CN116756644A - Early warning method for icing fault of anemoclinograph of wind turbine generator - Google Patents
Early warning method for icing fault of anemoclinograph of wind turbine generator Download PDFInfo
- Publication number
- CN116756644A CN116756644A CN202311027649.8A CN202311027649A CN116756644A CN 116756644 A CN116756644 A CN 116756644A CN 202311027649 A CN202311027649 A CN 202311027649A CN 116756644 A CN116756644 A CN 116756644A
- Authority
- CN
- China
- Prior art keywords
- data
- health
- index
- icing
- anemometer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000036541 health Effects 0.000 claims abstract description 55
- 230000008014 freezing Effects 0.000 claims abstract description 26
- 238000007710 freezing Methods 0.000 claims abstract description 26
- 239000011248 coating agent Substances 0.000 claims abstract description 12
- 238000000576 coating method Methods 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims description 28
- 238000004364 calculation method Methods 0.000 claims description 21
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 6
- 238000004519 manufacturing process Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 2
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Emergency Management (AREA)
- Wind Motors (AREA)
Abstract
An ice coating fault early warning method for an anemoclinograph of a wind turbine generator comprises the following steps: SCADA data of the wind generating set in a healthy operation period is extracted; calculating the freezing index of each time point; acquiring a health boundary of health data; obtaining a health state index; calculating a threshold value of the state index; and then the SCADA data to be tested is compared with the threshold value after being processed, and a fault alarm is sent out. The method can accurately alarm the icing fault.
Description
Technical Field
The invention relates to the field of wind power equipment state monitoring and fault diagnosis, in particular to an early warning method for icing faults of an anemometer of a wind turbine.
Background
The anemoscope is an indispensable device in each wind turbine, and comprises an anemoscope and a wind vane, wherein the anemoscope firstly converts wind speed into the rotating speed of a vertical rotating shaft through a wind cup, and then measures the rotating speed of the vertical rotating shaft through a digital method; the wind direction instrument is aligned with incoming wind by the tail wing plate and drives the vertical rotating shaft to rotate, and then the angle sensor is used for measuring the wind direction. The wind turbine controls the pitch angle and yaw angle of the wind turbine through the measured wind speed and wind direction, so that the wind turbine achieves higher wind energy utilization efficiency. The accuracy of the anemometer data is therefore particularly critical to fan operation.
However, the anemoscope is usually above the cabin and exposed outdoors, and in rainy and snowy weather, an icing phenomenon may exist, so that a vertical rotation shaft of the anemoscope and the wind vane is frozen together with the base, and cannot normally rotate, and further, the measured data is abnormal, so that certain economic loss is generated. And the data in the icing state is similar to the normal shutdown data, and whether the unit anemoclinograph is in the icing state can not be judged through a common classification algorithm.
The prior art CN116341892A discloses a fan icing and off-grid risk early warning method and system based on relative humidity, and the method comprises the following steps: acquiring weather forecast information of each prediction time point; substituting the predicted humidity information into a first calculation model fitted according to historical data to obtain the visibility of any predicted moment point with the predicted temperature less than zero, and substituting the obtained visibility into a second calculation model fitted according to the historical data to obtain the liquid water content of the predicted moment point; substituting the calculated liquid water content into a third calculation model fitted according to historical data to obtain the maximum number of icing and net-removing of the wind power plant fan blade corresponding to the predicted moment point; and meanwhile, substituting the calculated liquid water content into a fourth calculation model fitted according to historical data to obtain the expected number of the wind power plant fan blade icing and net-removing fans corresponding to the predicted moment point.
The prior art CN116362382A discloses a short-term power prediction method and a system based on a wind farm icing state, which belong to the technical field of wind farm power prediction, and the method comprises the following steps: e text data of each fan of the wind power plant, wind tower data and historical ice coating thickness of the fans are obtained; comparing the actual output power of the fan with the initial predicted output power obtained by prediction calculation according to the wind tower data, and establishing a power-assisted prediction model; acquiring numerical meteorological data and geographic data of each fan, and correcting the numerical meteorological data; calculating an inertia coefficient, a retention coefficient and a freezing coefficient of the fan, and establishing an icing thickness calculation model; introducing fan icing conditions, calculating the association degree of the corrected meteorological data and the icing thickness by using a gray association degree method, correcting an icing thickness calculation model, and further calculating the icing thickness; inputting the icing thickness of the fan and the initial predicted output power into a power-assisted prediction model, and determining the final predicted output power; and reporting the final predicted output power.
