A method and system for identifying fan power characteristic curve

Abstract

The present invention provides a method and system for identifying a wind turbine power characteristic curve. The present invention can identify an abnormal interval by identifying the wind turbine power characteristic curve, and discretize the wind turbine power characteristic curve based on its nonlinear characteristics. The curve is identified by analyzing internal discrete points, and the discrete points are divided into intervals according to wind speed. The abnormal points of the wind turbine are judged by identifying multiple intervals, thereby reducing the amount of data required to process a large number of discrete points. In addition, by constructing an operating status diagram, the working conditions of the wind turbine at different wind speeds can be easily seen, and abnormal points or abnormal intervals can be intuitively identified.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power analysis, and in particular to a method and system for identifying a wind turbine power characteristic curve. Background Art

[0002] The wind speed input to a wind turbine is the main factor affecting its output active power. Therefore, the internal characteristics of the wind turbine can be ignored, and only the relationship between the input wind speed and the output active power is considered. The curve that describes the corresponding relationship between wind speed and the output active power of the wind turbine is called the power characteristic curve of the wind turbine. The actual operating power curve of the wind turbine is obtained based on the actual operating data of the wind turbine collected and recorded during operation. Generally speaking, the working conditions of the wind turbine at different wind speeds can be identified through the power characteristic curve, but the power characteristic curve of the wind turbine is nonlinear, so tiny abnormal points or abnormal sections of the curve are difficult to detect directly by observing the characteristic curve. Therefore, it is necessary to identify the power characteristic curve and determine the abnormal points of the wind turbine to facilitate subsequent maintenance.

[0003] The identification of traditional fan power characteristic curve is based on recorded data, but the recorded data can only be triggered under major faults, which makes it difficult to meet the needs of online identification of different working conditions. The current research on fan power characteristic curve mainly focuses on the correction and calibration of fan power characteristic curve. For example, a fan power characteristic curve identification method and system provided by publication number CN105740595A adopts fan operating power curve segmentation and genetic algorithm calculation, which can significantly improve the conformity between the fitting curve and the actual curve, so that the obtained fan power curve can more accurately reflect the fan characteristics. However, it is still difficult to directly identify abnormal points and abnormal wind speed ranges through the calibrated curve. Therefore, it is very necessary to develop a method for directly identifying abnormal points of fan operation through the curve for the fan power characteristic curve. Summary of the Invention

[0004] The object of the present invention is to provide a method and system for identifying a wind turbine power characteristic curve to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for identifying a wind turbine power characteristic curve, wherein the identification method comprises:

[0007] S1: Divide the power characteristic curve into several segments according to wind speed, and discretize each segment into sets of the same number of points. Each set is:

[0008] C i ={…(v ij-1 、p ij-1)、(v ij 、p ij )、(v ij+1 、p ij+1 )…}

[0009] Where C i is the discrete point set of the i-th wind speed interval, v ij 、p ij represent the wind speed and power of the jth point in the i-th wind speed interval respectively;

[0010] S2: Calculate the average power value within a wind speed interval. There are n sets of data in each wind speed interval. The average power value within the wind speed interval is:

[0011]

[0012] Where, is the average power value in the i-th wind speed interval;

[0013] According to the power average value, the deviation value of each point in each wind speed interval from the power average value and the standard deviation of each wind speed interval are calculated as follows:

[0014]

[0015]

[0016] Where Δp ij is the deviation value of the jth point in the i-th wind speed interval, σ i is the standard deviation of the i-th wind speed interval;

[0017] S3: Compare the deviation value of each point in each wind speed interval with the standard deviation of the corresponding interval, and eliminate abnormal data in each interval. The elimination criteria are:

[0018] |Δp ij |>2σ i ;

[0019] S4: Based on the Logistic curve, a theoretical power equation is constructed in each wind speed space. The constructed theoretical power equation is:

[0020]

