computer program is executed by the processor, the power characteristic determination method described above is implemented.

[0009] The power characteristic determination method and apparatus for wind turbine generator sets according to the embodiments of this disclosure can achieve at least one of the following technical effects: The power characteristic determination apparatus and method according to the embodiments of this disclosure can achieve at least one of the following technical effects: It can combine the operating data and working status data of the wind turbine generator set to determine the power characteristic difference of the wind turbine generator set before and after applying a predetermined control strategy, excluding working states that are not practically meaningful for evaluating the working performance of the wind turbine generator set (e.g., fault state, reduced power operation state, etc.); it can also determine the power characteristic noise range of the wind turbine generator set, and then use the power characteristic difference outside the power characteristic noise range as a statistically significant power characteristic difference for multi-dimensional analysis, thereby comprehensively evaluating the power capacity difference of the wind turbine generator set and improving data processing for random observation noise.

[0010] Other aspects and / or advantages of the general concept of this disclosure will be set forth in part in the description which follows, and in part will be clear from the description or may be learned by practice of the general concept of this disclosure. Attached Figure Description

[0011] The above and other objects and features of this disclosure will become clearer from the following description taken in conjunction with the accompanying drawings.

[0012] Figure 1 This is a flowchart of a method for determining the power characteristics of a wind turbine generator set according to an embodiment of the present disclosure.

[0013] Figure 2 This is another flowchart of a method for determining the power characteristics of a wind turbine generator set according to an embodiment of the present disclosure.

[0014] Figure 3 This is a schematic diagram of environmental space matching according to an embodiment of the present disclosure.

[0015] Figure 4 This is a graph used to analyze power characteristic differences according to embodiments of the present disclosure.

[0016] Figure 5 This is a graph used to analyze power characteristic differences according to another embodiment of the present disclosure.

[0017] Figure 6 This is a block diagram of a power characteristic determination device for a wind turbine generator set according to an embodiment of the present disclosure.

[0018] Figure 7 This is a schematic diagram of a computing device according to an embodiment of the present disclosure. Detailed Implementation

[0019] In existing technologies, the power characteristics of wind turbine generators are typically rigorously evaluated or tested according to the IEC (International Electrotechnical Commission) power characteristic testing standards. Completing this evaluation or test requires at least three months, is time-consuming, places strict requirements on the wind turbine tower and test site, and is costly. Furthermore, the power characteristic evaluation or test primarily certifies the power characteristic curve, power factor, annual power generation, and average power generation over a period of time, but does not include a comprehensive evaluation and continuous tracking analysis of parameters such as the power curve. For wind farms with uneven or special terrain conditions, it is impossible to establish a suitable wind turbine tower around some of them. Therefore, standard testing methods are not suitable for long-term monitoring of the power characteristics of all wind turbine generators.

[0020] Some wind farms use a method that compares the design power curve with the test power curve, simply averaging the data in segments to obtain the wind speed-power curve of the wind turbine generator. Since the design power curve is calculated based on ideal conditions, it differs significantly from the actual power curve, making the monitoring of power characteristics in actual operation meaningless. Existing methods lack comparison of external environmental differences during evaluation or testing, and also lack processing of random observation noise.

[0021] This disclosure proposes a method and apparatus for determining the power characteristics of wind turbine generator sets. It overcomes the shortcomings of existing technologies by analyzing differences in power characteristics of the generator sets through a Bayesian regression framework and a stochastic process model. This effectively assesses the differences and uncertainties in the power characteristics of wind turbine generator sets, achieving an objective evaluation of the relative differences in their power characteristics. Combined with practical application needs, this method has practical statistical significance and value. Furthermore, it provides a practical and effective data processing method and power characteristic evaluation method for determining the differences in the power characteristics of wind turbine generator sets. The power characteristic determination method and apparatus of this disclosure can be applied to similar scenarios such as evaluating the power output of wind turbine generator sets.

[0022] The following description, in conjunction with the accompanying drawings, provides specific embodiments to aid the reader in gaining a comprehensive understanding of the methods, apparatus, and / or systems described herein. However, upon understanding this disclosure, various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but may be altered as will become clear upon understanding this disclosure, except for operations that must occur in a specific order. Furthermore, for clarity and conciseness, descriptions of features known in the art may be omitted.

[0023] The features described herein may be implemented in different forms and should not be construed as limited to the examples described herein. Rather, the examples described herein are provided only to illustrate some of the many feasible ways of implementing the methods, apparatus, and / or systems described herein, which will become clear upon understanding the disclosure of this application.

[0024] As used herein, the term “and / or” includes any one of the associated listed items and any combination of any two or more.

[0025] Although terms such as “first,” “second,” and “third” may be used herein to describe various components, assemblies, regions, layers, or parts, these components, assemblies, regions, layers, or parts should not be limited by these terms. Rather, these terms are used only to distinguish one component, assembly, region, layer, or part from another. Thus, without departing from the teaching of the examples described herein, the first component, first assembly, first region, first layer, or first part referred to as the first component, first assembly, first region, first layer, or first part may also be referred to as the second component, second assembly, second region, second layer, or second part.

