CN112431726A - Method for monitoring bearing state of gearbox of wind turbine generator - Google Patents

Method for monitoring bearing state of gearbox of wind turbine generator Download PDF

Info

Publication number
CN112431726A
CN112431726A CN202011316083.7A CN202011316083A CN112431726A CN 112431726 A CN112431726 A CN 112431726A CN 202011316083 A CN202011316083 A CN 202011316083A CN 112431726 A CN112431726 A CN 112431726A
Authority
CN
China
Prior art keywords
gearbox
sample
bearing
temperature
data
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.)
Pending
Application number
CN202011316083.7A
Other languages
Chinese (zh)
Inventor
王忠杰
张一飞
韩斌
胡照宇
甘勇
孙世辉
赵勇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan Clean Energy Branch Of Huaneng International Power Co ltd
Xian Thermal Power Research Institute Co Ltd
Huaneng Power International Inc
Original Assignee
Hunan Clean Energy Branch Of Huaneng International Power Co ltd
Xian Thermal Power Research Institute Co Ltd
Huaneng Power International Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hunan Clean Energy Branch Of Huaneng International Power Co ltd, Xian Thermal Power Research Institute Co Ltd, Huaneng Power International Inc filed Critical Hunan Clean Energy Branch Of Huaneng International Power Co ltd
Priority to CN202011316083.7A priority Critical patent/CN112431726A/en
Publication of CN112431726A publication Critical patent/CN112431726A/en
Pending legal-status Critical Current

Links

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
    • F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics

Abstract

The invention discloses a method for monitoring the condition of a bearing of a gearbox of a wind turbine generator, which comprises the steps of selecting variables by adopting a Relieff characteristic selection algorithm, building an improved noise reduction self-coding network, establishing a relation model between the temperature of the bearing of the gearbox and the influence variables of the bearing, reconstructing a modeling variable in a monitoring stage by using the model, and predicting the temperature of the bearing of the gearbox. Calculating according to a modeling variable reconstruction error of normal operation data of the wind turbine generator to obtain an exponential weighted moving average control chart threshold; if the EWMA control chart statistic of the monitored unit is smaller than the threshold value, the unit operates normally; if the temperature exceeds the threshold value, an alarm for the temperature abnormity of the bearing of the gearbox is sent out. The method is used for analyzing the temperature data of the bearing of the gearbox, and the purposes of artificial intelligent monitoring and fault early warning of the temperature of the bearing of the gearbox of the wind turbine generator are efficiently and accurately achieved. Example analysis verifies the utility and versatility of the invention.

