CN116557230A - A method and system for online evaluation of power abnormalities of wind farm units - Google Patents

Abstract

The invention discloses a wind farm unit power abnormality online assessment method and system, comprising the following steps: dividing wind speed steady-state working condition data, taking the intermediate value of a wind speed interval as the active power value of a fan, and obtaining a power curve of the active power of each single fan through the divided wind speed interval; further obtaining a power curve of active power of the full-field cluster unit; and judging the relation between the relative deviation of the active power of the single fan and the active power of the cluster unit in each wind speed interval and the threshold value, and judging the abnormal level of the fan power. According to the invention, on the basis of not adding additional sensors, fan power abnormality alarm, quantitative evaluation of the degradation degree of the power generation performance and identification of the power generation low-efficiency unit are realized, and the period of fan power abnormality evaluation and low-efficiency unit identification is shortened. The wind turbine generator system with abnormal power can be positioned, so that maintenance personnel can conveniently overhaul the wind turbine generator system, and the operation and maintenance cost of the wind turbine generator system for fixed inspection is reduced.

Description

A method and system for online evaluation of power abnormalities of wind farm units

technical field

The invention belongs to the technical field of wind power generation, and relates to an online evaluation method and system for abnormal power of a wind farm unit.

Background technique

After a period of operation of wind turbines, their power generation performance will be degraded to varying degrees due to the influence of changes in surrounding environmental conditions, the design structure of their own body, and the service life of components. Evaluating the power generation performance of wind turbines mainly depends on comparing the difference between the actual power curve of the wind turbine and the designed wind power curve. The acquisition of the actual power curve of the wind turbine generally requires on-site wind turbine power curve testing. During the wind turbine power curve test, it is necessary to use a large number of electrical measurement equipment such as airborne radar, power transmitter, and current sensor to obtain real-time meteorological data and fan power data around the wind turbine. The measurement results are affected by changes in meteorological conditions, and it takes a long time. It takes time to measure the full wind speed section, and the purchase and installation costs of a large number of measuring equipment are relatively high. Obviously, the wind turbine power curve test has a long test cycle and high test cost, which cannot meet the requirements of real-time online evaluation of wind turbine power generation performance.

Contents of the invention

The purpose of the present invention is to solve the problems of long test period and high test cost in the power curve test of wind turbines in the prior art, which cannot meet the real-time online evaluation requirements of wind turbine power generation performance, and provide an online evaluation method and system for abnormal power of wind farms .

In order to achieve the above object, the present invention adopts the following technical solutions to achieve:

An online evaluation method for abnormal power of wind farm units, comprising:

Obtain the operation data of the wind turbines in the whole field, and preprocess the operation data of the wind turbines in the whole field, and obtain the wind speed and steady-state working condition data;

Divide the wind speed steady-state working condition data to obtain several wind speed intervals;

Based on the active power of the fan in the wind speed interval, the active power value of the fan corresponding to the middle value of the wind speed interval is obtained;

Based on the divided wind speed intervals, obtain the active power values of the fans at each wind speed in the full wind speed range of n fans in the whole field, and then obtain the power curve of the active power of each single fan;

Based on the average value of active power of n wind turbines in each same wind speed range, the power curve of the active power of the whole field cluster unit is obtained;

Judging whether the relative deviation between the active power of a single fan and the active power of the cluster unit in each wind speed range is greater than the threshold, if it is greater, it is normal; if it is less than, based on the size of the relative deviation, the abnormal level of fan power is judged.

A further improvement of the present invention is:

Further, the operation data of the wind turbines in the whole field include: wind speed, wind direction, active power of wind turbines, generator speed, impeller speed and yaw angle per unit time, and the operation data of each measuring point is indexed in chronological order to form a A multidimensional timing matrix.

Further, preprocess the operating data of the wind turbines in the whole field to obtain the wind speed and steady-state working condition data, specifically: eliminate outliers and abnormal values and filter the steady-state working condition data;

Eliminate outliers and abnormal values, specifically: based on the direct screening method, eliminate the operating data of wind turbines below the cut-in wind speed and above the cut-out wind speed, and remove the operating data of the active power of the fan, the speed of the generator, and the speed of the impeller that are less than zero.

