CN-121993366-A - Intelligent wind generating set clearance analysis system and early warning method based on multi-parameter coupling

CN121993366ACN 121993366 ACN121993366 ACN 121993366ACN-121993366-A

Abstract

The invention relates to the technical field of wind generating set operation monitoring, in particular to a multi-parameter coupling-based intelligent analysis system and an early warning method for the clearance of a wind generating set, comprising the following steps of S1, importing main control SCADA operation data and completing field identification, cleaning and standardization to form a clearance operation data set; S2, combining a clearance threshold, a multi-parameter coupling relation and a historical baseline to identify and count clearance abnormality to generate an abnormal result set, S3, constructing a time sequence diagram, a scatter diagram, a histogram and a linkage change diagram based on the abnormal result to present abnormal characteristics, S4, summarizing the abnormal result and the map to generate an analysis report, and outputting early warning information according to the type and duration of the abnormality. According to the invention, the quick identification and closed-loop treatment of the clearance abnormality of the wind generating set are realized, the running safety of the set is effectively improved, and the risk of tower sweeping is reduced.

Inventors

LI ZHENJING
LIU YEPENG
ZHAO WEIDONG
LI MING
ZHAO JIANGBO
SUN XIN
HE JUN

Assignees

夏津田辰新能源有限公司

Dates

Publication Date: 20260508
Application Date: 20260309

Claims (9)

1. The intelligent pre-warning method for the clearance of the wind generating set based on multi-parameter coupling is characterized by comprising the following steps of: S1, acquiring operation data of a wind generating set, which is derived by a master control SCADA system, wherein the operation data comprises a clearance value, a generator rotating speed, a blade pitch angle and a wind speed, the operation data is imported in a CSV format, automatic field identification and unified analysis are performed on the operation data, and a standardized clearance operation data set is formed through data cleaning, missing value filling and data format standardization processing; S2, based on the standardized clearance operation data set, according to a preset clearance safety threshold, and by combining a multiparameter coupling relation between wind speed and rotation speed and a clearance value and a historical operation data base line, automatically analyzing operation data, screening and marking clearance safety threshold triggering abnormality, invalid numerical value abnormality, adjacent clearance difference value abnormality and abnormal pitch harvest abnormality after threshold triggering, and counting occurrence times and distribution characteristics of various abnormalities to generate an abnormality analysis result set; s3, respectively constructing corresponding maps based on the abnormal analysis result set, wherein the maps comprise a time sequence diagram of a clearance value and the rotation speed of the generator, a scattered point distribution diagram of the rotation speed of the generator and the clearance value, a histogram corresponding to the difference value of adjacent rotation speeds and a linkage change line diagram with the clearance value, the pitch angle of the blade and the wind speed under a threshold trigger condition; s4, comprehensively arranging the abnormal analysis result set and the corresponding atlas, generating a standardized analysis report comprising a data profile, an abnormal detail, a risk assessment conclusion and a processing suggestion, determining the severity of the abnormality according to the type of the abnormality and the total duration, and outputting corresponding clearance safety early warning information.
2. The intelligent pre-warning method for the headroom of a wind turbine generator system based on multi-parameter coupling according to claim 1, wherein the step S1 comprises: S11, collecting a clearance value by a laser range finder arranged on a tower, collecting the rotating speed of a generator by a rotating speed sensor arranged at the shaft end of the generator, collecting the pitch angle of blades by a pitch angle position sensor arranged in a pitch system, collecting the wind speed by an anemograph arranged at the top of a nacelle, synchronously recording operation data by a master control SCADA system according to a uniform time stamp, and exporting the operation data in a CSV format file to form the original operation data of the wind turbine generator ; S12, the original operation data of the unit exported by the main control SCADA system is processed After being introduced, is opposite to The header fields are automatically identified, data columns corresponding to the clearance value, the generator rotating speed, the blade pitch angle and the wind speed are determined through field matching rules, the identified data columns are analyzed and aligned according to a unified time index, and the numerical value fields are converted into a unified numerical value format to form an analysis operation data set with consistent field meaning and time synchronization ; S13, analyzing the operation data set Performing data cleaning and normalization processes to form a normalized headroom data set 。
3. The intelligent pre-warning method for the headroom of a wind turbine generator system based on multi-parameter coupling according to claim 1, wherein the step S2 comprises: S21, calling a preset clearance safety threshold based on a standardized clearance operation data set, and establishing a comparison standard for the current unit operation state by combining a historical operation data base line, wherein the clearance safety threshold is used for representing the safety boundaries of a clearance system under different risk levels, and the historical operation data base line is used for representing the distribution characteristics of the clearance value, the generator rotating speed, the blade pitch angle and the wind speed of the unit under a historical stable working condition; S22, under the constraint of the clearance safety threshold and the historical operation data base line, performing automatic analysis on a standardized clearance operation data set, and performing exception screening and marking by combining a multi-parameter coupling relation between wind speed and rotation speed and clearance values, wherein the exception screening and marking comprises screening and marking the clearance safety threshold triggering exception, screening and marking an invalid numerical value exception returned by a clearance system and calculating and marking an adjacent clearance difference value exception, and after the clearance safety threshold triggering, marking abnormal harvest exception after the clearance safety threshold triggering based on a comparison relation of whether a pitch angle change and a clearance value return to a normal range, so as to form exception marking data comprising an exception type label; S23, summarizing and counting the abnormal marking data, respectively calculating the occurrence times, occurrence time periods and distribution characteristics of various types of abnormalities in the running process of the unit, and outputting an abnormal analysis result set, wherein the abnormalities comprise abnormality types, abnormality occurrence time periods, abnormality times statistics, clearance values related to the abnormalities, generator rotating speed, blade pitch angles and wind speeds.
4. The intelligent pre-warning method for the headroom of a wind turbine generator system based on multi-parameter coupling according to claim 3, wherein the step S21 comprises: s211, calling a pre-configured headroom safety threshold based on the standardized headroom operation data set, wherein the headroom safety threshold comprises a normal threshold Early warning threshold Threshold of risk , wherein, ; S212, screening operation data under stable working conditions from the historical operation data, and constructing a historical operation data base line The operation data under the stable working condition meets the following conditions And is also provided with , wherein, Is the first A normalized null value corresponding to the bar history operating data, Is the first Standardized generator rotational speed corresponding to the bar history operating data, The rotational speed variation is the rotational speed variation at the adjacent time, Is a rotational speed stability threshold; s213, based on historical operation data base line under stable working condition Respectively calculating headroom values Rotation speed of generator Pitch angle of blade And wind speed And is used for constructing a comparison benchmark of the current unit running state.
5. The intelligent pre-warning method for the headroom of a wind turbine generator system based on multi-parameter coupling according to claim 4, wherein the step S22 comprises: s221, running the data set at the standardized headroom Calling a headroom safety threshold, carrying out threshold triggering judgment on the headroom value of each time point, and marking the headroom safety threshold triggering abnormality At the same time, identifying the invalid value returned by the clearance system and marking the abnormality of the invalid value ; S222, after eliminating the operation data with invalid values, calculating the clearance difference value of adjacent time points Determining whether adjacent clearance difference values are abnormal under the distribution characteristic constraint of historical operation data base lines, and identifying and marking the adjacent clearance difference value abnormality by combining the coupling relation of wind speed, rotating speed and clearance values ; S223, when detecting that the headroom safety threshold triggers abnormality When the trigger time is taken as the starting point, the observation window is extracted , In order to view the length of the window, For unifying time index intervals, comparing and checking the change of the pitch angle and the return condition of the clearance value in an observation window, and if the change of the pitch angle conforming to the trend of the harvest is not generated or the clearance value is not returned to the normal range, marking that the abnormal harvest is not generated after the clearance safety threshold is triggered ; S224, generating abnormal mark data 。
6. The intelligent pre-warning method for the headroom of a wind turbine generator system based on multi-parameter coupling according to claim 5, wherein S23 comprises: s231, marking data for abnormality Index by time Ordering and labeling for each type of abnormality The label value of the abnormal label in the adjacent time point is 1 and the time interval is not more than Continuously merging the abnormal records of the plurality of classes into the same abnormal event segment, thereby obtaining an abnormal event segment set corresponding to each class of abnormality , wherein, Is the first The number of class exception event segments, Is the first Class exception first Start time and end time of each abnormal event segment; S232, aiming at each type of abnormality Based on its abnormal event segment set Calculating the occurrence times, occurrence time and duration of the abnormal event, and forming an abnormal distribution statistical result, wherein the occurrence times of the abnormal event are defined as the number of abnormal event segments Each abnormal event segment has a duration of And starting time of each abnormal event segment And the end time The abnormal occurrence time period is recorded, so that the time distribution characteristics of various abnormal units in the running process are obtained; S233, for each event segment of each type of exception From normalized headroom run data sets Intercepting the operation data segment in the corresponding time window to form an abnormal associated data segment And will be of exception type Time period of occurrence of abnormality Number of occurrence of abnormality Abnormality related data fragment Structured summarization is carried out, and an abnormal analysis result set is output 。
7. The intelligent pre-warning method for the headroom of a wind turbine generator system based on multi-parameter coupling according to claim 6, wherein the step S3 comprises: S31, based on the anomaly analysis result set, extracting operation data fragments related to an abnormal event from a standardized clearance operation data set according to the anomaly type and the corresponding anomaly occurrence time period, and sorting the clearance value, the generator rotating speed, the blade pitch angle and the wind speed according to time sequence to form a map data set for map construction; S32, respectively constructing multi-type maps for representing the running characteristics and abnormal behaviors of the unit based on the map data set, wherein the multi-type maps specifically comprise the following steps: Constructing a time sequence diagram of the change of the clearance value and the rotating speed of the generator along with time, and representing the time sequence evolution relation of the clearance and the rotating speed; Constructing a scattered point distribution diagram of the rotating speed and the null value of the generator, and representing the related distribution characteristics between the rotating speed and the null value of the generator; Calculating the rotating speed difference value of the generator at the adjacent time index, and constructing a corresponding histogram for representing the rotating speed variation amplitude and the abnormal fluctuation condition thereof; S33, synchronously drawing a linkage change line diagram of the clearance value, the blade pitch angle and the wind speed parameter in a corresponding time window aiming at an abnormal event triggered by the clearance safety threshold, and visually presenting the control response behavior and the parameter coupling characteristic of the unit under the abnormal working condition by comparing the change trend of each operation data before and after the clearance safety threshold is triggered.
8. The intelligent pre-warning method for the headroom of a wind turbine generator system based on multi-parameter coupling according to claim 7, wherein the step S4 comprises: s41, merging the anomaly analysis result set and a corresponding map according to the dimension of the unit, extracting and summarizing a data profile and anomaly details, wherein the data profile comprises a statistical time range, the number of data records and operation data, the anomaly details comprise anomaly types, anomaly occurrence time periods, the number of anomaly event periods and anomaly associated data fragments, a time sequence diagram of a clearance value and a generator rotating speed, a scattered point distribution diagram of the generator rotating speed and the clearance value, a histogram corresponding to an adjacent rotating speed difference value and a linkage change line diagram with the clearance value, blade pitch angle and wind speed under a threshold trigger condition are used as map accessories to be associated and bound with the anomaly details, and a structured report data packet for report generation is formed; S42, based on the structured report data packet, aiming at each type of abnormality, according to the occurrence times of the abnormality Duration with abnormal event segment Calculating an abnormality severity score And determining the severity of the abnormality based on the severity classification threshold Including low, medium, and high, expressed as: ; Wherein, the Is the first A mid-level threshold for class anomalies, Is the first An abnormally high level threshold; S43, processing suggestions are given according to the abnormality type and the abnormality severity, the corresponding map is used as an attachment insertion report, and clearance safety early warning information comprising a unit identifier, the abnormality type, an abnormality occurrence time period, the abnormality severity and the processing suggestions is output.
9. The intelligent wind generating set clearance analysis system based on multi-parameter coupling is used for realizing the intelligent wind generating set clearance early warning method based on multi-parameter coupling as claimed in any one of claims 1 to 8, and is characterized by comprising the following modules: the data summarizing and sorting module is used for acquiring the operation data of the wind generating set, which is derived by the main control SCADA system, including a clearance value, a generator rotating speed, a blade pitch angle and a wind speed, importing the operation data in a CSV format, executing automatic field recognition and unified analysis, and forming a standardized clearance operation data set through data cleaning, missing value filling and format standardization processing; The abnormal screening and counting module is used for automatically analyzing the standardized clearance operation data set under the constraint of a preset clearance safety threshold, a wind speed-rotating speed-clearance value multi-parameter coupling relation and a historical operation data base line, screening and marking clearance safety threshold triggering abnormality, invalid numerical value abnormality, adjacent clearance difference abnormality and abnormal pitch harvest abnormality after threshold triggering, and counting the occurrence times and distribution characteristics of various abnormalities to generate an abnormal analysis result set; The map drawing module is used for constructing a time sequence diagram of the clearance value and the generator rotating speed, a scattered point distribution diagram of the generator rotating speed and the clearance value, a histogram corresponding to adjacent rotating speed difference values and a linkage change line diagram with the clearance value, the blade pitch angle and the wind speed under the threshold triggering condition based on the abnormal analysis result set so as to represent the coupling change relation between the unit operating parameters before and after the occurrence of the abnormality and present the abnormal evolution trend; and the report generation and early warning output module is used for comprehensively arranging the abnormal analysis result set and the corresponding map to generate a standardized analysis report comprising a data profile, an abnormal detail, a risk assessment conclusion and a processing suggestion, determining the severity of the abnormality according to the type of the abnormality and the total duration, and outputting corresponding clearance safety early warning information.

Description

Intelligent wind generating set clearance analysis system and early warning method based on multi-parameter coupling Technical Field The invention relates to the technical field of operation monitoring of wind generating sets, in particular to a wind generating set clearance intelligent analysis system and an early warning method based on multi-parameter coupling. Background With the development of the wind generating set to the directions of large capacity, high tower and long blades, the deflection and swing amplitude of the blades in the complex wind condition and variable working condition operation process are continuously increased, the clearance safety margin between the blades and the tower is gradually reduced, and the clearance operation safety becomes an important factor for influencing the reliability and operation safety of the wind generating set. In order to prevent the risk of tower sweeping, the existing wind turbine generator sets generally adopt a clearance measuring device arranged at a tower or a nacelle, and operation parameters such as a clearance value, a rotating speed, a pitch angle, a wind speed and the like are connected into a master control SCADA system for monitoring so as to realize basic perception of the running state of the wind turbine generator sets. The existing clearance monitoring and risk identification means mainly comprise a single clearance threshold value or simple alarm logic, generally only judges whether an instantaneous clearance value is out of limit, lacks system analysis on coupling relations among multiple operation parameters such as wind speed, rotating speed, pitch angle and the like, and is difficult to accurately distinguish transient disturbance from persistent risk hidden danger. Meanwhile, in the prior art, the recognition of the clearance abnormality is dependent on manual offline investigation or experience judgment, and quantitative statistics and grading evaluation on occurrence frequency, duration and evolution process of the abnormality are lacked, so that early warning results do not have interpretability and disposal guidance, and the problems of false alarm, missing alarm or response lag are easy to occur. In addition, existing methods often lack the ability to output analysis results in the form of visual maps and standardized reports, which is detrimental to the operation and maintenance personnel in quickly understanding the cause of the abnormality and taking targeted treatment measures. Disclosure of Invention The invention provides a multi-parameter coupling-based intelligent analysis system and an early warning method for the clearance of a wind generating set, which are characterized in that a clearance safety threshold, a multi-parameter coupling relation and a historical operation data base line are introduced by carrying out unified analysis and standardization processing on multi-source operation data led out by a master control SCADA system, automatic identification, statistical analysis and severity grading are carried out on multi-type clearance anomalies, and the accurate identification and grading early warning of the clearance operation risk of the wind generating set are realized by combining mapping display and standardized report output, so that the accuracy and the interpretability of the clearance anomaly identification are improved, the pertinence and the timeliness of operation and maintenance decisions are enhanced, the risk of tower sweeping is reduced, and the safe and stable operation of the wind generating set is ensured. A wind generating set clearance intelligent early warning method based on multi-parameter coupling comprises the following steps: S1, acquiring operation data of a wind generating set, which is derived by a master control SCADA system, wherein the operation data comprises a clearance value, a generator rotating speed, a blade pitch angle and a wind speed, the operation data is imported in a CSV format, automatic field identification and unified analysis are performed on the operation data, and a standardized clearance operation data set is formed through data cleaning, missing value filling and data format standardization processing; S2, based on the standardized clearance operation data set, according to a preset clearance safety threshold, and by combining a multiparameter coupling relation between wind speed and rotation speed and a clearance value and a historical operation data base line, automatically analyzing operation data, screening and marking clearance safety threshold triggering abnormality, invalid numerical value abnormality, adjacent clearance difference value abnormality and abnormal pitch harvest abnormality after threshold triggering, and counting occurrence times and distribution characteristics of various abnormalities to generate an abnormality analysis result set; s3, respectively constructing corresponding maps based on the abnormal analysis result set, wherein the ma