Although the prior art proposes to pre-warn ice coating, the pre-warn accuracy is low and the calculation is complex.
Disclosure of Invention
Based on the problems, the invention provides an icing fault early warning method for the anemoclinograph of the wind turbine. According to the invention, a freezing index is designed according to the icing characteristic of the anemograph, frozen data and healthy data are effectively distinguished in a state space, and fault early warning is carried out through the direction and Euclidean distance of the boundary between a state point and the healthy data set.
The invention comprises the following steps: step 1: SCADA data of the wind generating set in a healthy operation period is extracted, abnormal data points are screened out, and average downsampling is carried out;
step 2: calculating the freezing index of each time point by using the SCADA data processed in the step 1;
step 3: performing maximum-minimum normalization on the processed wind speed and active power data, then taking the processed wind speed and active power data and the frozen index as input, training a single classifier to obtain health data, and obtaining a health boundary of the health data;
step 4: calculating the shortest Euclidean distance between the health data and the health boundary, and using signs to represent the inside and the outside of the boundary and using the signs as health state indexes;
step 5: calculating a threshold value of the state index by using the health state index;
step 6: during testing, the SCADA data to be tested is subjected to data preprocessing according to the step 1 and the step 2, abnormal data points are deleted, and freezing indexes are calculated;
step 7: during testing, carrying out maximum-minimum normalization on wind speed and active power data in the preprocessing data in the step 6, substituting the data into a model together with freezing index data, calculating Euclidean distance between each state point and a health boundary, and adding signs to represent the inside and outside of the boundary to obtain the health state index of the testing data;
step 8: during testing, the health state indexes of the test data obtained in the step 7 are smoothed by a moving average method;
step 9: and (3) during testing, according to the threshold value calculated in the step (5), when 3 values of the smoothed health state index in the step (8) are out of the threshold value range in a time sequence, the anemoclinograph is considered to have icing faults, and an alarm is sent out.
Further, in the step 1, abnormal data with all 0 measuring points are screened out during data preprocessing, and the data are processed according to the following stepstThe duration is downsampled on average.
Further, in step 2, the calculation formula of the freezing index is as followsWhereintempIs the absolute value of the temperature outside the cabin at the current moment;angleis the absolute value of the difference between the current moment and the previous moment of the wind direction;windis the current wind speed.
Further, the freezing index in step 2 may be used to distinguish the anemometer icing status data from the health period data.
Further, in step 2, the ice indicator calculation formula is determined by the icing conditions, the icing characteristics of the anemometer and the icing characteristics of the anemometer.
Further, in step 2, the larger the freezing index is when the status point is more in line with the icing feature, the smaller the freezing index is when the status point is less in line with the icing feature.
Further, in step 3, the boundary of the health dataset is obtained through calculation by a vector machine algorithm.
Further, in step 4, the euclidean distance between the state point and the boundary of the health data set is taken, and the sign is added to indicate that the state point is inside or outside the boundary, so that the state point is used as a health state index for quantifying the icing state of the anemometer.
Further, in step 5, the health status indicator threshold calculation formula is:whereinμ、σRespectively represent the mean value and standard deviation of the health state indexes after the smoothing treatment in the health running period,kto control the constant of the threshold range size, λ is the weighting coefficient in an exponentially weighted moving average,nis the number of training data.
Further, the test data are subjected to the processes of preprocessing, calculating the freezing index, calculating the health state index, moving average smoothing and the like in sequence, the smoothed health state index is compared with the threshold value calculated in the step 5, and when 3 points which are continuous in time sequence exceed the threshold value range, the anemograph is considered to have icing faults, and an alarm is sent out.
The beneficial effects of the invention are as follows: compared with the prior art, the invention utilizes SCADA operation data, constructs the freezing index according to the characteristics of icing data, distinguishes the icing data of the anemoclinograph from the health data in a state space, and provides an early warning model aiming at the icing fault of the anemoclinograph, thereby effectively diagnosing the icing of the anemoclinograph and avoiding certain economic loss. According to the method, the health state index threshold value is calculated according to the training data, and then the test data is tested according to the threshold value, so that whether the icing fault occurs to the anemometer is judged according to whether the test data exceeds the threshold value.
Drawings
FIG. 1 shows a flow chart of a wind turbine anemometer icing fault early warning method.
Fig. 2 shows a graph of the annual ice-covering test result of the anemometer of the wind field No. 6 unit.
Fig. 3 shows a test result diagram of the anemorumbometer of the wind field No. 6 unit, which is iced for 10 months 1 day to 10 months 7 days.
Fig. 4 shows a graph of the annual ice-covering test result of the anemometer of the No. 14 unit in a certain wind farm.
Fig. 5 shows a test result diagram of the anemorumbometer of the No. 14 wind farm unit on ice for 10 months 1 day to 10 months 7 days.
Fig. 6 shows a graph of the annual ice-covering test result of the anemometer of the No. 15 wind farm.
Fig. 7 shows a test result diagram of the anemorumbometer of the No. 15 wind farm for icing for 10 months 1 day to 10 months 7 days.
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be understood that the process equipment or devices not specifically identified in the examples below are all conventional in the art. Furthermore, it is to be understood that the reference to one or more method steps in this disclosure does not exclude the presence of other method steps before or after the combination step or the insertion of other method steps between these explicitly mentioned steps, unless otherwise indicated; it should also be understood that the combined connection between one or more devices/means mentioned in the present invention does not exclude that other devices/means may also be present before and after the combined device/means or that other devices/means may also be interposed between these two explicitly mentioned devices/means, unless otherwise indicated. Moreover, unless otherwise indicated, the numbering of the method steps is merely a convenient tool for identifying the method steps and is not intended to limit the order of arrangement of the method steps or to limit the scope of the invention in which the invention may be practiced, as such changes or modifications in their relative relationships may be regarded as within the scope of the invention without substantial modification to the technical matter.
The following detailed description of the preferred embodiments of the invention is provided in connection with the accompanying drawings so that the advantages and features of the invention will be more readily understood by those skilled in the art, and thus the scope of the invention is to be more clearly defined.
As shown in FIG. 1, the method is a flow chart of an icing fault early warning method of the anemometer of the wind turbine. The embodiment data is SCADA data from a wind farm wind turbine generator of a certain actual 5.2MW model, the data is one piece per minute, and the time is from 1 month in 2022 to 3 months in 2023. The maintenance work list displays that three units of anemometers No. 6, no. 14 and No. 15 on 10 months of 2022 and 3 days of 3 days have icing faults.
S1: SCADA data of the wind generating set in a healthy operation period is extracted, wherein the SCADA data comprise wind speed, wind direction, active power, outside cabin temperature and the like, and then pretreatment is carried out: screening out abnormal data with each measuring point of 0, and carrying out average downsampling on the data according to 10 minutes, so as to avoid the influence of the abnormal data and the data which does not change in a short time on the calculation of the freezing index;
s2: after S1 pretreatment, the freezing index of each time point is calculated according to the wind speed, the wind direction and the outside cabin temperature data, and the calculation formula is as followsWhereintempIs the absolute value of the average value of the outside cabin temperature within 10 minutes at present;angleis the absolute value of the difference between the current 10 minute wind direction average and the previous 10 minute wind direction average;windis the current average value of the wind speed of 10 minutes;
s3: constructing a single-classification support vector machine mathematical model based on a Gaussian kernel function, taking 0.001 for tolerance of abnormal values of decision boundaries, carrying out maximum-minimum normalization on wind speeds and active powers processed by S1 and S2, and then taking the maximum-minimum normalization and the freezing indexes as input training models to obtain boundaries of health data distribution;
s4: calculating the shortest Euclidean distance between training data and a health boundary, adding signs to represent the inside and outside of the boundary, and taking the shortest Euclidean distance as a health state index;
s5: calculating a threshold value from the health state index of the training data, wherein the calculation formula of the health state index threshold value is as follows:whereinμ、σRespectively represent the mean value and standard deviation of the health state indexes after the smoothing treatment in the health running period,kto control the constant of the magnitude of the threshold range,ktaking 8, λ as the weighting coefficient in the exponentially weighted moving average, λ taking 0.2,nis the number of training data. Calculating to obtain a threshold value of the No. 6 machine set of 0.578, a threshold value of the No. 14 machine set of 0.272 and a threshold value of the No. 15 machine set of 0.132;
s6: during testing, the SCADA data to be tested is subjected to data preprocessing according to S1 and S2, abnormal data points are deleted, and freezing indexes are calculated;
s7: in the test, carrying out maximum-minimum normalization on the wind speed and active power data processed in the step S6, substituting the data into a model together with the freezing index data, calculating Euclidean distance between each state point and the health boundary obtained in the step S3, and adding signs to represent the inside and outside of the boundary to obtain the health state index of the test data;
s8: in the test, the health state index of the test data obtained in the step S7 is downsampled according to the average of half an hour, then moving average smoothing is carried out, and the length of a moving window is 6;
s9: when the smoothed health state index exceeds the threshold value range in a time sequence continuous mode, the anemograph is considered to have icing faults, and the early warning result is obtained according to the threshold value obtained through calculation in the step S5 and the smoothed health state index obtained in the step S8. 2-7 are graphs of test results of No. 6, no. 14 and No. 15, the No. 6 machine model 2022, 10 month, 3 days, 8:00:00 gives an alarm, the production management system 2022, 10 month, 3 days, 16:47:00 gives an alarm, and the alarm is given by 8 hours 47 in advance; the model No. 14 machine set 2022, 10 month, 3 day, 10:30:00 gives an alarm, the production management system 2022, 10 month, 3 day, 17:31:00 gives an alarm, and the alarm is given by 7 hours 1 in advance manually; the model of crew No. 15 2022, 10 month 3, 10:30:00, alerts, and the production management system 2022, 10 month 3, 13:15:00, issues alerts manually 2 hours in advance 45.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The ice coating fault early warning method for the anemoclinograph of the wind turbine generator is characterized by comprising the following steps of:
step 1: SCADA data of the wind generating set in a healthy operation period is extracted, abnormal data points are screened out, and average downsampling is carried out;
step 2: calculating the freezing index of each time point by using the SCADA data processed in the step 1;
step 3: performing maximum-minimum normalization on the processed wind speed and active power data, then taking the processed wind speed and active power data and the frozen index as input, training a single classifier to obtain health data, and obtaining a health boundary of the health data;
step 4: calculating the shortest Euclidean distance between the health data and the health boundary, and using signs to represent the inside and the outside of the boundary and using the signs as health state indexes;
step 5: calculating a threshold value of the state index by using the health state index;
step 6: during testing, the SCADA data to be tested is subjected to data preprocessing according to the step 1 and the step 2, abnormal data points are deleted, and freezing indexes are calculated;
step 7: during testing, carrying out maximum-minimum normalization on wind speed and active power data in the preprocessing data in the step 6, substituting the data into a model together with freezing index data, calculating Euclidean distance between each state point and a health boundary, and adding signs to represent the inside and outside of the boundary to obtain the health state index of the testing data;
step 8: during testing, the health state indexes of the test data obtained in the step 7 are smoothed by a moving average method;
step 9: and (3) during testing, according to the threshold value calculated in the step (5), when 3 values of the smoothed health state index in the step (8) are out of the threshold value range in a time sequence, the anemoclinograph is considered to have icing faults, and an alarm is sent out.
2. The method for pre-warning ice coating faults of anemoclinographs of wind turbine sets according to claim 1, wherein abnormal data of all 0 measuring points are screened out during data preprocessing in step 1, and the data are processed according to the datatThe duration is downsampled on average.
3. The method for pre-warning ice coating faults of anemoclinographs of wind turbine sets according to claim 1, wherein in step 2, a freezing index calculation formula isWhereintempIs the absolute value of the temperature outside the cabin at the current moment;angleis the absolute value of the difference between the current moment and the previous moment of the wind direction;windis the current wind speed.
4. The method for early warning ice coating faults of the anemometer of the wind turbine generator set according to claim 1, wherein the freezing index in the step 2 can be used for distinguishing ice coating state data of the anemometer from health period data.
5. The method for early warning of icing faults of the anemometer of the wind turbine generator set according to claim 1 is characterized in that in the step 2, a freezing index calculation formula is determined by icing production conditions, icing characteristics of the anemometer and icing characteristics of the anemometer.
6. The method for pre-warning the icing fault of the anemometer of the wind turbine generator set according to claim 1 is characterized in that in the step 2, when the state point is more in accordance with the icing characteristic, the freezing index is larger, and when the state point is less in accordance with the icing characteristic, the freezing index is smaller.
7. The method for early warning ice coating faults of the anemometer of the wind turbine generator set according to claim 1 is characterized in that in step 3, a healthy dataset boundary is calculated through a vector machine algorithm.
8. The method for early warning ice coating faults of the anemometer of the wind turbine generator set according to claim 1 is characterized in that in step 4, euclidean distance between a state point and a boundary of a health data set is taken, and signs are added to indicate that the state point is located inside or outside the boundary, so that the state point is taken as a health state index for quantifying the ice coating state of the anemometer.
9. The method for pre-warning ice coating faults of the anemometer of the wind turbine generator set according to claim 1, wherein in step 5, a health state index threshold value calculation formula is as follows:whereinμ、σRespectively represent the mean value and standard deviation of the health state indexes after the smoothing treatment in the health running period,kto control the constant of the threshold range size, λ is the weighting coefficient in an exponentially weighted moving average,nis the number of training data.
10. The method for early warning the icing fault of the anemometer of the wind turbine generator set according to claim 1, wherein the test data are subjected to the pretreatment, the calculation of the freezing index, the calculation of the health state index and the moving average smoothing process in sequence, the smoothed health state index is compared with the threshold value calculated in the step 5, and when 3 points which are continuous in time sequence exceed the threshold value range, the anemometer is considered to have the icing fault, and an alarm is sent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311027649.8A CN116756644B (en) | 2023-08-16 | 2023-08-16 | Early warning method for icing fault of anemoclinograph of wind turbine generator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311027649.8A CN116756644B (en) | 2023-08-16 | 2023-08-16 | Early warning method for icing fault of anemoclinograph of wind turbine generator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116756644A true CN116756644A (en) | 2023-09-15 |
| CN116756644B CN116756644B (en) | 2023-11-21 |
Family
ID=87948136
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311027649.8A Active CN116756644B (en) | 2023-08-16 | 2023-08-16 | Early warning method for icing fault of anemoclinograph of wind turbine generator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116756644B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809287A (en) * | 2016-03-10 | 2016-07-27 | 云南大学 | High-voltage transmission line icing process integrated prediction method |
| CN108343566A (en) * | 2018-03-28 | 2018-07-31 | 长沙理工大学 | Blade icing fault online monitoring method and system based on running state of wind turbine generator |
| CN109595130A (en) * | 2018-12-25 | 2019-04-09 | 济中能源技术服务(上海)有限公司 | A kind of fan blade icing failure prediction method and system |
| CN112784373A (en) * | 2021-01-19 | 2021-05-11 | 河北大学 | Fault early warning method for wind turbine generator gearbox |
| CN112832960A (en) * | 2020-12-23 | 2021-05-25 | 国电南瑞南京控制***有限公司 | Fan blade icing detection method based on deep learning and storage medium |
| WO2022015729A1 (en) * | 2020-07-15 | 2022-01-20 | Regents Of The University Of Minnesota | Control system for wind turbines in cold climates |
| CN115095487A (en) * | 2022-06-30 | 2022-09-23 | 宁波大学 | Wind turbine state monitoring method based on multi-source heterogeneous SCADA data |
| CN116541753A (en) * | 2023-07-07 | 2023-08-04 | 三峡智控科技有限公司 | Wind turbine generator blade icing early warning method |
-
2023
- 2023-08-16 CN CN202311027649.8A patent/CN116756644B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809287A (en) * | 2016-03-10 | 2016-07-27 | 云南大学 | High-voltage transmission line icing process integrated prediction method |
| CN108343566A (en) * | 2018-03-28 | 2018-07-31 | 长沙理工大学 | Blade icing fault online monitoring method and system based on running state of wind turbine generator |
| CN109595130A (en) * | 2018-12-25 | 2019-04-09 | 济中能源技术服务(上海)有限公司 | A kind of fan blade icing failure prediction method and system |
| WO2022015729A1 (en) * | 2020-07-15 | 2022-01-20 | Regents Of The University Of Minnesota | Control system for wind turbines in cold climates |
| CN112832960A (en) * | 2020-12-23 | 2021-05-25 | 国电南瑞南京控制***有限公司 | Fan blade icing detection method based on deep learning and storage medium |
| CN112784373A (en) * | 2021-01-19 | 2021-05-11 | 河北大学 | Fault early warning method for wind turbine generator gearbox |
| CN115095487A (en) * | 2022-06-30 | 2022-09-23 | 宁波大学 | Wind turbine state monitoring method based on multi-source heterogeneous SCADA data |
| CN116541753A (en) * | 2023-07-07 | 2023-08-04 | 三峡智控科技有限公司 | Wind turbine generator blade icing early warning method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116756644B (en) | 2023-11-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109958588B (en) | Icing prediction method, icing prediction device, storage medium, model generation method and model generation device | |
| AU2019413696B2 (en) | System and method for evaluating models for predictive failure of renewable energy assets | |
| CN106321368B (en) | Marine wind electric field wake losses measurement method based on operation blower SCADA data | |
| US11460003B2 (en) | Wind turbine damage detection system using machine learning | |
| CN110006552B (en) | Method for detecting abnormal temperature of equipment | |
| CN111680454A (en) | Fan blade icing fault prediction method based on double attention mechanism | |
| US20230265832A1 (en) | Load control method and apparatus for wind turbine generator system | |
| CN107728059A (en) | A kind of pitch-controlled system state evaluating method | |
| WO2018059259A1 (en) | Method and system of yaw control of wind turbines in a wind turbine farm | |
| Papatzimos et al. | Data insights from an offshore wind turbine gearbox replacement | |
| CN111340307B (en) | Method for predicting wind power generation power of fan and related device | |
| CN114593020A (en) | Blade icing monitoring and early warning method and device for wind turbine generator and electronic equipment | |
| CN116756644B (en) | Early warning method for icing fault of anemoclinograph of wind turbine generator | |
| CN116878886B (en) | Method for monitoring abrasion of water pump bearing of cooling tower | |
| CN116557230A (en) | Wind power plant unit power abnormality online assessment method and system | |
| CN115374653B (en) | NSET model-based wind driven generator temperature early warning method and related device | |
| CN108825452B (en) | Method and device for determining blade icing of wind generating set | |
| Canovas Lotthagen | Defining, analyzing and determining power losses-due to icing on wind turbine blades | |
| CN115456041A (en) | Equipment fault early warning method and device, computing equipment and storage medium | |
| Souza et al. | Evaluation of data based normal behavior models for fault detection in wind turbines | |
| DK201670502A1 (en) | Wind turbine and a method of operating a wind turbine | |
| CN116754794A (en) | Wind turbine generator anemometer fault identification method and system based on least square method | |
| Roy | Performance assesment of SCADA based wind turbine: condition monitoring approaches | |
| CN117631090B (en) | Cold tide identification method and device | |
| CN113822344B (en) | Method for monitoring state of front bearing of generator of wind turbine generator based on data driving |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |