Offshore wind power wind speed-power data preprocessing method and system

Abstract

The invention discloses an offshore wind power wind speed-power data preprocessing method and system, and the method comprises the steps: obtaining the time series data of the actually measured wind speed and power, and obtaining a suspicious power limiting abnormal data section in the time series data through combining with a time series segmentation and trend consistency measurement method; marking a data section of which the actual measurement power is in a first form for the suspicious power limiting abnormal data section, and carrying out rated power section identification marking on the actual measurement power time sequence data; matching the rated power section with the data section in the first form, and judging whether the distance between the end point of the rated power section and the starting point of the corresponding data section in the first form is smaller than the distance between the end points or not; and if the judgment condition is met, the data section in the first form is marked as the cut-out shutdown event data of the wind turbine generator, and the suspicious power-limiting abnormal data section is corrected, so that the identification effect of the power-limiting abnormal data and the cut-out shutdown event data is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of offshore wind power, and in particular to a method and system for preprocessing offshore wind power wind speed-power data. Background Art

[0002] At present, my country's offshore wind power installed capacity is huge and growing rapidly, and it is the key development direction of the new power system of the modern energy system. However, due to the frequent occurrence of external extreme weather such as severe convective weather and typhoons, superimposed with complex dynamic changes and harsh internal equipment operating environment of the grid-connected system, the grid-connected operation of offshore wind power faces the dual challenges of meteorological conditions and system status. In order to ensure the safety of offshore wind power grid connection, it is urgent to study the technology for building output loss models under extreme weather conditions. Under extreme weather conditions, the operating environment of offshore wind farms is complex and changeable. The wind speed and power data collection process may be interfered with and noisy by various factors, resulting in various problems with the collected original wind speed and power data, which in turn affects the data quality of the output loss model under extreme weather.

[0003] However, most of the existing wind power wind speed-power data preprocessing methods identify abnormal data based on two-dimensional scatter plots and then remove them. It is difficult to identify complete sequence abnormal data, resulting in incomplete removal of abnormal data. In addition, existing data preprocessing methods often remove wind turbine shutdown event data, which is not conducive to the construction of output loss models under extreme weather conditions. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an offshore wind power wind speed-power data preprocessing method and system to solve the problem that it is difficult to identify complete sequence abnormal data in the prior art and wind turbine shutdown event data is often eliminated, thereby affecting the quality of input source data for construction of technical related models such as remote fault diagnosis, status warning, risk assessment and deduction under extreme weather conditions.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for preprocessing offshore wind power wind speed-power data, comprising:

[0008] Obtaining time series data of measured wind speed and power, and combining time series segmentation and trend consistency measurement methods to obtain suspicious power-limiting abnormal data segments in the time series data;

[0009] For the suspected power-limiting abnormal data segment, mark the data segment where the measured power is in the first form, and perform a rated power segment identification mark on the measured power time series data;

[0010] Matching the rated power segment with the data segment in the first form, and determining whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the endpoint distance;

[0011] If the judgment condition is met, the data segment in the first form is marked as wind turbine shutdown event data, and the suspected power restriction abnormal data segment is corrected.

[0012] As a preferred solution of the offshore wind power wind speed-power data preprocessing method described in the present invention, the time series segmentation and trend consistency measurement method includes:

[0013] Preprocessing the wind farm's full-field measured wind speed time series data and full-field measured power time series data to obtain normalized full-field measured wind speed time series data and normalized full-field measured power time series data;

[0014] Calculate the difference between the normalized full-field measured wind speed time series data and the normalized full-field measured power time series data at each time point.

[0015] As a preferred solution of the offshore wind power wind speed-power data preprocessing method of the present invention, it also includes:

[0016] The positions where the difference sign changes are marked as interlaced points, and the time series data are segmented using the interlaced points;

[0017] Calculate the correlation coefficient between the normalized full-field measured wind speed time series data and the normalized full-field measured power time series data;

[0018] If the section correlation coefficient is less than the correlation threshold, the section is marked as a suspected power-cutoff section.

[0019] As a preferred solution of the offshore wind power wind speed-power data preprocessing method of the present invention, the data segment marking the measured power in the first form includes:

[0020] Traverse the measured power time series of the entire wind farm, mark the points that are less than the low output threshold as the first mark, and mark the points that do not meet the threshold as the second mark;

[0021] The first consecutive marked points are regarded as a low-output constant section, and the starting point and the ending point of the low-output constant section are recorded.