[0026] The terminology used herein is for the purpose of describing various examples only and is not intended to limit disclosure. Unless the context clearly indicates otherwise, the singular form is intended to include the plural form as well. The terms “comprising,” “including,” and “having” indicate the presence of the described features, quantities, operations, components, elements, and / or combinations thereof, but do not preclude the presence or addition of one or more other features, quantities, operations, components, elements, and / or combinations thereof.

[0027] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains upon understanding this disclosure. Unless expressly defined herein, terms (such as those defined in a general dictionary) shall be interpreted as having a meaning consistent with their meaning in the context of the relevant field and in this disclosure, and shall not be interpreted in an idealized or overly formalistic manner.

[0028] Furthermore, in the description of the examples, detailed descriptions of well-known related structures or functions will be omitted when it is believed that such detailed descriptions would lead to a vague interpretation of this disclosure.

[0029] According to embodiments of this disclosure, after a wind turbine is put into operation, the power characteristics determination method or apparatus of the wind turbine can be implemented by the wind farm operator or the wind turbine control entity. For example, the power characteristics of the wind turbine can be continuously monitored through cloud servers, controllers, data centers, cloud data platforms, etc., of the wind farm or the wind turbine, to achieve a comprehensive evaluation and continuous tracking analysis of the power characteristics. In particular, for wind farms with non-flat terrain or special terrain conditions, even if it is impossible to set up suitable wind measurement towers around some wind turbines, or if standard testing methods are not suitable for long-term monitoring of the power characteristics of all wind turbines, the power characteristics determination method or apparatus of this disclosure can still be used to continuously track and analyze the power characteristics of the wind turbines, and accurately predict the production capacity of the wind turbines by analyzing changes in power characteristics.

[0030] Figure 1This is a flowchart of a method for determining the power characteristics of a wind turbine generator set according to an embodiment of the present disclosure. According to exemplary embodiments of the present disclosure, the power characteristics of a wind turbine generator set can be monitored over a long period by operations such as identifying the power characteristic assessment object, collecting relevant data, cleaning and / or aggregating the collected data, matching the environmental space, predicting output power based on an environmental test set, determining the power characteristic noise range, and determining the power difference rate. In embodiments of the present disclosure, the environmental space represents the operating data vector space of the wind turbine generator set in its operating environment. The term "environmental set" hereinafter refers to a collection of operating data in the environmental space (e.g., environmental set 1 represents first operating data, environmental set 2 represents second operating data), and "environmental test set" refers to a collection of operating data in the environmental space used for testing.

[0031] like Figure 1 As shown, in step S11, first operating data of the wind turbine generator set before the application of the predetermined control strategy and second operating data after the application of the predetermined control strategy are acquired. The first operating data includes first environmental data and first output power data of the wind turbine generator set, and the second operating data includes second environmental data and second output power data of the wind turbine generator set. According to exemplary embodiments of this disclosure, for example, but not limited to, the predetermined control strategy may include control strategies or technical measures for increasing the output power or power generation of the wind turbine generator set, control strategies or technical measures for optimizing or improving the performance of the wind turbine generator set, technical measures for increasing power and efficiency, and other preset control strategies or technical measures.

[0032] According to exemplary embodiments of this disclosure, wind turbine generators in a predetermined wind farm can be selected for power characteristic evaluation. To determine the difference in power characteristics of the wind turbine generators before and after applying a predetermined control strategy, the spatiotemporal spaces of the wind turbine generators before and after applying the predetermined control strategy can be determined. For ease of explanation, the operating state before applying the predetermined control strategy can be referred to as the baseline state, and the operating state after applying the predetermined control strategy can be referred to as the state to be evaluated.

[0033] In the embodiments of this disclosure, the operating data in the baseline state can be referred to as the first operating data, and the operating data in the state to be evaluated can be referred to as the second operating data. In order to facilitate the determination of the power characteristic difference before and after the application of the predetermined control strategy, the start time of the predetermined control strategy can be taken as the starting point after the application of the predetermined control strategy (that is, the working state after the starting point is the state to be evaluated), and the time point before the start of the application of the predetermined control strategy can be taken as the end point before the application of the predetermined control strategy (that is, the working state before the end point is the baseline state), but the implementation period of the predetermined control strategy can be ignored.

[0034] For example, historical operational data under baseline and evaluation conditions can be obtained through various data sources. Data sources may include, but are not limited to, wind turbine operation databases, wind farm data centers, or cloud data platforms. For instance, the data structure and accuracy of operational data obtained from data sources can be referenced to the IEC 61400-25 information model. The data acquisition frequency for the first and second operational data can be no less than a predetermined frequency (e.g., once every 10 minutes). The first and second operational data can respectively include all external environmental variable data, output power data, and status data indicating the operating state before and after applying a predetermined control strategy. The duration of the first and second operational data can be greater than or equal to a predetermined duration (e.g., no less than 72 consecutive hours). The amount of data collected for the first operational data can be no less than the amount of data collected for the second operational data. The data fields for the first and second operational data can respectively include an environmental vector field and its corresponding output power field. The environmental vector may include environmental data items such as the wind turbine's environmental wind speed, wind direction, turbulence intensity, and environmental temperature. Each data entry consists of environmental data at the same timestamp and corresponding output power data.

[0035] After obtaining the first and second running data and before performing the environment space matching described below, the first and second running data may be cleaned and / or aggregated.

[0036] According to embodiments of this disclosure, when cleaning the first operating data and the second operating data, operating data under predetermined abnormal states can be removed from the first operating data and the second operating data, respectively, thereby retaining the first operating data and the second operating data when the wind turbine is in normal operation and free power generation state. For example, the predetermined abnormal states may include the fault state and / or reduced power operation state of the wind turbine. By cleaning the first operating data and the second operating data based on the operating state data of the wind turbine (e.g., indicating normal operation state, free power generation state, fault state, reduced power operation state, etc.), operating states that are not practically meaningful for evaluating the operating performance of the wind turbine can be eliminated.

[0037] According to embodiments of this disclosure, when aggregating first running data and second running data, the first running data and second running data can be aggregated according to the collection frequency when the first running data and second running data are acquired, so that the time of the first running data and the second running data is synchronized. For example, for running data with a collection frequency greater than a predetermined threshold (e.g., once every 10 minutes), the running data can be aggregated into an average value within a predetermined time period corresponding to the predetermined threshold (e.g., an average value within 10 minutes).

[0038] Reference Figure 1 In step S12, based on the first environmental data and the second environmental data, the environmental space intersection of the first running data and the second running data is determined by performing environmental space matching on the first running data and the second running data.

[0039] According to embodiments of this disclosure, the first environmental data and the second environmental data correspond to various environmental data types, such as environmental wind speed, environmental wind direction, turbulence intensity, and environmental temperature.

[0040] For each type of environment data, environmental space matching (also known as scene matching) is performed between the first running data and the second running data. The parts of the first running data and the second running data with the same environmental vector distribution are determined as the environmental space intersection.

[0041] For example, based on the distribution of environmental vectors, environmental spatial matching is performed between the first and second running data. Starting from data points in the second running data, matching is performed with data points in the first running data whose environmental vectors are distributed in the same way. Successfully matched data points form a subset of the first and second running data. These two subsets can be denoted as the data with completed environmental spatial matching, i.e., the environmental spatial intersection. This environmental spatial matching can be performed in the order of environmental wind speed, environmental temperature, turbulence intensity, and environmental wind direction.

[0042] The number of data points in the first and second running data subsets after matching is equal. For example, the first and second running data subsets can represent the sample dataset of the baseline state and the sample dataset of the state to be evaluated, respectively. The sample dataset of the baseline state can be denoted as D1 = (X1, y1), and the sample dataset of the state to be evaluated can be denoted as D2 = (X2, y2), where X1 and X2 can represent the corresponding environment vectors, and y1 and y2 can represent the corresponding output power vectors, as shown in the following equation:

[0043] X1 = [x 1,1 ,x 1,2 ,...,x 1,n1 ] T

[0044] y1=[y 1,1 ,y 1,2 ,...,y 1,n1 ] T (1)

[0045] x 1，i =[V 1，i T 1，i TI 1，i WD1，i Where i = 1, 2, ..., n1 (2)

[0046] X2=[x 2,1 ,x 2,2 ,...,x 2,n2 ] T

[0047] y2=[y 2,1 ,y 2,2 ,...,y 2,n2 ] T (3)

[0048] x 2，i =[V 2，i T 2，i TI 2，i WD 2，i ], where i = 1, 2, ..., n2 (4)

[0049] Where the number of vectors n2 = n1, x 1，i and x 2，i V represents the elements in the environment vectors X1 and X2, respectively. 1，i T 1，i TI 1，i WD 1，i They represent x respectively 1，i The element corresponding to ambient wind speed, ambient temperature, turbulence intensity, and ambient wind direction, V 2，i T 2，i TI 2，i WD 2，i They represent x respectively 2，i The elements in the middle correspond to ambient wind speed, ambient temperature, turbulence intensity, and ambient wind direction.

[0050] In this way, for each type of environmental data, the environmental space of the first operating data and the second operating data can be matched based on the distribution of the environmental vector to determine the intersection of the environmental space of the first operating data and the second operating data. This allows us to consider the differences in power characteristics under different environmental conditions and optimize the evaluation of power characteristics under different environmental conditions.

[0051] Refer again Figure 1 In step S13, based on the intersection of the environmental space, the first predicted power of the wind turbine generator before the application of the predetermined control strategy and the second predicted power after the application of the predetermined control strategy are predicted.

[0052] According to exemplary embodiments of this disclosure, it can be understood by referring to Figure 2 The described method determines a first predicted power of a wind turbine generator before the application of a predetermined control strategy and a second predicted power after the application of the predetermined control strategy.

[0053] Figure 2 This is another flowchart of a method for determining the power characteristics of a wind turbine generator set according to an embodiment of this disclosure. Figure 2 As shown, in step S21, an environmental test set is determined from the intersection of environmental spaces. For example, the environmental test set can be determined based on a first subset of running data and a second subset of running data (i.e., the intersection of environmental spaces).

[0054] Figure 3 This is a schematic diagram of environmental space matching according to an embodiment of the present disclosure. (Refer to...) Figure 3 In the environmental space, by performing environmental space matching between environmental set 1 (i.e., the first operational data) under the baseline state and environmental set 2 (i.e., the second operational data) under the state to be evaluated, the environmental space intersection (i.e., ...) can be determined. Figure 3 The overlapping portion of the square and circled parts). An equally spaced environmental test set can be constructed from the intersection of the environmental spaces (i.e., Figure 3 (The set of multiple data points indicated by the test vector in the dataset).

[0055] For example, different predetermined intervals can be set in the environmental spaces corresponding to different environmental data types. For instance, the predetermined interval for environmental temperature can be no higher than a predetermined temperature interval threshold (e.g., 3 degrees Celsius), and the predetermined interval for environmental wind speed can be no higher than a predetermined wind speed interval threshold (e.g., 0.5 meters per second). In this way, an environmental test set X0 can be obtained, and the elements x in the environmental test set X0... 0，i As shown in the following formula:

[0056] x 0,i =[V 0,i ,T 0,i TI 0,i WD 0,i ], i = 1, 2, ..., n0 (5)

[0057] Among them, V 0，i T 0，i TI 0，i WD 0，i They represent x respectively 0，i The elements in the equation correspond to ambient wind speed, ambient temperature, turbulence intensity, and ambient wind direction, where n0 represents x. 0，i The number of vectors in the array.

[0058] The number of vectors n0 in the environmental test set can be adjusted according to the size of the power characteristic noise range described below. For example, if the power characteristic noise range is too large, the number of vectors n0 can be appropriately reduced. Thus, the power characteristic noise range described below can be adjusted by adjusting the size of the environmental test set.

[0059] In step S22, based on the environmental test set, the Gaussian process regression algorithm is used to predict the first predicted power of the wind turbine generator before the application of the predetermined control strategy and the second predicted power after the application of the predetermined control strategy, wherein the first predicted power and the second predicted power correspond to the same environmental vector distribution.

[0060] According to an exemplary embodiment of this disclosure, the predicted power is calculated using a multivariate Gaussian regression mean model based on an environmental test set X0, under the baseline state and the state to be evaluated.

[0061] y 0，1 * y 0，2 * They represent x respectively 0，i The difference between the first and second predicted powers under the baseline and the state to be evaluated can be considered as the difference in power characteristics, as shown in the following equation:

[0062] y 0，1 * =K(X0,X1)[K(X1,X1)+σ 2 I n1 ] -1 y1 (6)

[0063] y 0，2 * =K(X0,X2)[K(X2,X2)+σ 2 I n2 ] -1 y2 (7)

[0064] Where K(·) represents the covariance matrix and I represents the standard identity matrix.

[0065]

[0066]

[0067] The covariance function k(·) can be a radial basis function for vector x. a and x b , Where, σ 2 Take (x) a ,x b The covariance of ) is half of 1 / 2, that is, σ 2 =cor(x) a ,x b ) / 2.

[0068] By predicting the first predicted power of a wind turbine generator before and the second predicted power after applying a predetermined control strategy based on an environmental test set, it is possible to comprehensively analyze the differences in power characteristics of the wind turbine generator before and after applying the predetermined control strategy under different environmental backgrounds.

[0069] The power characteristic determination method according to this disclosure further includes: determining the power characteristic noise range of the wind turbine generator set based on the first predicted power and the second predicted power.

[0070] For example, based on the first and second predicted power, the upper and lower limits of the power characteristic noise range of the wind turbine generator are determined using the Karhunen-Loève expansion method and the chi-square distribution method.

[0071] In the embodiments of this disclosure, the power characteristic difference can be a power characteristic difference outside the power characteristic noise range. Power characteristics within the power characteristic noise range have many uncertainties, and analyzing power characteristic differences within the power characteristic noise range has no practical statistical significance. However, power characteristic differences outside the power characteristic noise range have practical statistical significance. Therefore, analysis and evaluation can be performed only on power characteristic differences outside the power characteristic noise range.

[0072] In embodiments of this disclosure, the confidence level (1-α) corresponding to the power characteristic noise range can be determined and adjusted according to the actual power characteristic evaluation requirements. For example, the value of α in the confidence level can be set and adjusted according to the actual evaluation requirements (e.g., 5%). The power characteristic noise range can be determined based on the first predicted power, the second predicted power, and the confidence level.

[0073] In one embodiment of this disclosure, the size of the power characteristic noise range can be set to be less than a predetermined range threshold. The predetermined range threshold can be set according to the rated power of the wind turbine generator set. For example, the predetermined range threshold can be set as the rated power multiplied by a predetermined factor (e.g., 0.02).

[0074] The difference df between the power characteristics under the reference state and the power characteristics under the state to be evaluated can be calculated by the following formula:

[0075] df = y 0,2 * -y 0,1 * (10)

[0076] The upper limit ub and lower limit lb of the power characteristic noise interval can be determined using the Karhunen-Loève expansion method and the chi-square distribution method. The power characteristic difference outside the power characteristic noise interval is considered as a statistically significant difference.

[0077]

[0078]

[0079] Where Max(·) represents the function for finding the maximum value, Min(·) represents the function for finding the minimum value, z is an independent standard random vector of length n0, U is the eigenvector matrix of C(X0), and Λ is a triangular matrix constructed from the eigenvalues ​​of C(X0).

[0080] For the j-th vector x in the environmental test set 0,j The difference in power characteristics between the baseline state and the state to be evaluated can be expressed as df j :

[0081]

[0082]

[0083] Where K(·) denotes the covariance matrix, It is an eigenvalue of C(X0), (u k ) j The eigenvector u of C(X0) is a unit-normalized eigenvector. k The j-th element, z, is an independent standard random vector of length n0, z k It is the k-th element in vector z.

[0084] Furthermore, the above equation satisfies the following condition:

[0085]

[0086] in, It is the inverse cumulative distribution function of the chi-square distribution with n0 degrees of freedom, z j It is the j-th element in vector z.

[0087] The power characteristic difference Df corresponding to the environmental test set can be expressed as follows:

[0088]

[0089] In the above expression, U is the eigenvector matrix of C(X0), and Λ is the triangular matrix constructed from the eigenvalues ​​of C(X0).

[0090] As described above, the upper limit ub and lower limit lb of the power characteristic noise range of the wind turbine generator can be determined using the Karhunen-Loève expansion method and the chi-square distribution method based on the first and second predicted power. After determining the power characteristic noise range, the power characteristic differences outside the power characteristic noise range (i.e., exceeding the upper limit ub and / or falling below the lower limit lb) can be determined. Optionally, firstly, the power characteristic differences of the wind turbine generator before and after applying a predetermined control strategy can be determined based on the first and second predicted power; then, the power characteristic differences outside the determined power characteristic noise range can be determined in conjunction with the determined power characteristic noise range.

[0091] Refer again Figure 1 In step S14, the power characteristic difference of the wind turbine generator set before and after applying the predetermined control strategy is determined based on the first predicted power and the second predicted power. In the embodiments of this disclosure, the power characteristic difference of the wind turbine generator set before and after applying the predetermined control strategy can reflect the capacity difference caused by the predetermined control strategy, reflecting the impact of the predetermined control strategy on wind power generation capacity. By determining and analyzing the power characteristic difference, the capacity of the wind turbine generator set can be accurately predicted.

[0092] According to embodiments of this disclosure, the power characteristic difference includes: a power difference value and / or a power difference rate between a first predicted power and a second predicted power. For example, the power difference rate may include an average difference rate between the first predicted power and the second predicted power and / or a weighted difference rate based on environmental data.

[0093] In embodiments of this disclosure, the average difference rate df% can represent the ratio of the power difference between the first predicted power and the second predicted power to the first predicted power, as shown in the following formula:

[0094]

[0095] The average difference rate (df%) can be used to measure the overall power difference of wind turbine generators.

[0096] Furthermore, the weighted difference rate df can be used based on the probability distribution (also known as frequency proportion) of different types of environmental data. w The percentage (%) is used to measure the power differences of wind turbine generators across different environmental data dimensions. This can serve as a statistical analysis result of the power generation differences among wind turbine generators, reflecting the degree of difference in their production capacity. The weighted difference rate (df) can be calculated using the following formula. w %:

[0097]

[0098] Among them, y1,i * and y 2,i * The first predicted power y under the reference state can be represented respectively. 0,1 * The i-th component and the second predicted power y under the state to be evaluated 0,2 * The i-th component, and all of them correspond to the i-th environmental component x. 0,i p i x represents 0,i The probability distribution in local environmental data.

[0099] Figure 4 This is a graph used to analyze power characteristic differences according to embodiments of the present disclosure. Figure 5 This is a graph used to analyze power characteristic differences according to another embodiment of the present disclosure.

[0100] Reference Figure 4 and Figure 5 This example uses environmental wind speed as an example of environmental data type, but the invention is not limited to this; similar statistical methods can also be used to determine the power characteristic differences of other environmental data types. For example... Figure 4 and Figure 5 As shown, the horizontal axis represents wind speed (m / s), and the vertical axis represents the weighted difference rate.

[0101] exist Figure 4 In the example shown, the weighted difference rate of the wind turbine generator is distributed within the power characteristic noise range. Since the power characteristic difference within the power characteristic noise range is uncertain and has no practical statistical significance, it can be ignored. Therefore, in this example, the power characteristic difference of the wind turbine generator before and after applying the predetermined control strategy can be considered negligible.

[0102] exist Figure 5 In the example shown, the weighted difference rate of the wind turbine generator is distributed both within and outside the power characteristic noise interval. The power characteristic difference within the power characteristic noise interval is negligible due to its uncertainty and lack of practical statistical significance. However, the power characteristic difference outside the power characteristic noise interval is statistically significant for evaluating the power characteristic difference of the wind turbine generator before and after applying a predetermined control strategy; therefore, the portion outside the power characteristic noise interval can be considered the statistical difference interval. Furthermore, power characteristic difference parameters such as power difference values ​​and average difference rates can be further analyzed for the statistical difference interval. For example, a power characteristic evaluation report for the statistical difference interval can be generated.

[0103] As described above, by using the power characteristic determination method of wind turbine generator set according to the embodiments of the present disclosure, the power characteristic difference of wind turbine generator set before and after the application of a predetermined control strategy can be determined by combining the operating data and working status data of the wind turbine generator set; the power characteristic noise range of the wind turbine generator set can also be determined, and then the power characteristic difference outside the power characteristic noise range can be used as a statistically significant power characteristic difference for multi-dimensional analysis, thereby comprehensively evaluating the production capacity difference of the wind turbine generator set.

[0104] According to embodiments of this disclosure, the capacity increase within a predetermined time period after applying a predetermined control strategy can be predicted based on at least one of similar power characteristic difference parameters such as power difference value, average difference rate, and weighted difference rate, thereby determining the capacity change and economic value brought about by the predetermined control strategy. Furthermore, similar power characteristic difference parameters such as power difference value, average difference rate, and weighted difference rate can be provided to the wind farm or wind turbine operator to determine the economic value brought about by the predetermined control strategy. For example, the increase in annual power generation after applying the predetermined control strategy can be calculated based on the weighted difference rate, the annual power generation before applying the predetermined control strategy, and the data quality deviation rate of the power characteristic difference, wherein the data quality deviation rate of the power characteristic difference can represent the percentage deviation of the pre-set weighted difference rate. Thus, the capacity change and economic value brought about by the predetermined control strategy can be estimated using the power characteristic difference parameters obtained according to the power characteristic determination method of this disclosure.

[0105] The following reference Figure 6 The apparatus for determining power characteristics according to this disclosure is described. Figure 6 This is a block diagram of a power characteristic determination device 3 for a wind turbine generator set according to an embodiment of the present disclosure.

[0106] In embodiments of this disclosure, the power characteristic determination device 3 may be located in the central controller of the wind farm, in the processor of the wind turbine generator set, or in any other processing device that communicates with the wind turbine generator set.

[0107] The power characteristic determination device 3 may include an acquisition unit 31. The acquisition unit 31 may acquire first operating data of the wind turbine generator set before the application of a predetermined control strategy and second operating data of the wind turbine generator set after the application of the predetermined control strategy. The first operating data includes first environmental data and first output power data of the wind turbine generator set, and the second operating data includes second environmental data and second output power data of the wind turbine generator set.

[0108] The power characteristic determination device 3 may include a preprocessing unit 35, which may clean and / or aggregate the first operating data and the second operating data before performing the environmental space matching. The preprocessing unit 35 may remove operating data under predetermined abnormal states from the first operating data and the second operating data, respectively; wherein, the predetermined abnormal states include the fault state and / or power reduction operating state of the wind turbine generator.

[0109] The power characteristic determination device 3 may include an environmental space intersection determination unit 32. The environmental space intersection determination unit 32 may determine the environmental space intersection between the first operating data and the second operating data by performing environmental space matching between the first operating data and the second operating data based on the first environmental data and the second environmental data.

[0110] According to embodiments of this disclosure, the first environmental data and the second environmental data correspond to multiple environmental data types. The environmental space intersection determination unit 32 can, for each of the multiple environmental data types, perform environmental space matching between the first running data and the second running data, and determine the portion of the first running data and the second running data with the same environmental vector distribution as the environmental space intersection.

[0111] The power characteristic determination device 3 may include a power prediction unit 33, which can predict a first predicted power of the wind turbine generator before the application of a predetermined control strategy and a second predicted power after the application of the predetermined control strategy based on the environmental spatial intersection. According to embodiments of this disclosure, the power prediction unit 33 can determine an environmental test set in the environmental spatial intersection; based on the environmental test set, it uses a Gaussian process regression algorithm to predict the first predicted power of the wind turbine generator before the application of the predetermined control strategy and the second predicted power after the application of the predetermined control strategy, wherein the first predicted power and the second predicted power correspond to the same environmental vector distribution.

[0112] The power characteristic determination device 3 may include a power characteristic difference determination unit 34. The power characteristic difference determination unit 34 can determine the power characteristic difference of the wind turbine generator before and after applying a predetermined control strategy based on the first predicted power and the second predicted power.

[0113] The power characteristic determination device 3 may include a power characteristic noise range determination unit 36. The power characteristic noise range determination unit 36 ​​can determine the upper and lower limits of the power characteristic noise range of the wind turbine generator set based on the first predicted power and the second predicted power, using the Karhunen-Loève expansion method and the chi-square distribution method, wherein the power characteristic difference is the power characteristic difference outside the power characteristic noise range.

[0114] According to embodiments of this disclosure, the power characteristic difference includes: a power difference value and / or a power difference rate between a first predicted power and a second predicted power. For example, the power difference rate includes the average difference rate between the first predicted power and the second predicted power and / or a weighted difference rate based on environmental data.

[0115] For reference Figures 1 to 5 The operation of each unit in the power characteristic determination apparatus 3 is described in accordance with the power characteristic determination method of this disclosure, and will not be repeated here for the sake of brevity.

[0116] According to embodiments of the present disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed, implements the power characteristic determination method according to embodiments of the present disclosure.