Description

Method for monitoring bearing state of gearbox of wind turbine generator
Technical Field
The invention belongs to the field of wind turbine generator gearbox state monitoring, and particularly relates to a wind turbine generator gearbox bearing temperature state monitoring method.
Background
In recent years, the air environment of partial regions in China is gradually worsened, severe haze weather is frequent, the adjustment of the traditional energy structure mainly based on fossil fuels such as coal and petroleum is urgently needed, and the development of renewable energy sources scientifically and efficiently is urgent. Wind power generation is an important component of renewable energy, the development of the wind power generation is rapid in China, and the accumulated installed capacity and the newly-increased installed capacity are in the top of the world.
The wind turbine generator has severe operating conditions, such as large external temperature difference change, random wind speed change and the like. The fault rate of the wind turbine generator is high due to uncertain external factors, so that the later operation and maintenance cost of the wind power plant is high.
The gearbox is one of the important parts of the wind turbine. The wind turbine generator gearbox has the characteristics of speed change and load change during operation. With the change of the wind speed, the rotating speed and the load of each stage of the gearbox change at any time, which brings great challenges for the application of the traditional state monitoring method to the gearbox of the wind turbine generator.
The traditional gearbox bearing fault diagnosis technology, such as vibration analysis, oil analysis and the like, achieves certain results. The wind speed changes randomly, so that the rotating speed and the load of each stage of bearing of the gearbox of the wind turbine generator set change time instead of the stable working condition that the rotating speed is not changed. The current vibration analysis technology is low in fault diagnosis accuracy and high in false alarm and false alarm rate under the time-varying complex working condition of variable rotating speed and variable load of a gearbox bearing. The gear box oil analysis technology is used for collecting a gear box oil sample during the shutdown of the wind turbine generator, analyzing the water content, the number of metal particles and the diameter of lubricating oil in a laboratory to diagnose the state of a gear box bearing, but the oil analysis can only be used for off-line diagnosis, and the on-line real-time monitoring and diagnosis of the gear box bearing cannot be realized. And a multilayer forward neural network is adopted to model and monitor the temperature of the bearing of the gearbox of the wind turbine generator, but the forward neural network has a simple structure and low modeling precision, so that the abnormal change of the temperature of the bearing of the gearbox is difficult to be timely and accurately pre-warned.
Disclosure of Invention
In order to overcome the problems in the prior art, the invention aims to provide a wind turbine generator gearbox bearing state monitoring method, which is a wind turbine generator gearbox bearing temperature state monitoring method based on an improved noise reduction self-coding network. For the temperature data of the bearing of the gearbox, the noise reduction self-coding has the characteristics of quickly finishing the characteristic extraction, dimension reduction and modeling. The model is established by combining the noise reduction self-coding with the Particle Swarm (PSO), and the self-adaptive optimization of the structural parameters of the noise reduction self-coding is carried out by utilizing the particle swarm, so that the characteristic learning capability of the model can be improved, and more accurate signal reconstruction can be obtained. Meanwhile, the temperature reconstruction error of the gearbox bearing is monitored by using an exponential weighted moving average control chart, and the control chart method can accurately detect the abnormal change of the gearbox bearing temperature and timely give out early warning.
In order to achieve the purpose, the invention adopts the following technical scheme:
a method for monitoring the condition of a bearing of a gearbox of a wind turbine generator specifically comprises the following steps:
step 1, selecting a gear box bearing temperature modeling variable by adopting a Relieff characteristic selection algorithm
The method comprises the following steps that the temperature of a bearing of a gearbox of the wind turbine generator is influenced by a plurality of parameter variables of the wind turbine generator, in order to determine influence factors of the temperature of the bearing of the gearbox, input modeling variables of a temperature model of the bearing of the gearbox are determined, and modeling variables are selected from hundreds of operating parameters of the wind turbine generator by adopting a Relieff characteristic selection algorithm;
the input of the Relieff feature selection algorithm is a sample set D, the sample set D is divided into two types of normal historical operating data of the bearing temperature of the gearbox and overtemperature historical operating data of the bearing temperature of the gearbox, and the output is a feature contribution weight vector W corresponding to an original variable:
Figure BDA0002791458650000031
wherein:
m is the number of samples;
n is the number of original wind turbine generator variables;
the ReliefF feature selection algorithm is as follows:
(1) initializing a feature contribution weight vector W to be zero, and taking M samples from the sample set, wherein M is less than M;
(2) cycle i ═ 1:
(3) randomly selecting a sample R in a sample set Di
(4) Find and RiNearest neighbor sample of homogeneous H
(5) Find and RiNearest neighbor samples P of different classes
(6) Cycle a 1: N
(7) Updating the feature contribution weight corresponding to each original variable
W[A]＝W[A]-diff(A,Ri,H)/m+diff(A,Ri,P)/m
(8) End of cycle
(9) End of cycle
Wherein:
w [ A ] -the feature contribution weight corresponding to the original variable A;
diff(A,Rih) -sample RiThe distance from the sample H in the characteristic direction of the variable A;
diff(A,Rip) -sample RiThe distance from the sample P in the characteristic direction of the original variable A;
after the feature contribution weight vector W corresponding to each original variable is obtained by calculation by adopting the algorithm, for a certain original variable A, if the feature contribution weight is:
W[A]＞δ
selecting an original variable A as a modeling variable of the bearing temperature of the gearbox;
wherein: delta-set feature weight threshold;
step 2, building an improved noise reduction self-coding network model and training the model
Step 2.1, constructing a training sample and a verification sample of a gearbox bearing temperature model
In step 1, the number of modeling variables selected by a Relieff feature selection algorithm is L, and the L modeling variables and the bearing temperature of a gearbox are selected from each historical data of a period when the bearing temperature of the gearbox of the wind turbine generator is normal to form a sample:
S(t)＝[x1(t) x2(t) … xL(t),Y(t)]
wherein:
s (t) -gearbox bearing temperature sample at time t;
x1(t) x2(t) … xL(t) -values at time t of the L variables selected using the Relieff feature selection algorithm;
y (t) -the value of the gearbox bearing temperature at time t;
all samples consisting of historical data of the bearing temperature of the gearbox in a certain normal period are according to the following ratio of 3: 1, dividing the ratio into a training sample set and a verification sample set;
step 2.2, an improved noise reduction self-coding network model is set up, and the model comprises:
an encoder:
Figure BDA0002791458650000051
a decoder: z ═ f (W)2·h+b2)
Wherein:
h-hidden layer feature;
Figure BDA0002791458650000052
-performing a random noise addition process on the input sample data x to obtain a noisy input;
z — the reconstructed output of input sample data x;
f () -neural network excitation function, using nonlinear function sigmoid function;
W1-encoding the weight matrix;
b1-encoding the bias vector;
W2-decoding the weight matrix;
b2-decoding the bias vector;
optimizing and selecting the number of hidden layer nodes of the noise reduction self-coding network and the random zero setting proportion of input data by using a swarm intelligent optimization algorithm-particle swarm optimization algorithm, and optimizing the structural parameters to obtain the minimum loss function;
2.3, training the improved self-coding network model built in the step 2.2 by adopting a training sample set of a bearing temperature model of the gearbox, wherein the training algorithm adopts an error back propagation algorithm;
step 3, verifying the improved noise reduction self-coding network model, and determining a temperature alarm threshold of a bearing of the gearbox according to a reconstruction error of verification data;
after the training of the improved noise reduction self-coding network model is finished, sending the verification sample into the improved noise reduction self-coding network model; verifying that the number of samples is NVGear box bearing temperature model pair NVThe temperature prediction value sequence of each verification sample is
Figure BDA0002791458650000053
Verify the actual gearbox bearing temperature of the sample as
Figure BDA0002791458650000054
The model prediction residual of the i-th verification sample is
Figure BDA0002791458650000061
Using mean absolute value as a percentage of error epsilonMAPEThe modeling accuracy of the improved noise reduction self-coding network model is measured as follows:
Figure BDA0002791458650000062
wherein:
NV-verifying the number of samples;
yi-actual gearbox bearing temperature value of the ith validation sample;
Figure BDA0002791458650000063
-gearbox bearing temperature model predicted values for the ith validation sample;
the reconstruction error of the verification data is the Euclidean distance between the noise reduction self-coding network output reconstruction value of the verification data and the true value:
Figure BDA0002791458650000064
wherein:
RE-reconstruction error;
zi-noise reducing the reconstructed output value from the encoded network;
xi-denoising the self-encoded network raw input values;
l is the number of the temperature modeling variables of the bearing of the gearbox;
calculating an exponentially weighted moving average control map statistic l based on the validation datakComprises the following steps:
lk＝λREk+(1-λ)lk-1
wherein:
k is a time series;
λ -reconstruction error RE of kth validation samplekWeighting the current exponent to weight the moving average control map statistic, λ ∈ (0, 1)]Taking lambda as 0.2;
according to the verification sample set, calculating an upper limit threshold and a lower limit threshold of an exponential weighted moving average control chart as follows:
Figure BDA0002791458650000071
wherein:
μRE-mean of all validation sample reconstruction errors;
σRE-standard deviation of all validation sample reconstruction errors;
k is a set coefficient, and the value is 3;
UCL-Upper control map threshold;
CL-mean of validation set RE;
LCL-lower control map threshold;
step 4, after the steps are completed, switching to a monitoring stage; when a bearing of the gearbox fails or is abnormal, the relationship between the temperature of the bearing of the gearbox and the influence factors of the bearing of the gearbox changes, and the bearing deviates from the improved noise reduction self-coding network model, so that the reconstruction error of the monitoring data model changes; collecting the operation data of the monitored unit in real time, forming a monitoring sample sequence, sending the monitoring sample sequence into an improved noise reduction self-coding network model, and calculating the exponential weighted moving average control chart statistic l of the monitoring sample sequenceci：
Figure BDA0002791458650000072
lci＝λREci+(1-λ)lci-1
Wherein:
lci-monitoring control chart statistics of the sample sequence;
REc-monitoring the data for reconstruction errors;
zci-a reconstructed output value of the monitoring data;
xci-raw monitoring data input values;
exponentially weighted moving average control map statistic l when monitoring a sequence of samplesciWhen satisfying the following formula, the gearbox bearing temperature is unusual, sends the unusual warning of gearbox bearing temperature:
Figure BDA0002791458650000081
compared with the prior art, the invention has the following advantages:
1) the noise reduction self-coding network is applied to monitoring the temperature state of the bearing of the gearbox of the wind turbine generator;
2) the noise reduction self-coding and Particle Swarm (PSO) are combined to establish a model, and the particle swarm is used for self-adaptive optimization of the noise reduction self-coding structure parameters, so that data can be effectively trained, the model characteristic learning capability is improved, and more accurate signal reconstruction is obtained.
3) The hidden danger of the bearing of the gearbox can be found as early as possible by adopting an Exponential Weighted Moving Average (EWMA) control chart according to the reconstruction error of the noise reduction self-coding network model, and preventive measures are taken, so that the operation reliability of the wind turbine generator is improved, and the maintenance cost is reduced.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Fig. 2 is a block diagram of noise reduction self-encoding.
FIG. 3a shows predicted and actual bearing temperature values for the 7 month gearbox of the PSO-SDAE model.
FIG. 3b is the residual between predicted and actual values of the PSO-SDAE model 7 month gearbox bearing temperature.
FIG. 4a shows the predicted value and the actual value of the PSO-SDAE model 8 month bearing temperature over-temperature.
FIG. 4b is the residual between the predicted value and the actual value of the PSO-SDAE model 8 month bearing temperature over-temperature.
FIG. 5 is an exponentially weighted moving average EWMA control map for month 8 of PSO-SDAE.
Detailed Description
The invention is described in detail below with reference to the figures and examples.
With a gearbox of a 2MW unit in a certain wind power plant as a research object, selecting operation data recorded by an SCADA system of the unit in a level of 1 minute, as shown in FIG. 1, the method for monitoring the temperature state of the bearing of the gearbox of the wind power unit comprises the following steps:
step 1, setting a feature weight threshold value delta to be 0.012, and selecting 12 variables meeting the requirements through a Relieff algorithm, which is specifically shown in the following table 1.
Table 1: selecting variables for gearbox bearing temperature modeling
Figure BDA0002791458650000091
And 2, constructing an improved noise reduction self-coding network model, wherein the noise reduction self-coding network structure is shown in figure 2, and training the model by using the operation data of the unit in 5 to 6 months as a training data set to obtain model parameters.
And 3, verifying the improved noise reduction self-coding network model, and determining the alarm threshold of the bearing of the gearbox according to the reconstruction error of the verification data.
And (3) sending the gear box bearing temperature normal verification sample of 7 months into an improved noise reduction self-coding network, wherein the obtained predicted value of the gear box bearing temperature is shown in figure 3a, and the temperature prediction residual error is shown in figure 3 b. As can be seen from FIGS. 3a and 3b, the prediction accuracy of the improved noise reduction self-coding network on the temperature of the bearing of the gearbox is very high, and the absolute value of the prediction residual is mostly within 5 degrees. Calculated epsilon of validation sampleMAPE＝2.14％。
And 4, in the monitoring stage, transmitting the data of 8 months with abnormal bearing temperature of the gearbox into a noise reduction self-coding network model. The predicted values of the temperature of the model calculation monitoring samples are shown in fig. 4a, and the residual prediction of the temperature is shown in fig. 4 b. In fig. 4a, due to the fault of the bearing of the gearbox, the temperature of the gearbox is abnormally increased, the model predicted value and the actual value of the monitoring sample gradually deviate obviously after the 957 th sample, and the improved noise reduction self-coding network model temperature prediction residual error in fig. 4b is continuously increased. In fig. 5, at 957 th monitoring sample, the EWMA control map statistic exceeds the upper limit of the gearbox bearing temperature alarm threshold UCL, and the system issues a gearbox bearing temperature anomaly alarm. The effectiveness of the method is verified.

Claims (1)

1. A wind turbine generator gearbox bearing state monitoring method is characterized by comprising the following steps: the method specifically comprises the following steps:
step 1, selecting a gear box bearing temperature modeling variable by adopting a Relieff characteristic selection algorithm
The method comprises the following steps that the temperature of a bearing of a gearbox of the wind turbine generator is influenced by a plurality of parameter variables of the wind turbine generator, in order to determine influence factors of the temperature of the bearing of the gearbox, input modeling variables of a temperature model of the bearing of the gearbox are determined, and modeling variables are selected from hundreds of operating parameters of the wind turbine generator by adopting a Relieff characteristic selection algorithm;
the input of the Relieff feature selection algorithm is a sample set D, the sample set D is divided into two types of normal historical operating data of the bearing temperature of the gearbox and overtemperature historical operating data of the bearing temperature of the gearbox, and the output is a feature contribution weight vector W corresponding to an original variable:
Figure FDA0002791458640000011
wherein:
m is the number of samples;
n is the number of original wind turbine generator variables;
the ReliefF feature selection algorithm is as follows:
(1) initializing a feature contribution weight vector W to zero, and taking m samples from the sample set,
m＜M；
(2) cycle i ═ 1:
(3) randomly selecting a sample R in a sample set Di
(4) Find and RiNearest neighbor sample of homogeneous H
(5) Find and RiNearest neighbor samples P of different classes
(6) Cycle a 1: N
(7) Updating the feature contribution weight corresponding to each original variable
W[A]＝W[A]-diff(A,Ri,H)/m+diff(A,Ri,P)/m
(8) End of cycle
(9) End of cycle
Wherein:
w [ A ] -the feature contribution weight corresponding to the original variable A;
diff(A,Rih) -sample RiThe distance from the sample H in the characteristic direction of the variable A;
diff(A,Rip) -sample RiThe distance from the sample P in the characteristic direction of the original variable A;
after the feature contribution weight vector W corresponding to each original variable is obtained by calculation by adopting the algorithm, for a certain original variable A, if the feature contribution weight is:
W[A]＞δ
selecting an original variable A as a modeling variable of the bearing temperature of the gearbox;
wherein: delta-set feature weight threshold;
step 2, building an improved noise reduction self-coding network model and training the model
Step 2.1, constructing a training sample and a verification sample of a gearbox bearing temperature model
In step 1, the number of modeling variables selected by a Relieff feature selection algorithm is L, and the L modeling variables and the bearing temperature of a gearbox are selected from each historical data of a period when the bearing temperature of the gearbox of the wind turbine generator is normal to form a sample:
S(t)＝[x1(t) x2(t)…xL(t),Y(t)]
wherein:
s (t) -gearbox bearing temperature sample at time t;
x1(t) x2(t)…xL(t) -values at time t of the L variables selected using the Relieff feature selection algorithm;
y (t) -the value of the gearbox bearing temperature at time t;
all samples consisting of historical data of the bearing temperature of the gearbox in a certain normal period are according to the following ratio of 3: 1, dividing the ratio into a training sample set and a verification sample set;
step 2.2, an improved noise reduction self-coding network model is set up, and the model comprises:
an encoder:
Figure FDA0002791458640000031
a decoder: z ═ f (W)2·h+b2)
Wherein:
h-hidden layer feature;
Figure FDA0002791458640000032
-performing a random noise addition process on the input sample data x to obtain a noisy input;
z — the reconstructed output of input sample data x;
f () -neural network excitation function, using nonlinear function sigmoid function;
W1-encoding the weight matrix;
b1-encoding the bias vector;
W2-decoding the weight matrix;
b2-decoding the bias vector;
optimizing and selecting the number of hidden layer nodes of the noise reduction self-coding network and the random zero setting proportion of input data by using a swarm intelligent optimization algorithm-particle swarm optimization algorithm, and optimizing the structural parameters to obtain the minimum loss function;
2.3, training the improved self-coding network model built in the step 2.2 by adopting a training sample set of a bearing temperature model of the gearbox, wherein the training algorithm adopts an error back propagation algorithm;
step 3, verifying the improved noise reduction self-coding network model, and determining a temperature alarm threshold of a bearing of the gearbox according to a reconstruction error of verification data;
after the training of the improved noise reduction self-coding network model is finished, the verification sample is sent to the improved noise reduction self-coding network model(ii) a Verifying that the number of samples is NVGear box bearing temperature model pair NVThe temperature prediction value sequence of each verification sample is
Figure FDA0002791458640000041
Verify the actual gearbox bearing temperature of the sample as
Figure FDA0002791458640000042
The model prediction residual of the i-th verification sample is
Figure FDA0002791458640000043
Using mean absolute value as a percentage of error epsilonMAPEThe modeling accuracy of the improved noise reduction self-coding network model is measured as follows:
Figure FDA0002791458640000044
wherein:
NV-verifying the number of samples;
yi-actual gearbox bearing temperature value of the ith validation sample;
Figure FDA0002791458640000045
-gearbox bearing temperature model predicted values for the ith validation sample;
the reconstruction error of the verification data is the Euclidean distance between the noise reduction self-coding network output reconstruction value of the verification data and the true value:
Figure FDA0002791458640000046
wherein:
RE-reconstruction error;
zi-noise reducing the reconstructed output value from the encoded network;
xi-denoising the self-encoded network raw input values;
l is the number of the temperature modeling variables of the bearing of the gearbox;
calculating an exponentially weighted moving average control map statistic l based on the validation datakComprises the following steps:
lk＝λREk+(1-λ)lk-1
wherein:
k is a time series;
λ -reconstruction error RE of kth validation samplekWeighting the current exponent to weight the moving average control map statistic, λ ∈ (0, 1)]Taking lambda as 0.2;
according to the verification sample set, calculating an upper limit threshold and a lower limit threshold of an exponential weighted moving average control chart as follows:
Figure FDA0002791458640000051
wherein:
μRE-mean of all validation sample reconstruction errors;
σRE-standard deviation of all validation sample reconstruction errors;
k is a set coefficient, and the value is 3;
UCL-Upper control map threshold;
CL-mean of validation set RE;
LCL-lower control map threshold;
step 4, after the steps are completed, switching to a monitoring stage; when a bearing of the gearbox fails or is abnormal, the relationship between the temperature of the bearing of the gearbox and the influence factors of the bearing of the gearbox changes, and the bearing deviates from the improved noise reduction self-coding network model, so that the reconstruction error of the monitoring data model changes; collecting the operation data of the monitored unit in real time, forming a monitoring sample sequence, sending the monitoring sample sequence into an improved noise reduction self-coding network model, and calculating the exponential weighted moving average control chart statistic l of the monitoring sample sequenceci：
Figure FDA0002791458640000061
lci＝λREci+(1-λ)lci-1
Wherein:
lci-monitoring control chart statistics of the sample sequence;
REc-monitoring the data for reconstruction errors;
zci-a reconstructed output value of the monitoring data;
xci-raw monitoring data input values;
exponentially weighted moving average control map statistic l when monitoring a sequence of samplesciWhen satisfying the following formula, the gearbox bearing temperature is unusual, sends the unusual warning of gearbox bearing temperature:
Figure FDA0002791458640000062
CN202011316083.7A 2020-11-22 2020-11-22 Method for monitoring bearing state of gearbox of wind turbine generator Pending CN112431726A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011316083.7A CN112431726A (en) 2020-11-22 2020-11-22 Method for monitoring bearing state of gearbox of wind turbine generator

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011316083.7A CN112431726A (en) 2020-11-22 2020-11-22 Method for monitoring bearing state of gearbox of wind turbine generator

Publications (1)

Publication Number Publication Date
CN112431726A true CN112431726A (en) 2021-03-02

Family

ID=74693383

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011316083.7A Pending CN112431726A (en) 2020-11-22 2020-11-22 Method for monitoring bearing state of gearbox of wind turbine generator

Country Status (1)

Country Link
CN (1) CN112431726A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113221453A (en) * 2021-04-30 2021-08-06 华风数据(深圳)有限公司 Fault monitoring and early warning method for output shaft of gearbox of wind turbine generator
CN113776818A (en) * 2021-08-10 2021-12-10 中国长江三峡集团有限公司 Offshore wind turbine generator system gear system fault diagnosis method based on Park transformation
CN114046228A (en) * 2021-10-26 2022-02-15 华能(浙江)能源开发有限公司清洁能源分公司 Wind turbine generator abnormity diagnosis method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070294768A1 (en) * 2006-01-31 2007-12-20 Deutsche Telekom Ag Method and system for detecting malicious behavioral patterns in a computer, using machine learning
AU2010249990A1 (en) * 2009-05-22 2011-11-03 Bio-Rad Laboratories, Inc. System and method for automatic quality control of clinical diagnostic processes
CN106682688A (en) * 2016-12-16 2017-05-17 华南理工大学 Pile-up noise reduction own coding network bearing fault diagnosis method based on particle swarm optimization
CN110044554A (en) * 2019-04-16 2019-07-23 重庆大学 A kind of online test method of the metal pressure container leakage based on acoustic emission signal

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070294768A1 (en) * 2006-01-31 2007-12-20 Deutsche Telekom Ag Method and system for detecting malicious behavioral patterns in a computer, using machine learning
AU2010249990A1 (en) * 2009-05-22 2011-11-03 Bio-Rad Laboratories, Inc. System and method for automatic quality control of clinical diagnostic processes
CN106682688A (en) * 2016-12-16 2017-05-17 华南理工大学 Pile-up noise reduction own coding network bearing fault diagnosis method based on particle swarm optimization
CN110044554A (en) * 2019-04-16 2019-07-23 重庆大学 A kind of online test method of the metal pressure container leakage based on acoustic emission signal

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
刘旭斌等: "基于深度学习与SPRT的风电机组齿轮箱轴承状态监测", 动力工程学报 *
唐刚;秦朝任;: "基于声学信号瓶颈特征的滚动轴承故障诊断", 复旦学报(自然科学版) *
赵洪山;刘辉海;: "基于深度学习网络的风电机组主轴承故障检测", 太阳能学报 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113221453A (en) * 2021-04-30 2021-08-06 华风数据(深圳)有限公司 Fault monitoring and early warning method for output shaft of gearbox of wind turbine generator
CN113776818A (en) * 2021-08-10 2021-12-10 中国长江三峡集团有限公司 Offshore wind turbine generator system gear system fault diagnosis method based on Park transformation
CN113776818B (en) * 2021-08-10 2022-10-04 中国长江三峡集团有限公司 Offshore wind turbine generator system gear system fault diagnosis method based on Park transformation
CN114046228A (en) * 2021-10-26 2022-02-15 华能(浙江)能源开发有限公司清洁能源分公司 Wind turbine generator abnormity diagnosis method and system
CN114046228B (en) * 2021-10-26 2023-11-07 华能(浙江)能源开发有限公司清洁能源分公司 Wind turbine generator abnormality diagnosis method and system

Similar Documents

Publication Publication Date Title
CN110907066B (en) Wind turbine generator gearbox bearing temperature state monitoring method based on deep learning model
CN112431726A (en) Method for monitoring bearing state of gearbox of wind turbine generator
CN111537219B (en) Fan gearbox performance detection and health assessment method based on temperature parameters
CN111581597A (en) Wind turbine generator gearbox bearing temperature state monitoring method based on self-organizing kernel regression model
CN107291991B (en) Early defect early warning method for wind turbine generator based on dynamic network sign
CN111415070A (en) Wind turbine generator gearbox oil temperature over-temperature fault early warning method based on SCADA data
CN112733283A (en) Wind turbine generator component fault prediction method
Zhan et al. Health assessment methods for wind turbines based on power prediction and mahalanobis distance
CN112994101B (en) Neural network-based wind power plant generated power post-evaluation and monitoring method
Tao et al. Abnormal detection of wind turbine based on SCADA data mining
CN115222141A (en) Multivariable time sequence abnormity detection method, system, medium, equipment and terminal
Zhang et al. A novel fault diagnosis method for wind turbine based on adaptive multivariate time-series convolutional network using SCADA data
CN111814849A (en) DA-RNN-based wind turbine generator key component fault early warning method
Wang et al. An adaptive condition monitoring method of wind turbines based on multivariate state estimation technique and continual learning
Du et al. A SOM based Anomaly detection method for wind turbines health management through SCADA data
Dao et al. Operational condition monitoring of wind turbines using cointegration method
CN108709426B (en) Sintering machine air leakage fault online diagnosis method based on frequency spectrum characteristic bilateral detection method
Nuñez-Montoya et al. Development of a wind turbine digital-twin for failure prognosis: first results
Zhang Comparison of data-driven and model-based methodologies of wind turbine fault detection with SCADA data
CN113158535A (en) Wind turbine generator main bearing temperature prediction method based on deep learning
Li et al. Research on digital twin and collaborative cloud and edge computing applied in operations and maintenance in wind turbines of wind power farm
Ou et al. Fault Prediction Model of Wind Power Pitch System Based on BP Neural Network
Ristić et al. A short overview on Industry 4.0 in maintenance of hydropower plants
Encalada-Dávila et al. Wind turbine multi-fault detection based on SCADA data via an AutoEncoder
Guo et al. Early Fault Detection of Wind Turbine Gearbox Based on Adam-trained LSTM

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