[0022] As a preferred solution of the offshore wind power wind speed-power data preprocessing method of the present invention, it also includes:

[0023] The starting point of each low-output constant section is set as the starting point for finding the steep drop section, the corresponding step size and window length are set, the data is traversed from the starting point forward, the corresponding data segment is linearly fitted, and the slope of the data segment is calculated;

[0024] Compare the slope of the data segment with the slope threshold, if the slope is less than the slope threshold, it is a steep drop segment, and return the corresponding data segment in the first form;

[0025] If there is no steep drop in the window length, the search ends.

[0026] As a preferred solution of the offshore wind power wind speed-power data preprocessing method of the present invention, the rated power section identification mark of the measured power time series data includes:

[0027] Traverse the time series of the measured power of the entire wind farm, mark the points that meet the rated power range first, and mark the points that do not meet the rated power range second;

[0028] The first consecutive marked points are regarded as the rated power section, and the starting point and the ending point of the rated power section are recorded.

[0029] As a preferred solution of the offshore wind power wind speed-power data preprocessing method of the present invention, matching the rated power segment with the data segment in the first form includes:

[0030] Marking the sequence number of each rated power segment and the data segment in the first form according to the starting point size of each rated power segment and the data segment in the first form;

[0031] For each rated power segment, starting from the rated power end point, searching for the data segment with the smallest sequence number among the data segments in the first form whose sequence numbers are greater than the sequence number of the current rated power segment;

[0032] Determine whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the endpoint distance;

[0033] If it is smaller, the data segment in the first form is marked as wind turbine shutdown event data, and the suspicious power-limiting abnormal data segment is corrected.

[0034] In a second aspect, the present invention provides an offshore wind power wind speed-power data preprocessing system, comprising:

[0035] A data acquisition module is used to acquire the time series data of measured wind speed and power, and obtain the suspected power-limiting abnormal data segment in the time series data by combining the time series segmentation and trend consistency measurement method;

[0036] An identification and marking module is used to mark the data segment where the measured power is in the first form for the suspected power restriction abnormal data segment, and to perform a rated power segment identification mark on the measured power time series data;

[0037] A matching module, used for matching the rated power segment with the data segment in the first form, and determining whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the end point distance;

[0038] The judgment module is used to mark the data segment in the first form as wind turbine shutdown event data if the judgment condition is met, and to correct the suspicious power-limiting abnormal data segment.

[0039] In a third aspect, the present invention provides a computing device, comprising:

[0040] Memory and processor;

[0041] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the offshore wind power wind speed-power data preprocessing method are implemented.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the offshore wind power wind speed-power data preprocessing method.

[0043] Compared with the prior art, the present invention has the following beneficial effects: by combining the time series segmentation based on the staggered points and the trend consistency measurement method, the present invention can more accurately identify the suspicious power-limiting abnormal data segments in the full-field time series, thereby improving the accuracy of data preprocessing, while eliminating the abnormal wind speed-power data of wind power, retaining the data of the shutdown event under extreme weather conditions, which is crucial for building an offshore wind power output loss model under extreme weather conditions. The present invention uses the connected domain method and the slope method to identify and mark the steep drop section-low output constant section and the rated power section, thereby improving the effect of identifying the abnormal power-limiting data and the shutdown event data. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0045] Figure 1A schematic diagram of the overall process logic of the offshore wind power wind speed-power data preprocessing method according to an embodiment of the present invention;

[0046] Figure 2 A sorting diagram of the sequence numbers of each rated power segment and the data segment in the first form in the offshore wind power wind speed-power data preprocessing method according to an embodiment of the present invention;

[0047] Figure 3 A data segment schematic diagram of a first form of an offshore wind power wind speed-power data preprocessing method according to an embodiment of the present invention;

[0048] Figure 4 This is an effect diagram of power-limiting abnormal data identification using an offshore wind power wind speed-power data preprocessing method according to an embodiment of the present invention;

[0049] Figure 5 This is a diagram showing the effect of identifying shutdown event data in a method for preprocessing offshore wind power wind speed-power data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the above-mentioned purposes, features and advantages of the present invention more understandable, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0051] Example 1

[0052] Reference Figure 1 , as an embodiment of the present invention, provides a method for preprocessing offshore wind power wind speed-power data, comprising:

[0053] S100: obtaining time series data of measured wind speed and power, and combining time series segmentation and trend consistency measurement method to obtain suspicious power-limiting abnormal data segments in the time series data;

[0054] S200: for the suspected power-limiting abnormal data segment, mark the data segment where the measured power is in the first form, and perform rated power segment identification marking on the measured power time series data;

[0055] S300: Matching the rated power segment with the data segment in the first form, and determining whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the endpoint distance;

[0056] S400: If the judgment condition is met, the data segment in the first form is marked as wind turbine shutdown event data, and the suspected power restriction abnormal data segment is corrected.

