Wind generating set clearance intelligent analysis system and early warning method based on multi-parameter coupling

Abstract

The invention relates to the technical field of wind generating set operation monitoring, in particular to a wind generating set clearance intelligent analysis system and early warning method based on multi-parameter coupling, and the method comprises the following steps: S1, importing master control SCADA operation data, completing field identification, cleaning and standardization, and forming a clearance operation data set; s2, in combination with a clearance threshold, a multi-parameter coupling relationship and a historical baseline, recognizing and counting clearance anomalies, and generating an anomaly result set; s3, constructing a time sequence diagram, a scatter diagram, a histogram and a linkage change diagram based on an abnormal result, and presenting abnormal features; and S4, summarizing an abnormal result and a map to generate an analysis report, and outputting early warning information according to an abnormal type and duration. According to the method, rapid identification and closed-loop disposal of the clearance abnormity of the wind generating set are achieved, the operation safety of the set is effectively improved, and the tower sweeping risk is reduced.

Description

Technical Field

[0001] This invention relates to the field of wind turbine generator operation monitoring technology, and in particular to a wind turbine generator airspace intelligent analysis system and early warning method based on multi-parameter coupling. Background Technology

[0002] As wind turbines develop towards larger capacity, taller towers, and longer blades, the deflection and oscillation amplitudes of the blades during operation in complex wind conditions and varying operating conditions are constantly increasing. This leads to a gradual decrease in the safety margin of clearance between the blades and the tower, making clearance safety a crucial factor affecting the reliability and operational safety of wind turbines. To mitigate the risk of tower sweeping, existing wind turbines typically employ clearance measurement devices installed on the tower or nacelle. These devices are then connected to the main control SCADA system for monitoring, including clearance values, engine speed, pitch angle, and wind speed, to achieve basic awareness of the unit's operating status.

[0003] Existing methods for monitoring and identifying airspace risks primarily rely on single airspace thresholds or simple alarm logic. They typically only assess whether instantaneous airspace values ​​exceed limits, lacking systematic analysis of the coupling relationships between multiple operating parameters such as wind speed, engine speed, and propeller pitch angle. This makes it difficult to accurately distinguish between transient disturbances and persistent risks. Furthermore, current technologies rely heavily on manual offline investigation or experience-based judgment for identifying airspace anomalies, lacking quantitative statistics and tiered assessments of anomaly frequency, duration, and evolution. This results in warnings lacking interpretability and guidance, leading to false alarms, missed alarms, or delayed responses. Moreover, existing methods often lack the ability to output analysis results in the form of intuitive graphs and standardized reports, hindering maintenance personnel from quickly understanding the causes of anomalies and taking targeted countermeasures. Summary of the Invention

[0004] This invention provides a wind turbine generator airspace intelligent analysis system and early warning method based on multi-parameter coupling. By uniformly analyzing and standardizing multi-source operating data exported from the main control SCADA system, and introducing airspace safety thresholds, multi-parameter coupling relationships, and historical operating data baselines, it automatically identifies, statistically analyzes, and classifies the severity of various types of airspace anomalies. Combined with graphical display and standardized report output, it achieves accurate identification and graded early warning of airspace operation risks of wind turbine generators, thereby improving the accuracy and interpretability of airspace anomaly identification, enhancing the pertinence and timeliness of operation and maintenance decisions, reducing tower sweep risks, and ensuring the safe and stable operation of wind turbine generators.

[0005] The intelligent airspace warning method for wind turbine generators based on multi-parameter coupling includes the following steps: S1. Obtain the wind turbine generator operation data exported from the main control SCADA system, including clearance value, generator speed, blade pitch angle and wind speed, and import it in CSV format. Perform automatic field recognition and unified parsing on the operation data. Through data cleaning, missing value filling and data format standardization processing, a standardized clearance operation dataset is formed. S2, based on the standardized airspace operation dataset, according to the preset airspace safety threshold, and combined with the multi-parameter coupling relationship between wind speed, rotation speed and airspace value and the historical operation data baseline, the operation data is automatically analyzed, and the airspace safety threshold triggering anomalies, invalid value anomalies, adjacent airspace difference anomalies, and anomalies of failure to properly retract the paddle after the threshold is triggered are filtered and marked. The occurrence frequency and distribution characteristics of various anomalies are statistically analyzed to generate an anomaly analysis result set. S3. Based on the anomaly analysis result set, construct corresponding graphs, including a time series graph of net clearance value and generator speed, a scatter plot of generator speed and net clearance value, a bar chart of adjacent speed difference, and a line graph of linkage change of net clearance value, blade pitch angle and wind speed under threshold triggering conditions. S4. The anomaly analysis result set and corresponding graphs are comprehensively organized to generate a standardized analysis report including data overview, anomaly details, risk assessment conclusions and handling suggestions. The severity of the anomaly is determined according to the anomaly type and total duration, and the corresponding airspace safety warning information is output.

[0006] Optionally, S1 includes: S11. During the operation of the wind turbine generator set, the clearance value is collected by a laser rangefinder mounted on the tower, the generator speed is collected by a speed sensor installed on the generator shaft, the blade pitch angle is collected by a pitch angle position sensor installed in the pitch system, and the wind speed is collected by an anemometer installed on the top of the nacelle. The main control SCADA system synchronously records the operating data according to a unified timestamp and exports it as a CSV file to form the original operating data of the unit. ; S12, exporting the unit's raw operating data from the main control SCADA system. After importing, for The header fields are automatically identified, and data columns corresponding to clearance values, generator speeds, blade pitch angles, and wind speeds are determined through field matching rules. The identified data columns are then parsed and aligned according to a unified time index, and numeric fields are converted to a unified numeric format, resulting in a parsed and synchronized dataset with consistent field meanings and time synchronization. ; S13, parsing the running dataset Perform data cleaning and standardization to create a standardized airspace operation dataset. .

