Wind turbine generator blade stall diagnosis method and system based on data distribution

Abstract

The invention discloses a wind turbine generator blade stall diagnosis method and system based on data distribution. The method comprises the following steps: S1, obtaining a complete paddle opening power generation data set D of a wind turbine generator; s2, using a DBSCAN spatial density clustering algorithm to screen the complete propeller opening power generation data set D, screening out a normal data cluster N, and obtaining an initial lower boundary of the normal data cluster N; s3, performing boundary smoothing on the initial lower boundary of the normal data cluster N to obtain a smoothed lower boundary; and S4, judging whether the data points of the wind turbine generator are in a stall condition or not based on the smoothed lower boundary. The method has the advantages of low cost, high diagnosis precision, high feasibility and the like.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of wind power, and in particular to a method and system for diagnosing wind turbine blade stall based on data distribution. Background Art

[0002] like Figure 1 As shown in the figure, blade stall refers to the phenomenon that when the wind speed increases to a certain level and the blade attack angle of the wind turbine generator set (i.e. the angle between the airflow direction and the blade chord) exceeds a certain critical value during wind power generation, the flow condition on the back of the blade begins to deteriorate, the boundary layer is damaged, and a vortex zone appears at the back end of the blade. At this time, the lift of the blade drops sharply, while the resistance increases significantly, resulting in the blade being unable to continue to provide sufficient lift to maintain a normal rotation speed, which in turn causes a decrease in the power generation and efficiency of the wind turbine set. Summary of the invention

[0003] In view of the technical problems existing in the prior art, the present invention provides a wind turbine blade stall diagnosis method and system based on data distribution with low cost, high diagnostic accuracy and high feasibility.

[0004] In order to solve the above technical problems, the technical solution proposed by the present invention is:

[0005] A wind turbine blade stall diagnosis method based on data distribution comprises the following steps:

[0006] S1, obtaining a wind turbine generator set fully powered-up data set D;

[0007] S2. Using the DBSCAN spatial density clustering algorithm to screen the full propeller power generation data set D, screen out the normal data cluster N, and obtain the initial lower boundary of the normal data cluster N;

[0008] S3, performing boundary smoothing on the initial lower boundary of the normal data cluster N to obtain a smoothed lower boundary;

[0009] S4. Determine whether the data point of the wind turbine generator set is in a stall condition based on the smoothed lower boundary.

[0010] Preferably, the specific process of step S2 is:

[0011] S201, divide the full-propeller power generation data set D into m bins according to wind speed i ,i=1,2,3...m;

[0012] S202, for each wind speed bin i Sort by active power and perform clustering calculation using ε neighborhood and MinPts minimum number of points as parameters;

[0013] S203, select the cluster with the largest number of samples as the wind speed bin i Normal data cluster N i ;

[0014] S204, bin all wind speeds i The normal data cluster set {N1,N2,N3...N m} is defined as the initial normal data cluster N.

[0015] Preferably, in step S201, the specific process of obtaining the ε neighborhood is:

[0016] S1.1. The full-propeller power generation data set D is divided into m bins according to wind speed i ,i=1,2,3...m;

[0017] S1.2, for each bin i Calculate the upper and lower limits of active power L up ,L low , filter bin i Active power between L up ,L low The samples between get:

[0018] G i = {g ij |L low <g ij <L up} (1)

[0019] S1.3, G i Sort in ascending order of active power, then perform difference calculation on the sorted powers to obtain the difference d between two adjacent active power data. ij ,j=1,2,3..., take d ij , j=1,2,3..., we get ε i ;

[0020] 1.4. Take the collective {ε i The 80% quantile of |i=1,2,3...,m} is taken as the ε neighborhood value.

[0021] Preferably, in step S201, the process of obtaining the minimum number of points MinPts is:

[0022] S2.1. The full-propeller power generation data set D is divided into m bins according to wind speed. i ,i=1,2,3...m;

[0023] S2.2, for each bin i Calculate the upper and lower limits of active power L up ,Llow , upper and lower limit distance R i =L up -L low , total number of samples Count i ;

[0024] S2.3, take R i ,i=1,2,3...m, 80% quantile, get R, let:

[0025]

[0026] S2.4. Get MinPts i , the minimum value of i=1,2,3... is used as the minimum point number parameter of MinPts.