[0057] It should be noted that, in the time series segmentation and trend consistency measurement method based on intersection points, the present invention can more accurately identify suspicious power-limiting abnormal data segments. Compared with the traditional power-limiting abnormal data identification method based on two-dimensional scatter plots, it can identify complete sequence abnormal data and ensure the integrity of each suspicious power-limiting abnormal data segment.

[0058] In the embodiment of the present application, the above step S100 includes the following sub-steps A1-A2;

[0059] In A1: preprocessing the wind farm's full-field measured wind speed time series data and full-field measured power time series data to obtain normalized full-field measured wind speed time series data and normalized full-field measured power time series data;

[0060] In A2: Calculate the difference between the normalized full-field measured wind speed time series data and the normalized full-field measured power time series data at each time point.

[0061] Specifically, the MinMaxScaler in the sklearn library in Python is used to preprocess the wind speed time series data and the power time series data of the entire field respectively; a time series segmentation and trend consistency measurement method based on staggered points is used to find and mark all the suspicious power-limiting abnormal data segments in the entire field time series;

[0062] In an optional implementation, the preprocessing method may be minimum-maximum normalization, which adjusts the numerical range of the data to a specific interval, usually [0,1], but may also be any other interval;

[0063] The preprocessing method can also be Z-score standardization, which converts each feature value of the data into a score indicating how many standard deviations the value is from the mean;

[0064] You can also scale the data to the range of [0,1] or [-1,1] for regularization.

[0065] In the embodiment of the present application, the minimum-maximum normalization is used for preprocessing, and the minimum-maximum normalization is expressed as:

[0066]

[0067] Among them, Z is a value in the original data, Z ′ is the normalized value, Z minis the minimum value in the original data, Z max is the maximum value in the original data.

[0068] It should be noted that the minimum-maximum normalization algorithm is simple, easy to understand and implement, and the normalized data maintains the distribution and shape of the original data, but only changes the scale of the data.

[0069] In the embodiment of the present application, after completing steps A1-A2, the above step S100 may further include the following sub-steps A3-A5:

[0070] In A3: the positions where the difference sign changes are marked as crossover points, and the time series data are segmented using the crossover points;

[0071] In A4: calculating the correlation coefficient between the normalized full-field measured wind speed time series data and the normalized full-field measured power time series data;

[0072] In A5: if the section correlation coefficient is less than the correlation threshold, the section is marked as a suspected power-cutoff section.

[0073] Specifically, the difference time series is traversed, and the difference signs of two adjacent time points are compared. If the sign changes from positive to negative or from negative to positive, it is considered that there is a sign change at this time point. The positions where all signs change are found, and the time indexes or timestamps of these positions are recorded.

[0074] For each section, the correlation coefficient of the normalized measured power time series and the normalized measured wind speed time series is calculated;

[0075] The correlation threshold is set according to the actual situation;

[0076] In an optional implementation, the correlation coefficient may be a Pearson correlation coefficient, which is an indicator for measuring the degree of linear correlation between two variables, and its value is between -1 and 1, where a value of 1 indicates a perfect positive correlation, -1 indicates a perfect negative correlation, and 0 indicates no linear correlation;

[0077] The correlation coefficient may also be the Spearman rank correlation coefficient, which is used to evaluate the correlation between the rank orders of two variables, or the cosine similarity, which is used to measure the similarity between two non-zero vectors, determined by calculating the cosine of the angle between them.

[0078] In the embodiment of the present application, the Pearson correlation coefficient is expressed as:

[0079]

[0080] Among them, P i ′ is the ith value of the normalized power data, Vi ′ is the i-th value of the normalized wind speed data, is the average value of the normalized power data, is the average value of the normalized wind speed data, and n is the total number of data points in the time series;

[0081] It should be noted that by calculating the Pearson correlation coefficient of the full-field measured wind speed time series data and the full-field measured power time series data, those suspicious power-limiting abnormal data segments that are inconsistent with the overall trend can be quickly identified, thereby improving the efficiency of the data cleaning process.

[0082] In the embodiment of the present application, the step S200 of marking the data segment in which the measured power is in the first form includes the following sub-steps B1-B2:

[0083] In B1: traverse the time series of the measured power of the entire wind farm, mark the points that are less than the low output threshold as the first mark, and mark the points that do not meet the threshold as the second mark;

[0084] In B2: the first consecutive marked points are regarded as a low-output constant section, and the starting point and the ending point of the low-output constant section are recorded.

[0085] Specifically, the low output threshold is set according to the actual situation, and the connected domain method is used to detect the low output constant section and the slope method is combined to detect the steep drop section;

[0086] In an optional implementation, the first mark may be 1, and the second mark may be 0, then the mark that satisfies the condition is 1, and the mark that does not satisfy the condition is 0;

[0087] The first mark can also be True, and the second mark can also be False, then the mark that meets the condition is True, and the mark that does not meet the condition is False; it can also be other symbols or numbers;

[0088] In the embodiment of the present application, the time series of the actual power measured in the entire wind farm is traversed, and the points that are less than the low output threshold are marked as 1, and those that do not meet the threshold are marked as 0, and the continuous points marked as 1 are regarded as the low output constant area;

[0089] In the embodiment of the present application, after completing steps B1-B2, the above step S200 may further include the following sub-steps B3-B5;

[0090] In B3: the starting point of each low output constant section is set as the starting point for finding the steep drop section, the corresponding step size and window length are set, the data is traversed from the starting point forward, the corresponding data segment is linearly fitted, and the slope of the data segment is calculated;

[0091] In B4: the slope of the data segment is compared with the slope threshold. If the slope is less than the slope threshold, it is a steep drop segment, and the corresponding data segment in the first form is returned;

[0092] In B5: If the window length does not have a steep drop, the search ends.

[0093] Specifically, the slope threshold is set according to the actual situation. If the data segment in the first form is a steep drop segment-low output constant segment, then if the slope is less than the slope threshold, it means that the steep drop segment has been found, and the corresponding steep drop segment-low output constant segment is returned;

[0094] It should be noted that by accurately identifying the low-output constant section and the steep drop section, the cut-out and shutdown events of wind turbines can be marked more accurately, thereby retaining small sample event data under extreme weather conditions, providing a good foundation for improving the adaptability and stability of wind farms in the face of severe convective weather, typhoons and other extreme weather conditions.

[0095] In the embodiment of the present application, after completing steps B3-B5, the above step S200 may further include the following sub-steps B6-B7;

[0096] In B6: traverse the time series of the measured power of the entire wind farm, mark the points that meet the rated power range first, and mark the points that do not meet the rated power range second;

[0097] In B7: The first consecutive marked points are regarded as a rated power section, and the starting point and the ending point of the rated power section are recorded.

[0098] Specifically, the measured power time series of the entire wind farm is traversed, and the points that meet the rated power range are marked as 1, and those that do not meet the rated power range are marked as 0. The continuous points marked as 1 are regarded as a rated power section, and the starting and ending points of these rated power sections are recorded.

[0099] In the embodiment of the present application, the matching of the rated power segment with the data segment in the first form in the above step S300 includes the following sub-steps C1-C2:

[0100] In C1: marking the sequence number of each rated power segment and the data segment in the first form according to the starting point size of each rated power segment and the data segment in the first form;

[0101] In C2: for each rated power segment, starting from the rated power end point, find the data segment with the smallest sequence number among the data segments in the first form whose sequence numbers are greater than the sequence number of the current rated power segment;

[0102] In C3: determining whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the endpoint distance;

[0103] In C4: if it is less than, the data segment in the first form is marked as wind turbine shutdown event data, and the suspected power restriction abnormal data segment is corrected.

[0104] Specifically, Figure 2 As shown, the sequence numbers of each rated power segment and the data segment in the first form are marked from small to large according to the starting point size of each rated power segment and the data segment in the first form, and a schematic diagram of a steep drop section-low output constant section is obtained, as shown in FIG. Figure 3 shown.

[0105] Mark the serial number of each rated power section and steep drop section-low output constant section. For each rated power section, start from its end point and find the first steep drop section-low output constant section whose number is greater than the current rated power section number. If the end point of each rated power section and the starting point between the corresponding steep drop section-low output constant section are less than the end point distance, the steep drop section-low output constant section that meets the conditions is marked as the wind turbine shutdown event data, and the section mark of the suspected power restriction abnormal data is corrected.

[0106] It should be noted that by matching the rated power section and the subsequent steep drop section-low output constant section, the cut-out shutdown event of the wind turbine can be identified more accurately, which is crucial for analyzing the behavior of the wind turbine under specific conditions. By detecting the time interval between the end point of the rated power section and the starting point of the steep drop section-low output constant section, the behavioral changes of the wind turbine after reaching the rated power can be better understood.

[0107] The above is a schematic scheme of an offshore wind power wind speed-power data preprocessing method of this embodiment. It should be noted that the technical scheme of the offshore wind power wind speed-power data preprocessing system and the technical scheme of the offshore wind power wind speed-power data preprocessing method described above belong to the same concept. For details not described in detail in the technical scheme of the offshore wind power wind speed-power data preprocessing system in this embodiment, please refer to the description of the technical scheme of the offshore wind power wind speed-power data preprocessing method described above.

[0108] The offshore wind power wind speed-power data preprocessing system in this embodiment includes:

[0109] A data acquisition module is used to acquire the time series data of measured wind speed and power, and obtain the suspected power-limiting abnormal data segment in the time series data by combining the time series segmentation and trend consistency measurement method;

[0110] An identification and marking module is used to mark the data segment where the measured power is in the first form for the suspected power restriction abnormal data segment, and to perform a rated power segment identification mark on the measured power time series data;

[0111] A matching module, used for matching the rated power segment with the data segment in the first form, and determining whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the end point distance;

[0112] The judgment module is used to mark the data segment in the first form as wind turbine shutdown event data if the judgment condition is met, and to correct the suspicious power-limiting abnormal data segment.

[0113] This embodiment further provides a computing device, which is applicable to offshore wind power wind speed-power data preprocessing, and includes:

[0114] Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the offshore wind power wind speed-power data preprocessing method proposed in the above embodiment.

[0115] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the offshore wind power wind speed-power data preprocessing method proposed in the above embodiment is implemented.

[0116] The storage medium proposed in this embodiment and the method for preprocessing offshore wind power wind speed-power data proposed in the above embodiment belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0117] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (Random Access Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0118] Example 2

[0119] Refer to Table 1 and Figure 1-Figure 2 This embodiment is different from the first embodiment and provides a verification test of an offshore wind power wind speed-power data preprocessing method to verify and illustrate the technical effects used in this method.

[0120] Collect the full-field measured wind speed time series data and full-field measured power time series data of an offshore wind farm in East China for 365 days from January 1, 2023 to December 31, 2023 as sample data;

[0121] The accuracy and recall rate are used to measure the effect of identifying abnormal power-limiting data and shutdown event data. The accuracy rate is expressed as:

[0122]

[0123] The recall rate is expressed as:

[0124]

[0125] Among them, P is the precision, R is the recall rate, TP is the event data sample that is correctly identified, FP is the event data sample that is incorrectly identified, and FN is the event data sample that actually occurred but was not identified.

[0126] Table 1 Accuracy and recall of power-limiting data recognition and shutdown event data recognition

[0127] Power restriction data identification Cut-out downtime event data recognition Accuracy (%) 95.23％ 98.95％ Recall rate (%) 99.12％ 99.33％

[0128] As shown in Table 1, whether it is power-limiting data recognition or shutdown event data recognition, the corresponding accuracy and recall rates are relatively high, indicating the effectiveness of the data preprocessing and recognition method in this embodiment.

[0129] like Figure 2 As shown in the figure, the normalized measured power and normalized measured wind speed change curves over time, as well as the power restriction abnormal period identified and marked by the method of this embodiment (as shown in the rectangular box in the figure). Under normal working conditions, an increase in wind speed usually leads to an increase in power output, that is, the trend of the wind speed curve and the power curve are generally consistent. Figure 2 As shown, in the power-limiting abnormal period identified and marked by this embodiment, the trend direction difference between the normalized measured power and the normalized measured wind speed curve is relatively large, indicating that the method of this embodiment can accurately identify the power-limiting abnormal data.

[0130] like Figure 3 As shown, the changes of the measured power and the measured wind speed over time and the cut-out shutdown period identified and marked by the method of this embodiment (as shown in the rectangular box position in the figure) are shown. The figure shows two situations in which the measured power is first maintained at the rated power, and then drops rapidly and then remains at a low output (i.e., the first form). Both situations correspond to the cut-out shutdown events of the wind turbine set, and both cut-out shutdown events can be accurately identified.

[0131] In general, relevant power-cutting data and shutdown event data can be accurately identified.

[0132] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for preprocessing offshore wind power wind speed-power data, characterized in that: include: Obtaining time series data of measured wind speed and power, and combining time series segmentation and trend consistency measurement methods to obtain suspicious power-limiting abnormal data segments in the time series data; For the suspected power-limiting abnormal data segment, mark the data segment where the measured power is in the first form, and perform a rated power segment identification mark on the measured power time series data; Matching the rated power segment with the data segment in the first form, and determining whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the endpoint distance; If the judgment condition is met, the data segment in the first form is marked as wind turbine shutdown event data, and the suspected power restriction abnormal data segment is corrected.

2. The offshore wind power wind speed-power data preprocessing method according to claim 1, characterized in that: Time series segmentation and trend consistency measures include: Preprocessing the wind farm's full-field measured wind speed time series data and full-field measured power time series data to obtain normalized full-field measured wind speed time series data and normalized full-field measured power time series data; Calculate the difference between the normalized full-field measured wind speed time series data and the normalized full-field measured power time series data at each time point.

3. The offshore wind power wind speed-power data preprocessing method according to claim 2, characterized in that: Also includes: The positions where the difference sign changes are marked as interlaced points, and the time series data are segmented using the interlaced points; Calculate the correlation coefficient between the normalized full-field measured wind speed time series data and the normalized full-field measured power time series data; If the section correlation coefficient is less than the correlation threshold, the section is marked as a suspected power-cutoff section.

4. The offshore wind power wind speed-power data preprocessing method according to claim 2 or 3, characterized in that: The data segment indicating that the measured power is in the first form includes: Traverse the measured power time series of the entire wind farm, mark the points that are less than the low output threshold as the first mark, and mark the points that do not meet the threshold as the second mark; The first consecutive marked points are regarded as a low-output constant section, and the starting point and the ending point of the low-output constant section are recorded.

5. The offshore wind power wind speed-power data preprocessing method according to claim 4, characterized in that: Also includes: The starting point of each low-output constant section is set as the starting point for finding the steep drop section, the corresponding step size and window length are set, the data is traversed from the starting point forward, the corresponding data segment is linearly fitted, and the slope of the data segment is calculated; Compare the slope of the data segment with the slope threshold, if the slope is less than the slope threshold, it is a steep drop segment, and return the corresponding data segment in the first form; If there is no steep drop in the window length, the search ends.

6. The offshore wind power wind speed-power data preprocessing method according to claim 5, characterized in that: The rated power section identification marking of the measured power time series data includes: Traverse the time series of the measured power of the entire wind farm, mark the points that meet the rated power range first, and mark the points that do not meet the rated power range second; The first consecutive marked points are regarded as the rated power section, and the starting point and the ending point of the rated power section are recorded.

7. The offshore wind power wind speed-power data preprocessing method according to claim 6, characterized in that: Matching the rated power segment with the data segment in the first form includes: Marking the sequence number of each rated power segment and the data segment in the first form according to the starting point size of each rated power segment and the data segment in the first form; For each rated power segment, starting from the rated power end point, searching for the data segment with the smallest sequence number among the data segments in the first form whose sequence numbers are greater than the sequence number of the current rated power segment; Determine whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the endpoint distance; If it is smaller, the data segment in the first form is marked as wind turbine shutdown event data, and the suspicious power restriction abnormal data segment is corrected.

8. A system using the offshore wind power wind speed-power data preprocessing method according to any one of claims 1 to 7, characterized in that: include: A data acquisition module is used to acquire the time series data of measured wind speed and power, and obtain the suspected power-limiting abnormal data segment in the time series data by combining the time series segmentation and trend consistency measurement method; An identification and marking module is used to mark the data segment where the measured power is in the first form for the suspected power restriction abnormal data segment, and to perform a rated power segment identification mark on the measured power time series data; A matching module, used for matching the rated power segment with the data segment in the first form, and determining whether the distance between the end point of the rated power segment and the start point of the corresponding data segment in the first form is less than the end point distance; The judgment module is used to mark the data segment in the first form as wind turbine shutdown event data if the judgment condition is met, and to correct the suspicious power-limiting abnormal data segment.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the offshore wind power wind speed-power data preprocessing method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the offshore wind power wind speed-power data preprocessing method as claimed in any one of claims 1 to 7.

Images