CN112052426A - Temperature rise fault early warning method for fan variable pitch motor - Google Patents
Temperature rise fault early warning method for fan variable pitch motor Download PDFInfo
- Publication number
- CN112052426A CN112052426A CN202010906883.8A CN202010906883A CN112052426A CN 112052426 A CN112052426 A CN 112052426A CN 202010906883 A CN202010906883 A CN 202010906883A CN 112052426 A CN112052426 A CN 112052426A
- Authority
- CN
- China
- Prior art keywords
- early warning
- sigma
- temperature difference
- wind turbine
- turbine generator
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Algebra (AREA)
- Wind Motors (AREA)
Abstract
The invention discloses a temperature rise fault early warning method for a variable pitch motor of a fan, which comprises the following steps of: the method comprises the following steps: the method comprises the steps of firstly, obtaining historical measurement data of relevant measuring points of a wind turbine generator in a period of time, and obtaining effective historical measurement data after preprocessing; step two: calculating the temperature difference between every two three pitch motors of the wind turbine generator in the first step to obtain historical temperature difference data; step three: constructing an abnormality detection model based on TF-IDF by using a symbolization characterization technology; and the like. According to the early warning method, the early warning of the potential fault of the variable pitch motor is realized by mining mass historical measurement data of the wind turbine generator and analyzing the operation rule of the wind turbine generator. According to the invention, the wind turbine generator can set the fault early warning threshold value of the variable pitch motor according to the self operating characteristics, so that the fault early warning is more accurate, and the early warning accuracy is improved.
Description
Technical Field
The invention relates to the field of wind power generation fault early warning, in particular to a temperature rise fault early warning method for a variable pitch motor of a fan.
Background
In recent years, with the high-speed development of wind power generation technology, on one hand, the single-machine capacity of a wind turbine generator is larger and larger, and a wind power plant is constructed and connected to the grid on a large scale; on the other hand, the continuous increase of the running time of the fan leads the fatigue failure time and the maintenance cost of the unit to be obviously increased. The variable pitch system is an important transmission chain control device as a core system of the wind generating set, has a complex structure and high failure rate, and is an important part influencing the generating capacity and energy control of the fan and even the safety of the set. However, the heat is easily increased due to aging and decline of the performance of the variable pitch system, increase of the work load and the like, if the heat dissipation of the variable pitch system is abnormal, the variable pitch motor is further deteriorated, the motor is overheated, the motor is locked and the like, and the problems of asynchronous variable pitch control, reduction of wind energy utilization and even more serious safety are caused. Therefore, the potential heat dissipation fault of the variable pitch motor can be timely and accurately found, and the key problem of safe and economic operation of the wind power plant is solved.
At present, in the early warning scheme aiming at the heat dissipation fault of the variable pitch motor, a threshold value method is mostly adopted, and logic judgment is carried out by setting a single threshold value, so that the early warning purpose is achieved. The scheme does not consider the working condition of the equipment, is easy to give a false alarm, is not beneficial to the stable operation of the wind turbine generator, and does not utilize the historical data of the equipment per se and induce the operation characteristics, so that the potential fault is difficult to find. In order to save maintenance cost and improve the operation efficiency of a unit, a fault early warning method for a variable pitch motor considering both the self operation characteristics and the operation condition of equipment needs to be researched urgently, and powerful support is provided for realizing cost reduction and efficiency improvement of a wind power plant.
Disclosure of Invention
Aiming at the problems, the invention aims to provide a temperature rise fault early warning method for a variable pitch motor of a fan, which can solve the problem that early warning cannot be timely and accurately given in the prior art.
The invention provides a temperature rise fault early warning method for a variable pitch motor of a fan, which comprises the following steps of:
the method comprises the following steps: the method comprises the steps of firstly, obtaining historical measurement data of relevant measuring points of a wind turbine generator in a period of time, and obtaining effective historical measurement data after preprocessing;
step two: calculating the temperature difference between every two three pitch motors of the wind turbine generator in the first step to obtain historical temperature difference data;
step three: constructing an abnormality detection model based on TF-IDF by using a symbolization characterization technology;
step four: training the historical temperature difference data in the second step by using the anomaly detection model to obtain a characteristic vector and a weight of a temperature difference baseline;
step five: acquiring real-time measurement data of relevant measuring points of the wind turbine generator in the first step in a period of time, and preprocessing the real-time measurement data to obtain effective real-time measurement data;
step six: calculating the temperature difference between every two three pitch motors of the wind turbine generator set in the fifth step to obtain actual temperature difference data;
step seven: calculating the characteristic vector and the weight of the actual temperature difference data by using an anomaly detection model in the third step;
step eight: calculating the space distance between the characteristic vector of the actual temperature difference data in the step seven and the characteristic vector of the temperature difference base line in the step four;
step nine: and judging whether the calculation result in the step eight is early-warning according to a 3 sigma principle, and if so, obtaining an early-warning level.
The wind turbine generator is influenced by the external environment in the operation process, so that the obtained data is inaccurate, and the real and stable operation state of the wind turbine generator cannot be expressed. For example, the sensor is interfered by strong electromagnetic signals, which causes the data in a period of time to be larger or smaller; as another example, a sensor failure or communication interruption results in failure to transmit real data, resulting in failure to obtain real data. Therefore, in the information of each measuring point of the wind turbine generator, a rough normal interval needs to be obtained by combining the interval set when the equipment is shipped, and a judgment interval is built in advance.
And calculating the temperature difference between every two of the three pitch motors of the wind turbine generator in the second step and the sixth step to obtain historical temperature difference data and current temperature difference data, wherein the historical temperature difference data and the current temperature difference data are key indexes for judging the heat dissipation fault of the pitch motors. The temperature difference between every two pitch motors is a key index for judging the heat dissipation fault of the pitch motors, because the temperatures of the pitch motors of the three blades keep a certain relation (the relation is not necessarily a fixed value or specific distribution) under a stable working condition, if the relation changes, the heat dissipation is considered to be abnormal.
Further, the measurement points related to the wind turbine generator in the first step are specifically the temperatures of three pitch motors and the angles of three blades of the wind turbine generator.
Further, the historical measurement data of a period of time in the step one is specifically measurement data of 3 to 6 months in history.
Further, the preprocessing in the first step specifically includes screening stable working conditions for the obtained data, removing singular values and logic abnormal values, and reserving a data section capable of expressing the real and stable running state of the equipment.
Further, the step of training the temperature difference baseline characteristic vector and the weight for the historical measurement data by using the anomaly detection model in the fourth step comprises: s1: for each wind turbine generator, combining the three groups of temperature difference data calculated in the second step; s2: discretizing the combined temperature difference data at an interval of 0.5 to obtain a baseline characteristic in a specific range; s3: and obtaining a baseline characteristic vector weight TFIDF and a baseline characteristic vector IDF by using the anomaly detection model in the third step, wherein the calculation formula is as follows:
the formula I is as follows:
the formula II is as follows:
the formula III is as follows: TFIDF ═ TFi×IDFi
Wherein n isi，jRepresents the number of times a certain temperature value appears in the window, sigmaknk，jRepresents the total number of temperature values contained within the window, | D | represents the total number of windows, | { j: t is ti∈djAnd | represents the number of windows containing the temperature value plus one.
Further, in the eighth step, according to the results of the fourth step and the seventh step, the spatial distance between the feature vector of the actual temperature difference data in the seventh step and the feature vector of the temperature difference baseline in the fourth step is calculated, and the calculation formula is defined as:
the formula four is as follows:
wherein, IDF1 represents the feature vector of the temperature difference baseline obtained by training, and IDF2 represents the feature vector of the actual temperature difference data obtained by training.
Further, the 3 σ rule in the ninth step means that the normal distribution diagram is divided into 4 probability ranges by ± 1 σ, ± 2 σ, and ± 3 σ on both sides of the symmetry axis x ═ μ, in units of the standard deviation σ, and the probability ranges are respectively: the range of-2 sigma to mu plus 2 sigma, -3 sigma to mu plus 3 sigma, -4 sigma to mu plus 4 sigma and-4 sigma to mu plus 4 sigma; the last 3 ranges are considered as early warning ranges.
Further, in the ninth step, whether the calculation result in the eighth step is early-warning is judged according to a 3 sigma principle, and if the calculation result is early-warning, an early-warning level is output; whether the calculation result is early-warned or not is determined according to whether the calculation result is distributed in an early-warning range or not; if the early warning is carried out, outputting the early warning level, namely mapping the early warning range into different early warning levels, and outputting the corresponding early warning level according to the early warning range distributed by the calculation result, wherein the early warning level is as follows:
a: the probability of the distribution in-3 sigma is less than or equal to mu and less than or equal to +3 sigma is 0.9973, and the early warning level is light outside 2 sigma;
b: the probability of the distribution in-4 sigma is less than or equal to mu and less than or equal to +4 sigma is 0.9999, and the early warning level is moderate outside 3 sigma;
c: the early warning level is severe and is distributed in the range of-4 sigma to more than mu and more than +4 sigma.
Compared with the prior art, the invention has the following beneficial effects:
(1) according to the early warning method, the early warning of the potential fault of the variable pitch motor is realized by mining mass historical measurement data of the wind turbine generator and analyzing the operation rule of the wind turbine generator.
(2) According to the invention, the wind turbine generator can set the fault early warning threshold value of the variable pitch motor according to the self operating characteristics, so that the fault early warning is more accurate, and the early warning accuracy is improved.
(3) The invention can continuously and iteratively train historical measurement data along with the operation of the wind turbine generator, and dynamically adjust the early warning threshold value of the single wind turbine generator so as to adapt to the change of the operation working condition.
Drawings
Fig. 1 is a flowchart of a fault warning method provided in this embodiment;
FIG. 2 is a schematic diagram of historical measurement data according to the present embodiment;
FIG. 3 is a schematic diagram of actual measurement data according to the present embodiment;
fig. 4 and 5 are schematic diagrams illustrating calculation in steps two to eight of the present embodiment;
FIG. 6 is a schematic diagram illustrating the calculation of step eight in the present embodiment;
FIG. 7 is a line graph showing temperature difference data according to the present embodiment;
fig. 8 is a diagram of a temperature difference data box according to the present embodiment.
Detailed Description
The present invention is described in detail below with reference to examples, but it should be understood that the scope of the present invention is not limited by the specific embodiments.
As shown in fig. 1, the temperature rise fault early warning method for the variable pitch motor of the fan provided by the invention comprises the following steps:
the method comprises the following steps: the method comprises the steps of firstly, obtaining historical measurement data of relevant measuring points of a wind turbine generator in a period of time, and obtaining effective historical measurement data after preprocessing;
step two: calculating the temperature difference between every two three pitch motors of the wind turbine generator in the first step to obtain historical temperature difference data;
step three: constructing an abnormality detection model based on TF-IDF by using a symbolization characterization technology;
step four: training the historical temperature difference data in the second step by using the anomaly detection model to obtain a characteristic vector and a weight of a temperature difference baseline;
step five: acquiring real-time measurement data of relevant measuring points of the wind turbine generator in the first step in a period of time, and preprocessing the real-time measurement data to obtain effective real-time measurement data;
step six: calculating the temperature difference between every two three pitch motors of the wind turbine generator set in the fifth step to obtain actual temperature difference data;
step seven: calculating the characteristic vector and the weight of the actual temperature difference data by using an anomaly detection model in the third step;
step eight: calculating the space distance between the characteristic vector of the actual temperature difference data in the step seven and the characteristic vector of the temperature difference base line in the step four;
step nine: and judging whether the calculation result in the step eight is early-warning according to a 3 sigma principle, and if so, obtaining an early-warning level.
The wind turbine generator is influenced by the external environment in the operation process, so that the obtained data is inaccurate, and the real and stable operation state of the wind turbine generator cannot be expressed. For example, the sensor is interfered by strong electromagnetic signals, which causes the data in a period of time to be larger or smaller; as another example, a sensor failure or communication interruption results in failure to transmit real data, resulting in failure to obtain real data. Therefore, in the information of each measuring point of the wind turbine generator, a rough normal interval needs to be obtained by combining the interval set when the equipment is shipped, and a judgment interval is built in advance.
Wherein, 3 σ is divided into 4 probability ranges, which are respectively: (μ -2 σ, μ +2 σ), (μ -3 σ, μ +3 σ), (μ -4 σ, μ +4 σ), and the last 3 ranges are considered as early warning ranges.
And calculating the temperature difference between every two of the three pitch motors of the wind turbine generator in the second step and the sixth step to obtain historical temperature difference data and current temperature difference data, wherein the historical temperature difference data and the current temperature difference data are key indexes for judging the heat dissipation fault of the pitch motors. The temperature difference between every two pitch motors is a key index for judging the heat dissipation fault of the pitch motors, because the temperatures of the pitch motors of the three blades keep a certain relation (the relation is not necessarily a fixed value or specific distribution) under a stable working condition, if the relation changes, the heat dissipation is considered to be abnormal.
Further, the measurement points related to the wind turbine generator in the first step are specifically the temperatures of three pitch motors and the angles of three blades of the wind turbine generator.
Further, the historical measurement data of a period of time in the step one is specifically measurement data of 3 to 6 months in history.
Further, the preprocessing in the first step specifically includes screening stable working conditions for the obtained data, removing singular values and logic abnormal values, and reserving a data section capable of expressing the real and stable running state of the equipment.
Further, the step of training the temperature difference baseline characteristic vector and the weight for the historical measurement data by using the anomaly detection model in the fourth step comprises: s1: for each wind turbine generator, combining the three groups of temperature difference data calculated in the second step; s2: discretizing the combined temperature difference data at an interval of 0.5 to obtain a baseline characteristic in a specific range; s3: and obtaining a baseline characteristic vector weight TFIDF and a baseline characteristic vector IDF by using the anomaly detection model in the third step, wherein the calculation formula is as follows:
the formula I is as follows:
the formula II is as follows:
the formula III is as follows: TFIDF ═ TFi×IDFi
Wherein n isi，jRepresents the number of times a certain temperature value appears in the window, sigmaknk，jRepresents the total number of temperature values contained within the window, | D | represents the total number of windows, | { j: t is ti∈djAnd | represents the number of windows containing the temperature value plus one.
Further, in the eighth step, according to the results of the fourth step and the seventh step, the spatial distance between the feature vector of the actual temperature difference data in the seventh step and the feature vector of the temperature difference baseline in the fourth step is calculated, and the calculation formula is defined as:
the formula four is as follows:
wherein, IDF1 represents the feature vector of the temperature difference baseline obtained by training, and IDF2 represents the feature vector of the actual temperature difference data obtained by training.
Further, the 3 σ rule in the ninth step means that the normal distribution diagram is divided into 4 probability ranges by ± 1 σ, ± 2 σ, and ± 3 σ on both sides of the symmetry axis x ═ μ, in units of the standard deviation σ, and the probability ranges are respectively: the range of-2 sigma to mu plus 2 sigma, -3 sigma to mu plus 3 sigma, -4 sigma to mu plus 4 sigma and-4 sigma to mu plus 4 sigma; the last 3 ranges are considered as early warning ranges.
Further, in the ninth step, whether the calculation result in the eighth step is early-warning is judged according to a 3 sigma principle, and if the calculation result is early-warning, an early-warning level is output; whether the calculation result is early-warned or not is determined according to whether the calculation result is distributed in an early-warning range or not; if the early warning is carried out, outputting the early warning level, namely mapping the early warning range into different early warning levels, and outputting the corresponding early warning level according to the early warning range distributed by the calculation result, wherein the early warning level is as follows:
a: the probability of the distribution in-3 sigma is less than or equal to mu and less than or equal to +3 sigma is 0.9973, and the early warning level is light outside 2 sigma;
b: the probability of distribution in-4 sigma is less than or equal to mu and less than or equal to +4 sigma is 0.9999, and the early warning level is moderate outside 3 sigma:
c: the early warning level is severe and is distributed in the range of-4 sigma to more than mu and more than +4 sigma.
Specifically, a plurality of 2MW direct-drive wind generation sets of a certain wind power plant in the south are selected as implementation cases.
1) As shown in fig. 2, a measurement data of 3 months and 3 days of the unit in 6 months from 3 months to 9 months in 2019 is obtained as a historical measurement data, and as shown in fig. 3, a measurement data of 22 days of the unit in 9 months in 2019 is taken as an actual measurement data. The method comprises the steps that specifically collected temperature measuring point measurement data and blade angle measurement data of the variable pitch motor are removed, data sections with null data, singular values and communication interruption are removed, and effective historical measurement data are obtained; and the measured data comprises pmtl, pmt2 and pmt3 which respectively represent the temperature of the 3 pitch motors, and ps2 and ps3 which respectively represent the blade angle.
2) And as shown in fig. 4 and 5, the calculation process between the second step and the eighth step is realized by using the Scala language programming.
3) Based on the effective data in 1), early warning calculation is carried out by using the program in 2).
The early warning result is shown in fig. 6, the #50 fan gives an early warning, the early warning level is 3(0 represents normal, 1 represents mild, 2 represents moderate, and 3 represents severe), the early warning is represented as severe, and the distance between the current temperature difference characteristic vector and the training baseline characteristic vector is 100. And then, verifying whether the early warning result is accurate or not in a big data exploration mode, namely whether the measured data of the early warning unit is in an abnormal field or not.
And drawing box line graphs and line graphs of the temperature difference data of the variable pitch motors of all the units by using actual measurement data of 9, 9 and 22 days in 2019. The box line graph is shown in fig. 8, the line graph is shown in fig. 7, and it is obvious that the data performance of the early warning unit is greatly different from that of the normal unit, which indicates that the potential fault exists in the pitch motor of the early warning unit.
The above disclosure is only for the specific embodiment of the present invention, but the embodiment of the present invention is not limited thereto, and any variations that can be made by those skilled in the art should fall within the scope of the present invention.
Claims (8)
1. A temperature rise fault early warning method for a variable pitch motor of a fan is characterized by comprising the following steps:
the method comprises the following steps: the method comprises the steps of firstly, obtaining historical measurement data of relevant measuring points of a wind turbine generator in a period of time, and obtaining effective historical measurement data after preprocessing;
step two: calculating the temperature difference between every two three pitch motors of the wind turbine generator in the first step to obtain historical temperature difference data;
step three: constructing an abnormality detection model based on TF-IDF by using a symbolization characterization technology;
step four: training the historical temperature difference data in the second step by using the anomaly detection model to obtain a characteristic vector and a weight of a temperature difference baseline;
step five: acquiring real-time measurement data of relevant measuring points of the wind turbine generator in the first step in a period of time, and preprocessing the real-time measurement data to obtain effective real-time measurement data;
step six: calculating the temperature difference between every two three pitch motors of the wind turbine generator set in the fifth step to obtain actual temperature difference data;
step seven: calculating the characteristic vector and the weight of the actual temperature difference data by using an anomaly detection model in the third step;
step eight: calculating the space distance between the characteristic vector of the actual temperature difference data in the step seven and the characteristic vector of the temperature difference base line in the step four;
step nine: and judging whether the calculation result in the step eight is early-warning according to a 3 sigma principle, and if so, obtaining an early-warning level.
2. The fan variable pitch motor temperature rise fault early warning method according to claim 1, characterized in that: the measurement points related to the wind turbine generator in the step one specifically refer to the temperatures of three pitch motors and the angles of three blades of the wind turbine generator.
3. The fan variable pitch motor temperature rise fault early warning method according to claim 1, characterized in that: the historical measurement data of a period of time in the step one is specifically measurement data of 3 to 6 months in history.
4. The fan variable pitch motor temperature rise fault early warning method according to claim 1, characterized in that: the preprocessing in the first step specifically includes screening stable working conditions for the obtained data, eliminating singular values and logic abnormal values, and reserving a data section capable of expressing the real and stable running state of the equipment.
5. The fan variable pitch motor temperature rise fault early warning method according to claim 1, characterized in that: the step of training the temperature difference baseline characteristic vector and the weight of the historical measurement data by using the anomaly detection model in the fourth step comprises the following steps: s1: for each wind turbine generator, combining the three groups of temperature difference data calculated in the second step; s2: discretizing the combined temperature difference data at an interval of 0.5 to obtain a baseline characteristic in a specific range; s3: and obtaining a baseline characteristic vector weight TFIDF and a baseline characteristic vector IDF by using the anomaly detection model in the third step, wherein the calculation formula is as follows:
the formula I is as follows:
the formula II is as follows:
the formula III is as follows: TFIDF ═ TFi×IDFi
Wherein n isi，jRepresents the number of times a certain temperature value appears in the window, sigmaknk，jRepresenting inclusion in windowsThe total number of temperature values, | D | represents the total number of windows, | { j: t is ti∈djAnd | represents the number of windows containing the temperature value plus one.
6. The fan variable pitch motor temperature rise fault early warning method according to claim 1, characterized in that: in the step eight, according to the results of the step four and the step seven, the spatial distance between the feature vector of the actual temperature difference data in the step seven and the feature vector of the temperature difference baseline in the step four is calculated, and the calculation formula is defined as:
the formula four is as follows:
wherein, IDF1 represents the feature vector of the temperature difference baseline obtained by training, and IDF2 represents the feature vector of the actual temperature difference data obtained by training.
7. The fan variable pitch motor temperature rise fault early warning method according to claim 1, characterized in that: the 3 σ rule in the step nine means that, taking the standard deviation σ as a unit, the two sides of the symmetry axis X ═ μ of the normal distribution diagram are divided into 4 probability ranges by ± 1 σ, ± 2 σ, and ± 3 σ, respectively, which are: the range of-2 sigma to mu plus 2 sigma, -3 sigma to mu plus 3 sigma, -4 sigma to mu plus 4 sigma and-4 sigma to mu plus 4 sigma; the last 3 ranges are considered as early warning ranges.
8. The blower variable pitch motor temperature rise fault early warning method according to claim 7, characterized in that: in the ninth step, whether the calculation result in the eighth step is early-warned is judged according to a 3 sigma principle, and if the calculation result is early-warned, an early-warning level is output; whether the calculation result is early-warned or not is determined according to whether the calculation result is distributed in an early-warning range or not; if the early warning is carried out, outputting the early warning level, namely mapping the early warning range into different early warning levels, and outputting the corresponding early warning level according to the early warning range distributed by the calculation result, wherein the early warning level is as follows:
a: the probability of the distribution in-3 sigma is less than or equal to mu and less than or equal to +3 sigma is 0.9973, and the early warning level is light outside 2 sigma;
b: the probability of the distribution in-4 sigma is less than or equal to mu and less than or equal to +4 sigma is 0.9999, and the early warning level is moderate outside 3 sigma;
c: the early warning level is severe and is distributed in the range of-4 sigma to more than mu and more than +4 sigma.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010906883.8A CN112052426A (en) | 2020-09-01 | 2020-09-01 | Temperature rise fault early warning method for fan variable pitch motor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010906883.8A CN112052426A (en) | 2020-09-01 | 2020-09-01 | Temperature rise fault early warning method for fan variable pitch motor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112052426A true CN112052426A (en) | 2020-12-08 |
Family
ID=73607968
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010906883.8A Pending CN112052426A (en) | 2020-09-01 | 2020-09-01 | Temperature rise fault early warning method for fan variable pitch motor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112052426A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115263680A (en) * | 2022-07-04 | 2022-11-01 | 华能吉林发电有限公司镇赉风电厂 | Abnormal temperature rise early warning method for variable pitch motor by combining TF-IDF model and LSTM model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104807644A (en) * | 2015-04-14 | 2015-07-29 | 北京中恒博瑞数字电力科技有限公司 | Fault early warning method and system for wind generation set variable-pitch system |
| CN105760617A (en) * | 2016-03-07 | 2016-07-13 | 华北电力大学（保定） | Calculation method applied to multi-parameter fault prediction and judgment indexes of wind generating set |
| CN107609574A (en) * | 2017-08-18 | 2018-01-19 | 上海电力学院 | Wind turbines fault early warning method based on data mining |
| CN109840666A (en) * | 2017-11-29 | 2019-06-04 | 中国电力科学研究院有限公司 | A kind of model building method and system for predicting that the following Wind turbines break down |
| CN110926809A (en) * | 2019-06-19 | 2020-03-27 | 山西大学 | Big data analysis-based wind turbine generator transmission chain fault early warning method |
-
2020
- 2020-09-01 CN CN202010906883.8A patent/CN112052426A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104807644A (en) * | 2015-04-14 | 2015-07-29 | 北京中恒博瑞数字电力科技有限公司 | Fault early warning method and system for wind generation set variable-pitch system |
| CN105760617A (en) * | 2016-03-07 | 2016-07-13 | 华北电力大学（保定） | Calculation method applied to multi-parameter fault prediction and judgment indexes of wind generating set |
| CN107609574A (en) * | 2017-08-18 | 2018-01-19 | 上海电力学院 | Wind turbines fault early warning method based on data mining |
| CN109840666A (en) * | 2017-11-29 | 2019-06-04 | 中国电力科学研究院有限公司 | A kind of model building method and system for predicting that the following Wind turbines break down |
| CN110926809A (en) * | 2019-06-19 | 2020-03-27 | 山西大学 | Big data analysis-based wind turbine generator transmission chain fault early warning method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115263680A (en) * | 2022-07-04 | 2022-11-01 | 华能吉林发电有限公司镇赉风电厂 | Abnormal temperature rise early warning method for variable pitch motor by combining TF-IDF model and LSTM model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104951851B (en) | A kind of Wind turbines State Forecasting Model method for building up based on gray relative regression support vector machine | |
| EP2518309B1 (en) | Method, system and computer program product for dynamic rule engine for a wind turbine farm | |
| EP4531277A2 (en) | System and method for predicting failure of components using temporal scoping of sensor data | |
| CN111950505B (en) | A SSA-AANN-based wind turbine sensor state assessment method | |
| CN108072524B (en) | Wind turbine generator gearbox bearing fault early warning method | |
| EP3741991B1 (en) | Method for dynamic real-time optimization of the performance of a wind park and wind park | |
| Zhang et al. | A novel fault diagnosis method for wind turbine based on adaptive multivariate time-series convolutional network using SCADA data | |
| CN113007041A (en) | Wind turbine generator blade icing detection system and detection method | |
| Mahmoud et al. | Designing and prototyping the architecture of a digital twin for wind turbine | |
| CN116306139A (en) | Intelligent monitoring method and system for service life of wind turbine blade | |
| CN114810507A (en) | Aerogenerator trouble early warning system based on data analysis | |
| Dong et al. | One data-driven vibration acceleration prediction method for offshore wind turbine structures based on extreme gradient boosting | |
| CN112052426A (en) | Temperature rise fault early warning method for fan variable pitch motor | |
| CN112101646A (en) | Temperature rise fault early warning method for fan water cooling system | |
| CN116624341A (en) | A method, system and device for diagnosing abnormal power of a wind turbine | |
| Liu et al. | Analysis and comparison of turbulence models on wind turbine performance using SCADA data and machine learning technique | |
| Han et al. | Yaw system restart strategy optimization of wind turbines in mountain wind farms based on operational data mining and multi-objective optimization | |
| Dinh et al. | Implementation of digital twin-assisted condition monitoring and fault diagnosis for wind turbines | |
| CN112578232B (en) | Lightning early warning method and lightning early warning equipment of wind generating set | |
| CN111178601B (en) | Wind turbine generator power prediction method based on meteorological data post-processing | |
| EP4045791A1 (en) | Method and an apparatus for computer-implemented monitoring of a wind turbine | |
| Pandit et al. | Comparison of binned and Gaussian Process based wind turbine power curves for condition monitoring purposes | |
| CN107218180B (en) | A kind of wind power generating set driving unit fault alarm method based on vibration acceleration measurement | |
| Ma et al. | A fault prediction framework for Doubly‐fed induction generator under time‐varying operating conditions driven by digital twin | |
| CN116591910A (en) | Wind turbine generator operation and maintenance method and device, electronic equipment and storage medium |
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 | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20230818 |
|
| AD01 | Patent right deemed abandoned |