[0117] In embodiments of this disclosure, the computer-readable storage medium may carry one or more programs, which, when executed, can achieve reference... Figures 1 to 5 The described steps are as follows: First operating data of the wind turbine generator set before the application of a predetermined control strategy and second operating data after the application of the predetermined control strategy, wherein the first operating data includes first environmental data and first output power data of the wind turbine generator set, and the second operating data includes second environmental data and second output power data of the wind turbine generator set; based on the first environmental data and the second environmental data, environmental spatial matching is performed between the first operating data and the second operating data to determine the environmental spatial intersection of the first operating data and the second operating data; based on the environmental spatial intersection, a first predicted power of the wind turbine generator set before the application of the predetermined control strategy and a second predicted power of the wind turbine generator set after the application of the predetermined control strategy are predicted; based on the first predicted power and the second predicted power, the power characteristic difference of the wind turbine generator set before and after the application of the predetermined control strategy is determined.

[0118] Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In embodiments of this disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a computer program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable storage medium can be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof. A computer-readable storage medium can be included in any apparatus; it can also exist independently without being assembled into that apparatus.

[0119] The above has been combined Figures 1 to 5 A method for determining power characteristics according to embodiments of the present disclosure has been described. Next, in conjunction with... Figure 7 A computing device according to embodiments of the present disclosure will be described.

[0120] Figure 7 This is a schematic diagram of a computing device 4 according to an embodiment of the present disclosure.

[0121] Reference Figure 7 The computing device 4 according to an embodiment of the present disclosure may include a memory 41 and a processor 42. A computer program 43 is stored in the memory 41. When the computer program 43 is executed by the processor 42, it implements the power characteristic determination method according to an embodiment of the present disclosure.

[0122] In embodiments of this disclosure, when the computer program 43 is executed by the processor 42, reference can be implemented. Figures 1 to 5The described power characteristic determination method involves: acquiring first operating data of a wind turbine generator set before the application of a predetermined control strategy and second operating data after the application of the predetermined control strategy, wherein the first operating data includes first environmental data and first output power data of the wind turbine generator set, and the second operating data includes second environmental data and second output power data of the wind turbine generator set; determining the environmental spatial intersection of the first operating data and the second operating data by performing environmental spatial matching on the first operating data and the second operating data based on the environmental spatial intersection; predicting the first predicted power of the wind turbine generator set before the application of the predetermined control strategy and the second predicted power after the application of the predetermined control strategy based on the first predicted power and the second predicted power; and determining the power characteristic difference of the wind turbine generator set before and after the application of the predetermined control strategy based on the first predicted power and the second predicted power.

[0123] Figure 7 The computing device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0124] The above has been referred to Figures 1 to 7 A method for determining the power characteristics of a wind turbine generator set, a power characteristics determination apparatus, a computer-readable storage medium, and a computing device according to embodiments of the present disclosure are described. However, it should be understood that: Figure 6 The power characteristic determination device and its units shown can be configured as software, hardware, firmware, or any combination thereof to perform specific functions. Figure 7 The computing device shown is not limited to the components shown above, but some components may be added or removed as needed, and the above components may also be combined.

[0125] The power characteristic determination apparatus and method according to the embodiments of this disclosure can achieve at least one of the following technical effects: it can combine the operating data and working status data of the wind turbine generator set to determine the power characteristic difference of the wind turbine generator set before and after the application of a predetermined control strategy, and exclude working states that are not of practical significance for evaluating the working performance of the wind turbine generator set (e.g., fault state, power reduction operation state, etc.); it can also determine the power characteristic noise range of the wind turbine generator set, and then use the power characteristic difference outside the power characteristic noise range as a statistically significant power characteristic difference for multi-dimensional analysis, thereby comprehensively evaluating the power capacity difference of the wind turbine generator set and improving data processing for random observation noise.

[0126] Control logic or functions performed by various components or controllers in a control system can be represented by flowcharts or similar diagrams in one or more accompanying figures. These figures provide representative control strategies and / or logic, which can be implemented using one or more processing strategies (such as event-driven, interrupt-driven, multitasking, multithreading, etc.). Therefore, the individual steps or functions shown may be performed in the order shown, in parallel, or in some cases omitted. Although not always explicitly shown, those skilled in the art will recognize that one or more steps or functions shown may be repeatedly performed depending on the specific processing strategy used.

[0127] Although this disclosure has been shown and described with reference to preferred embodiments, those skilled in the art will understand that various modifications and variations may be made to these embodiments without departing from the spirit and scope of this disclosure as defined by the claims.

Claims

1. A method for determining the power characteristics of a wind turbine generator set, characterized in that, The power characteristic determination method includes: Acquire first operating data of the wind turbine generator set before applying a predetermined control strategy and second operating data after applying the predetermined control strategy, wherein the first operating data includes first environmental data and first output power data of the wind turbine generator set, and the second operating data includes second environmental data and second output power data of the wind turbine generator set. Based on the first environmental data and the second environmental data, the environmental space intersection of the first operating data and the second operating data is determined by performing environmental space matching on the first operating data and the second operating data. Based on the intersection of the environmental space, the Gaussian process regression algorithm is used to predict the first predicted power of the wind turbine generator before the application of the predetermined control strategy and the second predicted power after the application of the predetermined control strategy. Based on the first predicted power and the second predicted power, the power characteristic difference of the wind turbine generator before and after applying the predetermined control strategy is determined. The power characteristic determination method further includes: determining the upper and lower limits of the power characteristic noise range of the wind turbine generator set using the Karhunen-Loève expansion method and the chi-square distribution method based on the first predicted power and the second predicted power, wherein the power characteristic difference is the power characteristic difference outside the power characteristic noise range.