Further, the steady-state working condition data is screened, specifically: based on the quartile method and the 3σ criterion method, the steady-state working condition data is screened; the calculation method of the quartile method processing data is as follows:

Divide the multi-dimensional time-series matrix from the minimum wind speed to the maximum wind speed at an interval of 0.1m/s, and sort the active power values of the fans in each interval from small to large to obtain a power value sequence [p ₁ , p ₂ , p ₃ ,..., p _n ];

Calculate the lower quartile Q ₁ , upper quartile Q ₃ , and interquartile range I _QR :

Set the normal value range of the operating data to [N ₁ , N ₂ ], the lower bound N ₁ of the normal value interval, and the upper bound N ₂ of the normal value interval are calculated according to formula (4):

If the operating data is not within the normal range, it is considered as an abnormal value and eliminated;

The screening of steady-state working condition data based on the 3σ criterion method is specifically: dividing the wind speed from the minimum value to the maximum value in intervals of 0.1m/s, and for the operation data in each wind speed interval, it is assumed that the fan operation data x is subject to a normal distribution, then

R(|x-μ|>3σ)≤0.003 (5)

In the formula, μ and σ represent the mathematical expectation and standard deviation of the normal population, respectively; data values greater than μ+3σ or less than μ-3σ are regarded as outliers and eliminated.

Further, divide the wind speed steady-state working condition data to obtain several wind speed intervals, specifically:

Divide the wind speed steady-state working condition data at fixed intervals from the minimum wind speed to the maximum wind speed, and sort the active power values of the fans in each interval from small to large to obtain a power value sequence [P ₁ , P ₂ , P ₃ , . . . , P _n ].

Further, based on the active power of the fan in the wind speed interval, the active power value of the fan corresponding to the middle value of the wind speed interval is obtained, specifically: taking the average value of the active power of the fan in the wind speed interval as the active power of the fan corresponding to the middle value of the wind speed interval value.

Further, based on the active power of the fan in the wind speed interval, the active power value of the fan corresponding to the middle value of the wind speed interval is obtained, which also includes: for the absence of active power data of the fan in some wind speed intervals, based on the interpolation method, the wind speed The active power value of the fan corresponding to the wind speed value in the center of the interval; the calculation formula is as follows:

Among them, P _c is the active power value of the fan obtained by interpolation, P ₂ and P ₁ are the active power values of the fan at the upper and lower wind speed intervals corresponding to the missing value of active power, and v ₂ , v ₁ , and v _c are the values corresponding to the missing value of active power. The upper bound value, lower bound value and central value of the wind speed range.

Further, the threshold is specifically:

Among them, P _i is the active power of a single fan, It is the cluster power of n wind turbines in the whole field under the same wind field, the same model, and the same wind speed.

Further, based on the size of the relative deviation, determine the abnormal level of the fan power, specifically: set the active power relative deviation threshold as a; set the fan power abnormality as three levels; respectively, abnormal level three, abnormal level two and abnormal level one ;The alarm thresholds of the third level of abnormality, the second level of abnormality and the first level of abnormality are that the relative deviation between the power of a single unit and the active power of a cluster unit is lower than a, b, and c respectively; among them, a>b>c;

An online evaluation system for abnormal power of wind farm units, including:

A pre-processing module, the pre-processing module obtains the operation data of the wind turbines in the whole field, and preprocesses the operation data of the wind turbines in the whole field, and obtains the wind speed steady-state working condition data;

A division module, the division module divides the wind speed steady-state working condition data to obtain several wind speed intervals;

A first acquisition module, the first acquisition module acquires the active power value of the fan corresponding to the middle value of the wind speed interval based on the active power of the fan in the wind speed interval;

The second acquisition module, based on the divided wind speed interval, the second acquisition module acquires the fan active power values of the n fans in the whole field at each wind speed in the full wind speed segment, and then acquires the active power of each single fan curve;

The third acquisition module, the third acquisition module obtains the power curve of the active power of the cluster units in the whole field based on the average value of active power values of n fans in each same wind speed interval;

Judging module, the judging module judges whether the relative deviation between the active power of a single fan and the active power of the cluster unit at each wind speed is greater than a threshold value, if greater, it is normal; exception level.