[0027] Preferably, in step S3, the process of boundary smoothing is:

[0028] Assume the lower boundary after the nth smoothing is:

[0029] L n (v),n=1,2,3... (3)

[0030] The lower boundary of the nth smoothing is calculated from the lower boundary of the n-1th smoothing, that is:

[0031]

[0032] Where L n-1 (v) is the lower boundary curve after the n-1th smoothing, v i For bin i ,i=1,2,3...m average wind speed.

[0033] Preferably, the cutoff condition for lower boundary smoothing is:

[0034]

[0035] where n cut is the smoothing cutoff number, p cut is a constant.

[0036] Preferably, in step S4, when the data point i is below the lower boundary and i is not a noise point, the data point i is a stall sample, otherwise it is another abnormal sample.

[0037] The present invention also discloses a computer program product, comprising a computer program, wherein the computer program executes the steps of the method described above when executed by a processor.

[0038] The present invention further discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are executed.

[0039] The present invention also discloses a wind turbine blade stall diagnosis based on data distribution, comprising a memory and a processor connected to each other, wherein a computer program is stored on the memory, and when the computer program is run by the processor, the steps of the method described above are executed.

[0040] Compared with the prior art, the advantages of the present invention are:

[0041] The present invention considers the spatial distribution characteristics of the data, firstly performs preliminary screening on the wind turbine fully-opened power generation data set D through the DBSCAN spatial density clustering algorithm to obtain the initial normal data cluster N, then performs boundary smoothing on the data cluster N to obtain the smoothed lower boundary, and then uses the smoothed lower boundary to determine whether the data point is stalled or not. The above scheme has the following advantages:

[0042] 1) DBSCAN plus boundary smoothing provides a new perspective

[0043] The DBSCAN algorithm is a widely used clustering algorithm. It divides the data points in the data set into several categories based on the spatial distribution density of the data. After initially screening the normal categories, it uses the boundary smoothing method to determine the lower boundary of the normal data, and then identifies whether the unit is stalled based on the lower boundary. Compared with other methods, this scheme is based on the existing wind turbine SCADA data and does not require the addition of new sensors (blade pressure, deformation, torsion angle, etc.); in addition, the DBSCAN plus boundary smoothing method fully explores the distribution law of wind turbine SCADA data, providing a new perspective for blade stall diagnosis.

[0044] 2) Improved the feasibility and accuracy of blade stall diagnosis

[0045] Since this solution does not require the addition of new sensors, there is no need to invest in new hardware costs for both existing and newly built wind farms, so this solution can be deployed and implemented without obstacles in both existing and newly built wind farms, and the solution is highly feasible. Since this solution is based on the distribution rules of a large amount of historical data, the analysis process is clear and each step of the operation has a clear basis and purpose, which helps to improve the accuracy of diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of blade stall phenomenon in the prior art.

[0047] Figure 2 The present invention is a flow chart of a method for diagnosing stall of a wind turbine blade in an embodiment.

[0048] Figure 3 This is a normal unit diagnostic result diagram in the present invention.

[0049] Figure 4 This is a diagram showing the diagnosis results of the blade stall unit in the present invention. DETAILED DESCRIPTION

[0050] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0051] like Figure 2 As shown, the wind turbine blade stall diagnosis method based on data distribution provided by the embodiment of the present invention comprises the steps of:

[0052] S1. Obtain a wind turbine fully-opened power generation data set D; wherein the fully-opened power generation data set D is collected when the wind turbine is in a wind turbine fully-opened power generation condition; wherein the wind turbine fully-opened power generation condition refers to a condition in which the wind turbine is in a power generation state and the wind turbine blades are in an optimal pitch angle state set by a main control program to obtain wind energy to the maximum extent;

[0053] S2. Using the DBSCAN spatial density clustering algorithm to screen the full propeller power generation data set D, screen out the normal data cluster N, and obtain the initial lower boundary of the normal data cluster N;

[0054] S3, performing boundary smoothing on the initial lower boundary of the normal data cluster N to obtain a smoothed lower boundary;

[0055] S4. Determine whether the data point of the wind turbine generator set is in a stall condition based on the smoothed lower boundary.

[0056] Specifically, DBSCAN: Density-Based Spatial Clustering of Applications with Noise, is a density-based clustering algorithm that is widely used in the fields of data mining and machine learning. It organizes data by dividing the points in the data set into core points, boundary points, and noise points to form clusters.

[0057] The DBSCAN algorithm is based on two main parameters: ε (neighborhood size) and MinPts (minimum number of points). For each point in the dataset, the algorithm checks whether its ε-neighborhood contains at least MinPts points; if so, the point is considered a core point and may be expanded into a cluster. The algorithm starts from any unvisited core point and builds a cluster by continuously adding core points in its ε-neighborhood until no more core points can be added. Points that are not core points but belong to the ε-neighborhood of a core point are considered to be the boundary points of the cluster, and points that do not belong to any cluster are considered to be noise points.

[0058] Specifically, in step S2, the process of DBSCAN clustering and screening the initial normal data clusters is:

[0059] S201, divide the full-propeller power generation data set D into m bins according to wind speed i ,i=1,2,3...m;

[0060] S202, for each wind speed bin i Sort by active power and perform clustering calculation using ε neighborhood and MinPts minimum number of points as parameters;

[0061] S203, select the cluster with the largest number of samples as the wind speed bin i Normal data cluster N i ;

[0062] S204, all wind speed bins i The normal data cluster set {N1,N2,N3...N m} is defined as the initial normal data cluster N.

[0063] Specifically, the process of determining DBSCAN parameters is:

[0064] (1) Steps for determining the ε neighborhood

[0065] S1.1. The full-propeller power generation data set D is divided into m bins according to wind speed i ,i=1,2,3...m;

[0066] S1.2, for each bin i Calculate the upper and lower limits of active power L up ,L low , filter bin i Active power between L up ,L low The sample between g ij , and get G i :

[0067] G i = {g ij |L low <g ij <L up} (1)

[0068] S1.3, G i Sort in ascending order of active power, then perform difference calculation on the sorted powers to obtain the difference d between two adjacent active power data. ij ,j=1,2,3..., take d ij , j=1,2,3..., we get ε i ;

[0069] S1.4, take the collective {ε i The 80% quantile of |i=1,2,3...,m} is taken as the ε neighborhood value.

[0070] (2) Steps to determine the minimum number of points in MinPts

[0071] S2.1. The full-propeller power generation data set D is divided into m bins according to wind speed. i ,i=1,2,3...m;

[0072] S2.2, for each bin i Calculate the upper and lower limits of active power L up ,L low , upper and lower limit distance R i =L up -L low , total number of samples Count i ;

[0073] S2.3, take R i ,i=1,2,3...m, 80% quantile, get R, let:

[0074]

[0075] S2.4. Get MinPts i , the minimum value of i=1,2,3... is used as the minimum point parameter of MinPts.

[0076] Specifically, the full-propeller power generation data set D is divided into m bins according to wind speed: i ,i=1,2,3...m; let each bin i The average wind speed is v i ; N i The minimum active power is P i_min ; In the wind speed-power rectangular coordinate system, the point (v i ,P i ) is defined as the initial lower boundary L0(v) of the initial normal data cluster N.

[0077] Since the rate of change of the initial lower boundary L0(v) with respect to the wind speed v often changes suddenly, it is visually presented as an obvious "sawtooth" shape. Based on the analysis experience of a large amount of actual operation data of wind farms, it can be found that the normal operation data of wind turbines should have a smooth boundary in the spatial distribution, so the initial lower boundary L0(v) needs to be smoothed.

[0078] The boundary smoothing method is:

[0079] Let the lower boundary L after the nth smoothing ben (v) is:

[0080] L n (v),n=1,2,3... (3)

[0081] The lower boundary of the nth smoothing is calculated from the lower boundary of the n-1th smoothing, that is:

[0082]

[0083] Where L n-1 (v) is the lower boundary curve after the n-1th smoothing, v i For bin i ,i=1,2,3...m average wind speed.

[0084] The number of times the boundary is smoothed depends on the data itself. Generally speaking, after a certain number of boundary smoothings are performed, the change in the lower boundary will become very small if the smoothing process continues.

[0085] Therefore, the cutoff condition for lower boundary smoothing is set as:

[0086]

[0087] Where n cut is the smoothing cutoff number, p cut is a constant.

[0088] Specifically, in step S4, when the data point i is below the lower boundary and i is not a noise point, the data point i is a stall sample, otherwise it is another abnormal sample.

[0089] In the specific application, a mountain wind farm in the south is selected as the sample set. Some information of the wind farm is shown in Table 1.

[0090] Table 1 Basic information of the case wind farm

[0091]

[0092] The wind farm is mainly based on its comprehensive and balanced composition. Specifically, it covers positive samples (i.e. blade stall samples) and negative samples (i.e. blade non-stall samples). This feature is crucial for building an accurate and reliable analysis model. The coexistence of positive and negative samples not only ensures that multiple dimensions of the problem are covered during the analysis process, but also effectively improves the recognition ability and generalization performance of the model when processing actual data.

[0093] Normal unit diagnostic results are as follows Figure 3 As shown, the diagnosis results of the blade stall unit are as follows Figure 4 The red dotted line is the lower boundary of the normal data scatter points, and the scatter points are the data points of fully open propeller power generation.

[0094] Figure 3 The data points of normal units fully opening for power generation are all above the lower boundary, and it can be seen intuitively that the unit has no blade stall phenomenon.

[0095] Figure 4 The data points of fully-opened power generation are close to the rated power of the unit, the power characteristics of the wind turbine unit are significantly reduced, and a large number of data points are below the lower boundary. According to the algorithm diagnosis, the data below the lower boundary are determined to be stall sample data.

[0096] The present invention considers the spatial distribution characteristics of the data, firstly performs preliminary screening on the wind turbine fully-opened power generation data set D through the DBSCAN spatial density clustering algorithm to obtain the initial normal data cluster N, then performs boundary smoothing on the data cluster N to obtain the smoothed lower boundary, and then uses the smoothed lower boundary to determine whether the data point is stalled or not. The above scheme has the following advantages:

[0097] 1) DBSCAN plus boundary smoothing provides a new perspective

[0098] The DBSCAN algorithm is a widely used clustering algorithm. It divides the data points in the data set into several categories based on the spatial distribution density of the data. After initially screening the normal categories, it uses the boundary smoothing method to determine the lower boundary of the normal data, and then identifies whether the unit is stalled based on the lower boundary. Compared with other methods, this scheme is based on the existing wind turbine SCADA data and does not require the addition of new sensors (blade pressure, deformation, torsion angle, etc.); in addition, the DBSCAN plus boundary smoothing method fully explores the distribution law of wind turbine SCADA data, providing a new perspective for blade stall diagnosis.

[0099] 2) Improved the feasibility and accuracy of blade stall diagnosis

[0100] Since this solution does not require the addition of new sensors, there is no need to invest in new hardware costs for both existing and newly built wind farms, so this solution can be deployed and implemented without obstacles in both existing and newly built wind farms, and the solution is highly feasible. Since this solution is based on the distribution rules of a large amount of historical data, the analysis process is clear and each step of the operation has a clear basis and purpose, which helps to improve the accuracy of diagnosis.

[0101] The present invention also discloses a computer program product, including a computer program, which executes the steps of the above method when executed by a processor. The present invention further discloses a computer-readable storage medium, which stores a computer program, which executes the steps of the above method when executed by a processor. The present invention also discloses a wind turbine blade stall diagnosis based on data distribution, including an interconnected memory and a processor, the memory stores a computer program, and the computer program executes the steps of the above method when executed by the processor. The product, medium and system of the present invention correspond to the above method and also have the advantages described in the above method.

[0102] The present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiment when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable storage media include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. The memory is used to store computer programs and / or modules, and the processor implements various functions by running or executing computer programs and / or modules stored in the memory, and calling data stored in the memory. The memory may include high-speed random access memory and may also include non-volatile memory, such as a hard disk, an internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0103] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A wind turbine blade stall diagnosis method based on data distribution, characterized in that: Includes steps: S1, obtaining a wind turbine generator set fully powered-up data set D; S2. Using the DBSCAN spatial density clustering algorithm to screen the full propeller power generation data set D, screen out the normal data cluster N, and obtain the initial lower boundary of the normal data cluster N; S3, performing boundary smoothing on the initial lower boundary of the normal data cluster N to obtain a smoothed lower boundary; S4. Determine whether the data point of the wind turbine generator set is in a stall condition based on the smoothed lower boundary.

2. The wind turbine blade stall diagnosis method based on data distribution according to claim 1, characterized in that: The specific process of step S2 is: S201, divide the full-propeller power generation data set D into m bins according to wind speed i ,i=1,2,3...m; S202, for each wind speed bin i Sort by active power and perform clustering calculation using ε neighborhood and MinPts minimum number of points as parameters; S203, select the cluster with the largest number of samples as the wind speed bin i Normal data cluster N i ; S204, bin all wind speeds i The normal data cluster set {N1,N2,N3...N m } is defined as the initial normal data cluster N.

3. The wind turbine blade stall diagnosis method based on data distribution according to claim 2, characterized in that: In step S201, the specific process of obtaining the ε neighborhood is: S1.

1. The full-propeller power generation data set D is divided into m bins according to wind speed i ,i=1,2,3...m; S1.2, for each bin i Calculate the upper and lower limits of active power L up ,L low , filter bin i Active power between L up ,L low The samples between get: G i ={g ij |L low <g ij <L up } (1) S1.3, G i Sort in ascending order of active power, then perform difference calculation on the sorted powers to obtain the difference d between two adjacent active power data. ij ,j=1,2,3..., take d ij , j=1,2,3..., we get ε i ; S1.4, take the collective {ε i The 80% quantile of |i=1,2,3...,m} is taken as the ε neighborhood value.

4. The wind turbine blade stall diagnosis method based on data distribution according to claim 2, characterized in that: In step S201, the process of obtaining the minimum number of points MinPts is as follows: S2.

1. The full-propeller power generation data set D is divided into m bins according to wind speed. i ,i=1,2,3...m; S2.2, for each bin i Calculate the upper and lower limits of active power L up ,L low , upper and lower limit distance R i =L up -L low , total number of samples Count i ; S2.3, take R i ,i=1,2,3...m, 80% quantile, get R, let: S2.

4. Get MinPts i , the minimum value of i=1,2,3... is used as the minimum point number parameter of MinPts.

5. The wind turbine blade stall diagnosis method based on data distribution according to any one of claims 1 to 4, characterized in that: In step S3, the boundary smoothing process is: Assume the lower boundary after the nth smoothing is: L n (v),n=1,2,3... (3) The lower boundary of the nth smoothing is calculated from the lower boundary of the n-1th smoothing, that is: Where L n-1 (v) is the lower boundary curve after the n-1th smoothing, v i For bin i ,i=1,2,3...m average wind speed.

6. The wind turbine blade stall diagnosis method based on data distribution according to claim 5, characterized in that: The cutoff condition for lower boundary smoothing is: Where n cut is the smoothing cutoff number, p cut is a constant.

7. The wind turbine blade stall diagnosis method based on data distribution according to any one of claims 1 to 4, characterized in that: In step S4, when data point i is below the lower boundary and i is not a noise point, data point i is a stall sample, otherwise it is another abnormal sample.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are performed.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program performs the steps of the method according to any one of claims 1 to 7.

10. A wind turbine blade stall diagnosis system based on data distribution, comprising a memory and a processor connected to each other, wherein a computer program is stored in the memory, characterized in that: When the computer program is executed by a processor, the computer program performs the steps of the method according to any one of claims 1 to 7.

Images