[0007] Optionally, S2 includes: S21. Based on the standardized airspace operation dataset, the preset airspace safety threshold is called, and a comparison benchmark is established for the current unit operation status in combination with the historical operation data baseline. The airspace safety threshold is used to characterize the safety boundary of the airspace system under different risk levels, and the historical operation data baseline is used to characterize the distribution characteristics of the airspace value, generator speed, blade pitch angle and wind speed of the unit under historical stable operating conditions. S22, under the constraints of the airspace safety threshold and the historical operation data baseline, perform automatic analysis on the standardized airspace operation dataset, and perform anomaly screening and marking by combining the multi-parameter coupling relationship between wind speed, rotational speed and airspace value, including screening and marking anomalies triggered by the airspace safety threshold, screening and marking anomalies of invalid values ​​returned by the airspace system, and calculating and marking anomalies of adjacent airspace difference values. After the airspace safety threshold is triggered, based on the comparison relationship between the pitch angle change and whether the airspace value returns to the normal range, mark the anomaly of not properly retracting the paddle after the airspace safety threshold is triggered, forming anomaly marking data including anomaly type labels; S23, summarize and statistically analyze the abnormality marker data, calculate the occurrence frequency, occurrence time period and distribution characteristics of various abnormalities during unit operation, and output the abnormality analysis result set. The abnormalities include abnormality type, abnormality occurrence time period, abnormality frequency statistics, and abnormality-related net clearance value, generator speed, blade pitch angle and wind speed.

[0008] Optionally, S21 includes: S211, based on the standardized airspace operation dataset, invoke the pre-configured airspace safety thresholds, which include normal thresholds. Warning threshold and danger threshold ,in, ; S212, Select operating data under stable operating conditions from historical operating data to construct a historical operating data baseline. The operating data under the stable operating conditions satisfy and ,in, For the first The standardized net air value corresponding to each historical operational data point. For the first The standardized generator speed corresponding to each historical operating data point. This represents the change in rotational speed between adjacent time points. This is the speed stability threshold; S213, based on historical operating data baseline under stable operating conditions Calculate the net air value separately. Generator speed Blade pitch angle and wind speed The statistical characteristics are used to construct a benchmark for the current operating status of the unit.

[0009] Optionally, S22 includes: S221, in the standardized clearance operation dataset The system calls the airspace safety threshold, performs threshold trigger determination on the airspace value at each time point, and marks the airspace safety threshold trigger anomaly. At the same time, invalid values ​​returned by the air clearance system are identified and marked as invalid values ​​as abnormal. ; S222, After removing invalid values ​​from the running data, calculate the net air difference between adjacent time points. Under the constraints of the distribution characteristics of historical operational data baseline, it determines whether adjacent clearance differences are abnormal. At the same time, it identifies and marks abnormal adjacent clearance differences by combining the coupling relationship between wind speed, air speed, and clearance value. ; S223, when an abnormality is detected in the airspace safety threshold. At that time, the observation window is extracted starting from the trigger moment. , To observe the window length, To standardize the time indexing interval, the changes in pitch angle and the return of clearance value are compared and verified within the observation window. If the pitch angle does not show a change consistent with the pitch recovery trend or the clearance value does not return to the normal range, it is marked as an abnormality of not properly recovering the pitch after the clearance safety threshold is triggered. ; S224, Generate anomaly marker data .

[0010] Optionally, S23 includes: S231, marking abnormal data Indexed by time Sort the data and assign each type of anomaly label accordingly. Set the label value of the anomaly label in adjacent time points to 1 and the time interval to no more than 1. The abnormal records are continuously merged into the same abnormal event segment, thus obtaining the set of abnormal event segments corresponding to each type of abnormality. ,in, For the first The number of exception event segments, For the first The first class of exceptions The start and end times of each abnormal event segment; S232, for each type of exception Based on its abnormal event segment set Calculate the number of anomalies, their occurrence time, and duration, and generate anomaly distribution statistics, where the number of anomalies is defined as the number of anomaly event segments. The duration of each abnormal event segment is And the start time of each abnormal event segment. and the end time The time period of occurrence of the anomaly is recorded to obtain the temporal distribution characteristics of various anomalies during unit operation. S233, for each event segment of each type of exception From the standardized net air operation dataset Extract the runtime data segment within the corresponding time window to form an anomaly-related data segment. and exception type Time period of abnormal occurrence Number of abnormal occurrences and abnormal related data fragments Perform structured aggregation and output anomaly analysis result set. .

[0011] Optionally, S3 includes: S31. Based on the anomaly analysis result set, according to the anomaly type and its corresponding anomaly occurrence time period, extract the operation data fragments associated with the anomaly event from the standardized airspace operation dataset, and organize the airspace value, generator speed, blade pitch angle and wind speed in chronological order to form a map dataset for map construction. S32, Based on the aforementioned map dataset, construct multiple types of maps to characterize the unit's operational characteristics and abnormal behaviors, specifically including: Construct a time series plot of the change in headroom value and generator speed over time to characterize the temporal evolution relationship between headroom and speed; Construct a scatter plot of generator speed versus net clearance to characterize the correlation between the two. Calculate the generator speed difference at adjacent time indices and construct the corresponding bar chart to characterize the speed change amplitude and its abnormal fluctuations; S33, in response to abnormal events that trigger the airspace safety threshold, simultaneously plots a line graph showing the linked changes in airspace value, blade pitch angle, and wind speed parameters within the corresponding time window. By comparing the changing trends of various operating data before and after the airspace safety threshold is triggered, the control response behavior and parameter coupling characteristics of the unit under abnormal operating conditions are presented intuitively.

[0012] Optionally, S4 includes: S41, the anomaly analysis result set and corresponding graphs are merged according to the unit dimension, and the data overview and anomaly details are extracted and summarized. The data overview includes the statistical time range, the number of data records and the operating data. The anomaly details include the anomaly type, the time period of the anomaly occurrence, the number of anomaly event segments and the anomaly-related data segments. The time series graph of net clearance value and generator speed, the scatter plot of generator speed and net clearance value, the bar chart corresponding to the difference between adjacent speeds, and the line graph of the linkage change of net clearance value, blade pitch angle and wind speed under the threshold trigger condition are used as graph attachments and associated with the anomaly details to form a structured report data package for report generation. S42, based on the structured report data packet, for each type of anomaly, according to the number of times the anomaly occurred. Duration of the abnormal event segment Calculate the severity score of the abnormality. The severity of the anomaly is determined based on the severity grading threshold. This includes low, medium, and high, represented as: ; in, For the first The intermediate level threshold for class-specific anomalies, For the first High-level threshold for class-specific anomalies; S43 provides handling suggestions based on the type and severity of the anomaly, inserts the corresponding graph as an attachment into the report, and outputs airspace safety warning information, including unit identification, anomaly type, time period of anomaly occurrence, severity of anomaly, and handling suggestions.

[0013] The intelligent airspace analysis system for wind turbine generators based on multi-parameter coupling is used to implement the aforementioned intelligent airspace early warning method for wind turbine generators based on multi-parameter coupling, and includes the following modules: Data aggregation and organization module: acquires wind turbine operation data exported from the main control SCADA system, including clearance value, generator speed, blade pitch angle and wind speed, imports it in CSV format, performs automatic field recognition and unified parsing, and forms a standardized clearance operation dataset after data cleaning, missing value filling and format standardization. Anomaly screening and statistics module: Under the constraints of preset airspace safety threshold, multi-parameter coupling relationship of wind speed-rotation speed-airspace value and historical operation data baseline, the standardized airspace operation dataset is automatically analyzed, and the airspace safety threshold trigger anomalies, invalid value anomalies, adjacent airspace difference anomalies, and anomalies of failure to properly stop the paddle after threshold triggering are screened and marked. The occurrence frequency and distribution characteristics of various anomalies are statistically analyzed to generate an anomaly analysis result set. The graph plotting module constructs a time series graph of net clearance value and generator speed, a scatter plot of generator speed and net clearance value, a bar chart of adjacent speed differences, and a line graph of the linkage change of net clearance value, blade pitch angle and wind speed under threshold triggering conditions based on the anomaly analysis result set, so as to characterize the coupling change relationship between unit operating parameters before and after the anomaly occurs and present the anomaly evolution trend. The report generation and early warning output module comprehensively organizes the anomaly analysis results set and corresponding graphs to generate a standardized analysis report that includes a data overview, anomaly details, risk assessment conclusions and handling suggestions. Based on the anomaly type and total duration, the module determines the severity of the anomaly and outputs corresponding airspace safety early warning information.

[0014] The beneficial effects of this invention are: This invention automatically identifies, aligns, fills in missing values, and standardizes the clearance values, generator speed, blade pitch angle, and wind speed exported from the main control SCADA system to construct a unified, standardized clearance operation dataset. It also introduces clearance safety thresholds and historical operation data baselines as dual comparison benchmarks to achieve automated and batch analysis of the unit's clearance operation status. This avoids the problems of low efficiency and easy omissions caused by relying on experience judgment and item-by-item verification in traditional manual inspection methods, and significantly improves the analysis efficiency and consistency of clearance operation data.

[0015] This invention, by combining the multi-parameter coupling relationship between wind speed, engine speed, and air clearance value on a standardized air clearance operation dataset, classifies and filters abnormalities triggered by air clearance safety thresholds, abnormalities of invalid values, abnormalities of adjacent air clearance differences, and abnormalities of failure to properly retract propellers after threshold triggering. It also merges continuous event segments and statistically analyzes the frequency and duration of occurrence, constructing a structured set of anomaly analysis results. This enables not only accurate identification of unit air clearance anomalies but also quantitative characterization from multiple dimensions such as frequency of occurrence, duration, and evolution process, thereby improving the accuracy and interpretability of identifying potential tower sweeping hazards and control anomalies.

[0016] This invention constructs multi-type maps based on anomaly analysis result sets and further maps the frequency and duration of anomalies to anomaly severity scores. It outputs airspace safety early warning information according to graded thresholds and generates standardized analysis reports containing data overview, anomaly details, risk assessment conclusions, and targeted handling suggestions. This achieves closed-loop management from anomaly discovery to risk assessment, handling suggestions, and early warning output, effectively supporting maintenance personnel in carrying out differentiated handling and priority decisions. It reduces the risk of tower sweeping accidents or unit power limitation operation caused by untimely handling of airspace anomalies, and improves the operational safety and maintenance management level of wind turbine units. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the early warning method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the system functional modules according to an embodiment of the present invention. Detailed Implementation

[0019] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Those skilled in the art may employ other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0020] like Figure 1 As shown, the intelligent airspace warning method for wind turbine generators based on multi-parameter coupling includes the following steps: S1. Obtain the wind turbine generator operation data exported from the main control SCADA system, including clearance value, generator speed, blade pitch angle and wind speed, and import it in CSV format. Perform automatic field recognition and unified parsing on the operation data. Through data cleaning, missing value filling and data format standardization processing, a standardized clearance operation dataset is formed. S2, based on a standardized airspace operation dataset, automatically analyzes the operation data according to the preset airspace safety threshold, combined with the multi-parameter coupling relationship between wind speed, rotational speed and airspace value, and the historical operation data baseline. It filters and marks airspace safety threshold triggering anomalies, invalid value anomalies, adjacent airspace difference anomalies, and anomalies of failure to properly retract the paddle after threshold triggering. It also statistically analyzes the occurrence frequency and distribution characteristics of various anomalies and generates an anomaly analysis result set. S3, based on the anomaly analysis result set, construct corresponding graphs, including time series graphs of net clearance value and generator speed, scatter distribution graphs of generator speed and net clearance value, bar graphs of adjacent speed differences, and line graphs of linkage changes of net clearance value, blade pitch angle and wind speed under threshold triggering conditions, to characterize the coupling change relationship between unit operating parameters before and after the anomaly occurs, so as to present the unit net clearance operation characteristics and anomaly evolution trend; S4 comprehensively organizes the anomaly analysis results set and corresponding graphs to generate a standardized analysis report that includes a data overview, anomaly details, risk assessment conclusions and handling suggestions. It also determines the severity of the anomaly based on the anomaly type and total duration and outputs corresponding airspace safety warning information.

[0021] S1 includes: S11. During the operation of the wind turbine generator set, the clearance value is collected by a laser rangefinder mounted on the tower, the generator speed is collected by a speed sensor installed on the generator shaft, the blade pitch angle is collected by a pitch angle position sensor installed in the pitch system, and the wind speed is collected by an anemometer installed on the top of the nacelle. The main control SCADA system synchronously records the operating data according to a unified timestamp and exports it as a CSV file to form the unit's original operating data. ; S12, exporting the unit's raw operating data from the main control SCADA system. After importing, for The header fields are automatically identified, and data columns corresponding to clearance values, generator speeds, blade pitch angles, and wind speeds are determined through field matching rules. The identified data columns are then parsed and aligned according to a unified time index, and numeric fields are converted to a unified numeric format, resulting in a parsed and synchronized dataset with consistent field meanings and time synchronization. Specifically, it includes: (1) The original operating data of the unit The header fields are automatically identified, and the corresponding net values ​​are determined according to preset field matching rules. Generator speed Blade pitch angle and wind speed Data columns; (2) For the identified net clearance value Generator speed Blade pitch angle and wind speed Parsing and alignment are performed using a unified time index, mapping data from different sampling time points to a unified time axis to obtain a time synchronization parameter sequence, represented as: ; in, Indicates parameters , , , Any parameter in, The sampling timestamp of the original data. For a unified time index, To standardize the time window width corresponding to the time index, The number of valid sampling points within the time window. For unified time indexing The parameter values ​​are as follows; (3) Net value after time alignment Generator speed Blade pitch angle and wind speed Convert to a uniform numerical format, construct a parsed execution dataset with consistent field meanings and synchronized time, represented as: ; in, The total number of time indexes; S13, parsing the running dataset Perform data cleaning and standardization to create a standardized airspace operation dataset. Specifically, it includes: (1) When a certain data point in the parsed dataset is missing in the time series, linear interpolation is used to fill the gap, as shown below: ; in, Time to be filled The values ​​of the running data parameters, , These are the closest known times before and after the missing point. , These are the known parameter values ​​at the corresponding times; (2) Perform Z-score standardization on each running data, as follows: ; ; ; ; in, , , , These represent the mean values ​​of headroom, rotational speed, pitch angle, and wind speed in the dataset, respectively. , , , These are the standard deviations of the corresponding parameters. , , , These are the standardized headroom, generator speed, blade pitch angle, and wind speed, respectively. (3) Output standardized airspace operation dataset ,in, For the first The timestamp of the running data, This represents the total number of data samples used in the operation.

[0022] S2 includes: S21, based on the standardized airspace operation dataset, calls the preset airspace safety threshold and combines it with the historical operation data baseline to establish a reference for the current unit operation status. The airspace safety threshold is used to characterize the safety boundary of the airspace system under different risk levels, and the historical operation data baseline is used to characterize the distribution characteristics of the unit's airspace value, generator speed, blade pitch angle and wind speed under historical stable operating conditions. S22, under the constraints of the airspace safety threshold and the historical operation data baseline, performs automatic analysis on the standardized airspace operation dataset, and performs anomaly screening and marking by combining the multi-parameter coupling relationship between wind speed, rotational speed and airspace value. This includes screening and marking anomalies triggered by the airspace safety threshold, screening and marking anomalies of invalid values ​​returned by the airspace system, and calculating and marking anomalies of adjacent airspace difference values. After the airspace safety threshold is triggered, based on the comparison relationship between the pitch angle change and whether the airspace value returns to the normal range, it marks the anomaly of not properly retracting the paddle after the airspace safety threshold is triggered, forming anomaly marking data including anomaly type labels. S23 summarizes and statistically analyzes the anomaly marker data, calculates the occurrence frequency, occurrence time period, and distribution characteristics of various anomalies during unit operation, and outputs anomaly analysis result set. Anomalies include anomaly type, anomaly occurrence time period, anomaly frequency statistics, and anomaly-related net clearance value, generator speed, blade pitch angle, and wind speed.

[0023] S21 includes: S211, based on the standardized airspace operation dataset, calls the pre-configured airspace safety thresholds, which include normal thresholds. Warning threshold and danger threshold ,in, ; S212, Select operating data under stable operating conditions from historical operating data to construct a historical operating data baseline. Operating data under stable conditions meets and ,in, For the first The standardized net air value corresponding to each historical operational data point. For the first The standardized generator speed corresponding to each historical operating data point. This represents the change in rotational speed between adjacent time points. This is the speed stability threshold; ; in, It is the 95th percentile statistic; S213, based on historical operating data baseline under stable operating conditions Calculate the net air value separately. Generator speed Blade pitch angle and wind speed The statistical characteristics are used to construct a benchmark for the current unit operating status, and are expressed as follows: ; ; in, This is the average of historical operating data. The standard deviation of historical operating data. For historical operating data, , This represents the number of samples under historical stable operating conditions.