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

The present invention divides the wind speed steady-state working condition data, and uses the middle value of the wind speed interval as the active power value of the fan, and obtains the power curve of the active power of each single fan through the divided wind speed interval; and then obtains the cluster unit of the whole field The power curve of the active power; judge the relationship between the relative deviation and the threshold value of the active power of a single fan and the active power of the cluster unit in each wind speed range, and judge the abnormal level of the fan power. The invention has low cost, realizes abnormal fan power alarm, quantitative evaluation of power generation performance degradation degree and identification of low-efficiency power generation units without adding additional sensors, and has the advantages of real-time online, fast and accurate, and greatly reduces fan power abnormalities Cycle of assessment and identification of inefficient units. The invention can locate wind turbines with abnormal power, which is convenient for accurately guiding the operation and maintenance personnel to timely check and repair the wind turbines, and effectively reduces the operation and maintenance cost of the regular inspection of the units.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and thus It should be regarded as a limitation on the scope, and those skilled in the art can also obtain other related drawings based on these drawings without creative work.

Fig. 1 is a kind of schematic flow chart of the abnormal online evaluation method of wind farm unit power of the present invention;

Fig. 2 is another kind of flow chart of the abnormal online evaluation method of wind farm unit power of the present invention;

Fig. 3 is the comparison diagram before and after the operation data preprocessing of the present invention;

Fig. 4 is the comparison chart of power curve abnormal unit and cluster power of the present invention; Wherein, Fig. 4 (a) is the curve comparison chart of the organic power of No. 3 blower fan and the organic power of cluster; Fig. 4 (b) is the graph of No. 7 blower fan The curve comparison graph of organic power and the organic power of the cluster; Figure 4(c) is the curve comparison graph of the organic power of No. 11 fan and the organic power of the cluster; The curve comparison diagram of power; Fig. 4 (e) is the curve contrast diagram of the organic power of No. 51 fan and the organic power of cluster; Fig. 4 (f) is the curve comparison diagram of the organic power of No. 62 fan and the organic power of cluster;

Fig. 5 is a schematic structural diagram of an online evaluation system for abnormal power of wind farm units according to the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. The components of the embodiments of the invention generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations.

Accordingly, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

It should be noted that like numerals and letters denote similar items in the following figures, therefore, once an item is defined in one figure, it does not require further definition and explanation in subsequent figures.

In the description of the embodiments of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", "horizontal", "inside" etc. is based on the orientation or positional relationship shown in the drawings , or the orientation or positional relationship that the product of the invention is usually placed in use is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation or be constructed in a specific orientation and operation, and therefore should not be construed as limiting the invention. In addition, the terms "first", "second", etc. are only used for distinguishing descriptions, and should not be construed as indicating or implying relative importance.

In addition, when the term "horizontal" appears, it does not mean that the part is required to be absolutely horizontal, but may be slightly inclined. For example, "horizontal" only means that its direction is more horizontal than "vertical", and it does not mean that the structure must be completely horizontal, but can be slightly inclined.

In the description of the embodiments of the present invention, it should also be noted that, unless otherwise specified and limited, the terms "setting", "installation", "connection" and "connection" should be interpreted in a broad sense, for example, It can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediary, and it can be the internal communication of two components. Those of ordinary skill in the art can understand the specific meanings of the above terms in the present invention according to specific situations.

The present invention is described in further detail below in conjunction with accompanying drawing:

Referring to Fig. 1, the present invention discloses an online evaluation method for abnormal power of wind farm units, including:

S101. Obtain the operation data of the wind turbines in the whole field, and perform preprocessing on the operation data of the wind turbines in the whole field, and obtain the wind speed steady-state working condition data.

The operation data of wind turbines in the whole field includes: wind speed, wind direction, active power of wind turbine, generator speed, impeller speed and yaw angle per unit time, and the operation data of each measuring point is indexed in chronological order to form a multi-dimensional time series matrix.

Preprocess the operating data of the wind turbines in the whole field to obtain the wind speed and steady-state working condition data, specifically: eliminate outliers and abnormal values and filter the steady-state working condition data;

Eliminate outliers and abnormal values, specifically: based on the direct screening method, eliminate the operating data of wind turbines below the cut-in wind speed and above the cut-out wind speed, and remove the operating data of the active power of the fan, the speed of the generator, and the speed of the impeller that are less than zero.

To screen the steady-state working data, specifically: based on the quartile method and the 3σ criterion method, to filter the steady-state working data; the calculation method of the quartile method to process the data is as follows:

Divide the multi-dimensional time-series matrix from the minimum wind speed to the maximum wind speed at an interval of 0.1m/s, and sort the active power values of the fans in each interval from small to large to obtain a power value sequence [p ₁ , p ₂ , p ₃ ,..., p _n ];

Calculate the lower quartile Q ₁ , upper quartile Q ₃ , and interquartile range I _QR :

I _QR ＝Q ₃ -Q ₁ (3)

Set the normal value range of the operating data to [N ₁ , N ₂ ], the lower bound N ₁ of the normal value interval, and the upper bound N ₂ of the normal value interval are calculated according to formula (4):

If the operating data is not within the normal range, it is considered as an abnormal value and eliminated;

The screening of steady-state working condition data based on the 3σ criterion method is specifically: dividing the wind speed from the minimum value to the maximum value in intervals of 0.1m/s, and for the operation data in each wind speed interval, it is assumed that the fan operation data x is subject to a normal distribution, then

R(|x-μ|>3σ)≤0.003 (5)

In the formula, μ and σ represent the mathematical expectation and standard deviation of the normal population, respectively; data values greater than μ+3σ or less than μ-3σ are regarded as outliers and eliminated.

S102. Divide the wind speed steady-state working condition data to obtain several wind speed intervals.

Divide the wind speed steady-state working condition data at fixed intervals from the minimum wind speed to the maximum wind speed, and sort the active power values of the fans in each interval from small to large to obtain a power value sequence [P ₁ , P ₂ , P ₃ , . . . , P _n ].

S103. Based on the active power of the fan in the wind speed range, acquire the active power value of the fan corresponding to the middle value of the wind speed range.

The average value of the active power of the fan in the wind speed interval is taken as the active power value of the fan corresponding to the middle value of the wind speed interval.

For the absence of fan active power data in some wind speed intervals, based on the interpolation method, the fan active power value corresponding to the wind speed value in the center of the wind speed interval is obtained; the calculation formula is as follows:

Among them, P _c is the active power value of the fan obtained by interpolation, P ₂ and P ₁ are the active power values of the fan at the upper and lower wind speed intervals corresponding to the missing value of active power, and v ₂ , v ₁ , and v _c are the values corresponding to the missing value of active power. The upper bound value, lower bound value and central value of the wind speed range.

S104. Based on the divided wind speed intervals, obtain the active power values of the wind turbines at each wind speed of the n wind turbines in the whole field in the full wind speed range, and then obtain the power curve of the active power of each single wind turbine.

Connect the active power values of the fans in each wind speed range of a single fan in sequence to obtain the power curve of the active power of a single fan; repeat this step to obtain the power curve of the active power of n fans in the whole field.

S105, based on the average active power values of the n wind turbines in each same wind speed range, obtain the power curve of the active power of the clustered units in the whole field.

The active power values of n wind turbines in each same wind speed interval are superimposed and averaged, and the average value is used as the active power value of the cluster base group in this wind speed interval. Repeat this step to obtain the average active power of the n wind turbines in each same wind speed range; then obtain the power curve of the active power of the cluster units in the whole field.

S106, judging whether the relative deviation between the active power of a single fan and the active power of the cluster unit at each wind speed is greater than a threshold, if greater, it is normal; if less, then based on the size of the relative deviation, determine the abnormal level of the fan power.

The thresholds are specifically:

Among them, P _i is the active power of a single fan, It is the cluster power of n wind turbines in the whole field under the same wind field, the same model, and the same wind speed.

Based on the size of the relative deviation, judge the abnormal level of the fan power, specifically: set the active power relative deviation threshold as a; Level 1, abnormal level 2 and abnormal level 1 alarm thresholds are that the relative deviation between the power of a single unit and the active power of a cluster unit is lower than a, b, and c respectively; among them, a>b>c;

Example:

As shown in Figure 2, the present invention discloses an online evaluation method for abnormal power of wind farm units, the steps are:

A. Acquisition and preprocessing of operating data of wind turbines in the whole site.

B. The fitting calculation of the power curve of the single unit and the cluster unit of the whole site.

C. Determination of abnormal alarm threshold of wind turbine power and identification of low-efficiency units.

Preferably, in step A, read the operating data of measuring points such as wind speed, wind direction, fan active power, generator speed, impeller speed, yaw angle within a period of time by wind turbine SCADA (data acquisition and monitoring control system), The running data of each measuring point is indexed in chronological order, and together constitute a multi-dimensional time series matrix.

Preferably, in step A, after obtaining the wind turbine SCADA (data acquisition and monitoring and control system) operating data, it is necessary to perform data preprocessing on the extracted operating data, including: outliers, abnormal values, steady-state conditions Data filtering and other preprocessing steps.

Referring to Fig. 3, preferably, the elimination of outliers and abnormal values in the operation data taken out in step A mainly includes: adopting the direct screening method to eliminate the operation data of the wind turbine set below the cut-in wind speed and above the cut-out wind speed, and to remove the active power of the wind turbine , generator speed, impeller speed less than zero operation data;

Preferably, after the outliers and outliers are eliminated, the operation data extracted in step A needs to be screened for steady-state working data. The methods used include: quartile method and 3σ criterion method. The calculation method of the quartile method to process the data is as follows:

Divide the wind speed interval: divide the multi-dimensional time series matrix from the minimum wind speed to the maximum wind speed at an interval of 0.1m/s, and sort the active power values of the fans in each interval from small to large to obtain a power value sequence [p ₁ , p ₂ , p ₃ ,..., p _n ].

Calculate the lower quartile Q ₁ , the upper quartile Q ₃ , and the interquartile range I _QR :

I _QR ＝Q ₃ -Q ₁ (3)

Set the normal value range of the operating data to [N ₁ , N ₂ ]: the lower bound N ₁ of the normal value interval and the upper bound N ₂ of the normal value interval are calculated according to the following formulas:

The 3σ criterion method is used to screen the steady-state working condition data: the wind speed is divided into 0.1m/s intervals from the minimum value to the maximum value, and for the operation data in each wind speed interval, it is assumed that the fan operation data x is subject to a normal distribution if

R(|x-μ|>3σ)≤0.003 (5)

In the formula, μ and σ represent the mathematical expectation and standard deviation of a normal population, respectively. At this time, the probability of data values greater than μ+3σ or less than μ-3σ appearing in the operating data values is very small. Therefore, according to the above formula, data values greater than μ+3σ or less than μ-3σ are regarded as outliers and eliminated.

Preferably, in step B, the fitting calculation of the power curves of the entire field single generating set and the entire field cluster generating set. The steps to calculate the power curve of n wind turbines in the whole wind field are as follows: the data obtained after data preprocessing and steady-state screening are equalized at intervals of 0.2m/s from the minimum value of wind speed v _min to the maximum value of v _max Divide to obtain m wind speed intervals, wherein the i-th wind speed interval is [v _min +0.1×(i-1), v _min +0.1×(i+1)]. The average value of the active power of the fan in each wind speed interval is calculated as the active power value of the fan corresponding to the middle value v _i of the wind speed interval. The minimum value of wind speed v _min , the middle value of the wind speed interval v _i , and the maximum value of wind speed v _max satisfy the following conditions, as shown in formula (6):

Preferably, in step B, the fitting calculation of the power curves of the entire field single generating set and the entire field cluster generating set. For some cases where the active power data of the fan is missing in some wind speed intervals, the active power value of the fan corresponding to the central wind speed value of the wind speed interval is calculated by using the interpolation method. As shown in formula (7):

In the formula, P _c is the active power value of the fan obtained by interpolation, P ₂ and P ₁ are the active power values of the fan at the top and bottom of the wind speed range corresponding to the missing value of active power, and v ₂ , v ₁ , and v _c are the missing values of active power Corresponding wind speed range upper bound value, lower bound value and center value.

Preferably, in step B, the fitting calculation of the power curves of the entire field single generating set and the entire field cluster generating set. After calculating the active power values of n fans in the whole wind field at each wind speed in the full wind speed section, draw the power curve of n fans in the whole field. The active power values of n wind turbines in the whole field under the same wind field, wind type and wind speed are averaged to obtain the active power value of the whole field cluster unit, and then the power curve of the whole field cluster unit can be drawn. Referring to Fig. 4, Fig. 4 (a) is a curve comparison diagram of the organic power of No. 3 fan and the organic power of the cluster; Fig. 4 (b) is a curve comparison diagram of the organic power of No. 7 fan and the organic power of the cluster; Fig. 4 (c) is a curve comparison chart of the organic power of No. 11 fan and the organic power of the cluster;

Figure 4(d) is a curve comparison chart of the organic power of No. 21 wind turbine and the cluster organic power; Figure 4(e) is a curve comparison chart of the organic power of No. 51 wind turbine and the cluster organic power; Figure 4(f) is The graph comparing the organic power of wind turbine No. 62 with the organic power of the cluster.

Preferably, in step C, the determination of the abnormal alarm threshold of the power of the wind turbine unit and the identification of the low-efficiency unit. Calculate the relative deviation value _δi of the active power between the single unit and the cluster unit at each wind speed as the determination condition for the abnormal power of the wind turbine unit.

In the formula, P _i is the active power of a single fan, It is the cluster power of n wind turbines in the whole field under the same wind field, the same model, and the same wind speed.

Preferably, in step C, the determination of the abnormal alarm threshold of the power of the wind turbine unit and the identification of the low-efficiency unit. Set the active power relative deviation threshold to -5%. When the relative deviation between the active power of a single fan and the active power of the cluster unit is less than -5%, it is determined that the unit is in an abnormal state, and the unit is identified as a low-efficiency unit for power generation and an alarm is issued. Set the three-level, two-level, and first-level alarm thresholds for fan power abnormalities to be less than -5%, -10%, and -15% for the relative deviation between the power of a single unit and the active power of a cluster unit, respectively.

Preferably, in step C, the determination of the abnormal alarm threshold of the power of the wind turbine unit and the identification of the low-efficiency unit. Further, in order to quantify the degree of degradation of wind turbine power generation performance, according to the relative deviation of the active power of a single unit and cluster units, the degree of degradation is classified as follows:

Referring to Fig. 5, the present invention discloses an online evaluation system for abnormal power of wind farm units, including:

A pre-processing module, the pre-processing module obtains the operation data of the wind turbines in the whole field, and preprocesses the operation data of the wind turbines in the whole field, and obtains the wind speed steady-state working condition data;

A division module, the division module divides the wind speed steady-state working condition data to obtain several wind speed intervals;

A first acquisition module, the first acquisition module acquires the active power value of the fan corresponding to the middle value of the wind speed interval based on the active power of the fan in the wind speed interval;

The second acquisition module, based on the divided wind speed interval, the second acquisition module acquires the fan active power values of the n fans in the whole field at each wind speed in the full wind speed segment, and then acquires the active power of each single fan curve;

The third acquisition module, the third acquisition module obtains the power curve of the active power of the cluster units in the whole field based on the average value of active power values of n fans in each same wind speed interval;

Judging module, the judging module judges whether the relative deviation between the active power of a single fan and the active power of the cluster unit at each wind speed is greater than a threshold value, if greater, it is normal; exception level.

The above are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. An online evaluation method for abnormal power of a wind farm unit, characterized in that it comprises: Obtain the operation data of the wind turbines in the whole field, and preprocess the operation data of the wind turbines in the whole field, and obtain the wind speed and steady-state working condition data; Divide the wind speed steady-state working condition data to obtain several wind speed intervals; Based on the active power of the fan in the wind speed interval, the active power value of the fan corresponding to the middle value of the wind speed interval is obtained; Based on the divided wind speed intervals, obtain the active power values of the fans at each wind speed in the full wind speed range of n fans in the whole field, and then obtain the power curve of the active power of each single fan; Based on the average value of active power of n wind turbines in each same wind speed range, the power curve of the active power of the whole field cluster unit is obtained; Judging whether the relative deviation between the active power of a single fan and the active power of the cluster unit in each wind speed range is greater than the threshold, if it is greater, it is normal; if it is less than, based on the size of the relative deviation, the abnormal level of fan power is judged. 2. The online evaluation method for abnormal power of wind farm units according to claim 1, characterized in that, the operation data of the whole field wind turbines includes: wind speed, wind direction, active power of wind turbines, generator speed, impeller speed per unit time and yaw angle, and the running data of each measuring point is indexed in chronological order to form a multi-dimensional time series matrix. 3. The online evaluation method for abnormal power of wind farm units according to claim 2, characterized in that, the preprocessing of the operation data of the whole field of wind turbines to obtain the wind speed steady-state working condition data is specifically: removing outliers and outliers and screen the steady-state data; The elimination of outliers and abnormal values is specifically: based on the direct screening method, the operation data of the wind turbine below the cut-in wind speed and the cut-out wind speed is removed, and the operation data of the active power of the fan, the speed of the generator, and the speed of the impeller are less than zero. 4. The online evaluation method for abnormal power of wind farm units according to claim 3, characterized in that the screening of steady-state data is specifically: based on the quartile method and the 3σ criterion method, the steady-state Situation data is screened; the calculation method of described quartile method processing data is as follows: Divide the multi-dimensional time-series matrix from the minimum wind speed to the maximum wind speed at an interval of 0.1m/s, and sort the active power values of the fans in each interval from small to large to obtain a power value sequence [p ₁ , p ₂ , p ₃ ,..., p _n ]; Calculate the lower quartile Q ₁ , upper quartile Q ₃ , and interquartile range I _QR : I _QR ＝Q ₃ -Q ₁ (3) Set the normal value range of the operating data to [N ₁ , N ₂ ], the lower bound N ₁ of the normal value interval, and the upper bound N ₂ of the normal value interval are calculated according to formula (4): N ₁ =Q ₁ -1.5I _QR (4) N ₂ =Q ₂ +1.5I _QR If the operating data is not within the normal range, it is considered as an abnormal value and eliminated; The screening of steady-state working condition data based on the 3σ criterion method is specifically: dividing the wind speed from the minimum value to the maximum value in intervals of 0.1m/s, and for the operation data in each wind speed interval, it is assumed that the fan operation data x is subject to a normal distribution, then P(|x-μ|>3σ)≤0.003 (5) In the formula, μ and σ represent the mathematical expectation and standard deviation of the normal population, respectively; data values greater than μ+3σ or less than μ-3σ are regarded as outliers and eliminated. 5. The method for online evaluation of wind farm unit power abnormality according to claim 1, wherein the wind speed steady-state working condition data is divided to obtain several wind speed intervals, specifically: Divide the wind speed steady-state working condition data at fixed intervals from the minimum wind speed to the maximum wind speed, and sort the active power values of the fans in each interval from small to large to obtain a power value sequence [P ₁ , P ₂ , P ₃ , . . . , P _n ]. 6. The method for online evaluation of wind farm unit power abnormality according to claim 5, wherein, based on the active power of the wind turbine in the wind speed interval, the active power value of the wind turbine corresponding to the intermediate value of the wind speed interval is obtained, specifically: The average value of the active power of the fan in the wind speed interval is used as the active power value of the fan corresponding to the middle value of the wind speed interval. 7. The method for online evaluation of wind farm unit power abnormality according to claim 6, wherein said obtaining the active power value of the fan corresponding to the intermediate value of the wind speed interval based on the active power of the fan in the wind speed interval further includes: In the case of missing fan active power data in some wind speed intervals, based on the interpolation method, the fan active power value corresponding to the wind speed value in the center of the wind speed interval is obtained; the calculation formula is as follows: Among them, P _c is the active power value of the fan obtained by interpolation, P ₂ and P ₁ are the active power values of the fan at the upper and lower wind speed intervals corresponding to the missing value of active power, and v ₂ , v ₁ , and v _c are the values corresponding to the missing value of active power. The upper bound value, lower bound value and central value of the wind speed range. 8. The method for online evaluation of wind farm unit power abnormality according to claim 7, wherein the threshold is specifically: Among them, P _i is the active power of a single fan, It is the cluster power of n wind turbines in the whole field under the same wind field, the same model, and the same wind speed. 9. The method for online evaluation of wind farm unit power abnormality according to claim 8, wherein the abnormal level of fan power is judged based on the size of the relative deviation, specifically: setting the active power relative deviation threshold as a; Set fan power abnormalities to three levels; respectively, abnormal level three, abnormal level two and abnormal one level; abnormal level three, abnormal level two and abnormal one level alarm thresholds are the relative deviations of the active power of a single unit power and cluster units below a, b, c; among them, a>b>c; 10. An online evaluation system for abnormal power of wind farm units, characterized in that it includes: A pre-processing module, the pre-processing module obtains the operation data of the wind turbines in the whole field, and preprocesses the operation data of the wind turbines in the whole field, and obtains the wind speed steady-state working condition data; A division module, the division module divides the wind speed steady-state working condition data to obtain several wind speed intervals; A first acquisition module, the first acquisition module acquires the active power value of the fan corresponding to the middle value of the wind speed interval based on the active power of the fan in the wind speed interval; The second acquisition module, based on the divided wind speed interval, the second acquisition module acquires the fan active power values of the n fans in the whole field at each wind speed in the full wind speed segment, and then acquires the active power of each single fan curve; The third acquisition module, the third acquisition module obtains the power curve of the active power of the cluster units in the whole field based on the average value of active power values of n fans in each same wind speed interval; Judging module, the judging module judges whether the relative deviation between the active power of a single fan and the active power of the cluster unit at each wind speed is greater than a threshold value, if greater, it is normal; exception level.