CN108343566A - A kind of blade icing On-line Fault monitoring method and system based on running of wind generating set state - Google Patents
A kind of blade icing On-line Fault monitoring method and system based on running of wind generating set state Download PDFInfo
- Publication number
- CN108343566A CN108343566A CN201810265635.2A CN201810265635A CN108343566A CN 108343566 A CN108343566 A CN 108343566A CN 201810265635 A CN201810265635 A CN 201810265635A CN 108343566 A CN108343566 A CN 108343566A
- Authority
- CN
- China
- Prior art keywords
- blade
- icing
- wind
- data
- degree
- 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 30
- 238000012544 monitoring process Methods 0.000 title claims abstract description 30
- 238000013528 artificial neural network Methods 0.000 claims abstract description 32
- 239000011248 coating agent Substances 0.000 claims abstract description 18
- 238000000576 coating method Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000010606 normalization Methods 0.000 claims description 33
- 238000010248 power generation Methods 0.000 claims description 23
- 238000003062 neural network model Methods 0.000 claims description 16
- 238000003745 diagnosis Methods 0.000 claims description 15
- 230000004927 fusion Effects 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 3
- 210000005036 nerve Anatomy 0.000 claims description 2
- 230000001737 promoting effect Effects 0.000 abstract description 2
- 238000012512 characterization method Methods 0.000 abstract 1
- 238000012549 training Methods 0.000 description 10
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 10
- 239000007788 liquid Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 239000013256 coordination polymer Substances 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000007774 longterm Effects 0.000 description 4
- 238000001556 precipitation Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000004781 supercooling Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 239000013078 crystal Substances 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 206010009866 Cold sweat Diseases 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000010297 mechanical methods and process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- F03D80/00—Details, components or accessories not provided for in groups F03D1/00 - F03D17/00
- F03D80/40—Ice detection; De-icing means
-
- 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
Abstract
The invention discloses a kind of blade icing On-line Fault monitoring method and system based on running of wind generating set state, this method utilizes the running state data that Wind turbines SCADA system obtains, the ice coating state characteristic index of blade is extracted after processing appropriate, the blade icing of input structure diagnoses neural network, output had not only characterized the icing degree of blade, but also the characteristic index of the icing position of characterization blade.This method is lossless one kind, dynamic, real-time monitoring, diagnosing method, and can diagnose blade icing failure in practice for engineering and provide theoretical foundation and technical support；Entire monitoring system, it is easy to operate, there is larger promotional value.
Description
Technical field
The invention belongs to field of wind power equipment, more particularly to a kind of blade icing based on running of wind generating set state
On-line Fault monitoring method and system.
Background technology
China has vast territory, and meteorological condition is complicated, land and sea in northern territory, the High aititude mountain of southern areas
The wind and snow frost in area etc., winter brings high risks to the operation of wind energy conversion system.The Wind turbines in clammy area in winter are run,
It often will appear serious icing phenomenon in winter.After wind power generation unit blade icing in work, the dynamic characteristics of blade can be destroyed
Cause blade overload and load uneven after the properties such as quality, the rigidity of ice incorporate blade material characteristic with aerodynamic characteristic,
So that its characteristic is not inconsistent with design original intention, causes wind energy conversion system to be unable to normal power generation, even result in serious safety accident.In time, accurate
The ice coating state for really detecting blade, convenient for taking timely, effective operation and maintenance measure to wind energy conversion system, to improving wind-powered electricity generation
The security reliability and economy of field operation, are of great significance.
Since the 1990s, people's extensive concern, research pair have been caused for the research of component icing detection
As being concentrated mainly on aviation field (such as aircraft), main measurement method includes：Mechanical Method, electric method, calorifics method, optical method,
Waveguide method etc..But the technology for being directly used in the monitoring of wind power generation unit blade icing at present is still fewer, ripe can be used for work
The actual wind power generation unit blade icing monitoring method of journey and the still rare document announcement of technology.
Invention content
The technical issues of present invention is for the monitoring of existing wind power generation unit blade icing and diagnosis aspect, providing one kind can
Quickly, the method and system of online, lossless, real-time monitoring and diagnosis wind power generation unit blade ice coating state.
A kind of blade icing On-line Fault monitoring method based on running of wind generating set state, includes the following steps：
Step 1：Acquire history Wind turbines data；
The Wind turbines data include sensing data and icing data；
Wherein, the sensing data includes temperature, relative air humidity, wind speed, wind direction, cabin direction, blade pitch
Angle, generator electrical power, the icing data include the icing degree in each area of blade corresponding with sensing data and entire
Blade icing degree；
Each area of blade includes tip region, middle part of blade region and root of blade region, and icing degree is followed successively by
Without icing [0.0~0.2], slight icing (0.2~0.4], general icing (0.4~0.6], more serious icing (0.6~0.8],
Serious icing (0.8~1], entire blade icing degree is tip region, middle part of blade region and root of blade region three
The mean value of icing degree；
Step 2：It after sensing data is amplified and is filtered, then is normalized, it is special to obtain normalization
Levy index；
Step 3：It builds blade icing and diagnoses neural network model；
Using the normalization characteristic index of history Wind turbines data as input data, corresponding icing data are as output
Data are trained neural network, obtain blade icing and diagnose neural network model；
Step 4：The sensing data of acquisition Wind turbines in real time, and handled according to step 2, obtain normalization characteristic
Index, by obtained normalization characteristic index input blade icing diagnose neural network model, obtain each area of real-time blade and
Whole ice coating state characterize data realizes the real-time monitoring to blade icing.
Further, the normalization characteristic index is by sensing data by carrying out following fusion treatment acquisition, including
Temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics and yaw corner characteristics；
1) temperature profile x1：
Wherein, t is the temperature data of temperature sensor acquisition at wind-powered machine unit hub height；
The environment temperature of Wind turbines working region is the key that influence blade icing natural cause, therefore, temperature profile
Index is to diagnose the core feature index of blade icing failure.The temperature range of Wind turbines work is -40 DEG C~50 DEG C, if wind
Currently practical operating temperature is t DEG C at motor group hub height, x1Value range [- 1~+1].
Long-term wind field experience have shown that, when environment temperature is less than 5 DEG C, unit can have the phenomenon that slight icing.Therefore,
x1When >=0.1, blade is not in icing phenomenon.
2) Humidity Features x2：x2=w；
Wherein, w is the moisture signal for the relative air humidity sensor acquisition being installed on outside wind turbine cabin, value model
It encloses for [0,1]；
The relative air humidity of Wind turbines working region is the another crucial natural cause for influencing blade icing.Work as environment
One timing of temperature, with the increase of relative air humidity, blade icing rate is also in the trend of monotone increasing.Humidity influences air
The density and diameter of middle super-cooling waterdrop, when super-cooling waterdrop strikes blade surface, the density and diameter of water droplet directly affect blade
The quality and area of icing.When relative air humidity is higher (85% or more), under low temperature environment, the liquid being relatively large in diameter is precipitated
Water droplet after being captured by blade, easily forms glaze ice.When relative air humidity is relatively low (45%~65%), the liquid of precipitation
Water content is reduced, and drop has not also just formd crystal, as hernel ice by blade capture at low ambient temperatures.When air is wet
When degree is less than 30%, it is difficult to there is liquid water droplets precipitation, the generation of icing phenomenon is not had generally.
x2Value be ranging from [0,1].Work as x2When ＜ 0.3, blade is not in icing phenomenon.
3) power-wind speed feature x3：
Wherein, P and P0The Wind turbines realtime power and rated power of the acquisition of generator electrical power sensor are indicated respectively,Indicate average value of the wind speed of the air velocity transducer acquisition of Wind turbines in one minute,Indicate the specified wind of Wind turbines
Speed；The wind speed of Wind turbines working region is the another crucial natural cause for influencing blade icing.Wind is responsible for the liquid in air
State water droplet is transported to blade surface and influences its Convective Heat Transfer, and wind speed is bigger, the amount of droplets of unit interval blade capture
Increase, and the heat exchange efficiency of blade and frozen surface is reinforced, at low ambient temperatures, blade is more prone to icing phenomenon.
Ignore the loss of transmission system, the output power of Wind turbines is：
Wherein, ρ is atmospheric density；R is wind wheel radius；For mean wind speed；CPFor power coefficient.
Under declared working condition in no icing, the output power of Wind turbines is：
After blade icing, lead to the power coefficient C of bladePIt reduces, is reduced so as to cause the output power of unit.Institute
With CPBe reflection blade whether the important indicator of icing.
It can obtain：
In formula, P0Indicate rated power,Indicate rated wind speed, CP0It is utilized for the optimal wind energy of unit under nominal power
Coefficient.
x3Value range be [0,1].When blade does not freeze, x3=0；It (at this time can not output work when blade heavy ice
Rate), x3=1；When blade is under different ice coating states, x3∈(0,1)。
4) blade pitch corner characteristics x4：
Wherein, β is propeller pitch angle theoretical value when wind power generation unit blade is run,β ' is the propeller pitch angle of wind power generation unit blade actual motion,
It is read by SCADA system；
Wind power generation unit blade operation propeller pitch angle is the principal states parameter for reflecting blade and whether having frozen.Studies have shown that
When unit output power reaches rated power P0And when remaining unchanged, there is very strong more than rated wind speed for propeller pitch angle and wind speed
It is non-linear, i.e., with mean wind speedFor independent variable, propeller pitch angle β has as dependent variable：
Using wind speed number more than the unit rated wind speed (9m/s) of 2MW of method pair of least square polynomial fit
According to being fitted, the nonlinear function model of following wind speed and propeller pitch angle is established：
Then, the theoretical value of propeller pitch angle can be indicated by following formula：
Wherein,Indicate average value of the wind speed of the air velocity transducer acquisition of Wind turbines in one minute,Indicate wind-powered electricity generation
The rated wind speed of unit；
When blade does not freeze, x4=0；When blade heavy ice, x4=1.Long-term wind power plant operating experience shows to work as wind
When speed is more than incision wind speed and is less than rated wind speed, the propeller pitch angle of not each unit is maintained at 0 °, specified defeated to reach
Going out power, the best propeller pitch angle of wind power generation unit blade is generally all in the range of 0 °~2 °, in cut-out wind speed, the paddle of unit
Elongation maximum does not exceed 40 °.Therefore, work as x4When ＜ 0, it is thought of as Wind turbines failure caused by non-icing factor.
5) yaw corner characteristics x5,
Wherein, θ and θ0Yaw angle of the Wind turbines under actual condition and declared working condition, θ are indicated respectively0=0 °, θ=γ-
α, α are the wind direction angle signal of wind direction angle transducer acquisition, and γ is the cabin deflection of cabin direction sensor acquisition.
Wind turbines yaw angle is the principal states parameter for reflecting blade and whether having frozen.Wind turbine is in actual moving process
In, it is influenced by unit sender, yaw angle belongs to normal phenomenon in the range of maintaining -8 °~8 °, yaw range is more than 8 °
It is contemplated that yawed since mistake caused by icing failure occurs for unit, some researches show that, one timing of wind speed, pneumatic equipment bladess
The power of wind wheel output and the cosine value of yaw angle are proportional.
x5Value range be [0,1].When blade does not freeze, x5=0；When blade heavy ice, wind-driven generator is defeated at this time
It is 0, x to go out power5=1.
Further, blade icing diagnosis neural network is three-decker BP neural network, wherein input layer knot
Points are 5, and output layer nodal point number is 4, and a hidden layer node number m is determined with following formula：
Wherein, mxIt is input layer nodal point number；myIt is output layer nodal point number；A is constant, and value range is [1,10]；
A takes different values that can influence the training effect of neural network, value is carried out to a using the method for exhaustion, until meeting god
Until being required through neural network accuracy；
Made successively with the temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics, yaw corner characteristics
For input layer, with the icing degree of entire blade, tip region icing degree, middle part of blade region icing degree and leaf
Piece root area icing degree absolute value is successively as output node layer；
The icing degree absolute value in each region of blade is calculated using following formula and is obtained：
Wherein, δ indicates that the practical ice covering thickness of blade, L are the length of blade, and N indicates intermediate variable, y2、y3、y4Respectively
Indicate tip region icing degree absolute value, middle part of blade region icing degree absolute value and root of blade region icing degree
Absolute value.
Due to size and structure according to the difference of the unit capacity and difference of pneumatic equipment bladess, practical ice sheet is used merely
Thickness absolute value has great limitation, icing degree absolute value y to describe pneumatic equipment bladess icing degreei, to describe
The practical icing degree of blade；
Further, using the normalization characteristic index of history Wind turbines data as input data, corresponding icing number
According to as output data, when being trained to neural network, used training data further includes the sample number under representative condition
According to；
Sample data under the representative condition is as shown in table 1：
Sample data under 1 representative condition of table
Wherein, y1For the icing degree of entire blade, y2For tip region icing degree, y3For middle part of blade region icing
Degree and y4For root of blade region icing degree.
A kind of blade icing On-line Fault monitoring system based on running of wind generating set state, including：
History Wind turbines data acquisition module, the sensing data for acquiring Wind turbines and icing data；
Wherein, the sensing data includes temperature, relative air humidity, wind speed, wind direction, cabin direction, blade pitch
Angle, generator electrical power, the icing data include the icing degree in each area of blade corresponding with sensing data and entire
Blade icing degree；
Each area of blade includes tip region, middle part of blade region and root of blade region, and icing degree is followed successively by
Without icing [0.0~0.2], slight icing (0.2~0.4], general icing (0.4~0.6], more serious icing (0.6~0.8],
Serious icing (0.8~1], entire blade icing degree is tip region, middle part of blade region and root of blade region three
The mean value of icing degree；
After sensing data is amplified and is filtered, then place is normalized in Wind turbines data processing module
Reason, obtains normalization characteristic index；
Blade icing diagnoses neural network model and builds module, for referring to the normalization characteristic of history Wind turbines data
It is denoted as input data, corresponding icing data are trained neural network as output data, obtain the diagnosis of blade icing
Neural network model；
Real-time monitoring modular, to acquiring the sensing data of Wind turbines in real time, by the Wind turbines data processing
Resume module obtains normalization characteristic index, and obtained normalization characteristic index input blade icing is diagnosed neural network mould
Type obtains each area of real-time blade and whole ice coating state characterize data, realizes the real-time monitoring to blade icing.
Further, further include fusion treatment module, to the normalization characteristic index by carrying out sensing data
Following fusion treatment obtains, and obtains temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics and yaw angle
Feature；
1) temperature profile x1：
Wherein, t is the temperature data of temperature sensor acquisition at wind-powered machine unit hub height；
2) Humidity Features x2：x2=w；
Wherein, w is the moisture signal for the relative air humidity sensor acquisition being installed on outside wind turbine cabin, value model
It encloses for [0,1]；
3) power-wind speed feature x3：
Wherein, P and P0The Wind turbines realtime power and rated power of the acquisition of generator electrical power sensor are indicated respectively,Indicate average value of the wind speed of the air velocity transducer acquisition of Wind turbines in one minute,Indicate the specified wind of Wind turbines
Speed；
4) blade pitch corner characteristics x4：
Wherein, β is propeller pitch angle theoretical value when wind power generation unit blade is run,β ' is the propeller pitch angle of wind power generation unit blade actual motion,
It is read by SCADA system；
5) yaw corner characteristics x5,
Wherein, θ and θ0Yaw angle of the Wind turbines under actual condition and declared working condition, θ are indicated respectively0=0 °, θ=γ-
α, α are the wind direction angle signal of wind direction angle transducer acquisition, and γ is the cabin deflection of cabin direction sensor acquisition.
Further, blade icing diagnosis neural network is three-decker BP neural network, wherein input layer knot
Points are 5, and output layer nodal point number is 4, and a hidden layer node number m is determined with following formula：
Wherein, mxIt is input layer nodal point number；myIt is output layer nodal point number；A is constant, and value range is [1,10]；
A takes different values that can influence the training effect of neural network, value is carried out to a using the method for exhaustion, until meeting god
Until being required through neural network accuracy；
Made successively with the temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics, yaw corner characteristics
For input layer, with the icing degree of entire blade, tip region icing degree, middle part of blade region icing degree and leaf
Piece root area icing degree absolute value is successively as output node layer；
The icing degree absolute value in each region of blade is calculated using following formula and is obtained：
Wherein, δ indicates that the practical ice covering thickness of blade, L are the length of blade, and N indicates intermediate variable, y2、y3、y4Respectively
Indicate tip region icing degree absolute value, middle part of blade region icing degree absolute value and root of blade region icing degree
Absolute value.
Due to size and structure according to the difference of the unit capacity and difference of pneumatic equipment bladess, practical ice sheet is used merely
Thickness absolute value has great limitation, icing degree absolute value y to describe pneumatic equipment bladess icing degreei, to describe
The practical icing degree of blade；
Further, further include the display module being connected with real-time monitoring modular, for showing each state acquired letter
Number time-domain curve and extraction icing characteristic versus time curve.
Advantageous effect
The present invention provides a kind of blade icing On-line Fault monitoring method and system based on running of wind generating set state,
This method chooses the signal of the existing sensor acquisition in the part being mounted on Wind turbines, passes through theoretical calculation model and computer
Software and hardware system is extracted the characteristic index that can reflect blade ice coating state；Construct the nerve of diagnosis blade ice coating state
Network model retouches blade icing degree characteristic index, icing position by features described above index as the input layer variable of network
State output layer variable of the index as network；Using the operating states of the units data under the practical ice coating state of blade, in conjunction with some
Gross data sample constructs the training sample set of blade icing diagnostic network；Utilize constructed training sample data to institute
The diagnostic network of construction is trained；The status data of Wind turbines actual motion is inputted into trained neural network, you can
Diagnose the ice coating state of blade.
The program in terms of existing technologies, has the following advantages：
(1) sensor, measuring point quantity needed for are few, measuring point quantity 7 (temperature sensor 1, humidity sensor 1
It is a, air velocity transducer 1, wind transducer 1, cabin direction sensor 1, blade pitch angle transducer 3)；
(2) value volume and range of product of Wind turbines sensor in existing operation can not be increased；
(3) dynamic, real-time, non-destructive monitoring and the diagnosis for realizing wind power generation unit blade ice coating state, will not cause blade
New damage；
(4) it is analyzed, is learnt high using the blade icing diagnostic result accuracy of scheme of the present invention by experimental verification；
(5) system is entirely monitored, it is easy to operate, there is larger promotional value.
Description of the drawings
Fig. 1 is the diagnosis blade icing neural network structure schematic diagram of the present invention；
Fig. 2 is the topological structure schematic diagram of the present invention；
Fig. 3 is the principle schematic of the present invention in specific application；
Fig. 4 is the flow diagram of the method for the invention.
Specific implementation mode
Below in conjunction with drawings and examples, the present invention is described further.
As shown in Figure 3 and Figure 4, a kind of blade icing On-line Fault monitoring method based on running of wind generating set state, packet
Include following steps：
Step 1：Acquire history Wind turbines data；
The Wind turbines data include sensing data and icing data；
Wherein, the sensing data includes temperature, relative air humidity, wind speed, wind direction, cabin direction, blade pitch
Angle, generator electrical power, the icing data include the icing degree in each area of blade corresponding with sensing data and entire
Blade icing degree；
Each area of blade includes tip region, middle part of blade region and root of blade region, and icing degree is followed successively by
Without icing [0.0~0.2], slight icing (0.2~0.4], general icing (0.4~0.6], more serious icing (0.6~0.8],
Serious icing (0.8~1], entire blade icing degree is tip region, middle part of blade region and root of blade region three
The mean value of icing degree；
Step 2：It after sensing data is amplified and is filtered, then is normalized, it is special to obtain normalization
Levy index；
Step 3：It builds blade icing and diagnoses neural network model；
Using the normalization characteristic index of history Wind turbines data as input data, corresponding icing data are as output
Data are trained neural network, obtain blade icing and diagnose neural network model；
It is as shown in Figure 1 with neural network structure schematic diagram to diagnose blade icing.
Step 4：The sensing data of acquisition Wind turbines in real time, and handled according to step 2, obtain normalization characteristic
Index, by obtained normalization characteristic index input blade icing diagnose neural network model, obtain each area of real-time blade and
Whole ice coating state characterize data realizes the real-time monitoring to blade icing.
The normalization characteristic index by the way that sensing data is carried out following fusion treatment acquisition, including temperature profile,
Humidity Features, power-wind speed feature, blade pitch corner characteristics and yaw corner characteristics；
1) temperature profile x1：
Wherein, t is the temperature data of temperature sensor acquisition at wind-powered machine unit hub height；
The environment temperature of Wind turbines working region is the key that influence blade icing natural cause, therefore, temperature profile
Index is to diagnose the core feature index of blade icing failure.The temperature range of Wind turbines work is -40 DEG C~50 DEG C, if wind
Currently practical operating temperature is t DEG C at motor group hub height, x1Value range [- 1~+1].
Long-term wind field experience have shown that, when environment temperature is less than 5 DEG C, unit can have the phenomenon that slight icing.Therefore,
x1When >=0.1, blade is not in icing phenomenon.
2) Humidity Features x2：x2=w；
Wherein, w is the moisture signal for the relative air humidity sensor acquisition being installed on outside wind turbine cabin, value model
It encloses for [0,1]；
The relative air humidity of Wind turbines working region is the another crucial natural cause for influencing blade icing.Work as environment
One timing of temperature, with the increase of relative air humidity, blade icing rate is also in the trend of monotone increasing.Humidity influences air
The density and diameter of middle super-cooling waterdrop, when super-cooling waterdrop strikes blade surface, the density and diameter of water droplet directly affect blade
The quality and area of icing.When relative air humidity is higher (85% or more), under low temperature environment, the liquid being relatively large in diameter is precipitated
Water droplet after being captured by blade, easily forms glaze ice.When relative air humidity is relatively low (45%~65%), the liquid of precipitation
Water content is reduced, and drop has not also just formd crystal, as hernel ice by blade capture at low ambient temperatures.When air is wet
When degree is less than 30%, it is difficult to there is liquid water droplets precipitation, the generation of icing phenomenon is not had generally.
x2Value be ranging from [0,1].Work as x2When ＜ 0.3, blade is not in icing phenomenon.
3) power-wind speed feature x3：
Wherein, P and P0The Wind turbines realtime power and rated power of the acquisition of generator electrical power sensor are indicated respectively,Indicate average value of the wind speed of the air velocity transducer acquisition of Wind turbines in one minute, V0Indicate the specified wind of Wind turbines
Speed；The wind speed of Wind turbines working region is the another crucial natural cause for influencing blade icing.Wind is responsible for the liquid in air
State water droplet is transported to blade surface and influences its Convective Heat Transfer, and wind speed is bigger, the amount of droplets of unit interval blade capture
Increase, and the heat exchange efficiency of blade and frozen surface is reinforced, at low ambient temperatures, blade is more prone to icing phenomenon.
Ignore the loss of transmission system, the output power of Wind turbines is：
Wherein, ρ is atmospheric density；R is wind wheel radius；For mean wind speed；CPFor power coefficient.
Under declared working condition in no icing, the output power of Wind turbines is：
After blade icing, lead to the power coefficient C of bladePIt reduces, is reduced so as to cause the output power of unit.Institute
With CPBe reflection blade whether the important indicator of icing.
It can obtain：
In formula, P0Indicate rated power,Indicate rated wind speed, CP0It is utilized for the optimal wind energy of unit under nominal power
Coefficient.
x3Value range be [0,1].When blade does not freeze, x3=0；It (at this time can not output work when blade heavy ice
Rate), x3=1；When blade is under different ice coating states, x3∈(0,1)。
4) blade pitch corner characteristics x4：
Wherein, β is propeller pitch angle theoretical value when wind power generation unit blade is run,β ' is the propeller pitch angle of wind power generation unit blade actual motion,
It is read by SCADA system；
Wind power generation unit blade operation propeller pitch angle is the principal states parameter for reflecting blade and whether having frozen.Studies have shown that
When unit output power reaches rated power P0And when remaining unchanged, there is very strong more than rated wind speed for propeller pitch angle and wind speed
It is non-linear, i.e., with mean wind speedFor independent variable, propeller pitch angle β has as dependent variable：
Using wind speed number more than the unit rated wind speed (9m/s) of 2MW of method pair of least square polynomial fit
According to being fitted, the nonlinear function model of following wind speed and propeller pitch angle is established：
Then, the theoretical value of propeller pitch angle can be indicated by following formula：
Wherein,Indicate average value of the wind speed of the air velocity transducer acquisition of Wind turbines in one minute,Indicate wind
The rated wind speed of motor group；
When blade does not freeze, x4=0；When blade heavy ice, x4=1.Long-term wind power plant operating experience shows to work as wind
When speed is more than incision wind speed and is less than rated wind speed, the propeller pitch angle of not each unit is maintained at 0 °, specified defeated to reach
Going out power, the best propeller pitch angle of wind power generation unit blade is generally all in the range of 0 °~2 °, in cut-out wind speed, the paddle of unit
Elongation maximum does not exceed 40 °.Therefore, work as x4When ＜ 0, it is thought of as Wind turbines failure caused by non-icing factor.
5) yaw corner characteristics x5,
Wherein, θ and θ0Yaw angle of the Wind turbines under actual condition and declared working condition, θ are indicated respectively0=0 °, θ=γ-
α, α are the wind direction angle signal of wind direction angle transducer acquisition, and γ is the cabin deflection of cabin direction sensor acquisition.
Wind turbines yaw angle is the principal states parameter for reflecting blade and whether having frozen.Wind turbine is in actual moving process
In, it is influenced by unit sender, yaw angle belongs to normal phenomenon in the range of maintaining -8 °~8 °, yaw range is more than 8 °
It is contemplated that yawed since mistake caused by icing failure occurs for unit, some researches show that, one timing of wind speed, pneumatic equipment bladess
The power of wind wheel output and the cosine value of yaw angle are proportional.
x5Value range be [0,1].When blade does not freeze, x5=0；When blade heavy ice, wind-driven generator is defeated at this time
It is 0, x to go out power5=1.
The blade icing diagnosis neural network is three-decker BP neural network, and wherein input layer nodal point number is 5, defeated
It is 4 to go out layer nodal point number, and a hidden layer node number m is determined with following formula：
Wherein, mxIt is input layer nodal point number；myIt is output layer nodal point number；A is constant, and value range is [1,10]；
A takes different values that can influence the training effect of neural network, value is carried out to a using the method for exhaustion, until meeting god
Until being required through neural network accuracy；
Made successively with the temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics, yaw corner characteristics
For input layer, with the icing degree of entire blade, tip region icing degree, middle part of blade region icing degree and leaf
Piece root area icing degree absolute value is successively as output node layer；
The icing degree absolute value in each region of blade is calculated using following formula and is obtained：
Wherein, δ indicates that the practical ice covering thickness of blade, L are the length of blade, and N indicates intermediate variable, y2、y3、y4Respectively
Indicate tip region icing degree absolute value, middle part of blade region icing degree absolute value and root of blade region icing degree
Absolute value.
Due to size and structure according to the difference of the unit capacity and difference of pneumatic equipment bladess, practical ice sheet is used merely
Thickness absolute value has great limitation, icing degree absolute value y to describe pneumatic equipment bladess icing degreei, to describe
The practical icing degree of blade；
The icing degree absolute value in each region of blade is served only for model training；
Using the normalization characteristic index of history Wind turbines data as input data, corresponding icing data are as output
Data, when being trained to neural network, used training data further includes the sample data under representative condition；
Sample data under the representative condition is as shown in table 1：
Sample data under 1 representative condition of table
Wherein, y1For the icing degree of entire blade, y2For tip region icing degree, y3For middle part of blade region icing
Degree and y4For root of blade region icing degree.
As shown in Fig. 2, for the topology diagram of system of the present invention, a kind of blade based on running of wind generating set state
Icing On-line Fault monitors system, including：
History Wind turbines data acquisition module, the sensing data for acquiring Wind turbines and icing data；
Wherein, the sensing data includes temperature, relative air humidity, wind speed, wind direction, cabin direction, blade pitch
Angle, generator electrical power, the icing data include the icing degree in each area of blade corresponding with sensing data and entire
Blade icing degree；
Each area of blade includes tip region, middle part of blade region and root of blade region, and icing degree is followed successively by
Without icing [0.0~0.2], slight icing (0.2~0.4], general icing (0.4~0.6], more serious icing (0.6~0.8],
Serious icing (0.8~1], entire blade icing degree is tip region, middle part of blade region and root of blade region three
The mean value of icing degree；
After sensing data is amplified and is filtered, then place is normalized in Wind turbines data processing module
Reason, obtains normalization characteristic index；
Blade icing diagnoses neural network model and builds module, for referring to the normalization characteristic of history Wind turbines data
It is denoted as input data, corresponding icing data are trained neural network as output data, obtain the diagnosis of blade icing
Neural network model；
Real-time monitoring modular, to acquiring the sensing data of Wind turbines in real time, by the Wind turbines data processing
Resume module obtains normalization characteristic index, and obtained normalization characteristic index input blade icing is diagnosed neural network mould
Type obtains each area of real-time blade and whole ice coating state characterize data, realizes the real-time monitoring to blade icing.
The system further includes fusion treatment module, following by carrying out sensing data to the normalization characteristic index
Fusion treatment obtains, and it is special to obtain temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics and yaw angle
Sign；
The system further includes the display module being connected with real-time monitoring modular, for showing each status signal acquired
Time-domain curve and the icing characteristic versus time curve of extraction.
The function of scheme of the present invention realizes key is how to select the Wind turbines fortune comprising blade icing information
Row status signal establishes the mathematical model for the ice coating state characteristic parameter that blade is extracted from running of wind generating set status signal,
It is used in combination the blade icing fault status signal obtained in gross data and engineering as blade icing examining training sample, from these
Icing diagnostic characteristic set is built in sample data, and neural network is diagnosed with this characteristic set data training blade icing, from
And establish the initial neural network model of blade icing diagnosis.
After the present invention carries out analysis experiment in conjunction with a large amount of real data, proposition detects the existing sensor of Wind turbines
Operating state signal (including temperature sensor output signal, relative air humidity sensor output signal, air velocity transducer are defeated
The signal (mean wind speed signal) that goes out, wind transducer output signal, cabin direction sensor output signal, blade pitch angle pass
Sensor output signal, power of the assembling unit sensor output signal), monitoring system (SCADA) network built up by wind power plant,
Signal is sent to wind power plant monitoring center.User in wind farm local area network or can access any of Internet
Side enters wind power generation unit blade ice coating state monitoring and fault diagnosis system by Web modes, monitors wind power generation unit blade icing feelings
Condition.
Specific embodiment described herein is only an example for the spirit of the invention.Technology belonging to the present invention is led
The technical staff in domain can make various modifications or additions to the described embodiments or replace by a similar method
In generation, however, it does not deviate from the spirit of the invention or beyond the scope of the appended claims.
Claims (8)
1. a kind of blade icing On-line Fault monitoring method based on running of wind generating set state, which is characterized in that including following
Step：
Step 1：Acquire history Wind turbines data；
The Wind turbines data include sensing data and icing data；
Wherein, the sensing data includes temperature, relative air humidity, wind speed, wind direction, cabin direction, blade pitch angle, hair
Motor power, the icing data include that the icing degree in each area of blade corresponding with sensing data and entire blade cover
Ice degree；
Each area of blade includes tip region, middle part of blade region and root of blade region, and icing degree is followed successively by nothing and covers
Ice [0.0~0.2], slight icing (0.2~0.4], general icing (0.4~0.6], more serious icing (0.6~0.8], it is serious
Icing (0.8~1], entire blade icing degree is tip region, middle part of blade region and root of blade region three's icing
The mean value of degree；
Step 2：It after sensing data is amplified and is filtered, then is normalized, obtains normalization characteristic and refer to
Mark；
Step 3：It builds blade icing and diagnoses neural network model；
Using the normalization characteristic index of history Wind turbines data as input data, corresponding icing data are as output number
According to, neural network is trained, obtain blade icing diagnose neural network model；
Step 4：The sensing data of acquisition Wind turbines in real time, and handled according to step 2, it obtains normalization characteristic and refers to
Obtained normalization characteristic index input blade icing is diagnosed neural network model, obtains each area of real-time blade and whole by mark
Body ice coating state characterize data realizes the real-time monitoring to blade icing.
2. according to the method described in claim 1, it is characterized in that, the normalization characteristic index by by sensing data into
The following fusion treatment of row obtains, including temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics and yaw
Corner characteristics；
1) temperature profile x1：
Wherein, t is the temperature data of temperature sensor acquisition at wind-powered machine unit hub height；
2) Humidity Features x2：x2=w；
Wherein, w is the moisture signal for the relative air humidity sensor acquisition being installed on outside wind turbine cabin, and value range is
[0,1]；
3) power-wind speed feature x3：
Wherein, P and P0The Wind turbines realtime power and rated power of the acquisition of generator electrical power sensor are indicated respectively,Table
Show average value of the wind speed of the air velocity transducer acquisition of Wind turbines in one minute,Indicate the rated wind speed of Wind turbines；
4) blade pitch corner characteristics x4：
Wherein, β is propeller pitch angle theoretical value when wind power generation unit blade is run,
β ' is the propeller pitch angle of wind power generation unit blade actual motion, is read by SCADA system；
5) yaw corner characteristics x5,
Wherein, θ and θ0Yaw angle of the Wind turbines under actual condition and declared working condition, θ are indicated respectively0=0 °, θ=γ-α, α
For the wind direction angle signal of wind direction angle transducer acquisition, γ is the cabin deflection of cabin direction sensor acquisition.
3. according to the method described in claim 2, it is characterized in that, the blade icing diagnosis neural network is three-decker
BP neural network, wherein input layer nodal point number are 5, and output layer nodal point number is 4, and a hidden layer node number m is determined with following formula：
Wherein, mxIt is input layer nodal point number；myIt is output layer nodal point number；A is constant, and value range is [1,10]；
Using the temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics, yaw corner characteristics successively as defeated
Enter node layer, with the icing degree of entire blade, tip region icing degree, middle part of blade region icing degree and blade root
Portion region icing degree absolute value is successively as output node layer；
The icing degree absolute value in each region of blade is calculated using following formula and is obtained：
Wherein, δ indicates that the practical ice covering thickness of blade, L are the length of blade, and N indicates intermediate variable, y2、y3、y4It indicates respectively
Tip region icing degree absolute value, middle part of blade region icing degree absolute value and root of blade region icing degree are absolute
Value.
4. according to claim 2-3 any one of them methods, which is characterized in that the normalization of history Wind turbines data is special
Index is levied as input data, corresponding icing data are as output data, when being trained to neural network, used instruction
It further includes the sample data under representative condition to practice data；
Sample data under the representative condition is as shown in table 1：
Sample data under 1 representative condition of table
Wherein, y1For the icing degree of entire blade, y2For tip region icing degree, y3For middle part of blade region icing degree
And y4For root of blade region icing degree.
5. a kind of blade icing On-line Fault based on running of wind generating set state monitors system, which is characterized in that including：
History Wind turbines data acquisition module, the sensing data for acquiring Wind turbines and icing data；
Wherein, the sensing data includes temperature, relative air humidity, wind speed, wind direction, cabin direction, blade pitch angle, hair
Motor power, the icing data include that the icing degree in each area of blade corresponding with sensing data and entire blade cover
Ice degree；
Each area of blade includes tip region, middle part of blade region and root of blade region, and icing degree is followed successively by nothing and covers
Ice [0.0~0.2], slight icing (0.2~0.4], general icing (0.4~0.6], more serious icing (0.6~0.8], it is serious
Icing (0.8~1], entire blade icing degree is tip region, middle part of blade region and root of blade region three's icing
The mean value of degree；
Wind turbines data processing module after sensing data is amplified and is filtered, then is normalized, obtains
To normalization characteristic index；
Blade icing diagnoses neural network model and builds module, for making the normalization characteristic index of history Wind turbines data
For input data, corresponding icing data are trained neural network as output data, obtain blade icing diagnosis nerve
Network model；
Real-time monitoring modular, to acquiring the sensing data of Wind turbines in real time, by the Wind turbines data processing module
Processing, obtains normalization characteristic index, and obtained normalization characteristic index input blade icing is diagnosed neural network model, is obtained
Each area of real-time blade and whole ice coating state characterize data are obtained, realizes the real-time monitoring to blade icing.
6. system according to claim 2, which is characterized in that further include fusion treatment module, to the normalization characteristic
Index by the way that sensing data is carried out following fusion treatment acquisition, obtain temperature profile, Humidity Features, power-wind speed feature,
Blade pitch corner characteristics and yaw corner characteristics；
1) temperature profile x1：
Wherein, t is the temperature data of temperature sensor acquisition at wind-powered machine unit hub height；
2) Humidity Features x2：x2=w；
Wherein, w is the moisture signal for the relative air humidity sensor acquisition being installed on outside wind turbine cabin, and value range is
[0,1]；
3) power-wind speed feature x3：
Wherein, P and P0The Wind turbines realtime power and rated power of the acquisition of generator electrical power sensor are indicated respectively,Table
Show average value of the wind speed of the air velocity transducer acquisition of Wind turbines in one minute,Indicate the rated wind speed of Wind turbines；
4) blade pitch corner characteristics x4：
Wherein, β is propeller pitch angle theoretical value when wind power generation unit blade is run,
β ' is the propeller pitch angle of wind power generation unit blade actual motion, is read by SCADA system；
5) yaw corner characteristics x5,
Wherein, θ and θ0Yaw angle of the Wind turbines under actual condition and declared working condition, θ are indicated respectively0=0 °, θ=γ-α, α
For the wind direction angle signal of wind direction angle transducer acquisition, γ is the cabin deflection of cabin direction sensor acquisition.
7. system according to claim 6, which is characterized in that the blade icing diagnosis neural network is three-decker
BP neural network, wherein input layer nodal point number are 5, and output layer nodal point number is 4, and a hidden layer node number m is determined with following formula：
Wherein, mxIt is input layer nodal point number；myIt is output layer nodal point number；A is constant, and value range is [1,10]；
Using the temperature profile, Humidity Features, power-wind speed feature, blade pitch corner characteristics, yaw corner characteristics successively as defeated
Enter node layer, with the icing degree of entire blade, tip region icing degree, middle part of blade region icing degree and blade root
Portion region icing degree absolute value is successively as output node layer；
The icing degree absolute value in each region of blade is calculated using following formula and is obtained：
Wherein, δ indicates that the practical ice covering thickness of blade, L are the length of blade, and N indicates intermediate variable, y2、y3、y4It indicates respectively
Tip region icing degree absolute value, middle part of blade region icing degree absolute value and root of blade region icing degree are absolute
Value.
8. system according to claim 7, which is characterized in that further include the display module being connected with real-time monitoring modular,
The icing characteristic versus time curve of time-domain curve and extraction for showing each status signal acquired.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810265635.2A CN108343566B (en) | 2018-03-28 | 2018-03-28 | A kind of blade icing On-line Fault monitoring method and system based on running of wind generating set state |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810265635.2A CN108343566B (en) | 2018-03-28 | 2018-03-28 | A kind of blade icing On-line Fault monitoring method and system based on running of wind generating set state |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108343566A true CN108343566A (en) | 2018-07-31 |
| CN108343566B CN108343566B (en) | 2019-07-09 |
Family
ID=62957694
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810265635.2A Expired - Fee Related CN108343566B (en) | 2018-03-28 | 2018-03-28 | A kind of blade icing On-line Fault monitoring method and system based on running of wind generating set state |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108343566B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109209790A (en) * | 2018-10-09 | 2019-01-15 | 浙江运达风电股份有限公司 | A kind of wind electricity blade icing estimation method based on data modeling |
| CN109578226A (en) * | 2018-11-08 | 2019-04-05 | 明阳智慧能源集团股份公司 | Wind turbines icing condition detection method based on icing detector and field group's intercommunication |
| CN109751204A (en) * | 2019-02-18 | 2019-05-14 | 中国空气动力研究与发展中心低速空气动力研究所 | A kind of wind energy conversion system icing method for numerical simulation |
| CN111767505A (en) * | 2020-05-19 | 2020-10-13 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Method for monitoring icing rate of tangent tower wire of power transmission line |
| EP3730784A1 (en) * | 2019-04-25 | 2020-10-28 | Siemens Gamesa Renewable Energy A/S | A method for a computer-implemented analysis of operation data of one or more wind turbines |
| CN113027695A (en) * | 2019-12-24 | 2021-06-25 | 新疆金风科技股份有限公司 | Detection method and device for pitch angle abnormity of wind generating set |
| CN113255239A (en) * | 2021-07-12 | 2021-08-13 | 南京信息工程大学 | Deep learning method for predicting icing thickness of ultrahigh voltage transmission line tower |
| CN113847216A (en) * | 2021-10-14 | 2021-12-28 | 远景智能国际私人投资有限公司 | Method, device and equipment for predicting state of fan blade and storage medium |
| CN115711206A (en) * | 2022-11-14 | 2023-02-24 | 长春大学 | Aerogenerator blade icing state monitoring system based on clustering weight |
| CN116756644A (en) * | 2023-08-16 | 2023-09-15 | 华北电力大学 | Early warning method for icing fault of anemoclinograph of wind turbine generator |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102938021A (en) * | 2012-11-02 | 2013-02-20 | 云南大学 | Quantitative estimation and prediction method for icing load of power transmission line |
| CN103914740A (en) * | 2014-03-28 | 2014-07-09 | 云南电力试验研究院（集团）有限公司电力研究院 | Method for icing prediction and automatic correction of power transmission line based on data driving |
| CN104454386A (en) * | 2014-11-24 | 2015-03-25 | 北京金风科创风电设备有限公司 | Icing control method and device for wind generating set |
| CN105298761A (en) * | 2015-11-06 | 2016-02-03 | 周志宏 | Freezing early warning and control method for wind generating unit and device of freezing early warning and control method |
| CN105464912A (en) * | 2016-01-27 | 2016-04-06 | 国电联合动力技术有限公司 | Method and device for detecting freezing of wind generating set blades |
| CN105508152A (en) * | 2015-12-31 | 2016-04-20 | 北京金风科创风电设备有限公司 | Blade icing model construction method and icing state monitoring method and device |
| US20170058871A1 (en) * | 2015-08-27 | 2017-03-02 | General Electric Company | System and method for mitigating ice throw from a wind turbine rotor blade |
| CN107725286A (en) * | 2017-11-02 | 2018-02-23 | 北京三力新能科技有限公司 | A kind of wind power generating set based on inverse time lag control, which freezes, detects control method |
| CN107784395A (en) * | 2017-10-27 | 2018-03-09 | 云南电网有限责任公司电力科学研究院 | A kind of electric power line ice-covering thickness Forecasting Methodology based on LSTM artificial neural networks |
| US10062272B2 (en) * | 2014-07-23 | 2018-08-28 | Nordex Energy Gmbh | Method for testing a rotor blade ice detection system as well as rotor blade ice detection system and wind turbine for carrying out the method |
-
2018
- 2018-03-28 CN CN201810265635.2A patent/CN108343566B/en not_active Expired - Fee Related
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102938021A (en) * | 2012-11-02 | 2013-02-20 | 云南大学 | Quantitative estimation and prediction method for icing load of power transmission line |
| CN103914740A (en) * | 2014-03-28 | 2014-07-09 | 云南电力试验研究院（集团）有限公司电力研究院 | Method for icing prediction and automatic correction of power transmission line based on data driving |
| US10062272B2 (en) * | 2014-07-23 | 2018-08-28 | Nordex Energy Gmbh | Method for testing a rotor blade ice detection system as well as rotor blade ice detection system and wind turbine for carrying out the method |
| CN104454386A (en) * | 2014-11-24 | 2015-03-25 | 北京金风科创风电设备有限公司 | Icing control method and device for wind generating set |
| US20170058871A1 (en) * | 2015-08-27 | 2017-03-02 | General Electric Company | System and method for mitigating ice throw from a wind turbine rotor blade |
| CN105298761A (en) * | 2015-11-06 | 2016-02-03 | 周志宏 | Freezing early warning and control method for wind generating unit and device of freezing early warning and control method |
| CN105508152A (en) * | 2015-12-31 | 2016-04-20 | 北京金风科创风电设备有限公司 | Blade icing model construction method and icing state monitoring method and device |
| CN105464912A (en) * | 2016-01-27 | 2016-04-06 | 国电联合动力技术有限公司 | Method and device for detecting freezing of wind generating set blades |
| CN107784395A (en) * | 2017-10-27 | 2018-03-09 | 云南电网有限责任公司电力科学研究院 | A kind of electric power line ice-covering thickness Forecasting Methodology based on LSTM artificial neural networks |
| CN107725286A (en) * | 2017-11-02 | 2018-02-23 | 北京三力新能科技有限公司 | A kind of wind power generating set based on inverse time lag control, which freezes, detects control method |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109209790A (en) * | 2018-10-09 | 2019-01-15 | 浙江运达风电股份有限公司 | A kind of wind electricity blade icing estimation method based on data modeling |
| CN109578226A (en) * | 2018-11-08 | 2019-04-05 | 明阳智慧能源集团股份公司 | Wind turbines icing condition detection method based on icing detector and field group's intercommunication |
| CN109751204A (en) * | 2019-02-18 | 2019-05-14 | 中国空气动力研究与发展中心低速空气动力研究所 | A kind of wind energy conversion system icing method for numerical simulation |
| EP3730784A1 (en) * | 2019-04-25 | 2020-10-28 | Siemens Gamesa Renewable Energy A/S | A method for a computer-implemented analysis of operation data of one or more wind turbines |
| CN113027695A (en) * | 2019-12-24 | 2021-06-25 | 新疆金风科技股份有限公司 | Detection method and device for pitch angle abnormity of wind generating set |
| CN111767505B (en) * | 2020-05-19 | 2022-12-09 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Method for monitoring icing rate of tangent tower wire of power transmission line |
| CN111767505A (en) * | 2020-05-19 | 2020-10-13 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Method for monitoring icing rate of tangent tower wire of power transmission line |
| CN113255239A (en) * | 2021-07-12 | 2021-08-13 | 南京信息工程大学 | Deep learning method for predicting icing thickness of ultrahigh voltage transmission line tower |
| CN113255239B (en) * | 2021-07-12 | 2021-09-28 | 南京信息工程大学 | Deep learning method for predicting icing thickness of ultrahigh voltage transmission line tower |
| CN113847216A (en) * | 2021-10-14 | 2021-12-28 | 远景智能国际私人投资有限公司 | Method, device and equipment for predicting state of fan blade and storage medium |
| WO2023063887A3 (en) * | 2021-10-14 | 2023-07-20 | Envision Digital International Pte. Ltd. | Method and apparatus for predicting state of wind turbine blade, and device and storage medium therefor |
| CN113847216B (en) * | 2021-10-14 | 2023-09-26 | 远景智能国际私人投资有限公司 | Fan blade state prediction method, device, equipment and storage medium |
| CN115711206A (en) * | 2022-11-14 | 2023-02-24 | 长春大学 | Aerogenerator blade icing state monitoring system based on clustering weight |
| CN115711206B (en) * | 2022-11-14 | 2023-08-11 | 长春大学 | Wind driven generator blade icing state monitoring system based on clustering weight |
| CN116756644A (en) * | 2023-08-16 | 2023-09-15 | 华北电力大学 | Early warning method for icing fault of anemoclinograph of wind turbine generator |
| CN116756644B (en) * | 2023-08-16 | 2023-11-21 | 华北电力大学 | Early warning method for icing fault of anemoclinograph of wind turbine generator |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108343566B (en) | 2019-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108343566B (en) | A kind of blade icing On-line Fault monitoring method and system based on running of wind generating set state | |
| Gao et al. | A field study of ice accretion and its effects on the power production of utility-scale wind turbines | |
| CN105464912B (en) | A kind of method and apparatus of wind generator set blade icing detection | |
| CN101603500B (en) | Method and apparatus for measuring air flow condition at a wind turbine blade | |
| DK1906192T3 (en) | Apparatus for evaluating sensors and / or for controlling the operation of an apparatus which includes a sensor | |
| EP3139038B1 (en) | A method for estimating the surface condition of a rotating blade | |
| US7922449B2 (en) | Passive deicing for wind turbine blades | |
| CN102817780B (en) | Device and method for controlling freezing of wind generating set | |
| CN101825893A (en) | Centralized and remote control monitoring, and fault diagnosis system of wind turbine | |
| CN201820121U (en) | Centralized and remote monitoring and fault diagnosis system of wind turbine generator set | |
| CN102444544B (en) | The system and method for air distribution in wind turbine | |
| US20100135790A1 (en) | Wind turbine blade with foreign matter detection devices | |
| WO2012125842A2 (en) | Load shape control of wind turbines | |
| CN102182633B (en) | Independent paddle change control method and device of large-scale wind generator system | |
| CN102072965A (en) | Wind sensor system using blade signals | |
| CN102213143A (en) | Control device for a wind power plant and corresponding wind power plant | |
| US9494624B2 (en) | Method of monitoring a wind turbine | |
| CN107725286B (en) | A kind of wind power generating set icing detection control method based on inverse time lag control | |
| CN105041571B (en) | The intelligence control system and its control method of prediction of wind speed wind direction | |
| CN105138845B (en) | The method for obtaining wind-driven generator air speed value | |
| CN115059587B (en) | Internet of things real-time monitoring and early warning system based on 5G communication | |
| CN110263477B (en) | Blade tip speed ratio acquisition method of wind generating set | |
| Lu et al. | A status of study on icing of wind turbine blades | |
| WO2023165159A1 (en) | Online indirect measurement system and method for pitching and yawing moments of wind energy or tidal current energy generator set | |
| CN107656091B (en) | A kind of wind measurement method and its system based on air-blower control sensor |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190709 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |