CN116816615A - Online monitoring method and system for power of wind driven generator - Google Patents
Online monitoring method and system for power of wind driven generator Download PDFInfo
- Publication number
- CN116816615A CN116816615A CN202310734928.1A CN202310734928A CN116816615A CN 116816615 A CN116816615 A CN 116816615A CN 202310734928 A CN202310734928 A CN 202310734928A CN 116816615 A CN116816615 A CN 116816615A
- Authority
- CN
- China
- Prior art keywords
- power
- wind
- generator
- wind speed
- driven 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000013507 mapping Methods 0.000 claims abstract description 22
- 230000002159 abnormal effect Effects 0.000 claims abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 239000006185 dispersion Substances 0.000 claims description 3
- 230000005284 excitation Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000010248 power generation Methods 0.000 abstract description 6
- 230000005856 abnormality Effects 0.000 abstract description 4
- 238000012545 processing Methods 0.000 description 2
- 235000017166 Bambusa arundinacea Nutrition 0.000 description 1
- 235000017491 Bambusa tulda Nutrition 0.000 description 1
- 241001330002 Bambuseae Species 0.000 description 1
- 235000015334 Phyllostachys viridis Nutrition 0.000 description 1
- 239000011425 bamboo Substances 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Abstract
The invention relates to a method and a system for monitoring power of a wind driven generator on line, and belongs to the technical field of wind power generation. The method comprises the following steps: collecting operation data of the wind driven generator, wherein the operation data comprises wind speed and generator power; inputting the wind speed into a pre-obtained wind speed power mapping model to obtain a corresponding wind driven generator power predicted value; and calculating the reliability of the current power of the wind driven generator as a normal value according to the current power of the wind driven generator, the power predicted value of the power of the wind driven generator calculated according to the corresponding wind speed and the discrete degree of the power sample in the interval where the wind speed corresponding to the training set of the wind speed power mapping model is located, and if the reliability is smaller than a set threshold value, considering the power of the wind driven generator as abnormal, otherwise, considering the power of the wind driven generator as normal. According to the invention, the wind power generator data are collected in real time, the wind power generator data are analyzed on line, the power state of the wind power generator is given out, and the analysis conclusion is reached, so that the power abnormality of the wind power generator can be found in time, and the loss of a wind power plant is reduced.
Description
Technical Field
The invention relates to a method and a system for monitoring power of a wind driven generator on line, and belongs to the technical field of wind power generation.
Background
The wind power generation in the last 10 years is developed in blowout mode, the development of the wind power generation industry is further stimulated by the double-carbon target in 2020, the wind power plant appears like bamboo shoots in spring after raining, and the number of the accumulated installed wind power generation sets in the whole country exceeds 15 ten thousand by the end of 2020. The world wind energy market is expected to grow by 25% each year for the next 20 years. With the installed operation of a large number of wind driven generators, the monitoring and management of the generators are more important, and timely and accurate discovery of abnormal power of the wind driven generators and rapid fault removal are guarantees for maintaining efficient operation of wind power plants.
Patent document with publication number of CN116044677A discloses an intelligent early warning method and system for a wind generating set. According to the method, the real-time output power of the wind turbine generator is compared with the preset output power, whether the generation temperature of the wind turbine generator needs to be obtained is judged, when the real-time output power is smaller than the preset output power, whether the wind turbine generator fails or not is judged according to the obtained generation temperature, and the generation temperature is interfered by external environmental factors, so that the accuracy of a detection result is reduced.
Patent document with publication number of CN116221037A discloses a wind turbine monitoring method and device. According to the wind turbine generator monitoring method, a virtual environment is built by acquiring environment parameters, a virtual wind turbine generator corresponding to a real wind turbine generator is built in the virtual environment, output power data of the real wind turbine generator is compared with output power data of the virtual wind turbine generator, the difference value of the output power data of the real wind turbine generator and the output power data of the virtual wind turbine generator is compared with a preset fixed threshold value, so that whether the wind turbine generator works normally is judged, but compared with the real environment, deviation caused by sudden or uncontrollable factors still exists in the built virtual environment, and sudden change of simulation power data output by the virtual wind turbine generator also affects power comparison results to cause misjudgment.
Disclosure of Invention
The invention aims to provide a method and a system for monitoring power of a wind driven generator on line, which are used for solving the problem of low reliability of power monitoring of the existing wind driven generator.
In order to achieve the above object, the present invention provides a method comprising:
the invention discloses a technical scheme of an online monitoring method for wind driven generator power, which comprises the following steps:
1) Collecting operation data of the wind driven generator, wherein the operation data comprises wind speed and generator power;
2) Inputting the wind speed into a pre-obtained wind speed power mapping model to obtain a corresponding wind driven generator power predicted value;
3) And calculating the reliability of the current power of the wind driven generator as a normal value according to the current power of the wind driven generator, the power predicted value of the power of the wind driven generator calculated according to the corresponding wind speed and the discrete degree of the power sample in the interval where the wind speed corresponding to the training set of the wind speed power mapping model is located, and if the reliability is smaller than a set threshold value, considering the power of the wind driven generator as abnormal, otherwise, considering the power of the wind driven generator as normal.
According to the wind power generation system, wind power generator data are collected in real time, a wind speed power mapping model is obtained by taking wind speed and generator power historical data as training sets, the reliability of normal operation of a fan is calculated and is analyzed on line based on the discrete degree of a power sample in an interval where wind speed corresponding to the wind speed in the training sets of the wind speed power mapping model, a wind power generator power state analysis conclusion is timely given, and wind power generator power abnormality is found, so that wind power field loss is reduced.
Further, in step 3), the training set of the training wind speed power mapping model is divided into a plurality of continuous sets according to wind speeds, each set corresponds to a certain wind speed interval, and the variance of the power of the generator corresponding to all wind speeds in a certain wind speed interval in the training set is the degree of dispersion of the power samples of the corresponding wind speed interval.
According to the invention, a large amount of wind speed data in the training set is divided into continuous wind speed intervals, and the dispersion degree of the power samples is represented by the variance of the power of the generator in the intervals.
Further, the calculation formula of the confidence level p in the step 3) is as follows:
wherein, the average value of the power of the generator is P, the predicted value of the power of the generator is sigma 2 Is the power variance.
Further, the operation data in the step 1) is the average value of the wind speed data and the generator power data acquired according to a certain sampling period in a set time interval.
The average value in a certain time interval represents the running data of the current running of the fan, so that the misjudgment of on-line monitoring caused by data mutation due to interference is avoided.
Further, in step 1), after preprocessing the original data to delete the unusable data, calculating the average value to obtain the running data; the unavailable data includes shutdown, fault and power limit data.
The method improves the reliability of the running data of the wind driven generator by deleting the abnormal data, and further eliminates misjudgment caused by abnormal monitoring results.
Further, the wind speed power mapping model in the step 2) is obtained through training of the following steps: obtaining a certain number of wind speed-generator power data pairs in a normal state from the historical data of the wind driven generator as a training set; the wind speed is taken as input, the generator power is taken as output, and the neural network is trained by using the training set.
Further, the neural network is a 3-layer BP neural network, a sigmoid excitation function is adopted in an hidden layer, and 5 nodes are arranged in the hidden layer.
Further, the set threshold in step 3) is 0.01.
The technical scheme of the wind driven generator power online monitoring system comprises a processor, wherein the processor is used for executing instructions to realize the wind driven generator power online monitoring method according to any one of the above.
Drawings
FIG. 1 is a wind turbine power on-line monitoring system;
FIG. 2 is a schematic diagram of data processing of a data acquisition and analysis device;
FIG. 3 is a wind speed power mapping model.
Detailed Description
The present invention will be further described in detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent.
An embodiment of a method for monitoring power of a wind driven generator on line comprises the following steps:
the embodiment provides a wind driven generator power on-line monitoring method, as shown in fig. 1, wherein the wind driven generator power on-line monitoring method is mainly completed by data acquisition and analysis equipment, the data acquisition and analysis equipment reads wind driven generator data from a wind driven generator main control cabinet according to a certain sampling period, then preprocesses the data, and analyzes and compares the data to obtain a conclusion about whether the power is abnormal or not and sends the conclusion to a wind power plant monitoring terminal. The data processing principle is shown in fig. 2.
The on-line monitoring method for the power of the wind driven generator comprises the following steps:
firstly, preprocessing data after equipment collects original data of a wind driven generator: deleting unavailable data, wherein the unavailable data comprises shutdown, fault and power limit data; and deleting unavailable data, and then carrying out corresponding mean value calculation to obtain operation data: wind speed average value of current time period of wind driven generatorAnd output power mean>The mean time was set at 10 minutes.
Step two, inputting the wind speed average value calculated in the step one into a wind speed power mapping model to obtain a wind driven generator power prediction result P * 。
Third step, according to the current generator power and the calculated predicted value P of the generator power corresponding to the wind speed * And calculating the reliability p of the current power of the wind driven generator as a normal value according to the discrete degree of the power sample in the interval where the wind speed corresponding to the training set of the wind speed power mapping model is located, analyzing and outputting, and if the reliability p calculated by the following formula (1) is smaller than a certain threshold value (such as 0.01), considering the power of the wind driven generator to be abnormal, otherwise, considering the power of the motor to be normal.
The calculation formula of the confidence level p is as follows:
the wind speed power mapping model used in the second step comprises two parts: the wind speed power mapping module is used for a wind speed power variance list. Table 1 is a wind speed power variance list, and works in parallel with the wind speed power mapping module to form a wind speed power mapping model, as shown in FIG. 3.
The wind speed power mapping module adopts an artificial neural network for modeling, is set as a 3-layer BP neural network, adopts a sigmoid excitation function in an hidden layer, sets 5 nodes in the hidden layer, and has an average wind speed as an input signal and a power predicted value P of a generator as an output signal. The model making steps are as follows:
a. preprocessing off-line data of the wind driven generator in the past 1 month, removing unavailable data, and calculating average value of wind speed and power variable to obtain a certain number of data pairsThe wind speed average value and the generator power average value are respectively,the mean time was 10 minutes.
b. To be used forFor inputting value +.>For the target value, use data pair +.>And (3) training the neural network as a training set, and obtaining the wind speed power mapping module after training is completed.
The steps for obtaining the wind speed power variance list are as follows:
s1, the data pairs in the training set are matchedPress->Divided into n consecutive sets, wherein the data pair +.>Maximum->Rounding to n, all data pairs in the kth set satisfy +.>I.e. the data pairs in the kth set are data when the mean value of the wind speed is about equal to k.
S2, the wind speed power mapping module inputs the average wind speed k to obtain an expected value of output power
S3, if the kth set contains m data pairs, the power variance corresponding to the average wind speed k is as follows:
in the aboveIs the average power in the ith sample in the kth set.
The wind speed power variance list shown in table 1 can be obtained by performing the above steps.
TABLE 1
The invention provides a wind driven generator power online monitoring method which can collect wind driven generator data in real time and analyze the wind driven generator data online, so that the reliability of judging the power abnormality of a wind driven generator is improved, and the analysis conclusion of the power state of the wind driven generator can be timely given out and the power abnormality of the wind driven generator is found, thereby reducing the loss of a wind power plant.
Wind power generator power on-line monitoring system embodiment:
the embodiment provides an online monitoring system for power of a wind driven generator, as shown in fig. 1, which comprises a processor, wherein the processor is used for executing instructions to realize the online monitoring method for power of the wind driven generator in any one of the above steps. The system is described in detail in the above embodiment of the wind power generator power on-line monitoring method, and will not be described herein.
Claims (9)
1. The on-line monitoring method for the power of the wind driven generator is characterized by comprising the following steps of:
1) Collecting operation data of the wind driven generator, wherein the operation data comprises wind speed and generator power;
2) Inputting the wind speed into a pre-obtained wind speed power mapping model to obtain a corresponding wind driven generator power predicted value;
3) And calculating the reliability of the current power of the wind driven generator as a normal value according to the current power of the wind driven generator, the power predicted value of the power of the wind driven generator calculated according to the corresponding wind speed and the discrete degree of the power sample in the interval where the wind speed corresponding to the training set of the wind speed power mapping model is located, and if the reliability is smaller than a set threshold value, considering the power of the wind driven generator as abnormal, otherwise, considering the power of the wind driven generator as normal.
2. The method for on-line monitoring of wind power generator power according to claim 1, wherein in step 3), the training set of the training wind speed power mapping model is divided into a plurality of consecutive sets according to wind speeds, each set corresponds to a certain wind speed interval, and the variance of the generator power corresponding to all wind speeds in a certain wind speed interval in the training set is the degree of dispersion of the power samples of the corresponding wind speed interval.
3. The method for on-line monitoring of wind power generator power according to claim 1, wherein the calculation formula of the confidence level p in step 3) is as follows:
in the method, in the process of the invention,for the power average value of the generator, P * Sigma is the predicted value of generator power 2 Is the power variance.
4. The method for on-line monitoring of wind power generator power according to claim 1, wherein the operation data in step 1) is an average value of wind speed data and generator power data collected according to a certain sampling period in a set time interval.
5. The method for on-line monitoring of wind power generator power according to claim 4, wherein in step 1), after the preprocessing of deleting unavailable data is performed on the original data, the running data is obtained by calculating a mean value; the unavailable data includes shutdown, fault and power limit data.
6. The method for on-line monitoring of wind power generator power according to claim 1, wherein the wind speed power mapping model in step 2) is obtained by training the following steps: obtaining a certain number of wind speed-generator power data pairs in a normal state from the historical data of the wind driven generator as a training set; the wind speed is taken as input, the generator power is taken as output, and the neural network is trained by using the training set.
7. The method for on-line monitoring of wind power generator power according to claim 6, wherein the neural network is a 3-layer BP neural network, a hidden layer adopts a sigmoid excitation function, and 5 nodes are arranged in the hidden layer.
8. The method for on-line monitoring of wind power generator power according to claim 1, wherein the set threshold value in step 3) is 0.01.
9. A wind turbine power on-line monitoring system comprising a processor for executing instructions to implement a wind turbine power on-line monitoring method as claimed in any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310734928.1A CN116816615A (en) | 2023-06-20 | 2023-06-20 | Online monitoring method and system for power of wind driven generator |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310734928.1A CN116816615A (en) | 2023-06-20 | 2023-06-20 | Online monitoring method and system for power of wind driven generator |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116816615A true CN116816615A (en) | 2023-09-29 |
Family
ID=88142291
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310734928.1A Pending CN116816615A (en) | 2023-06-20 | 2023-06-20 | Online monitoring method and system for power of wind driven generator |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116816615A (en) |
-
2023
- 2023-06-20 CN CN202310734928.1A patent/CN116816615A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111680820B (en) | Distributed photovoltaic power station fault diagnosis method and device | |
CN111539553B (en) | Wind turbine generator fault early warning method based on SVR algorithm and off-peak degree | |
CN114723285B (en) | Power grid equipment safety evaluation prediction method | |
CN111209934A (en) | Fan fault prediction and alarm method and system | |
CN109356798B (en) | Wind driven generator gearbox state monitoring method based on co-integration analysis | |
CN116660672B (en) | Power grid equipment fault diagnosis method and system based on big data | |
CN113822418A (en) | Wind power plant power prediction method, system, device and storage medium | |
CN113268835A (en) | Pumped storage hydroelectric generating set health assessment method and device based on extended isolation forest and cloud model | |
CN115186935B (en) | Electromechanical device nonlinear fault prediction method and system | |
CN116816615A (en) | Online monitoring method and system for power of wind driven generator | |
CN111209914A (en) | Method for eliminating abnormal wind power data | |
CN115983507A (en) | Method and system for predicting cross-section broadband oscillation risk of transmission-end source power grid | |
CN114320773B (en) | Wind turbine generator system fault early warning method based on power curve analysis and neural network | |
CN113554203B (en) | Wind power prediction method and device based on high-dimensional meshing and LightGBM | |
CN111814325B (en) | Single-variable alarm threshold optimization method for non-stationary operation of coal-fired power generation equipment | |
CN115062007A (en) | Wind turbine generator set wind speed and power data cleaning method based on isolated forest algorithm | |
CN114708885A (en) | Fan fault early warning method based on sound signals | |
CN115326393A (en) | Wind turbine generator bearing pair fault diagnosis method based on temperature information | |
CN113431800A (en) | Method for distinguishing running state of fan in real time | |
EP3303835B1 (en) | Method for windmill farm monitoring | |
CN114607571A (en) | Offshore wind power gear box fault identification method and system based on lubricating system monitoring | |
CN113946977A (en) | Application method for early warning of fan variable pitch fault based on decision tree algorithm | |
Tingting et al. | Early warning method for power station auxiliary failure considering large-scale operating conditions | |
CN112682269A (en) | Wind turbine generator state monitoring method based on OC-RKELM | |
CN114033631B (en) | Online identification method for wind energy utilization coefficient of wind turbine generator |
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 |