[0021] Where, is the theoretical power of the jth point in the i-th wind speed interval, α, β, γ, and τ are the vector parameters of the theoretical power equation respectively;

[0022] S5: For each wind speed range, multiple groups of samples are selected. Each sample number constitutes a sample set. Each sample set consists of four sample points, and the number of sample points is the same, that is:

[0023] Y im ={(v ij-ε 、p ij-ε )、(v ij 、p ij )、(v ij+ε 、p ij+ε )、(v ij+2ε 、p ij+2ε )}

[0024] Where Y im represents the mth sample point set in the i-th wind speed interval, v ij-ε 、p ij-ε The wind speed and power of the j-εth point in the i-th wind speed interval, where ε represents the number of sample points.

[0025] S6: Substitute each sample set into the theoretical power equation, solve the values ​​of the α, β, γ, and τ vector parameters corresponding to each set, and calculate the average of multiple α, β, γ, and τ vector parameters. The average obtained is α i , β i , γ i , τ i , where α i , β i , γ i , τ i are respectively expressed as vector parameters of the theoretical power equation in the i-th wind speed interval;

[0026] S7: Calculate the average residual and standard deviation of each wind speed interval, and calculate the state upper limit and state lower limit of the entire power curve through the average residual and standard deviation. The calculation formula of the average residual and standard deviation is:

[0027]

[0028] Where, are the average residual and standard deviation in the i-th wind speed interval respectively. The calculation formula for the state upper limit and state lower limit of the entire power curve is:

[0029]

[0030] Where H and L are the upper and lower limits of the entire power curve, respectively, and N is the number of separated wind speed intervals.

[0031] S8: Construct an operating status diagram, use each wind speed interval as the horizontal axis, sort from low to high according to the average wind speed of each wind speed interval, and draw the average residual, state upper limit and state lower limit of each wind speed interval as the vertical axis in a diagram. If the average residual of the corresponding wind speed interval exceeds the state upper limit or state lower limit, it is judged that the operating status of the wind turbine in the wind speed interval is abnormal.

[0032] Preferably, the wind speed differences between adjacent discrete points in each wind speed interval in step S1 are the same, that is:

[0033] Δv i =v ij+1 -v ij =v ij -v ij-1

[0034] Where Δv i is the wind speed difference between adjacent discrete points in the i-th wind speed interval.

[0035] Preferably, the value of abnormal data removed in each wind speed interval in step S3 is the same. If the number of abnormal data removed in a certain wind speed interval is small, then the value of abnormal data removed in step S3 is the same as that in step S3. ij | Eliminate abnormal points in descending order until the number of eliminated abnormal points in each wind speed range is the same.

[0036] Preferably, the number intervals between the sample sets selected in the same wind speed interval in step S5 are all different, and the number of samples selected in each wind speed interval is not less than five groups.

[0037] Preferably, when calculating the average of multiple α, β, γ, and τ vector parameters in step S6, the highest and lowest values ​​of the corresponding vector parameters are removed, and the average of the remaining values ​​is calculated to obtain α i , β i , γ i , τ i The numerical value of .

[0038] The present invention further provides a system for identifying a wind turbine power characteristic curve. The system is applicable to the above-mentioned method for identifying a wind turbine power characteristic curve. The system comprises:

[0039] The curve separation and discretization module is used to separate the power characteristic curve into several curve segments according to the wind speed, and discretize each curve segment into a set of the same number of points;

[0040] The difference calculation module is used to obtain the average power value within a wind speed interval and, based on the average power value, calculate the deviation value of each point in each wind speed interval from the average power value and the standard deviation of each wind speed interval;

[0041] The abnormal data elimination module is used to compare the deviation value of each point in each wind speed interval with the standard deviation of the corresponding interval. If the absolute value of the deviation value is greater than twice the standard deviation, it is judged as abnormal data and eliminated;

[0042] Theoretical power equation construction module, used to construct theoretical power equations in various wind speed spaces based on the Logistic curve;

[0043] The sample selection module is used to select multiple groups of samples for each wind speed range. Each sample number constitutes a sample set. Each sample set consists of four sample points, and the number of sample points between each sample point is the same.

[0044] The vector parameter solving module is used to substitute each sample set into the theoretical power equation, solve the value of the vector parameter corresponding to each set, calculate the average of multiple vector parameters, and use the average as the vector parameter of the theoretical power equation in the corresponding wind speed range;

[0045] The upper and lower limit calculation module is used to calculate the average residual and standard deviation of each wind speed interval, and calculate the upper and lower limits of the state of the entire power curve through the average residual and standard deviation;

[0046] The operating status diagram construction module is used to use each wind speed interval as the horizontal axis, sort the wind speed intervals from low to high according to the average wind speed, and plot the average residual, state upper limit and state lower limit of each wind speed interval as the vertical axis in a graph. If the average residual of the corresponding wind speed interval exceeds the state upper limit or state lower limit, it is judged that the operating status of the wind turbine in the wind speed interval is abnormal.

[0047] Preferably, the wind speed differences between adjacent discrete points in each wind speed interval constructed by the curve separation discrete module are the same.

[0048] Preferably, the abnormal data elimination module eliminates the same abnormal data value in each wind speed interval. If the number of abnormal data eliminated in a certain wind speed interval is small, the abnormal data are eliminated in descending order of the absolute value of the deviation value until the number of abnormal points eliminated in each wind speed interval is the same.

[0049] Preferably, the sample sets selected by the sample selection module in the same wind speed interval are different in number and spacing between them, and the number of samples selected in each wind speed interval is not less than five groups.

[0050] Preferably, when the vector parameter solving module calculates the average of multiple vector parameters, the highest value and the lowest value of the corresponding vector parameters are removed, and the average of the remaining values ​​is calculated to obtain the average value of the corresponding vector parameters.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present invention can identify abnormal intervals by identifying the wind turbine power characteristic curve, and discretize the wind turbine power characteristic curve according to its nonlinear characteristics, identify the curve by analyzing internal discrete points, and divide the discrete points into intervals according to wind speed. The abnormal points of the wind turbine can be judged by identifying multiple intervals, thereby reducing the amount of data required to process a large number of discrete points. Moreover, by constructing an operating status diagram, the working conditions of the wind turbine at different wind speeds can be easily seen, and abnormal points or abnormal intervals can be intuitively identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 Schematic diagram of the process of the identification method of the present invention;

[0054] Figure 2 Schematic diagram of the structure of the identification system of the present invention. DETAILED DESCRIPTION

[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0056] Example:

[0057] See also Figure 1 , the present invention provides a technical solution:

[0058] A method for identifying a wind turbine power characteristic curve, wherein the identification method comprises:

[0059] S1: Divide the power characteristic curve into several segments according to wind speed. Each segment corresponds to a wind speed range. Discrete each segment into sets of the same number of points. Each set is:

[0060] C i ={…(v ij-1 、p ij-1 )、(v ij 、p ij )、(v ij+1 、p ij+1 )…}

[0061] Where C i is the discrete point set of the i-th wind speed interval, v ij 、p ijRespectively represent the wind speed and power of the jth point in the i-th wind speed interval. Among them, the wind speed difference of adjacent discrete points in each wind speed interval is the same, that is:

[0062] Δv i =v ij+1 -v ij =v ij -v ij-1

[0063] Where Δv i is the wind speed difference between adjacent discrete points in the i-th wind speed interval; the width of the wind speed interval corresponding to each curve segment can be the same or different. The denser the curve separation, the denser the corresponding wind speed space, and the more accurate the subsequent determination of the different curve segments.

[0064] By dividing the power characteristic curve into segments and discretizing them into a number of points, a second-order analysis sample is formed. First, there are samples of multiple intervals, and second, there are samples of points in each wind speed interval. Different from the nonlinear curve that is difficult to analyze, this embodiment makes it possible to identify the entire curve and reduces the amount of data required to process a large number of discrete points.

[0065] S2: Calculate the average power value within a wind speed range. Each wind speed range has n groups of data, i.e., n data points. The average power value within the wind speed range is:

[0066]

[0067] Where, is the average power value in the i-th wind speed interval;

[0068] According to the power average value, the deviation value of each point in each wind speed interval from the power average value and the standard deviation of each wind speed interval are calculated as follows:

[0069]

[0070] Where Δp ij is the deviation value of the jth point in the i-th wind speed interval, σ i is the standard deviation of the i-th wind speed interval.

[0071] Among them, the above steps are mainly for identifying abnormal data within the wind speed range. Abnormal data is not abnormal data of the fan operation, but abnormally collected data, or erroneous data. This step is mainly used to identify erroneous data to facilitate subsequent clearing, so that the validity of the data is guaranteed.

[0072] S3: Compare the deviation value of each point in each wind speed interval with the standard deviation of the corresponding interval, and eliminate abnormal data in each interval. The elimination criteria are:

[0073] |Δp ij |>2σ i .

[0074] Among them, the value of abnormal data removed in each wind speed interval is the same. If the number of abnormal data removed in a certain wind speed interval is small, then according to |Δp ij Eliminate abnormal points in descending order until the same number of points are eliminated in each wind speed interval, ensuring the coordination of data within each interval. By eliminating abnormal data within the wind speed interval, the validity of the data is guaranteed and erroneous data is avoided from being analyzed as abnormal data.

[0075] S4: Based on the Logistic curve, a theoretical power equation is constructed in each wind speed space. The constructed theoretical power equation is:

[0076]

[0077] Where, is the theoretical power of the jth point in the i-th wind speed interval. α, β, γ, and τ are the vector parameters of the theoretical power equation, which determine the shape of the curve corresponding to the theoretical power equation. By constructing the Logistic curve, each point is reparameterized. The reparameterized nonlinear function not only retains the S-shaped nonlinear characteristics, but also reduces the intensity of nonlinearity in some cases, which can play an important role in improving the fitting ability and fitting effect of the theoretical power of each point.

[0078] S5: For each wind speed range, multiple groups of samples are selected. Each sample number constitutes a sample set. Each sample set consists of four sample points, and the number of sample points is the same, that is:

[0079] Y im ={(v ij-ε 、p ij-ε )、(v ij 、p ij )、(v ij+ε 、p ij+ε )、(v ij+2ε 、p ij+2ε )};

[0080] Where Y im represents the mth sample point set in the oth wind speed interval, v ij-ε 、p ij-ε The wind speed and power of the j-εth point in the i-th wind speed interval, where ε represents the number of sample points.

[0081] Among them, the number intervals between sample sets selected in the same wind speed range are different, and the number of samples selected in each wind speed range is not less than five groups.

[0082] S6: Substitute each sample set into the theoretical power equation, solve the values ​​of the α, β, γ, and τ vector parameters corresponding to each set, and calculate the average of multiple α, β, γ, and τ vector parameters. The average obtained is α i , β i , γ i , τ i , where α i , β i , γ i , τ i are respectively represented as vector parameters of the theoretical power equation in the i-th wind speed interval.

[0083] When calculating the average of multiple α, β, γ, and τ vector parameters, the highest and lowest values ​​of the corresponding vector parameters are removed, and the remaining values ​​are averaged to obtain α. i , β i , γ i , τ i The value of α is obtained by using statistical methods i , β i , γ i , τ i The value of α can minimize the impact of outliers on the appropriate parameters. i , β i , γ i , τ i The value is closer to the actual value, reducing the impact of abnormal points.

[0084] S7: Calculate the average residual and standard deviation of each wind speed interval, and calculate the state upper limit and state lower limit of the entire power curve through the average residual and standard deviation. The calculation formula of the average residual and standard deviation is:

[0085]

[0086]

[0087] Where, are the average residual and standard deviation in the i-th wind speed interval respectively. The calculation formula for the state upper limit and state lower limit of the entire power curve is:

[0088]

[0089] Where H and L are the upper and lower limits of the entire power curve, respectively, and N is the number of separated wind speed intervals.

[0090] Among them, the average residual can reflect the degree of deviation of a certain wind speed range relative to the entire wind turbine power curve. The state upper limit and state lower limit of the entire power curve can be regarded as the deviation threshold of the entire wind turbine power curve. Once the average residual of a certain wind speed range exceeds the state upper limit or state lower limit, the wind speed range is considered to be an abnormal segment.

[0091] S8: Construct an operating status diagram, use each wind speed interval as the horizontal axis, sort from low to high according to the average wind speed of each wind speed interval, and draw the average residual, state upper limit and state lower limit in each wind speed interval as the vertical axis in a diagram. If the average residual of the corresponding wind speed interval exceeds the state upper limit or state lower limit, it is judged that the operating status of the wind turbine in the wind speed interval is abnormal.

[0092] By constructing an operating status diagram, it is very convenient to see the working conditions of the fan at different wind speeds, intuitively identify abnormal points or abnormal intervals, and directly identify the abnormal intervals of the fan power characteristic curve. If the wind speed interval of this embodiment is set small enough, the interval identification will be more accurate, but the corresponding calculation amount will increase. The width of the wind speed interval can be set according to actual needs.

[0093] See also Figure 2 The present invention further provides a system for identifying a wind turbine power characteristic curve. The system is applicable to the above-mentioned method for identifying a wind turbine power characteristic curve. The system comprises:

[0094] The curve separation discrete module is used to separate the power characteristic curve into several curve segments according to the wind speed, and discretize each curve segment into a set of the same number of points, wherein the wind speed difference between adjacent discrete points in each wind speed interval constructed by the curve separation discrete module is the same.

[0095] The difference calculation module is used to obtain the power average value within a wind speed interval and, based on the power average value, calculate the deviation value of each point in each wind speed interval from the power average value and the standard deviation of each wind speed interval.

[0096] The abnormal data elimination module is used to compare the deviation value of each point in each wind speed interval with the standard deviation of the corresponding interval. If the absolute value of the deviation value is greater than twice the standard deviation, it is judged as abnormal data and eliminated. The abnormal data elimination module eliminates the same abnormal data value in each wind speed interval. If the number of abnormal data eliminated in a certain wind speed interval is small, the data are eliminated in descending order according to the absolute value of the deviation value until the number of abnormal points eliminated in each wind speed interval is the same.

[0097] The theoretical power equation construction module is used to construct the theoretical power equation in each wind speed space based on the Logistic curve.

[0098] The sample selection module is used to select multiple groups of samples for each wind speed interval, each sample number constitutes a sample set, each sample set consists of four sample points, and the number intervals between each sample point are the same, wherein the sample sets selected by the sample selection module in the same wind speed interval are different, and the number of samples selected in each wind speed interval is not less than five groups.

[0099] The vector parameter solving module is used to substitute each sample set into the theoretical power equation, solve the value of the vector parameter corresponding to each set, calculate the average of multiple vector parameters, and use the obtained average as the vector parameter of the theoretical power equation in the corresponding wind speed range. When the vector parameter solving module calculates the average of multiple vector parameters, the highest and lowest values ​​of the corresponding vector parameters are removed, and the remaining values ​​are averaged to obtain the average value of the corresponding vector parameters.

[0100] The upper and lower limit calculation module is used to calculate the average residual and standard deviation of each wind speed interval, and calculate the state upper limit and state lower limit of the entire power curve through the average residual and standard deviation.

[0101] The operating status diagram construction module is used to use each wind speed interval as the horizontal axis, sort the wind speed intervals from low to high according to the average wind speed, and plot the average residual, state upper limit and state lower limit of each wind speed interval as the vertical axis in a graph. If the average residual of the corresponding wind speed interval exceeds the state upper limit or state lower limit, it is judged that the operating status of the wind turbine in the wind speed interval is abnormal.

[0102] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for identifying a wind turbine power characteristic curve, characterized in that: The identification method includes: S1: Divide the power characteristic curve into several segments according to wind speed, and discretize each segment into sets of the same number of points. Each set is: C i ={…(v ij-1 、p ij-1 )、(v ij 、p ij )、(v ij+1 、p ij+1 )…} Where C i is the discrete point set of the i-th wind speed interval, v ij 、p ij represent the wind speed and power of the jth point in the i-th wind speed interval respectively; S2: Calculate the average power value within a wind speed interval. There are n sets of data in each wind speed interval. The average power value within the wind speed interval is: Where, is the average power value in the i-th wind speed interval; According to the power average value, the deviation value of each point in each wind speed interval from the power average value and the standard deviation of each wind speed interval are calculated as follows: Where Δp ij is the deviation value of the jth point in the i-th wind speed interval, σ i is the standard deviation of the i-th wind speed interval; S3: Compare the deviation value of each point in each wind speed interval with the standard deviation of the corresponding interval, and eliminate abnormal data in each interval. The elimination criteria are: |Δp ij |>2σ i ; S4: Based on the Logistic curve, a theoretical power equation is constructed in each wind speed space. The constructed theoretical power equation is: Where, is the theoretical power of the jth point in the i-th wind speed interval, α, β, γ, and τ are the vector parameters of the theoretical power equation respectively; S5: For each wind speed range, multiple groups of samples are selected. Each sample number constitutes a sample set. Each sample set consists of four sample points, and the number of sample points is the same, that is: Y im ={(v ij-ε 、p ij-ε )、(v ij 、p ij )、(v ij+ε 、p ij+ε )、(v ij+2ε 、p ij+2ε )} Where Y im represents the mth sample point set in the i-th wind speed interval, v ij-ε 、p ij-ε The wind speed and power of the j-εth point in the i-th wind speed interval, where ε represents the number of sample points. S6: Substitute each sample set into the theoretical power equation, solve the values ​​of the α, β, γ, and τ vector parameters corresponding to each set, and calculate the average of multiple α, β, γ, and τ vector parameters. The average obtained is α i , β i , γ i , τ i , where α i , β i , γ i , τ i are respectively expressed as vector parameters of the theoretical power equation in the i-th wind speed interval; S7: Calculate the average residual and standard deviation of each wind speed interval, and calculate the state upper limit and state lower limit of the entire power curve through the average residual and standard deviation. The calculation formula of the average residual and standard deviation is: Where, are the average residual and standard deviation in the i-th wind speed interval respectively. The calculation formula for the state upper limit and state lower limit of the entire power curve is: Where H and L are the upper and lower limits of the entire power curve, respectively, and N is the number of separated wind speed intervals. S8: Construct an operating status diagram, use each wind speed interval as the horizontal axis, sort from low to high according to the average wind speed of each wind speed interval, and draw the average residual, state upper limit and state lower limit of each wind speed interval as the vertical axis in a diagram. If the average residual of the corresponding wind speed interval exceeds the state upper limit or state lower limit, it is judged that the operating status of the wind turbine in the wind speed interval is abnormal.

2. The method for identifying a wind turbine power characteristic curve according to claim 1, wherein: In step S1, the wind speed differences between adjacent discrete points in each wind speed interval are the same, that is: Δv i =v ij+1 -v ij =v ij -v ij-1 Where Δv i is the wind speed difference between adjacent discrete points in the i-th wind speed interval.

3. The method for identifying a wind turbine power characteristic curve according to claim 1, wherein: In step S3, the values ​​of abnormal data removed in each wind speed interval are the same. If the number of abnormal data removed in a certain wind speed interval is small, then the value of abnormal data removed in step S3 is the same as that in step S3. ij | Eliminate abnormal points in descending order until the number of eliminated abnormal points in each wind speed range is the same.

4. The method for identifying a wind turbine power characteristic curve according to claim 1, wherein: In step S5, the number of sample sets selected in the same wind speed interval and the intervals between them are all different, and the number of samples selected in each wind speed interval is no less than five groups.

5. The method for identifying a wind turbine power characteristic curve according to claim 4, wherein: When calculating the average of multiple α, β, γ, and τ vector parameters in step S6, the highest and lowest values ​​of the corresponding vector parameters are removed, and the average of the remaining values ​​is calculated to obtain α. i , β i , γ i , τ i The numerical value of .

6. A system for identifying a fan power characteristic curve, characterized by: The identification system is applicable to the method for identifying a wind turbine power characteristic curve according to any one of claims 1 to 5, and the identification system comprises: The curve separation and discretization module is used to separate the power characteristic curve into several curve segments according to the wind speed, and discretize each curve segment into a set of the same number of points; The difference calculation module is used to obtain the average power value within a wind speed interval and, based on the average power value, calculate the deviation value of each point in each wind speed interval from the average power value and the standard deviation of each wind speed interval; The abnormal data elimination module is used to compare the deviation value of each point in each wind speed interval with the standard deviation of the corresponding interval. If the absolute value of the deviation value is greater than twice the standard deviation, it is judged as abnormal data and eliminated; Theoretical power equation construction module, used to construct theoretical power equations in various wind speed spaces based on the Logistic curve; The sample selection module is used to select multiple groups of samples for each wind speed range. Each sample number constitutes a sample set. Each sample set consists of four sample points, and the number of sample points between each sample point is the same. The vector parameter solving module is used to substitute each sample set into the theoretical power equation, solve the value of the vector parameter corresponding to each set, calculate the average of multiple vector parameters, and use the average as the vector parameter of the theoretical power equation in the corresponding wind speed range; The upper and lower limit calculation module is used to calculate the average residual and standard deviation of each wind speed interval, and calculate the upper and lower limits of the state of the entire power curve through the average residual and standard deviation; The operating status diagram construction module is used to use each wind speed interval as the horizontal axis, sort the wind speed intervals from low to high according to the average wind speed, and plot the average residual, state upper limit and state lower limit of each wind speed interval as the vertical axis in a graph. If the average residual of the corresponding wind speed interval exceeds the state upper limit or state lower limit, it is judged that the operating status of the wind turbine in the wind speed interval is abnormal.

7. The system for identifying a wind turbine power characteristic curve according to claim 6, characterized in that: The wind speed differences between adjacent discrete points in each wind speed interval constructed by the curve separation discrete module are the same.

8. The system for identifying a wind turbine power characteristic curve according to claim 6, characterized in that: The abnormal data elimination module eliminates the same abnormal data value in each wind speed interval. If there are fewer abnormal data eliminated in a certain wind speed interval, the abnormal data are eliminated in descending order of the absolute value of the deviation value until the number of abnormal points eliminated in each wind speed interval is the same.

9. The system for identifying a wind turbine power characteristic curve according to claim 6, characterized in that: The sample sets selected by the sample selection module in the same wind speed interval and the number intervals between them are all different, and the number of samples selected in each wind speed interval is not less than five groups.

10. The system for identifying a wind turbine power characteristic curve according to claim 9, characterized in that: When the vector parameter solving module calculates the average of the plurality of vector parameters, the highest value and the lowest value of the corresponding vector parameters are removed, and the average of the remaining values ​​is calculated to obtain the average value of the corresponding vector parameters.

Images