2. The power characteristic determination method according to claim 1, characterized in that, The first environmental data and the second environmental data each correspond to multiple environmental data types. The step of determining the environmental spatial intersection of the first operating data and the second operating data by performing environmental spatial matching on the first operating data and the second operating data based on the first environmental data and the second environmental data includes: For each of the various environmental data types, environmental space matching is performed between the first running data and the second running data, and the portion of the first running data and the second running data with the same environmental vector distribution is determined as the environmental space intersection.

3. The power characteristic determination method according to claim 1, characterized in that, The prediction of the first predicted power of the wind turbine generator before the application of the predetermined control strategy and the second predicted power after the application of the predetermined control strategy, based on the intersection of the environmental space, includes: Determine the environmental test set from the intersection of the environmental spaces; Based on the environmental test set, the Gaussian process regression algorithm is used to predict the first predicted power of the wind turbine generator before the application of the predetermined control strategy and the second predicted power after the application of the predetermined control strategy, wherein the first predicted power and the second predicted power correspond to the same environmental vector distribution.

4. The power characteristic determination method according to any one of claims 1 to 3, characterized in that, The power characteristic difference includes: the power difference value and / or power difference rate between the first predicted power and the second predicted power.

5. The power characteristic determination method according to claim 4, characterized in that, The power difference rate includes the average difference rate between the first predicted power and the second predicted power and / or the weighted difference rate based on environmental data.

6. The method for determining power characteristics according to any one of claims 1 to 3, characterized in that, The power characteristic determination method further includes: Before performing the environmental space matching, the first running data and the second running data are cleaned and / or aggregated.

7. The power characteristic determination method according to claim 6, characterized in that, The cleaning of the first running data and the second running data includes: Remove the running data under the predetermined abnormal state from the first running data and the second running data respectively; The predetermined abnormal states include the fault state and / or reduced power operation state of the wind turbine generator set.

8. A device for determining the power characteristics of a wind turbine generator set, characterized in that, The power characteristic determination device includes: The acquisition unit is configured to acquire first operating data of the wind turbine generator set before the application of a predetermined control strategy and second operating data of the wind turbine generator set after the application of the predetermined control strategy. The first operating data includes first environmental data and first output power data of the wind turbine generator set, and the second operating data includes second environmental data and second output power data of the wind turbine generator set. The environmental space intersection determination unit is configured to determine the environmental space intersection of the first running data and the second running data by performing environmental space matching between the first running data and the second running data based on the first environmental data and the second environmental data. The power prediction unit is configured to predict the first predicted power of the wind turbine generator before the application of a predetermined control strategy and the second predicted power after the application of the predetermined control strategy based on the intersection of the environmental space and using a Gaussian process regression algorithm. The power characteristic difference determination unit is configured to determine the power characteristic difference of the wind turbine generator before and after applying a predetermined control strategy, based on the first predicted power and the second predicted power. The power characteristic determination device further includes a power characteristic noise interval determination unit, which is configured to determine the upper and lower limits of the power characteristic noise interval of the wind turbine generator set based on the first predicted power and the second predicted power, using the Karhunen-Loève expansion method and the chi-square distribution method, wherein the power characteristic difference is the power characteristic difference outside the power characteristic noise interval.

9. The power characteristic determination device according to claim 8, characterized in that, The first environmental data and the second environmental data each correspond to multiple environmental data types. The environmental space intersection determination unit is configured to: for each of the multiple environmental data types, perform environmental space matching between the first running data and the second running data, and determine the portion of the first running data and the second running data with the same environmental vector distribution as the environmental space intersection.

10. The power characteristic determination device according to claim 8, characterized in that, The power prediction unit is configured as follows: Determine the environmental test set from the intersection of the environmental spaces; Based on the environmental test set, the Gaussian process regression algorithm is used to predict the first predicted power of the wind turbine generator before the application of the predetermined control strategy and the second predicted power after the application of the predetermined control strategy, wherein the first predicted power and the second predicted power correspond to the same environmental vector distribution.

11. The power characteristic determination apparatus according to any one of claims 8 to 10, characterized in that, The power characteristic difference includes: the power difference value and / or power difference rate between the first predicted power and the second predicted power.

12. The power characteristic determination device according to claim 11, characterized in that, The power difference rate includes the average difference rate between the first predicted power and the second predicted power and / or the weighted difference rate based on environmental data.

13. The power characteristic determination apparatus according to any one of claims 8 to 10, characterized in that, The power characteristic determination device further includes a preprocessing unit configured to clean and / or aggregate the first operating data and the second operating data before performing the environmental space matching.

14. The power characteristic determination device according to claim 13, characterized in that, The preprocessing unit is configured as follows: Remove the running data under the predetermined abnormal state from the first running data and the second running data respectively; The predetermined abnormal states include the fault state and / or reduced power operation state of the wind turbine generator set.

15. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the power characteristic determination method as described in any one of claims 1 to 7.

16. A computing device, characterized in that, The computing device includes: processor; A memory storing a computer program that, when executed by a processor, implements the power characteristic determination method as described in any one of claims 1 to 7.

Images