CN108869174B - Nonlinear modeling wind driven generator blade natural frequency working condition compensation method - Google Patents
Nonlinear modeling wind driven generator blade natural frequency working condition compensation method Download PDFInfo
- Publication number
- CN108869174B CN108869174B CN201810622308.8A CN201810622308A CN108869174B CN 108869174 B CN108869174 B CN 108869174B CN 201810622308 A CN201810622308 A CN 201810622308A CN 108869174 B CN108869174 B CN 108869174B
- Authority
- CN
- China
- Prior art keywords
- natural frequency
- modeling
- data set
- working condition
- blade
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000012360 testing method Methods 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 14
- 239000011159 matrix material Substances 0.000 claims abstract description 11
- 238000012544 monitoring process Methods 0.000 claims description 8
- 238000010248 power generation Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 12
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03D—WIND MOTORS
- F03D7/00—Controlling wind motors
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Wind Motors (AREA)
Abstract
The invention discloses a wind power blade natural frequency working condition compensation method based on nonlinear modeling, which comprises the following steps of: 1) dividing the historical operation data set into a plurality of modeling data subsets according to the power P of the SCADA of the wind driven generator, and then dividing each modeling data subset intoModeling a training data set and a model testing data set; 2) normalizing the elements in each modeling data subset; 3) constructing a corresponding observation memory matrix D according to the normalized modeling training data set, and then testing a data set X according to the constructed observation memory matrix D and a corresponding modeltestConstructing a corresponding natural frequency prediction model; 4) and substituting the working condition parameters of the current unit into the corresponding natural frequency prediction model, performing subtraction operation on the actually measured current fixed frequency and the predicted current fixed frequency, and finally performing working condition compensation on the natural frequency of the blade by using the fixed frequency deviation value. The method can accurately realize the working condition compensation of the fixed frequency of the wind driven generator blade.
Description
Technical Field
The invention belongs to the field of natural frequency working condition compensation of a wind driven generator blade, and relates to a nonlinear modeling wind power blade natural frequency working condition compensation method.
Background
At present, the wind power generation industry in China is in a high-speed development period, and the number of wind power generators is increased year by year. The blades of the wind driven generator are key components in the unit, and the structural health of the blades is vital to the safe operation of the unit. Therefore, the online monitoring of the state of the wind driven generator blade has important engineering practical value. The natural frequency of the blade represents the characteristic of a blade structure, when the blade is damaged, such as cracks, the rigidity of the blade is reduced, and the natural frequency of the blade is reduced, so that the existing technology for monitoring the health state of the blade based on the natural frequency deviation of the blade is widely concerned by researchers at home and abroad, but the natural frequency of the blade is influenced by the operation conditions of a unit, such as wind speed, rotating speed, power, pitch angle, temperature and the like, and even in a normal health state, the natural frequency of the blade is in change, so that the health state of the blade cannot be evaluated by directly utilizing the actually measured natural frequency. Therefore, the problem of working condition compensation of the natural frequency of the blade needs to be solved.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a nonlinear modeling wind power blade natural frequency working condition compensation method which can realize the working condition compensation of the natural frequency of a wind driven generator blade.
In order to achieve the aim, the wind power blade natural frequency working condition compensation method based on the nonlinear modeling comprises the following steps:
1) selecting a historical operation data set when the wind turbine generator normally operates, dividing the historical operation data set into a plurality of modeling data subsets according to the power P of a SCADA (supervisory control and data acquisition) system of the wind turbine generator, and then dividing each modeling data subset into a modeling training data set and a model testing data set respectively;
2) normalizing the elements in each modeling data subset;
3) constructing a corresponding observation memory matrix D according to the normalized modeling training data set to construct an inherent frequency prediction model, and then utilizing the constructed observation memory matrix D and a corresponding model test data set X thereoftestTesting and verifying the natural frequency prediction model, wherein one modeling data subset corresponds to one natural frequency prediction model;
4) the method comprises the steps of obtaining the power of a current wind driven generator data acquisition and monitoring system SCADA and unit working condition parameters, searching a corresponding natural frequency prediction model according to the power of the current wind driven generator data acquisition and monitoring system SCADA, substituting the working condition parameters of the current unit into the corresponding natural frequency prediction model to obtain a predicted current fixed frequency, carrying out subtraction operation on the actually measured current fixed frequency and the predicted current fixed frequency, taking an operation result as a fixed frequency deviation value, and finally completing working condition compensation of the natural frequency of the wind driven generator blade by using the fixed frequency deviation value.
The specific operation of the step 1) is as follows:
selecting a historical operation data set when the wind turbine generator normally operates, wherein the historical operation data set comprises natural frequency, wind speed, power, rotating speed, pitch angle and temperature of a blade, when P is less than or equal to 0, the generator is in a shutdown state, and the working condition parameters of the generator influencing the fixed frequency of the blade are the wind speed and the temperature, and constructing a first modeling data subset by the natural frequency, the wind speed and the temperature data of the blade; when P is present>0 or P<PRated valueWhen the unit operates in the constant paddle state, working condition parameters influencing the fixed frequency of the unit are the natural frequency, the wind speed, the temperature and the power of the blade, and a second modeling data subset is constructed by the natural frequency, the wind speed, the temperature and the power data of the blade; when P is equal to PRated valueHour and machineAnd when the group runs in a variable pitch state, working condition parameters influencing the fixed frequency of the blade are the natural frequency, the wind speed, the temperature and the pitch angle of the blade, a third modeling data subset is constructed by the natural frequency, the wind speed, the temperature and the pitch angle of the blade, and then each modeling data subset is respectively divided into a modeling training data set and a model testing data set.
65% of the data in the subset of modeling data is divided into a modeling training data set and 35% of the data in the subset of modeling data is divided into a model test data set.
The expression of the observation memory matrix D is:
fixed frequency prediction result X output by natural frequency prediction modelpredictComprises the following steps:
the invention has the following beneficial effects:
according to the wind power generator blade natural frequency working condition compensation method based on the nonlinear modeling, during specific operation, a historical operation data set during normal operation of a wind power generator set is divided into a plurality of modeling data subsets, then a corresponding natural frequency prediction model is constructed according to each modeling data subset, during compensation, working condition parameters of the current set are only required to be substituted into the corresponding prediction fixed frequency model, then the difference is made between the predicted current fixed frequency and the actually measured fixed frequency, the difference result is used as a fixed frequency deviation value, the fixed frequency deviation value is not influenced by the working condition parameters, and finally the wind power generator blade natural frequency working condition compensation based on the nonlinear modeling is carried out according to the fixed frequency deviation value.
Drawings
FIG. 1 is a flow chart of the present invention for building a prediction model of a natural frequency prediction model;
FIG. 2 is a flow chart of the present invention for predicting the current natural frequency;
FIG. 3 is a diagram of the natural frequency prediction result in a unit shutdown state;
FIG. 4 is a diagram showing the results of the predicted relative errors of natural frequency, wind speed and temperature in the shutdown state of the unit;
FIG. 5 is a diagram of the natural frequency prediction result in the unit under the state of unchanged paddle operation;
FIG. 6 is a diagram showing the results of the predicted relative errors of natural frequency, wind speed, power and temperature under the operating state of the set of constant pitch;
FIG. 7 is a diagram of a natural frequency prediction result in a variable pitch operation state of a unit;
FIG. 8 is a diagram showing the results of the predicted relative errors of the natural frequency, the wind speed, the pitch angle and the temperature of the unit in the variable-pitch operating state.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
referring to fig. 1 and 2, the wind turbine blade natural frequency condition compensation method based on nonlinear modeling includes the following steps:
1) selecting a historical operation data set of the wind turbine generator set in normal operation for 1-3 months, dividing the historical operation data set into a plurality of modeling data subsets according to the power P of an SCADA (supervisory control and data acquisition) system of the wind turbine generator, and then dividing each modeling data subset into a modeling training data set and a model test data set respectively, wherein 65% of data in the modeling data subsets are divided into the modeling training data sets, and 35% of data in the modeling data subsets are divided into the model test data sets;
the specific operation of the step 1) is as follows: selecting a historical operation data set when the wind turbine generator normally operates, wherein the historical operation data set comprises natural frequency, wind speed, power, rotating speed, pitch angle and temperature of a blade, when P is less than or equal to 0, the generator is in a shutdown state, and the working condition parameters of the generator influencing the fixed frequency of the blade are the wind speed and the temperature, and constructing a first modeling data subset by the natural frequency, the wind speed and the temperature data of the blade; when P is present>0 or P<PRated valueIn the meantime, the unit operates in the constant paddle state, and the working condition parameters influencing the fixed frequency of the unit areConstructing a second modeling data subset by the natural frequency, the wind speed, the temperature and the power data of the blades; when P is equal to PRated valueWhen the unit operates in a variable pitch state, working condition parameters influencing the fixed frequency of the blades are the natural frequency, the wind speed, the temperature and the pitch angle of the blades, a third modeling data subset is constructed by the natural frequency, the wind speed, the temperature and the pitch angle of the blades, and then each modeling data subset is divided into a modeling training data set and a model testing data set.
2) Normalizing the elements in each modeling data subset;
3) constructing a corresponding observation memory matrix D according to the normalized modeling training data set to construct an inherent frequency prediction model, and then testing a data set X according to the constructed observation memory matrix D and the corresponding modeltestTesting and verifying the natural frequency prediction model, wherein one modeling data subset corresponds to one natural frequency prediction model, and the expression of the observation memory matrix D is as follows:
each column in D represents one normal state sample of the modeling data subset, which consists of n variables, D has a total of m sample sets,representing the calculation of the euclidean distance between the two vectors.
The fixed frequency prediction result X output by the natural frequency prediction modelpredictComprises the following steps:
and then checking whether the prediction error of the model test data set meets the requirement, setting the prediction relative error of the natural frequency to be less than or equal to 2%, setting the prediction relative error of the working condition parameter to be less than or equal to 5%, and if the prediction error meets the requirement, indicating that the constructed model meets the requirement, if the prediction error does not meet the requirement, continuing to correct and model, then repeating the steps until the prediction error meets the requirement and the modeling is finished.
4) The method comprises the steps of obtaining the power of a current wind driven generator data acquisition and monitoring system SCADA and unit working condition parameters, searching a corresponding natural frequency prediction model according to the power of the current wind driven generator data acquisition and monitoring system SCADA, substituting the working condition parameters of the current unit into the corresponding natural frequency prediction model to obtain a predicted current fixed frequency, carrying out subtraction operation on the actually measured current fixed frequency and the predicted current fixed frequency, taking an operation result as a fixed frequency deviation value, and finally completing working condition compensation of the natural frequency of the wind driven generator blade by using the fixed frequency deviation value.
Fig. 3 is a diagram of a prediction result of the natural frequency in the unit shutdown state, and the predicted value and the measured value in fig. 3 are all overlapped, so that the prediction is accurate. Fig. 4 is a diagram showing the result of the prediction relative errors of the natural frequency, the wind speed and the temperature in the shutdown state of the unit, wherein the maximum prediction error is only 0.012%, and the prediction error meets the requirement.
Fig. 5 is a diagram of a natural frequency prediction result in a unit constant-pitch running state, and the predicted value and the measured value in fig. 5 are all overlapped, so that prediction is accurate. FIG. 6 is a diagram showing the results of the predicted relative errors of the natural frequency, the wind speed, the temperature and the power of the unit under the operating state of the variable propeller, wherein the maximum predicted error is only-0.025%, and the predicted error meets the requirements.
Fig. 7 is a prediction result diagram of the natural frequency of the unit in the variable pitch operation state, and the predicted value and the measured value in fig. 7 are all overlapped, so that the prediction is accurate. FIG. 8 is a diagram showing the results of the predicted relative errors of the natural frequency, the wind speed, the temperature and the pitch angle of the unit in the variable-pitch operating state, wherein the maximum predicted error is only-0.1%, and the predicted error meets the requirements.
Claims (5)
1. A wind power blade natural frequency working condition compensation method based on nonlinear modeling is characterized by comprising the following steps:
1) selecting a historical operation data set when the wind turbine generator normally operates, dividing the historical operation data set into a plurality of modeling data subsets according to the power P of a SCADA (supervisory control and data acquisition) system of the wind turbine generator, and then dividing each modeling data subset into a modeling training data set and a model testing data set respectively;
2) normalizing the elements in each modeling data subset;
3) constructing a corresponding observation memory matrix D according to the normalized modeling training data set to construct an inherent frequency prediction model, and then utilizing the constructed observation memory matrix D and a corresponding model test data set X thereoftestTesting and verifying the natural frequency prediction model, wherein one modeling data subset corresponds to one natural frequency prediction model;
4) the method comprises the steps of obtaining the power of a current wind driven generator data acquisition and monitoring system SCADA and unit working condition parameters, searching a corresponding natural frequency prediction model according to the power of the current wind driven generator data acquisition and monitoring system SCADA, substituting the working condition parameters of the current unit into the corresponding natural frequency prediction model to obtain a predicted current fixed frequency, carrying out subtraction operation on the actually measured current fixed frequency and the predicted current fixed frequency, taking an operation result as a fixed frequency deviation value, and finally completing working condition compensation of the natural frequency of the wind driven generator blade by using the fixed frequency deviation value.
2. The nonlinear-modeled wind turbine blade natural frequency condition compensation method according to claim 1, characterized in that the specific operation of step 1) is as follows:
selecting a historical operation data set when the wind turbine generator normally operates, wherein the historical operation data set comprises natural frequency, wind speed, power, rotating speed, pitch angle and temperature of a blade, when P is less than or equal to 0, the generator is in a shutdown state, and the working condition parameters of the generator influencing the fixed frequency of the blade are the wind speed and the temperature, and constructing a first modeling data subset by the natural frequency, the wind speed and the temperature data of the blade; when P is present>0 and P<PRated valueIn the time, the unit operates in the constant paddle state, and the working condition parameters influencing the fixed frequency of the unit are the natural frequency, the wind speed, the temperature and the power of the blade, and the working condition parameters are the natural frequency, the wind speed, the temperature and the power of the bladeThe power data construct a second subset of modeling data, PRated valueThe rated power generation power of the wind generating set; when P is equal to PRated valueWhen the unit operates in a variable pitch state, working condition parameters influencing the fixed frequency of the blades are the natural frequency, the wind speed, the temperature and the pitch angle of the blades, a third modeling data subset is constructed by the natural frequency, the wind speed, the temperature and the pitch angle of the blades, and then each modeling data subset is divided into a modeling training data set and a model testing data set.
3. The nonlinear-modeled wind turbine blade natural frequency condition compensation method according to claim 2, characterized in that 65% of data in the modeling data subset is divided into a modeling training data set, and 35% of data in the modeling data subset is divided into a model test data set.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810622308.8A CN108869174B (en) | 2018-06-15 | 2018-06-15 | Nonlinear modeling wind driven generator blade natural frequency working condition compensation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810622308.8A CN108869174B (en) | 2018-06-15 | 2018-06-15 | Nonlinear modeling wind driven generator blade natural frequency working condition compensation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108869174A CN108869174A (en) | 2018-11-23 |
CN108869174B true CN108869174B (en) | 2020-06-19 |
Family
ID=64339289
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810622308.8A Active CN108869174B (en) | 2018-06-15 | 2018-06-15 | Nonlinear modeling wind driven generator blade natural frequency working condition compensation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108869174B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110298455B (en) * | 2019-06-28 | 2023-06-02 | 西安因联信息科技有限公司 | Mechanical equipment fault intelligent early warning method based on multivariate estimation prediction |
CN111080981B (en) * | 2019-12-30 | 2021-10-22 | 安徽容知日新科技股份有限公司 | Alarm method and alarm system of equipment and computing equipment |
CN111412115A (en) * | 2020-04-07 | 2020-07-14 | 国家电投集团广西电力有限公司 | Novel wind power tower cylinder state online monitoring method and system |
CN112594125A (en) * | 2020-11-29 | 2021-04-02 | 上海电机学院 | Automatic-shrinkage wind power generation blade and control method thereof |
CN113847212B (en) * | 2021-10-29 | 2023-05-02 | 中国华能集团清洁能源技术研究院有限公司 | Wind turbine generator blade natural frequency monitoring method |
CN115374653B (en) * | 2022-10-21 | 2022-12-20 | 宇动源(北京)信息技术有限公司 | NSET model-based wind driven generator temperature early warning method and related device |
Family Cites Families (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101033730B (en) * | 2007-01-25 | 2010-06-02 | 上海交通大学 | Control method for stably operating wind power field using double-fed asynchronous generator |
US8301311B2 (en) * | 2009-07-06 | 2012-10-30 | Siemens Aktiengesellschaft | Frequency-responsive wind turbine output control |
CN102478421B (en) * | 2010-11-24 | 2013-07-17 | 中国科学院工程热物理研究所 | Dynamic frequency analysis method of wind turbine blade and design method |
CN103629046B (en) * | 2012-08-20 | 2016-08-03 | 新疆金风科技股份有限公司 | Appraisal procedure, device and the wind-driven generator of a kind of wind-driven generator performance |
CN103967702B (en) * | 2014-04-25 | 2016-04-13 | 河海大学 | A kind of double-fed wind power generator full blast speed frequency response controlling method |
KR101541490B1 (en) * | 2014-04-29 | 2015-08-03 | 부산대학교 산학협력단 | Method for designing multilayer tuned liquid damper in floating wind turbine |
CN104005917B (en) * | 2014-04-30 | 2016-12-07 | 叶翔 | Method and system fan condition being predicted based on Bayesian inference mode |
CN104747368B (en) * | 2015-01-27 | 2017-11-07 | 风脉(武汉)可再生能源技术有限责任公司 | A kind of method and system of Wind turbines power optimization |
CN105134510A (en) * | 2015-09-18 | 2015-12-09 | 北京中恒博瑞数字电力科技有限公司 | State monitoring and failure diagnosis method for wind generating set variable pitch system |
CN105257470B (en) * | 2015-09-25 | 2018-03-20 | 南车株洲电力机车研究所有限公司 | A kind of Wind turbines wind direction compensation optimizing method and device |
CN105808829B (en) * | 2016-03-02 | 2018-10-30 | 西安交通大学 | A kind of turbomachinery Natural Frequency of Blade characteristic analysis method based on CPU+GPU heterogeneous Computings |
CN108092577B (en) * | 2016-11-23 | 2022-04-08 | 台达电子工业股份有限公司 | Wind power generation system and control method suitable for same |
CN106897717B (en) * | 2017-02-09 | 2020-11-03 | 同济大学 | Bayesian model correction method under multiple tests based on environmental excitation data |
CN107829885B (en) * | 2017-10-25 | 2020-04-07 | 西安锐益达风电技术有限公司 | Wind driven generator blade vibration monitoring and system considering environmental parameter correction |
-
2018
- 2018-06-15 CN CN201810622308.8A patent/CN108869174B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN108869174A (en) | 2018-11-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108869174B (en) | Nonlinear modeling wind driven generator blade natural frequency working condition compensation method | |
CN106815771B (en) | Long-term assessment method for wind farm load | |
CN110259646B (en) | Wind generating set component state early warning method based on historical data | |
CN111322205B (en) | Wind turbine generator set wind vane zero error identification method and correction method | |
CN105649875B (en) | Variable pitch control method and device of wind generating set | |
CN103219725A (en) | Wind power plant equivalent modeling method based on real-time operation data | |
Neustadter et al. | Method for evaluating wind turbine wake effects on wind farm performance | |
CN111340307B (en) | Method for predicting wind power generation power of fan and related device | |
CN111287911A (en) | Wind turbine fatigue load early warning method and system | |
CN111396248A (en) | Wind turbine generator set intelligent yaw control method based on short-term wind direction prediction | |
CN104074687A (en) | Load and performance testing method and device used for megawatt wind generation set | |
CN111311021A (en) | Theoretical power prediction method, device, equipment and storage medium for wind power plant | |
CN115906616A (en) | Method for calculating theoretical power and reactive power generation capacity of wind power plant | |
CN109798226B (en) | Wind turbine generator tower load prediction method and system | |
CN107221933B (en) | Probabilistic load flow calculation method | |
CN113468728A (en) | Variable pitch system fault prediction method based on neural network | |
CN116542030A (en) | Double-fed fan parameter identification method, system and equipment based on track sensitivity | |
CN115898787A (en) | Method and device for dynamically identifying static yaw error of wind turbine generator | |
CN109657380A (en) | A kind of double-fed fan motor field Dynamic Equivalence based on Extended Kalman filter | |
Sethi et al. | Vibration signal-based diagnosis of wind turbine blade conditions for improving energy extraction using machine learning approach | |
CN114295367A (en) | Wind turbine generator gearbox working condition online monitoring method | |
Sanchez et al. | Numerical modelling techniques to predict rotor imbalance problems in tidal stream turbines | |
CN115062653B (en) | Analysis maintenance system based on steam turbine of thermal power plant | |
CN112343776A (en) | Wind turbine generator blade efficiency diagnosis method based on field operation data | |
CN117910385A (en) | LightGBM-based wind speed wake restoration method, lightGBM-based wind speed wake restoration system, medium and electronic equipment |
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 |