[0024] S22 includes: S221, in the standardized clearance operation dataset The system calls the airspace safety threshold, performs threshold trigger determination on the airspace value at each time point, and marks the airspace safety threshold trigger anomaly. At the same time, invalid values ​​returned by the air clearance system are identified and marked as invalid values ​​as abnormal. , is represented as: ; ; in, An abnormality was triggered due to an airspace safety threshold; 1 indicates triggered, 0 indicates not triggered. for Standardized headroom value at any given time. The value is invalid or abnormal; 1 indicates invalid, and 0 indicates valid. This is a set of invalid values. S222, After removing invalid values ​​from the running data, calculate the net air difference between adjacent time points. Under the constraints of the distribution characteristics of historical operational data baseline, it determines whether adjacent clearance differences are abnormal. At the same time, it identifies and marks abnormal adjacent clearance differences by combining the coupling relationship between wind speed, air speed, and clearance value. , is represented as: ; ; ; in, The threshold for abnormal adjacent airspace differences. For threshold coefficient, , These are the historical operating data baselines under historical stable operating conditions. The mean and standard deviation; S223, when an abnormality is detected in the airspace safety threshold. At that time, the observation window is extracted starting from the trigger moment. , To observe the window length, To standardize the time indexing interval, the changes in pitch angle and the return of clearance value are compared and verified within the observation window. If the pitch angle does not show a change consistent with the pitch recovery trend or the clearance value does not return to the normal range, it is marked as an abnormality of not properly recovering the pitch after the clearance safety threshold is triggered. Specifically, it includes: (1) Pitch angle response: ; (2) Net void regression determination: ; (3) Judgment of abnormal oar recovery: ; in, This represents the change in propeller pitch angle within the window after threshold triggering. The minimum propeller response threshold, This is an indicator of whether the net air value has returned to the normal range; 1 indicates return, and 0 indicates no return. S224, Generate anomaly marker data .

[0025] S23 includes: S231, marking abnormal data Indexed by time Sort the data and assign each type of anomaly label accordingly. Set the label value of the anomaly label in adjacent time points to 1 and the time interval to no more than 1. The abnormal records are continuously merged into the same abnormal event segment, thus obtaining the set of abnormal event segments corresponding to each type of abnormality. ,in, For the first The number of exception event segments, For the first The first class of exceptions The start and end times of each abnormal event segment; S232, for each type of exception Based on its abnormal event segment set Calculate the number of anomalies, their occurrence time, and duration, and generate anomaly distribution statistics, where the number of anomalies is defined as the number of anomaly event segments. The duration of each abnormal event segment is And the start time of each abnormal event segment. and the end time The time period of occurrence of this anomaly is recorded, thus obtaining the temporal distribution characteristics of various anomalies during unit operation, represented as follows: ; S233, for each event segment of each type of exception From the standardized net air operation dataset Extract the runtime data segment within the corresponding time window to form an anomaly-related data segment. and exception type Time period of abnormal occurrence Number of abnormal occurrences and abnormal related data fragments Perform structured aggregation and output anomaly analysis result set. , is represented as: .

[0026] S3 includes: S31. Based on the anomaly analysis result set, according to the anomaly type and its corresponding anomaly occurrence time period, extract the operation data fragments associated with the anomaly event from the standardized net clearance operation dataset, and organize the net clearance value, generator speed, blade pitch angle and wind speed in chronological order to form a map dataset for map construction. S32, based on the graph dataset, constructs multiple types of graphs to characterize unit operating characteristics and abnormal behaviors, specifically including: Construct a time series plot of the change in headroom value and generator speed over time to characterize the temporal evolution relationship between headroom and speed; Construct a scatter plot of generator speed versus net clearance to characterize the correlation between the two. Calculate the generator speed difference at adjacent time indices and construct the corresponding bar chart to characterize the speed change amplitude and its abnormal fluctuations; S33, in response to abnormal events that trigger the airspace safety threshold, simultaneously plots a line graph showing the linked changes in airspace value, blade pitch angle, and wind speed parameters within the corresponding time window. By comparing the changing trends of various operating data before and after the airspace safety threshold is triggered, the control response behavior and parameter coupling characteristics of the unit under abnormal operating conditions are presented intuitively.

[0027] S4 includes: S41, merge the anomaly analysis result set and corresponding graphs by unit dimension, extract and summarize the data overview and anomaly details. The data overview includes the statistical time range, the number of data records and the operating data. The anomaly details include the anomaly type, the time period of the anomaly, the number of anomaly event segments and the anomaly-related data segments. The time series graph of net clearance value and generator speed, the scatter plot of generator speed and net clearance value, the bar chart corresponding to the difference between adjacent speeds, and the line graph of the linkage change of net clearance value, blade pitch angle and wind speed under the threshold trigger condition are used as graph attachments and associated with the anomaly details to form a structured report data package for report generation. S42, based on structured report data packets, for each type of exception, based on the number of exceptions occurring. Duration of the abnormal event segment Calculate the severity score of the abnormality. The severity of the anomaly is determined based on the severity grading threshold. This includes low, medium, and high, represented as: ; ; ; in, For the first The intermediate level threshold for class-specific anomalies, For the first High-level threshold for class anomalies, For the first The total duration of the anomaly within the statistical period. , These are the corresponding weight coefficients; ; ; in, , The first Mean and standard deviation of the severity scores for each class of abnormality. , These are the corresponding threshold coefficients. ; S43 provides handling suggestions based on the type and severity of the anomaly, inserts the corresponding graph as an attachment into the report, and outputs airspace safety warning information, including unit identification, anomaly type, time period of anomaly occurrence, severity of anomaly, and handling suggestions; Recommendations include: (a) Anomaly triggered by the airspace safety threshold: (1) Low severity: It is recommended to strengthen operational monitoring, with a focus on the changing trend of airspace clearance under high wind speed or high speed conditions; It is recommended to appropriately lower the upper limit of the rotational speed or optimize the control strategy parameters without affecting power generation. It is recommended to continue monitoring subsequent operational data to confirm whether the net air value has stably returned to the normal range.

[0028] (2) Moderate severity: It is recommended to check the pitch control logic of the unit and verify whether the control command is issued normally when the threshold is triggered; It is recommended to set temporary limits on the unit's operating parameters to reduce the duration of operation under high-risk conditions; It is recommended to arrange for maintenance personnel to conduct planned onboard inspections.

[0029] (3) High severity: It is recommended to immediately implement power limiting or shutdown measures for the generating unit; It is recommended to conduct an on-site inspection as soon as possible, focusing on the condition of the blades, tower, and clearance measurement device; It is recommended that the unit be prohibited from resuming high-load operation until the hidden dangers are eliminated.

[0030] (ii) Invalid numerical anomalies: (1) Low severity: It is recommended to check the data communication status and data acquisition continuity of the airspace system; It is recommended to filter out short-term invalid values ​​at the software level and continuously monitor the data recovery process.

[0031] (2) Moderate severity: It is recommended to check the working status and signal quality of the air clearance measurement device; It is recommended to clean and maintain the sensor lens or probe; It is recommended that the relevant data channels be reviewed.

[0032] (3) High severity: It is recommended that the clearance measurement device be inspected or replaced on-site immediately. Before the airspace data returns to normal, it is recommended that the units adopt a conservative operating strategy or shut down. It is recommended to conduct a thorough review and analysis of the operational data during the period of abnormality.

[0033] (iii) Abnormal difference in adjacent clearance: (1) Low severity: It is recommended to continuously monitor the trend of changes in the net airspace difference to determine whether it is a short-term disturbance; It is recommended to analyze whether the changes in wind speed and rotational speed are caused by environmental factors.

[0034] (2) Moderate severity: It is recommended to check the impeller's operating status and pay attention to whether there is any dynamic imbalance. It is recommended to verify the installation status and calibration of the measuring device; If necessary, it is recommended to adjust the unit's operating strategy to reduce the probability of anomalies.

[0035] (3) High severity: It is recommended to arrange an onboard inspection immediately, focusing on checking for blade damage, deformation, or foreign objects attached to the blades. It is recommended to suspend unit operation until the cause is confirmed; Conduct specific analysis on abnormal data to serve as a basis for subsequent maintenance decisions.

[0036] (iv) Abnormal failure to properly retract the paddle after threshold triggering: (1) Low severity: It is recommended to review the response logs of the pitch system to confirm whether there is any slight delay. It is recommended to closely monitor subsequent threshold-triggered events. (2) Moderate severity: It is recommended to inspect the pitch system actuators, control signals, and feedback loops; It is recommended to check and adjust the pitch system parameters as necessary. It is recommended to schedule planned maintenance.

[0037] (3) High severity: It is recommended to shut down the machine immediately. It is recommended to focus on inspecting the pitch system hardware, control modules, and safety protection logic; The unit must not be allowed to resume operation until the problem is completely resolved.

[0038] like Figure 2 As shown, the intelligent airspace analysis system for wind turbine generators based on multi-parameter coupling is used to implement the aforementioned intelligent early warning method for wind turbine generator airspace based on multi-parameter coupling, and includes the following modules: Data aggregation and organization module: acquires wind turbine operation data exported from the main control SCADA system, including clearance value, generator speed, blade pitch angle and wind speed, imports it in CSV format, performs automatic field recognition and unified parsing, and forms a standardized clearance operation dataset after data cleaning, missing value filling and format standardization. Anomaly Screening and Statistics Module: Under the constraints of preset airspace safety threshold, multi-parameter coupling relationship of wind speed-rotation speed-airspace value and historical operation data baseline, the module automatically analyzes the standardized airspace operation dataset, screens and marks anomalies triggered by airspace safety threshold, invalid value anomalies, adjacent airspace difference anomalies, and anomalies of failure to properly retract the paddle after threshold triggering, and generates an anomaly analysis result set by statistically analyzing the occurrence frequency and distribution characteristics of various anomalies. The graph plotting module constructs a time series graph of net clearance value and generator speed, a scatter plot of generator speed and net clearance value, a bar chart of adjacent speed differences, and a line graph of the linkage change of net clearance value, blade pitch angle and wind speed under threshold triggering conditions based on the anomaly analysis result set, so as to characterize the coupling change relationship between unit operating parameters before and after the anomaly occurs and present the anomaly evolution trend. The report generation and early warning output module comprehensively organizes the anomaly analysis results set and corresponding graphs to generate a standardized analysis report that includes a data overview, anomaly details, risk assessment conclusions and handling suggestions. Based on the anomaly type and total duration, the severity of the anomaly is determined, and corresponding airspace safety early warning information is output.

[0039] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0040] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent early warning of airspace clearance for wind turbine generators based on multi-parameter coupling, characterized in that, Includes the following steps: S1. Obtain the wind turbine generator operation data exported from the main control SCADA system, including clearance value, generator speed, blade pitch angle and wind speed, and import it in CSV format. Perform automatic field recognition and unified parsing on the operation data. Through data cleaning, missing value filling and data format standardization processing, a standardized clearance operation dataset is formed. S2, based on the standardized airspace operation dataset, according to the preset airspace safety threshold, and combined with the multi-parameter coupling relationship between wind speed, rotation speed and airspace value and the historical operation data baseline, the operation data is automatically analyzed, and the airspace safety threshold triggering anomalies, invalid value anomalies, adjacent airspace difference anomalies, and anomalies of failure to properly retract the paddle after the threshold is triggered are filtered and marked. The occurrence frequency and distribution characteristics of various anomalies are statistically analyzed to generate an anomaly analysis result set. S3. Based on the anomaly analysis result set, construct corresponding graphs, including a time series graph of net clearance value and generator speed, a scatter plot of generator speed and net clearance value, a bar chart of adjacent speed difference, and a line graph of linkage change of net clearance value, blade pitch angle and wind speed under threshold triggering conditions. S4. The anomaly analysis result set and corresponding graphs are comprehensively organized to generate a standardized analysis report including data overview, anomaly details, risk assessment conclusions and handling suggestions. The severity of the anomaly is determined according to the anomaly type and total duration, and the corresponding airspace safety warning information is output.

2. The intelligent early warning method for airspace clearance of wind turbine generator sets based on multi-parameter coupling according to claim 1, characterized in that, S1 includes: S11. During the operation of the wind turbine generator set, the clearance value is collected by a laser rangefinder mounted on the tower, the generator speed is collected by a speed sensor installed on the generator shaft, the blade pitch angle is collected by a pitch angle position sensor installed in the pitch system, and the wind speed is collected by an anemometer installed on the top of the nacelle. The main control SCADA system synchronously records the operating data according to a unified timestamp and exports it as a CSV file to form the original operating data of the unit. ; S12, exporting the unit's raw operating data from the main control SCADA system. After importing, for The header fields are automatically identified, and data columns corresponding to clearance values, generator speeds, blade pitch angles, and wind speeds are determined through field matching rules. The identified data columns are then parsed and aligned according to a unified time index, and numeric fields are converted to a unified numeric format, resulting in a parsed and synchronized dataset with consistent field meanings and time synchronization. ; S13, parsing the running dataset Perform data cleaning and standardization to create a standardized airspace operation dataset. .

3. The intelligent early warning method for airspace clearance of wind turbine generator sets based on multi-parameter coupling according to claim 1, characterized in that, S2 includes: S21. Based on the standardized airspace operation dataset, the preset airspace safety threshold is called, and a comparison benchmark is established for the current unit operation status in combination with the historical operation data baseline. The airspace safety threshold is used to characterize the safety boundary of the airspace system under different risk levels, and the historical operation data baseline is used to characterize the distribution characteristics of the airspace value, generator speed, blade pitch angle and wind speed of the unit under historical stable operating conditions. S22, under the constraints of the airspace safety threshold and the historical operation data baseline, perform automatic analysis on the standardized airspace operation dataset, and perform anomaly screening and marking by combining the multi-parameter coupling relationship between wind speed, rotational speed and airspace value, including screening and marking anomalies triggered by the airspace safety threshold, screening and marking anomalies of invalid values ​​returned by the airspace system, and calculating and marking anomalies of adjacent airspace difference values. After the airspace safety threshold is triggered, based on the comparison relationship between the pitch angle change and whether the airspace value returns to the normal range, mark the anomaly of not properly retracting the paddle after the airspace safety threshold is triggered, forming anomaly marking data including anomaly type labels; S23, summarize and statistically analyze the abnormality marker data, calculate the occurrence frequency, occurrence time period and distribution characteristics of various abnormalities during unit operation, and output the abnormality analysis result set. The abnormalities include abnormality type, abnormality occurrence time period, abnormality frequency statistics, and abnormality-related net clearance value, generator speed, blade pitch angle and wind speed.

4. The intelligent early warning method for airspace clearance of wind turbine generator sets based on multi-parameter coupling according to claim 3, characterized in that, S21 includes: S211, based on the standardized airspace operation dataset, invoke the pre-configured airspace safety thresholds, which include normal thresholds. Warning threshold and danger threshold ,in, ; S212, Select operating data under stable operating conditions from historical operating data to construct a historical operating data baseline. The operating data under the stable operating conditions satisfy and ,in, For the first The standardized net air value corresponding to each historical operational data point. For the first The standardized generator speed corresponding to each historical operating data point. This represents the change in rotational speed between adjacent time points. This is the speed stability threshold; S213, based on historical operating data baseline under stable operating conditions Calculate the net air value separately. Generator speed Blade pitch angle and wind speed The statistical characteristics are used to construct a benchmark for the current operating status of the unit.

5. The intelligent early warning method for airspace clearance of wind turbine generator sets based on multi-parameter coupling according to claim 4, characterized in that, S22 includes: S221, in the standardized clearance operation dataset The system calls the airspace safety threshold, performs threshold trigger determination on the airspace value at each time point, and marks the airspace safety threshold trigger anomaly. At the same time, invalid values ​​returned by the air clearance system are identified and marked as invalid values ​​as abnormal. ; S222, After removing invalid values ​​from the running data, calculate the net air difference between adjacent time points. Under the constraints of the distribution characteristics of historical operational data baseline, it determines whether adjacent clearance differences are abnormal. At the same time, it identifies and marks abnormal adjacent clearance differences by combining the coupling relationship between wind speed, air speed, and clearance value. ; S223, when an abnormality is detected in the airspace safety threshold. At that time, the observation window is extracted starting from the trigger moment. , To observe the window length, To standardize the time indexing interval, the changes in pitch angle and the return of clearance value are compared and verified within the observation window. If the pitch angle does not show a change consistent with the pitch recovery trend or the clearance value does not return to the normal range, it is marked as an abnormality of failure to properly recover the pitch after the clearance safety threshold is triggered. ; S224, Generate anomaly marker data .

6. The intelligent early warning method for airspace clearance of wind turbine generator sets based on multi-parameter coupling according to claim 5, characterized in that, S23 includes: S231, marking abnormal data Indexed by time Sort the data and assign each type of anomaly label accordingly. Set the label value of the anomaly label in adjacent time points to 1 and the time interval to no more than 1. The abnormal records are continuously merged into the same abnormal event segment, thus obtaining the set of abnormal event segments corresponding to each type of abnormality. ,in, For the first The number of exception event segments, For the first The first class of exceptions The start and end times of each abnormal event segment; S232, for each type of exception Based on its abnormal event segment set Calculate the number of anomalies, their occurrence time, and duration, and generate anomaly distribution statistics, where the number of anomalies is defined as the number of anomaly event segments. The duration of each abnormal event segment is And the start time of each abnormal event segment. and the end time The time period of occurrence of the anomaly is recorded to obtain the temporal distribution characteristics of various anomalies during unit operation. S233, for each event segment of each type of exception From the standardized net air operation dataset Extract the runtime data segment within the corresponding time window to form an anomaly-related data segment. and exception type Time period of abnormal occurrence Number of abnormal occurrences and abnormal related data fragments Perform structured aggregation and output anomaly analysis result set. .

7. The intelligent early warning method for airspace clearance of wind turbine generator sets based on multi-parameter coupling according to claim 6, characterized in that, S3 includes: S31. Based on the anomaly analysis result set, according to the anomaly type and its corresponding anomaly occurrence time period, extract the operation data fragments associated with the anomaly event from the standardized airspace operation dataset, and organize the airspace value, generator speed, blade pitch angle and wind speed in chronological order to form a map dataset for map construction. S32, Based on the aforementioned map dataset, construct multiple types of maps to characterize the unit's operational characteristics and abnormal behaviors, specifically including: Construct a time series plot of the change in headroom value and generator speed over time to characterize the temporal evolution relationship between headroom and speed; Construct a scatter plot of generator speed versus net clearance to characterize the correlation between the two. Calculate the generator speed difference at adjacent time indices and construct the corresponding bar chart to characterize the speed change amplitude and its abnormal fluctuations; S33, in response to abnormal events that trigger the airspace safety threshold, simultaneously plots a line graph showing the linked changes in airspace value, blade pitch angle, and wind speed parameters within the corresponding time window. By comparing the changing trends of various operating data before and after the airspace safety threshold is triggered, the control response behavior and parameter coupling characteristics of the unit under abnormal operating conditions are presented intuitively.

8. The intelligent early warning method for airspace clearance of wind turbine generator sets based on multi-parameter coupling according to claim 7, characterized in that, S4 includes: S41, the anomaly analysis result set and corresponding graphs are merged according to the unit dimension, and the data overview and anomaly details are extracted and summarized. The data overview includes the statistical time range, the number of data records and the operating data. The anomaly details include the anomaly type, the time period of the anomaly occurrence, the number of anomaly event segments and the anomaly-related data segments. The time series graph of net clearance value and generator speed, the scatter plot of generator speed and net clearance value, the bar chart corresponding to the difference between adjacent speeds, and the line graph of the linkage change of net clearance value, blade pitch angle and wind speed under the threshold trigger condition are used as graph attachments and associated with the anomaly details to form a structured report data package for report generation. S42, based on the structured report data packet, for each type of anomaly, according to the number of times the anomaly occurred. Duration of the abnormal event segment Calculate the severity score of the abnormality. The severity of the anomaly is determined based on the severity grading threshold. This includes low, medium, and high, represented as: ； in, For the first The intermediate level threshold for class-specific anomalies, For the first High-level threshold for class-specific anomalies; S43 provides handling suggestions based on the type and severity of the anomaly, inserts the corresponding graph as an attachment into the report, and outputs airspace safety warning information, including unit identification, anomaly type, time period of anomaly occurrence, severity of anomaly, and handling suggestions.

9. A wind turbine generator airspace intelligent analysis system based on multi-parameter coupling, used to implement the wind turbine generator airspace intelligent early warning method based on multi-parameter coupling as described in any one of claims 1-8, characterized in that, Includes the following modules: Data aggregation and organization module: acquires wind turbine operation data exported from the main control SCADA system, including clearance value, generator speed, blade pitch angle and wind speed, imports it in CSV format, performs automatic field recognition and unified parsing, and forms a standardized clearance operation dataset after data cleaning, missing value filling and format standardization. Anomaly screening and statistics module: Under the constraints of preset airspace safety threshold, multi-parameter coupling relationship of wind speed-rotation speed-airspace value and historical operation data baseline, the standardized airspace operation dataset is automatically analyzed, and the airspace safety threshold trigger anomalies, invalid value anomalies, adjacent airspace difference anomalies, and anomalies of failure to properly stop the paddle after threshold triggering are screened and marked. The occurrence frequency and distribution characteristics of various anomalies are statistically analyzed to generate an anomaly analysis result set. The graph plotting module constructs a time series graph of net clearance value and generator speed, a scatter plot of generator speed and net clearance value, a bar chart of adjacent speed differences, and a line graph of the linkage change of net clearance value, blade pitch angle and wind speed under threshold triggering conditions based on the anomaly analysis result set, so as to characterize the coupling change relationship between unit operating parameters before and after the anomaly occurs and present the anomaly evolution trend. The report generation and early warning output module comprehensively organizes the anomaly analysis results set and corresponding graphs to generate a standardized analysis report that includes a data overview, anomaly details, risk assessment conclusions and handling suggestions. Based on the anomaly type and total duration, the module determines the severity of the anomaly and outputs corresponding airspace safety early warning information.

Images