A wind turbine load jump correction method and system

Abstract

The application discloses a kind of wind generating set load jump correction method and system, comprising, S1, the time sequence load data of fan component is obtained;S2, whether time sequence load data is in preset range is judged, whether the unit is in normal power generation working condition is judged, if all satisfy then enter step S3;S3, average and standard deviation are calculated according to time sequence load data, and difference detection is carried out to judge whether it is single-point jump or multiple-point jump;S4, if it is single-point jump, then the jump value of jump point is corrected as the average of adjacent two points before and after the jump point, if it is multiple-point jump, then the jump value of each jump point is smoothed as the sum of previous point and standard deviation;S5, repeat execution step S3-step S5, until the requirement of smoothing effect is satisfied.The application can effectively solve the problem that sensor signal is disturbed by fan internal environment and jump, leading to inaccurate load value, affecting the safety judgment of unit.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind turbine generators, and particularly refers to a wind turbine generator load jump correction method and system. BACKGROUND

[0002] A wind turbine generator is a large and complex mechanical component. In order to monitor and evaluate the structural health of the wind turbine generator under wind load and operating load, to ensure the safety, reliability and performance of the system, mechanical load testing needs to be performed on the wind turbine to identify potential structural problems, fatigue cracks or strain concentration areas, so as to prevent equipment failure or damage. In addition, based on the analysis of the measured data, the system reliability can be improved, the downtime can be reduced, and the maintenance cost can be reduced.

[0003] The mechanical load testing process generally lasts for 3-6 months, during which the external conditions such as the environment of the wind turbine may change greatly, resulting in that during the testing process, the sensor signal often jumps with temperature or some interference conditions, resulting in that the corresponding load calculated is inaccurate and the testing data is invalid, the testing period is delayed, and the testing cost is increased. Therefore, the load signal needs to be reasonably corrected to ensure that the data is real and available. SUMMARY

[0004] The present application aims to overcome the shortcomings of the prior art, and provides a wind turbine generator load jump correction method, which can effectively solve the problem that the load signal of the wind turbine jumps due to the interference of the internal environment of the wind turbine, resulting in inaccurate load value and affecting the judgment of the safety of the unit.

[0005] Another object of the present application is to provide a wind turbine generator load jump correction system.

[0006] The object of the present application is achieved by the following technical scheme:

[0007] A wind turbine generator load jump correction method, characterized in that it comprises,

[0008] S1, acquiring time sequence load data of a wind turbine component;

[0009] S2, judging whether the time sequence load data is within a preset range, judging whether the unit is in a normal power generation working condition, and if both are satisfied, entering step S3;

[0010] S3, calculating the average value and the standard deviation according to the time sequence load data, and performing difference detection to judge whether it is a single-point jump or a multi-point jump;

[0011] S4, if it is a single-point jump, the jump value of the jump point is corrected to the average value of the two adjacent points before and after the jump point, and if it is a multi-point jump, the jump value of each jump point is smoothed to the sum of the previous point and the standard deviation;

[0012] S5, repeating steps S3-S5 until the smoothness requirement is met.

[0013] Further, in step S1, the following operations are specifically performed,

[0014] The fan component is a blade, a blade root load strain gauge is installed at the root of the blade, and the signal of the blade root load strain gauge is connected to the fan data acquisition system for data acquisition;

[0015] When the wind speed is less than 4 m / s, the fan nacelle is deviated clockwise and counterclockwise for 1 circle each for calibration, and when the wind speed is below 5 m / s, the blades are respectively calibrated for idling at a pitch angle of 0° and 90°, and the calculation relationship between the electrical signal and the load is obtained after calibration.

[0016] Further, in step S1, the following operations are specifically performed,

[0017] The fan component is a tower, a tower top load strain gauge and a tower bottom load strain gauge are installed at the top and bottom of the tower respectively, and the signals of the tower top load strain gauge and the tower bottom load strain gauge are connected to the fan data acquisition system for data acquisition;

[0018] When the wind speed is less than 4 m / s, the fan nacelle is deviated clockwise and counterclockwise for 1 circle each for calibration, and when the wind speed is below 5 m / s, the blades of the fan are respectively calibrated for idling at a pitch angle of 0° and 90°, and the calculation relationship between the electrical signal and the load is obtained after calibration.

[0019] Further, in step S2, the following operations are specifically performed,

[0020] The timing load data is determined by the fan industrial computer system whether it is within the preset range, and the main control data of the fan is determined by the industrial computer system whether the unit is in normal power generation condition, if both are satisfied, then step S3 is entered.

[0021] Further, in step S3, the following operations are specifically performed,

[0022] The average value X and the standard deviation S are calculated according to the timing load data, and then difference detection is performed, if multiple jump points Y i >X+3*S and the index of the jump point is not continuous, where i=1, 2, …, n, then it is identified as a single point jump, if there are two or more consecutive points and the number of jump points Y i The proportion does not exceed the preset proportion of the entire timing number of points, then it is identified as a multi-point jump.

[0023] Further, in step S4, the following operations are specifically performed,

[0024] If it is a single-point jump, determine the index position of the jump point in the timing, and then use the smoothing formula Y i =(Y i-1 +Y i+1 ) / 2, for the jump point Y i Correction is performed, where i=1, 2, ..., n, Y i-1 and Y i+1 Represents the jump point Y i Two adjacent points;

[0025] If it is a multi-point jump, each jump point Y i The jump value is smoothed to the previous adjacent point Y i-1 and the sum of the standard deviation S.

[0026] Another object of the present invention is achieved through the following technical solutions:

[0027] A load jump correction system for a wind turbine generator set, comprising:

[0028] An acquisition module is used to acquire time series load signals of fan components;

[0029] The fan data acquisition system is used to process the time series load signal into time series load data and transmit it to the industrial computer system;

[0030] The industrial computer system is used to determine whether the time series load data is within the preset range and whether the unit is in normal power generation conditions. If all conditions are met, the average value and standard deviation are calculated based on the time series load data, and differential detection is performed to determine whether it is a single-point jump or a multi-point jump, and then corrections are made to the jump points.

[0031] Furthermore, the wind turbine component is a blade, and the acquisition module includes a blade root load strain gauge arranged at the root of the blade.

[0032] Furthermore, the wind turbine component is a tower, and the acquisition module includes a top load strain gauge arranged at the top of the tower and a tower bottom load strain gauge arranged at the bottom of the tower.

[0033] Furthermore, the industrial computer system specifically performs the following operations:

[0034] Determine whether the time series load data is within the preset range, and determine whether the unit is in normal power generation conditions based on the main control data of the wind turbine. If both are met, calculate the average value X and standard deviation S based on the time series load data, and then perform differential detection:

[0035] If there are multiple jump points Y i >X+3*S and the index of the jump point is discontinuous, where i=1, 2, ..., n, it is identified as a single-point jump, and then the index position of the jump point in the timing is determined, and then according to the smoothing formula Yi = (Y i-1 + Y i+1 ) / 2, the jump point Y i is corrected, wherein i=1, 2, …, n, Y i-1 and Y i+1 respectively represent two points adjacent to the jump point Y i before and after;

[0036] If two or more consecutive points jump and the number of jump points Y i accounts for a preset proportion of the entire time sequence number of points, then the multiple-point jump is identified, and the jump value of each jump point Y i is smoothed to the sum of the previous adjacent point Y i-1 and the standard deviation S.

[0037] The difference detection and correction according to the time sequence load data are repeated until the smoothing effect requirement is met.

[0038] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0039] The present application can classify and process according to the jump type of the fan load signal that often occurs, can effectively remove the abnormal components in the signal, thereby improving the clarity and quality of the signal, and can also provide better signal smoothing effect, while maintaining the original characteristics of the signal, the processed signal is more stable and reliable, which helps to improve the accuracy of data processing and analysis, saves the time of on-site data collection and a large amount of manual maintenance cost. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 The flow chart of the method of the present application.

[0041] Figure 2 The side view of the fan.

[0042] Figure 3 The front view of the blade.

[0043] Figure 4 The schematic diagram before the blade load jump filtering.

[0044] Figure 5 The schematic diagram after the blade load jump filtering. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0046] Example 1:

[0047] like Figure 1 As shown, this embodiment provides a method for correcting load jump of a wind turbine generator set, including the steps of:

[0048] S1. Obtain the time series load data of the fan components; perform the following operations:

[0049] The wind turbine component is a blade 1 or a tower 3. When the load jump signal of the blade 1 needs to be corrected, a blade root load strain gauge 6 is installed at the root of the blade. The signal of the blade root load strain gauge is connected to the wind turbine data acquisition system 7 for data acquisition to obtain the time series load data of the blade.

[0050] When the load jump signal of the tower 3 needs to be corrected, a tower top load strain gauge 4 and a tower bottom load strain gauge 5 are installed at the top and bottom of the tower respectively, and the signals of the tower top load strain gauge and the tower bottom load strain gauge are connected to the wind turbine data acquisition system 7 for data acquisition to obtain the time series load data of the tower;

[0051] When the wind speed is less than 4m / s, the wind turbine nacelle 2 is yawed clockwise and counterclockwise by 1 circle each for calibration, such as Figure 3 As shown in the figure, when the wind speed is below 5m / s, the blades are calibrated at idling with pitch angles of 0° and 90° respectively. After the calibration is completed, the calculation relationship between the electrical signal and the load is obtained;

[0052] The calculation relationship is written into the industrial computer system 8 for load synthesis, and the output results are monitored in real time and exception processing is performed.

[0053] This embodiment takes the correction of the load jump signal of blade 1 as an example. Figure 2 shown.

[0054] S2, determine whether the time series load data is within the preset range, and determine whether the unit is in normal power generation condition. If all are satisfied, proceed to step S3; perform the following operations specifically:

[0055] The industrial computer system judges the time series load data transmitted by the wind turbine data acquisition system to determine whether the time series load data is within the preset range, and the industrial computer system determines whether the unit is in normal power generation conditions based on the main control data of the wind turbine. If all are met, it enters step S3, otherwise it only monitors without processing.

[0056] S3. Under the action of wind load, the blade vibration load value collected in the time sequence when the fan is running can be seen in the industrial computer system to form a load signal similar to a sine wave, such as Figure 4 Calculate the average value and standard deviation based on the time series load data, and perform differential detection to determine whether it is a single-point jump or a multi-point jump; perform the following operations specifically:

[0057] The industrial computer system calculates the average value X and standard deviation S based on the time series load data, and then performs differential detection. If multiple jump points Y appear i >X+3*S and the index of the jump point is discontinuous, where i=1, 2, ..., n, it is identified as a single-point jump. If there are two or more consecutive jumps and the jump point Y i If the number does not exceed the preset ratio of the total number of time series points (10% in this embodiment), it is identified as a multi-point hopping.

[0058] S4. If it is a single-point jump, the jump value of the jump point is corrected to the average value of the two adjacent points before and after the jump point. If it is a multi-point jump, the jump value of each jump point is smoothed to the sum of the previous point and the standard deviation. Specifically, the following operations are performed:

[0059] If it is a single-point jump, determine the index position of the jump point in the timing, and then use the smoothing formula Y i =(Y i-1 +Y i+1 ) / 2, for the jump point Y i Make corrections, Y i-1 and Y i+1 Represents the jump point Y i The curve obtained after filtering of two adjacent points is as follows Figure 5 As shown, the signal is repaired while maintaining its original characteristics;

[0060] If it is a multi-point jump, each jump point Y i The jump value is smoothed to the previous adjacent point Y i-1 and the sum of the standard deviation S.

[0061] S5. Repeat steps S3 to S5 until the smoothing effect requirement is met.

[0062] Example 2:

[0063] The embodiment provides a wind turbine load jump correction system, comprising,

[0064] The acquisition module is used for acquiring a time sequence load signal of a fan component; wherein the fan component is a blade or a tower drum; when the fan component is the blade, the system corrects a blade load jump signal; when the fan component is the tower drum, the system corrects a tower drum load jump signal; when the blade load jump signal is corrected, a blade root load strain gauge is installed at the root of the blade; when the tower drum load jump signal is corrected, a top load strain gauge is arranged at the top of the tower drum and a tower bottom load strain gauge is arranged at the bottom of the tower drum.

[0065] The wind turbine data acquisition system is used for processing the time sequence load signal into time sequence load data and transmitting the time sequence load data to the industrial computer system.

[0066] The industrial computer system is used for judging whether the time sequence load data is within a preset range, judging whether the unit is in a normal power generation working condition, calculating an average value and a standard deviation according to the time sequence load data if both are satisfied, and performing difference detection to judge whether it is a single point jump or a multiple point jump, and then correcting the jump point; the following operations are specifically performed,

[0067] The time sequence load data is judged whether it is within a preset range, and whether the unit is in a normal power generation working condition is judged according to the main control data of the wind turbine; if both are satisfied, the average value X and the standard deviation S are calculated according to the time sequence load data, and then difference detection is performed:

[0068] If multiple jump points Y i > X+3*S and the indexes of the jump points are not continuous, wherein i=1, 2, …, n, it is identified as a single point jump, the index position of the jump point in the time sequence is determined, and then the jump point Y i is corrected according to a smoothing processing formula Y i-1 =( Y i+1 +Y i ) / 2, wherein i=1, 2, …, n, Y i-1 and Y i+1 represent two adjacent points before and after the jump point Y i .

[0069] If there are two or more continuous jump points and the proportion of the number of the jump points Y i does not exceed a preset proportion of the number of points in the entire time sequence, it is identified as a multiple point jump, and then the jump value of each jump point Y i is smoothed to the sum of a previous adjacent point Y i-1 and the standard deviation S.

[0070] Difference detection and correction are repeatedly performed according to the time sequence load data until the smoothing effect requirement is met.

[0071] The above merely describes preferred embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can make equivalent substitutions or changes within the scope disclosed by the present application, according to the technical solution and inventive concept of the present application, and such substitutions or changes shall also fall within the protection scope of the present application.

Claims

1. A method for correcting load jump of a wind turbine generator set, characterized in that: include, S1. Obtaining time series load data of wind turbine components; Specifically, perform the following operations: When the wind turbine component is a blade, a blade root load strain gauge is installed at the root of the blade, and the signal from the blade root load strain gauge is connected to the wind turbine data acquisition system for data collection. When the wind speed is less than 4m / s, the wind turbine nacelle is yawed clockwise and counterclockwise by one circle each for calibration. When the wind speed is below 5m / s, the blade is calibrated at idling angles of 0° and 90° respectively. After the calibration is completed, the calculation relationship between the electrical signal and the load is obtained. When the wind turbine component is a tower, a top load strain gauge and a bottom load strain gauge are installed at the top and bottom of the tower, respectively. The signals of the top load strain gauge and the bottom load strain gauge are connected to the wind turbine data acquisition system for data collection. When the wind speed is less than 4m / s, the wind turbine nacelle is yawed clockwise and counterclockwise by one circle each for calibration. When the wind speed is below 5m / s, the wind turbine blades are calibrated for idling at pitch angles of 0° and 90° respectively. After the calibration is completed, the calculation relationship between the electrical signal and the load is obtained. S2, determine whether the time series load data is within the preset range, and determine whether the unit is in normal power generation condition. If all are satisfied, proceed to step S3; perform the following operations specifically: The industrial computer system of the wind turbine determines whether the time series load data is within a preset range, and the industrial computer system determines whether the unit is in normal power generation condition based on the main control data of the wind turbine. If both conditions are met, the process proceeds to step S3; S3. Calculate the average value and standard deviation based on the time series load data, and perform differential detection to determine whether it is a single-point jump or a multi-point jump; S4. If it is a single-point jump, the jump value of the jump point is corrected to the average value of the two adjacent points before and after the jump point. If it is a multi-point jump, the jump value of each jump point is smoothed to the sum of the previous point and the standard deviation. Specifically, the following operations are performed: If it is a single-point jump, determine the index position of the jump point in the timing, and then use the smoothing formula Y i =( Y i-1 +Y i+1 ) / 2, for the jump point Y i Correction is performed, where i=1, 2, ..., n, Y i-1 and Y i+1 Represents the jump point Y i Two adjacent points; If it is a multi-point jump, each jump point Y i The jump value is smoothed to the previous adjacent point Y i-1 and the sum of the standard deviation S; S5. Repeat steps S3 to S5 until the smoothing effect requirement is met.

2. The method for correcting load jump of a wind turbine generator set according to claim 1, characterized in that: In step S3, the following operations are specifically performed: Calculate the average value X and standard deviation S according to the time series load data, and then perform differential detection. If multiple jump points Y appear i >X+3*S and the index of the jump point is discontinuous, where i=1, 2, ..., n, it is identified as a single-point jump. If there are two or more consecutive jumps and the jump point Y i If the number does not exceed the preset ratio of the total number of time series points, it is identified as a multi-point jump.

3. A wind turbine load jump correction system, characterized in that: Used to implement the wind turbine load jump correction method according to any one of claims 1 to 2, include, An acquisition module is used to acquire time series load signals of fan components; The fan data acquisition system is used to process the time series load signal into time series load data and transmit it to the industrial computer system; The industrial computer system is used to determine whether the time series load data is within the preset range and whether the unit is in normal power generation conditions. If all conditions are met, the average value and standard deviation are calculated based on the time series load data, and differential detection is performed to determine whether it is a single-point jump or a multi-point jump, and then corrections are made to the jump points.

4. The wind turbine generator set load jump correction system according to claim 3, characterized in that: The fan component is a blade, and the acquisition module includes a blade root load strain gauge arranged at the root of the blade.

5. The wind turbine generator set load jump correction system according to claim 3, characterized in that: The wind turbine component is a tower, and the acquisition module includes a tower top load strain gauge arranged at the top of the tower and a tower bottom load strain gauge arranged at the bottom of the tower.

6. The wind turbine generator set load jump correction system according to claim 3, characterized in that: The industrial computer system specifically performs the following operations: Determine whether the time series load data is within the preset range, and determine whether the unit is in normal power generation conditions based on the main control data of the wind turbine. If both are met, calculate the average value X and standard deviation S based on the time series load data, and then perform differential detection: If there are multiple jump points Y i > X+3*S and the index of the jump point is discontinuous, where i=1, 2, ..., n, it is identified as a single-point jump, and then the index position of the jump point in the timing is determined, and then the smoothing formula Y is used. i =( Y i-1 +Y i+1 ) / 2, for the jump point Y i Correction is performed, where i=1, 2, ..., n, Y i-1 and Y i+1 Represents the jump point Y i Two adjacent points; If there are two or more consecutive jumps and the jump point Y i If the number of points does not exceed the preset ratio of the total number of points in the time series, it is identified as a multi-point jump, and each jump point Y i The jump value is smoothed to the previous adjacent point Y i-1 and the sum of the standard deviation S; Repeat differential detection and correction based on time series load data until the smoothing effect requirements are met.

Images