CN113780375A - Virtual-real interaction wind power plant wind power monitoring system based on digital twins and application - Google Patents
Virtual-real interaction wind power plant wind power monitoring system based on digital twins and application Download PDFInfo
- Publication number
- CN113780375A CN113780375A CN202110987657.1A CN202110987657A CN113780375A CN 113780375 A CN113780375 A CN 113780375A CN 202110987657 A CN202110987657 A CN 202110987657A CN 113780375 A CN113780375 A CN 113780375A
- Authority
- CN
- China
- Prior art keywords
- wind speed
- sensor
- wind
- virtual
- entity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 28
- 230000003993 interaction Effects 0.000 title claims abstract description 19
- 238000003745 diagnosis Methods 0.000 claims abstract description 104
- 230000002159 abnormal effect Effects 0.000 claims abstract description 75
- 238000013507 mapping Methods 0.000 claims abstract description 29
- 238000005259 measurement Methods 0.000 claims abstract description 15
- 230000002452 interceptive effect Effects 0.000 claims abstract description 12
- 239000007787 solid Substances 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 18
- 230000005856 abnormality Effects 0.000 claims description 16
- 238000005096 rolling process Methods 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000005070 sampling Methods 0.000 claims description 5
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 4
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 230000002547 anomalous effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 7
- 238000012423 maintenance Methods 0.000 description 7
- 238000012216 screening Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 6
- 230000002457 bidirectional effect Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 5
- 230000008447 perception Effects 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000013139 quantization Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 239000010687 lubricating oil Substances 0.000 description 3
- 230000006855 networking Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000000926 separation method Methods 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 238000012896 Statistical algorithm Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 235000012907 honey Nutrition 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000010977 unit operation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Evolutionary Biology (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- General Business, Economics & Management (AREA)
- Computational Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Operations Research (AREA)
- Mathematical Analysis (AREA)
- Computing Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Mathematical Optimization (AREA)
- Algebra (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Probability & Statistics with Applications (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
Abstract
The invention relates to a virtual-real interaction wind power plant wind power monitoring system based on digital twins and application thereof, wherein the system comprises: an array of solid sensors: the system is used for acquiring wind power data of each wind turbine generator in the wind power plant in real time; virtual sensor array: the sensor array is matched with the entity sensors in the entity sensor array one by one; an interactive platform: the system is used for carrying out association mapping on the entity sensor and the virtual sensor, and comprises abnormity diagnosis on wind data of the entity sensor, and prediction and setting on virtual wind speed of the virtual sensor. Compared with the prior art, the virtual sensor technical framework based on measurement-sharing-association-diagnosis-prediction-check can complete the abnormal diagnosis and wind speed mapping of the wind speed sensor, and the fault diagnosis strategy based on group cooperation can quickly and accurately position and sense abnormal states, so that the wind speed variable input into the unit control system is accurate and reliable under the condition of wind speed sensor wind measurement data loss or misalignment.
Description
Technical Field
The invention relates to the technical field of wind power generation, in particular to a virtual-real interaction wind power plant wind power monitoring system based on digital twins and application thereof.
Background
Under the drive of a double-carbon target, renewable energy represented by wind energy gradually becomes an important means for developing a novel energy industry and reducing carbon emission in China due to the advantages of cleanness, no pollution, abundant reserves, mature development technology, low cost and the like. Offshore wind power has the characteristics of no occupation of land resources, rich wind resources, low turbulence intensity and the like, is developed rapidly in China, and plays an important role in realizing a clean energy substitution process in China.
Wind speed is very important for fan optimal control and grid-connected scheduling, and misalignment of measured wind speed not only affects the generating efficiency of the wind turbine generator, but also can cause the generator to be damaged in severe cases. In practical engineering application, a sensor fault diagnosis module is arranged in a wind turbine SCADA system, the basic logic of diagnosis is to set an upper threshold value and a lower threshold value of wind speed, when a measured value of a sensor exceeds a threshold value range, the system gives an alarm, the sensor is in fault, and the wind turbine is in an operating state without an input wind speed signal. For non-destructive faults and evolution faults, the measurement data of the sensor may be out of alignment and influenced by the severity of the faults, the out-of-alignment wind speed may still be within the threshold range of diagnosis, the SCADA system cannot identify the fault state, the wind speed sensor continues to operate in a fault, the unit inputs the out-of-alignment wind speed data to the control system, and a corresponding control strategy is formulated.
In summary, the challenges of the operation and maintenance of the wind speed sensor can be summarized in the following two scenarios: (1) the SCADA system can identify the fault state of the sensor and give an alarm, but the operation and detection work is difficult to be carried out in time, and the fault state is difficult to be eliminated at the first time; (2) the sensor is operated in a misalignment mode, the system cannot timely and effectively identify the abnormal state of the sensor, and the wind turbine generator is controlled to operate based on the misalignment data. In addition, parameter perception of the wind power plant is realized by a large-scale sensor array, and a great amount of manpower and financial resources are required to be invested for one-by-one and comprehensive investigation and treatment of the sensor array.
For an offshore wind farm, the accessibility is limited by the weather conditions such as wind speed, wave height and the like, and people can be arranged to carry corresponding spare parts and take proper ships to enter the sea to perform maintenance tasks after a unit fails, so that the operation and maintenance cost of the offshore wind farm is high, and only the maintenance cost accounts for 20-35% of the total income of the wind turbine and is about 2 times of the maintenance cost of the onshore wind turbine. In addition, the capacity of a single machine of the offshore wind turbine generator is larger and larger, and the long-time failure and outage of the generator can cause serious economic loss in the face of abundant offshore wind energy resources. Therefore, the operation and maintenance of the existing offshore wind speed sensor face huge challenges from the aspects of economy and implementation.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a virtual-real interaction wind power plant wind power monitoring system based on digital twin and application thereof.
The purpose of the invention can be realized by the following technical scheme:
a virtual-real interaction wind power plant wind power monitoring system based on digital twinning comprises:
an array of solid sensors: the system is used for acquiring wind power data of each wind turbine generator in the wind power plant in real time;
virtual sensor array: the sensor array is matched with the entity sensors in the entity sensor array one by one;
an interactive platform: the system is used for carrying out association mapping on the entity sensor and the virtual sensor, and comprises abnormity diagnosis on wind data of the entity sensor, and prediction and setting on virtual wind speed of the virtual sensor.
Preferably, the interaction platform comprises:
a fault mutual diagnosis community division module: the system comprises a fault mutual diagnosis community, a fault mutual diagnosis community and a fault mutual diagnosis community, wherein the fault mutual diagnosis community is used for dividing entity sensors in an entity sensor array into a plurality of fault mutual diagnosis communities;
an abnormal state diagnostic module: a plurality of abnormal state diagnosis modules are arranged, one abnormal state diagnosis module corresponds to one fault mutual diagnosis community, abnormal state diagnosis is carried out in each fault mutual diagnosis community based on the measurement data of each entity sensor, and the entity sensor with the abnormality is determined;
a wind speed prediction and setting module: the method comprises the steps that a plurality of wind speed prediction and setting modules are arranged, one wind speed prediction and setting module corresponds to one fault mutual diagnosis community, measured wind speeds of entity sensors are directly related to virtual sensors for entity sensors without abnormity, and wind speed prediction and setting models of the virtual sensors are established for predicting and setting the virtual wind speeds for the entity sensors with abnormity.
Preferably, the fault mutual-diagnosis community dividing module screens the wind speed spatial correlation of the unit based on the unit yaw data of the unit corresponding to the entity sensor, screens a plurality of units with higher wind speed spatial correlation as associated units and divides the associated units into a fault mutual-diagnosis community, and further divides the corresponding entity sensor into a fault mutual-diagnosis community.
Preferably, the abnormal state diagnosis module includes:
wind speed abnormal period determination unit: determining a time interval with wind speed abnormal data based on the wind speed time sequence flow of the entity sensors in the fault mutual diagnosis community;
an abnormality sensor positioning unit: and locating the entity sensor with abnormality based on cross comparison of wind speed correlation in a time interval with wind speed abnormality data.
Preferably, the manner of determining the time interval with the wind speed abnormal time period determining unit includes:
acquiring the wind speed variance measured by each entity sensor in the fault mutual diagnosis community at each sampling time point;
determining a variable point variance threshold value based on a quantile algorithm by taking the wind speed variance as a variable point index for determining abnormity;
and determining the time interval when the wind speed variance is larger than the variable point variance threshold value as the time interval when the wind speed abnormal data exists.
Preferably, the method of locating the entity sensor having the abnormality by the abnormality sensor locating unit includes:
and calculating the correlation of the wind speeds measured by any two entity sensors in a rolling manner, selecting the entity sensor with the correlation coefficient smaller than a set threshold value, and determining the entity sensor with the largest occurrence frequency as abnormal.
Preferably, the wind speed prediction and setting model comprises a wind speed prediction model and a wind speed setting model which are sequentially cascaded, the wind speed prediction model predicts the wind speed of the abnormal entity sensor based on the measured wind speed of the upwind direction entity sensor with the optimal correlation with the wind speed of the abnormal entity sensor in the fault mutual diagnosis community, and the wind speed setting model sets the predicted wind speed based on the measured wind speed of the abnormal entity sensor to determine the final virtual wind speed.
Preferably, the wind speed prediction model comprises a bidirectional long-short term memory neural network model.
Preferably, the mathematical expression of the wind speed setting model is as follows:
wherein v ist virIndicating the set wind speed, v, at time tt mapIndicating the predicted wind speed at time t, vtActual wind speed, ξ, of an anomalous entity sensor representing the prediction at time tMAPEThe MAPE error of the wind speed prediction model is represented, and n represents the total number of moments in the prediction time series.
The utility model provides an application of false real interaction wind-powered electricity generation field wind speed monitoring system based on digit twin, false real interaction wind-powered electricity generation field wind speed monitoring system based on digit twin be used for wind speed monitoring and control wind turbine generator system operation, when the entity sensor in the entity sensor array takes place unusually, adopt the virtual wind speed control wind turbine generator system operation of the virtual sensor that corresponds with unusual entity sensor in the virtual sensor array, when unusual entity sensor resumes normally, adopt the measurement wind speed control wind turbine generator system operation of entity sensor.
Compared with the prior art, the invention has the following advantages:
(1) the method constructs the wind power plant digital twin body, and combines the multivariate mathematical statistics theory and the artificial intelligence algorithm to carry out digital virtual mapping on the running state of the physical equipment, thereby providing a new idea for monitoring the running state of the sensor and diagnosing faults.
(2) The system can complete the abnormal diagnosis and wind speed mapping of the wind speed sensor based on the virtual sensor technical framework of 'measurement-sharing-association-diagnosis-prediction-check', can quickly and accurately position and sense abnormal states based on a group cooperation fault diagnosis strategy, and ensures that the wind speed variable input into a unit control system is accurate and reliable under the condition of wind speed sensor wind measurement data loss or misalignment.
(3) The method can accurately sense the time sequence interval of the abnormal state based on the variable point algorithm, can effectively identify the spatial position of the fault sensor by utilizing cross comparison of multi-machine wind speed correlation, and realizes real-time diagnosis and fault identification of the running state of the sensor;
(4) the method can effectively mine the time-space characteristics of wind speed distribution based on a bidirectional long-time neural network of multi-machine shared mapping, effectively ensure the precision of the mapped wind speed by using the actually measured wind speed of a multi-association unit as the input of a model, input the set virtual wind speed into a unit control system, avoid potential safety hazards brought to unit operation by model error prediction, and guide the unit to optimally control the operation by using the predicted and set virtual wind speed.
Drawings
FIG. 1 is a structural block diagram of a virtual-real interaction wind power plant wind power monitoring system based on digital twins;
FIG. 2 is a schematic block diagram of a virtual sensor of the present invention;
FIG. 3 is a schematic diagram illustrating a fault inter-diagnosis community division result in an embodiment of the present invention;
FIG. 4 is a graph of measured wind speed of a wind speed sensor in a fault-diagnosis community according to an embodiment of the present invention;
FIG. 5 is a rolling statistical plot of velocity variance in the fault co-diagnosis community of FIG. 4;
FIG. 6 is a schematic view of a wind speed correlation rolling statistic;
FIG. 7 is a schematic diagram of a dynamic screening structure of a correlation line during wind speed prediction;
FIG. 8 is a virtual wind speed prediction result;
fig. 9 shows the virtual wind speed setting result.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. Note that the following description of the embodiments is merely a substantial example, and the present invention is not intended to be limited to the application or the use thereof, and is not limited to the following embodiments.
Examples
As shown in FIG. 1, the invention provides a virtual-real interaction wind power plant wind power monitoring system based on digital twin, which comprises:
an array of solid sensors: the system is used for acquiring wind power data of each wind turbine generator in the wind power plant in real time;
virtual sensor array: the sensor array is matched with the entity sensors in the entity sensor array one by one;
an interactive platform: the system is used for carrying out association mapping on the entity sensor and the virtual sensor, and comprises abnormity diagnosis on wind data of the entity sensor, and prediction and setting on virtual wind speed of the virtual sensor.
The invention researches and discusses the state recognition and wind speed perception of a wind power plant wind speed sensor, provides a six-dimensional virtual wind speed sensor technical framework driven by a digital twin technology on the basis of comprehensively considering the reliability, the economy and the engineering feasibility of the method, completes the interconnection, the intercommunication and the sharing and the fusion of multi-unit information on the basis of the traditional wind power plant sensor array and a machine networking platform, constructs a virtual digital twin body of a physical sensor by combining mathematical statistics and an artificial intelligence algorithm, excavates the space-time distribution characteristic of wind measurement data, realizes the real-time diagnosis of the running state of the wind speed sensor by the cross comparison of the running parameters of the multi-unit, corrects and trains and iterates a prediction model to complete the virtual perception prediction of the wind measurement data, and analyzing the historical error of the prediction model, setting and checking the virtual wind speed, inputting the set wind speed data into a control system, replacing missing data or fault data generated by the fault operation of the sensor, and guiding the operation control of the unit.
The virtual sensor is actually a digital virtual mapping of the physical sensor, the running state of the physical sensor is digitally described, feature extraction and prediction are carried out in a virtual space, real-time sensing, virtual-real cooperation and intelligent control over the wind power plant are achieved, and the diagnosis and sensing principle of the virtual sensor is shown in fig. 2. Through the depth fusion of the virtual sensor and the real sensor, the entity sensor can measure the operation parameters of the wind turbine generator in real time, data support is provided for the training and checking of the mapping model of the virtual sensor, the virtual sensor can carry out digital description and situation perception on the real-time operation state of the entity sensor, the virtual wind speed can be used as the redundant wind speed for controlling the operation of the wind turbine generator, and the redundant wind speed is input into a control system to guide the operation of the wind turbine generator when necessary.
The system of the present application is described in detail below, in which:
the entity sensor mainly comprises a wind power plant wind speed sensor array, and completes real-time collection of wind speed data.
The virtual entity integrates data mining methods such as a multivariate statistical algorithm and an artificial intelligence algorithm to describe the correlation characteristics and the operating characteristics of the physical entity from multiple space-time scales. The correlation characteristics are measured by taking the wind speed correlation among the units, measured data of a shared sensor array are transmitted by using a networking platform of a wind power plant, a unit community is divided on the basis of information interaction such as wind speed and direction, and the units with strong wind speed correlation are divided into a fault mutual diagnosis community. In addition, the fault inter-diagnosis community needs to be dynamically adjusted along with the change of parameters such as wind speed and wind direction, and high wind speed correlation among community units is ensured all the time. The operation characteristic analysis takes a fault inter-diagnosis community as a unit, the wind speed data distribution characteristics among the units in the community are dynamically counted, the distribution characteristics of the measured data of each sensor are mined, the operation state of the sensors is identified through cross comparison of the wind measuring data of the multiple units, for the fault state, the wind speed is sensed by using a multi-unit correlation prediction method, the predicted wind speed is input into a unit control system after being set, the fault data is replaced, and the unit is guided to operate. In the whole process, the fault inter-diagnosis community and the mapping model need to be updated and checked continuously, the real mapping relation of the physical entity is kept, and the consistency, the accuracy and the real-time performance of the fault inter-diagnosis community and the mapping model are ensured.
The connection is a network carrier composed of communication devices such as optical fibers, routers and servers, and forms a platform for information sharing and interaction of the wind power plant, which can be called as "machine networking". The timeliness and the accuracy of the data transmission of the wind turbine generator are guaranteed by utilizing the optical fiber communication distributed in the whole field of the wind power plant.
The interaction platform comprises:
a fault mutual diagnosis community division module: the system comprises a fault mutual diagnosis community, a fault mutual diagnosis community and a fault mutual diagnosis community, wherein the fault mutual diagnosis community is used for dividing entity sensors in an entity sensor array into a plurality of fault mutual diagnosis communities;
an abnormal state diagnostic module: a plurality of abnormal state diagnosis modules are arranged, one abnormal state diagnosis module corresponds to one fault mutual diagnosis community, abnormal state diagnosis is carried out in each fault mutual diagnosis community based on the measurement data of each entity sensor, and the entity sensor with the abnormality is determined;
a wind speed prediction and setting module: the method comprises the steps that a plurality of wind speed prediction and setting modules are arranged, one wind speed prediction and setting module corresponds to one fault mutual diagnosis community, measured wind speeds of entity sensors are directly related to virtual sensors for entity sensors without abnormity, and wind speed prediction and setting models of the virtual sensors are established for predicting and setting the virtual wind speeds for the entity sensors with abnormity.
The fault inter-diagnosis community division module, the abnormal state diagnosis module and the wind speed prediction and setting module are specifically explained as follows:
1. fault mutual diagnosis community division module
The fault mutual diagnosis community division module screens the wind speed spatial correlation of the units based on the unit yaw data of the units corresponding to the entity sensors, screens a plurality of units with higher wind speed spatial correlation as associated units and divides the associated units to a fault mutual diagnosis community, and further divides the corresponding entity sensors to the fault mutual diagnosis community.
The wind speed has spatial correlation, the wind speed spatial correlation among the units is used as a division standard, the wind generation units with the strong wind speed spatial correlation are divided into a fault mutual diagnosis community, and the same wind power plant can be divided into a plurality of fault mutual diagnosis communities to realize group cooperative mutual diagnosis. For the ultra-short-term wind speed mapping model, factors influencing the spatial correlation of wind speed are mainly caused by the change of atmospheric pressure, and random fluctuation of wind speed and wind direction is caused. The wind speed correlation strength among the wind turbine generators directly determines the accuracy of a wind speed prediction result based on the traditional spatial correlation, so that the seeds of the associated wind turbine generators are preferably the core part of a real-time prediction system of the wind turbine generators. Aiming at the real-time perception and prediction of the wind speed, the number of the associated seed units of the target unit is generally more than two, so that the reliability and the anti-interference performance of the wind speed prediction value can be ensured. The screening of the associated wind turbine generator can be carried out according to the following steps: (1) and (3) wind speed correlation time sequence analysis: because the wind speed correlation has strong time-varying property, the time sequence change of the wind speed correlation of the associated units is considered, and the change characteristics of the wind speed correlation coefficient of each alternative seed unit and the target unit within one day along with the time are analyzed according to the historical wind power operation data, so that a basis is provided for the subsequent dynamic screening of the seeds of the associated units; (2) screening a dynamic association seed set: and setting a threshold value of the associated unit according to the correlation coefficient of each wind turbine generator to be selected and the target unit at different time, and preferably selecting the associated seed unit when the wind speed correlation coefficient of the alternative unit reaches the threshold value range. According to the principle, the optimal results of the dynamic association seed set under different time of day can be obtained. Because the wind speed and the wind direction have strong time-varying characteristics, the fault inter-diagnosis community needs to be dynamically updated according to the measured data, a wind direction change threshold value is set, when the wind direction change exceeds the threshold value, the fault inter-diagnosis community needs to be divided again, and high correlation of the inter-diagnosis community is ensured to be kept all the time.
And a wind direction change threshold value is set, when the wind direction change exceeds the threshold value, the fault mutual diagnosis community needs to be divided again, and high correlation is kept among the fault mutual diagnosis community units all the time, so that the fault diagnosis result is more accurate.
2. Abnormal state diagnosis module
The abnormal state diagnosis module includes:
wind speed abnormal period determination unit: determining a time interval with wind speed abnormal data based on the wind speed time sequence flow of the entity sensors in the fault mutual diagnosis community;
an abnormality sensor positioning unit: and locating the entity sensor with abnormality based on cross comparison of wind speed correlation in a time interval with wind speed abnormality data.
The mode of the wind speed abnormal time period determining unit determining the time interval in which the wind speed abnormal data exists includes:
acquiring the wind speed variance measured by each entity sensor in the fault mutual diagnosis community at each sampling time point;
determining a variable point variance threshold value based on a quantile algorithm by taking the wind speed variance as a variable point index for determining abnormity;
and determining the time interval when the wind speed variance is larger than the variable point variance threshold value as the time interval when the wind speed abnormal data exists.
The mode of the abnormal sensor positioning unit for positioning the entity sensor with the abnormal sensor comprises the following steps:
and calculating the correlation of the wind speeds measured by any two entity sensors in a rolling manner, selecting the entity sensor with the correlation coefficient smaller than a set threshold value, and determining the entity sensor with the largest occurrence frequency as abnormal.
The specific steps of the variable point grouping method are as follows:
the wind speed sequence at a certain moment is recorded as follows:
Vt＝(v1,t,v2,t,......,vn,t)
in the formula, vn,tRepresenting the wind speed of the nth unit at time t.
The variance of each wind speed at the moment t is calculated as:
in the formula, StRepresents the sequence VtVariance of vt meanRepresents the sequence VtThe average wind speed of (2).
Further, the variance of the wind speed sequence in the time sequence to be diagnosed is respectively obtainedTo St1、St2……StNThe subscripts t1, t2 … … tN denote the sample times within the time series to be diagnosed, St1、St2……StNAccording to ascending order, the embodiment adopts a quartile algorithm to determine the variable point index parameter threshold, wherein the quartile refers to that one ordered data sample is evenly divided into four parts, and the numerical values of the positions of three dividing points are an upper quartile, a middle quartile and a lower quartile which are respectively marked as Q1、Q2、Q3. For a sequence sample containing abnormal data, the variance of the abnormal data sequence is far larger than that of normal sequence data, so that the upper quartile is selected as a boundary value for judging the abnormal data, and the data in a time interval with the variance value larger than the upper quartile is judged as the abnormal data.
The upper quartile is selected as a boundary value of the fault, actually, a criterion is formulated by utilizing the characteristic that the separation degree of the wind speed data among the fault inter-diagnosis intra-group units is increased due to fault wind speed data, and the time sequence interval of the fault data can be positioned by analyzing the variance statistic subsection condition of the wind speed data. In practical application, different services have different requirements on data quality, data scale, transmission speed and the like, and the fault boundary value formulation needs to be further perfected and refined by combining with the actual service category requirements.
In the fault mutual diagnosis community, the unit has high wind speed correlation, and the unit sensor with reduced wind speed correlation can be determined as a fault wind speed sensor. And rolling the wind speed correlation coefficients among the computer groups, selecting a plurality of groups of data with the minimum correlation coefficients, and counting the occurrence times of the associated units, wherein the wind speed sensor with the maximum occurrence times is the fault wind speed sensor. The wind speed correlation is determined by a rolling statistical method, namely, the rolling statistical method is to select a continuous wind speed sequence consisting of the point and n-1 wind speed sequences positioned in front of the point in time sequence, and further adopt Pearson coefficient quantization description in a quantization time sequence, wherein the description is as follows:
wherein r isiTo quantify the correlation, X, of the wind speed data measured by two wind speed sensors in time series ii,t、Yi,tCorresponding to the wind speed data X of two wind speed sensors at the sampling time t in the quantitative time sequence ii,avg、Yi,avgThe average value of the wind speed data of the two wind speed sensors in the quantization time sequence i corresponds to, and n is the total number of sampling moments in the quantization time sequence i.
3. Wind speed prediction and setting module
The wind speed prediction and setting model in the wind speed prediction and setting module comprises a wind speed prediction model and a wind speed setting model which are sequentially cascaded, the wind speed prediction model predicts the wind speed of an abnormal entity sensor based on the measured wind speed of an upper wind direction entity sensor with optimal correlation with the wind speed of the abnormal entity sensor in a fault mutual diagnosis community, and the wind speed setting model sets the predicted wind speed based on the measured wind speed of the abnormal entity sensor to determine the final virtual wind speed.
The wind speed prediction model comprises a bidirectional long-short term memory neural network model, the bidirectional long-short term memory neural network model takes the extracted measured wind speed of the optimal associated unit at each moment to form a measured wind speed sequence as input, the wind speed of the fault sensor as a target, the model is trained and tested in an off-line state through historical data, the measured wind speed of a time sequence interval needing to be predicted is input into BiD-LSTM-NN subjected to training and testing, and the output wind speed is the mapping wind speed of the position of the fault sensor.
Because the prediction model has certain errors, in order to improve the safety and reliability of the operation control strategy appointed by the unit depending on the virtual wind speed, the virtual wind speed needs to be set, the set wind speed is the upper limit value of the prediction error, and the mathematical expression of the wind speed setting model is as follows:
wherein v ist virIndicating the set wind speed, v, at time tt mapIndicating the predicted wind speed at time t, vtActual wind speed, ξ, of an anomalous entity sensor representing the prediction at time tMAPEThe MAPE error of the wind speed prediction model is represented, and n represents the total number of moments in the prediction time series.
The virtual-real interaction wind power plant wind speed monitoring system based on the digital twin is used for monitoring wind speed and controlling the operation of the wind turbine generator, when an entity sensor in the entity sensor array is abnormal, the virtual wind speed of a virtual sensor corresponding to the abnormal entity sensor in the virtual sensor array is used for controlling the operation of the wind turbine generator, and when the abnormal entity sensor is recovered to be normal, the wind speed measured by the entity sensor is used for controlling the operation of the wind turbine generator.
In the embodiment, a certain wind power plant in China is taken as a research object, original wind measurement data of 18 sets of wind speed sensors in 6 days, including 25 minutes in 17 pm of 2 months and 25 months in 2017, 3 months and 3 months in 3, 25 minutes in 17 pm of the embodiment, are randomly selected, the time resolution of the data is 10min, the number of samples of a single set is 864, the prevailing wind direction is northeast wind, and the relative wind direction angle alpha is 112-139 degrees (the clockwise included angle between the incoming wind direction and the righteast direction) in a selected time period. The embodiment completes construction of a relevant mapping model based on a PyTorch platform.
The selected 18 machine groups can be divided into { A11, A12, A13}, { A21, A22, A23, A24}, { A31, A32, A33, A34, A35}, { A41, A42, A43} and { A51, A52, A53}5 fault inter-diagnosis communities according to a fault inter-diagnosis community division method, and under the current prevailing wind direction, the wind speed spatial correlation among the machine groups in each community is kept high, and the fault inter-diagnosis community division result is shown in FIG. 3.
In the embodiment, one unit of the fault inter-diagnosis community is arbitrarily selected as a research object, and communities { A21, A22, A23 and A24}4 units are selected to be further researched. FIG. 4 shows the original wind speeds collected by the A21, A22, A23 and A24 wind turbine. It can be seen that the trend of the wind speed of 4 units changes in most of the time periodThe relevance remains high. There are also cases where the trend of change differs in a part of the time period, for example: the measured wind speed of the unit A24 has the condition of wind speed data missing before and after the 432 th time sequence point, and the measured value is displayed as 0; the measured wind speed of the unit A23 is different from the wind speed variation trend of other units before and after the 850 th time sequence point, and the wind speed correlation is low. In order to find the distribution characteristics of the wind speeds of the units, the wind speed separation degree between 4 units is calculated by using a variance-based variable point algorithm, a time sequence interval with a higher separation degree is marked, and the rolling statistical result of the wind speed variance between the units is shown in fig. 5. Quartile number of Q1＝0.377、Q2＝0.545、Q30.922, upper quartile Q3The data in the region above the red line, which is the discrimination boundary of the failure time-series section, is discriminated as abnormal data.
As mentioned above, under the prevailing wind direction, 4 units in the fault mutual diagnosis community keep higher wind speed spatial correlation, if a certain unit sensor in the fault mutual diagnosis community has a fault, the measured wind speed data has an abnormal value, the abnormal wind speed deviates from a normal value, so that the wind speed deviation degree between the units is increased suddenly, the wind speed variance is increased, and the correlation obtained by calculation according to the measured wind speed is reduced remarkably, therefore, the wind speed variance is used as the judgment basis of the fault state of the sensor, actually, the multivariate statistical distribution problem of the wind speed data under the normal state and the abnormal state is excavated by means of a statistical method, the normal-fault state change time point can be accurately identified based on the variable point quartile algorithm, the fault data can be accurately identified from the time scale, when the prevailing wind direction does not change in the unit in the same fault mutual diagnosis community, and when the wind speed variance is larger than a certain range, judging that the wind speed data in the time sequence interval possibly has an abnormal value, and judging that the wind speed sensor possibly operates in a fault mode and the collected data is inaccurate. Marking and screening out the wind speed variance larger than Q3The time sequence interval of (2), and determining that abnormal wind speed data may exist in the time sequence interval, for example: interval 349-380]、[422-441]、[804-852]And if the variance of the internal wind speed exceeds a set threshold value, preliminarily judging that abnormal data exist in the three intervals.
The variance-based variable point algorithm can determine a time sequence interval in which abnormal data are distributed, and a multi-machine correlation cross comparison method is adopted in the text for further positioning the spatial position of the wind speed sensor corresponding to the abnormal data. The wind speed correlation coefficients of 4 units in the fault inter-diagnosis community in the fault time sequence interval are calculated by rolling, and the rolling calculation results are shown in (a), (b) and (c) of FIG. 6 in sequence, taking three fault time sequence intervals of [ 349-. The correlation coefficient obtained in fig. 6 is a rolling calculation coefficient, the coefficient at each time point is calculated from the wind speed sequence formed by the first 14 points, taking the correlation coefficient of the #21 and #22 units at the 360 th time point in (a) of fig. 6 as an example, the rolling calculation method is to calculate the pearson coefficients of the two units at the 346 th and 360 th wind speed sequences, that is, the so-called rolling calculation.
As is apparent from fig. 6, the wind speed correlation between the 4 units is generally divided into two different levels, the wind speed correlation of the partial units is high, and the wind speed correlation of the partial units is also relatively low. In the timing interval 349-380]And [804-]Internal and external units A21、A22And A24The wind speed correlation between the units is kept at a relatively high level and the variation trends are similar, while the unit A23The wind speed correlation with other units is kept at a relatively low level, the correlation coefficient is less than 0 in most time periods, and the unit A is preliminarily judged23Wind speed sensor in time sequence interval 349-380]And [804-]Fault operation; at timing interval [422-]Set A21、A22And A23The wind speed correlation between the units is kept at a relatively high level, the variation trends are similar, and the unit A24The correlation between the wind speed and other units is kept at a relatively low level, the correlation coefficient is less than 0.4 in most time periods, and the unit A is preliminarily determined24The wind speed sensor operates in a fault.
The wind speed correlation among the unit groups within the 3 failure time sequence sections, the normal time sequence section (removing the wind speed correlation of the failure section), and the global wind speed correlation of the entire time sequence section were calculated, and the calculation results are shown in table 1.
TABLE 1 wind speed correlation coefficient
The wind speed dependency of the table represents the global wind speed dependency coefficient in the corresponding time sequence interval, denoted A21-A22At 349-]For example, the interval is calculated by selecting the wind speed sequences of two units #21 and #22 from 349 th time sequence point to 380 time sequence points and calculating according to the Pearson coefficient formula.
From table 1, in the three fault time sequence intervals, the correlation between the wind speed data acquired by the fault sensor and the wind speed data of other units is relatively low; calculating the correlation (normal interval) of the residual wind speed data after the fault wind speed is eliminated, wherein the wind speed correlation between the units is at a relatively high level, and the correlation coefficients are all above 0.85; the global wind speed correlation between the groups ([0-864 ]]) The distribution condition of the wind speed correlation is similar to that of a fault interval, the wind speed correlation between the normal operation units is kept at a higher level, and the correlation between the unit with abnormal data and other units is obviously lower. Based on this, the unit A is judged24In the time sequence interval [422-]A fault occurs, and measurement data is lost; unit A23In the interval 349-380]And [804-]Failure occurs and the measurement data is misaligned.
For unit A24The fault type is measured data loss, the fault starting time is 05 minutes at 17 pm in 2/28/2017, the fault duration is 190min, the fault is recovered to be normal at 15 minutes at 20 pm in 2/28/2017, the fault causes may be server crash and communication interruption, the real-time data collected by the sensor cannot be normally uploaded to the SCADA system, for the fault of the type, the fault diagnosis module of the SCADA system can quickly identify and locate the fault, the server is restarted and tested at a remote end, and the like, the server is recovered to be normal, the fault is relieved, and the data is recovered to be normal.
For unit A23The two faults occurred are similar in type, and the fault is represented by measurement data lower than that of the faultThe normal value is that the first fault starting time is 35 minutes at 03 morning of 28 days of 2 months in 2017, the fault duration time is 310min, the fault starting time is 45 minutes at 08 morning of 28 days of 2 months in 2017, the fault starting time is 25 minutes at 05 morning of 3 months in 2017, the fault duration time is 480min, the fault starting time is 25 minutes at 13 afternoon of 3 months in 2017, the fault monitoring system is searched for running records, fault alarming and action indication do not occur in the running period, the prevailing wind direction does not change in a large range, the unit does not operate in a yawing mode, the sensor is in a fault critical running state at the moment, the running state is between the normal state and the fault state, the sensor equipment is not damaged destructively, the fault does not cause the sensor to stop running, the unit is in a non-healthy running state, and the measured data is inaccurate. For example: partial sensor transmission bearing lubricating oil does not change or scribble the inequality for a long time, performance is attenuated gradually along with the increase of operating time, when the bad low temperature weather such as temperature dip, rainstorm, hail appear, lubricating oil performance worsens sharply, can't play abundant lubrication and guard action, transmission bearing appears the jam phenomenon, cause the data of gathering to be less than the normal value, wait that the weather condition resumes normally, lubricating oil performance resumes gradually, bearing transmission resistance reduces, sensor measured data keeps normal level or skew less, the trouble is relieved.
In order to repair and reconstruct abnormal data and avoid misoperation of a unit, the honey system establishes a bidirectional long-time and short-time neural network mapping model based on multi-machine association. The data set is divided into training data, test data and mapping data. The training data and the testing data are measured wind speed data of a plurality of wind direction units which are adjacent in time sequence and located in front of the abnormal data, and the training set data and the testing set data are as follows: and 3, dividing the proportion, and setting a training set and a test data set to train and adjust parameters of the model so as to prevent the model from being over-fitted in the mapping process. And selecting the measured wind speeds of other normal upwind related units in the fault time sequence interval by the mapping input parameters, and taking the mapping wind speed of the fault sensor as target data.
The dynamic screening of the associated units is shown in fig. 7, wherein (a) - (c) in fig. 7 correspond to the dynamic screening results of the associated units in the time sequence intervals [ 349-. In time sequence intervals [ 349-; in the time sequence interval [ 422-.
Inputting the actually measured wind speed of the associated unit into the two-way long-and-short-time neural network, calculating the virtual wind speed, wherein the mapping result is shown in fig. 8, and (a) - (c) in fig. 8 correspond to the abnormal entity sensor wind speed hidden-shooting results of the time sequence intervals [ 349-.
In order to test the error of the mapping model, the wind speed data in the normal period is selected to be input into the model, the average absolute percentage error MAPE of the mapping result is calculated, the error of the mapping model is calculated by using a test set, and the calculation result is 8.50%.
The wind speed is set based on the mapping error, the set virtual wind speed is shown in fig. 9, and as can be seen by comparing fig. 4 and fig. 9, the set wind speed data can keep higher consistency in spatial distribution, the spatial correlation of the wind speed among the units is obviously improved, the wind speed data does not have the conditions of deletion and mutation, most fault data can be repaired by applying the method provided by the invention, and the mapped virtual wind speed can reflect the spatial distribution characteristic of the wind speed.
And for the abnormal entity sensor, the set virtual wind speed is input into a unit control system instead of the measured wind speed in the fault interval, a corresponding optimized control strategy is formulated, and after the fault is eliminated in the maintenance window period, the original operation logic is recovered after the fault is eliminated through virtual-actual measurement comparison. In addition, the virtual sensor and the physical sensor are mutually verified in the operation process, under the normal operation state of the sensors, the virtual sensor inputs actual measurement data, and mapping model parameters and a mutual diagnosis community are dynamically adjusted and updated, so that the mapping result of the virtual sensor and the operation working condition of the physical sensor keep higher consistency, and the mapping error is minimum; when the physical sensor breaks down, the virtual sensor can be quickly and accurately positioned from a space-time angle, the virtual wind speed of the fault position is calculated in real time by using the trained mapping model and is input into the unit control system. The virtual sensor is actually a digital twin body of the physical sensor, multi-dimensional mapping and accurate identification of the real-time running state of the physical sensor are realized on the basis of an artificial intelligence algorithm and a data sharing fusion technology, real-time sensing and intelligent control of the running state of the wind power plant can be realized on the basis of a virtual-physical wind power plant synchronous running monitoring framework, and the safety, stability and economy of unit running are improved.
The above embodiments are merely examples and do not limit the scope of the present invention. These embodiments may be implemented in other various manners, and various omissions, substitutions, and changes may be made without departing from the technical spirit of the present invention.
Claims (10)
1. A virtual-real interaction wind power plant wind power monitoring system based on digital twinning is characterized by comprising:
an array of solid sensors: the system is used for acquiring wind power data of each wind turbine generator in the wind power plant in real time;
virtual sensor array: the sensor array is matched with the entity sensors in the entity sensor array one by one;
an interactive platform: the system is used for carrying out association mapping on the entity sensor and the virtual sensor, and comprises abnormity diagnosis on wind data of the entity sensor, and prediction and setting on virtual wind speed of the virtual sensor.
2. The system for monitoring wind speed of a virtual-real interactive wind farm based on digital twinning as claimed in claim 1, wherein the interactive platform comprises:
a fault mutual diagnosis community division module: the system comprises a fault mutual diagnosis community, a fault mutual diagnosis community and a fault mutual diagnosis community, wherein the fault mutual diagnosis community is used for dividing entity sensors in an entity sensor array into a plurality of fault mutual diagnosis communities;
an abnormal state diagnostic module: a plurality of abnormal state diagnosis modules are arranged, one abnormal state diagnosis module corresponds to one fault mutual diagnosis community, abnormal state diagnosis is carried out in each fault mutual diagnosis community based on the measurement data of each entity sensor, and the entity sensor with the abnormality is determined;
a wind speed prediction and setting module: the method comprises the steps that a plurality of wind speed prediction and setting modules are arranged, one wind speed prediction and setting module corresponds to one fault mutual diagnosis community, measured wind speeds of entity sensors are directly related to virtual sensors for entity sensors without abnormity, and wind speed prediction and setting models of the virtual sensors are established for predicting and setting the virtual wind speeds for the entity sensors with abnormity.
3. The system for monitoring wind speed in a virtual-real interaction wind power plant based on digital twins as claimed in claim 2, wherein the fault mutual-diagnosis community division module screens the wind speed spatial correlation of the units based on the unit yaw data of the unit corresponding to the entity sensor, screens a plurality of units with higher wind speed spatial correlation as associated units and divides the associated units into a fault mutual-diagnosis community, and further divides the corresponding entity sensor into a fault mutual-diagnosis community.
4. The system for monitoring wind speed of a virtual-real interactive wind farm based on digital twinning as claimed in claim 2, wherein the abnormal condition diagnosis module comprises:
wind speed abnormal period determination unit: determining a time interval with wind speed abnormal data based on the wind speed time sequence flow of the entity sensors in the fault mutual diagnosis community;
an abnormality sensor positioning unit: and locating the entity sensor with abnormality based on cross comparison of wind speed correlation in a time interval with wind speed abnormality data.
5. The system for monitoring wind speed of a virtual-real interactive wind farm based on digital twinning as claimed in claim 4, wherein the manner of the wind speed abnormal period determining unit determining the time interval with wind speed abnormal data comprises:
acquiring the wind speed variance measured by each entity sensor in the fault mutual diagnosis community at each sampling time point;
determining a variable point variance threshold value based on a quantile algorithm by taking the wind speed variance as a variable point index for determining abnormity;
and determining the time interval when the wind speed variance is larger than the variable point variance threshold value as the time interval when the wind speed abnormal data exists.
6. The system for monitoring wind speed of a virtual-real interactive wind farm based on digital twin according to claim 4, wherein the manner of locating the entity sensor with abnormality by the abnormality sensor locating unit comprises:
and calculating the correlation of the wind speeds measured by any two entity sensors in a rolling manner, selecting the entity sensor with the correlation coefficient smaller than a set threshold value, and determining the entity sensor with the largest occurrence frequency as abnormal.
7. The system for monitoring wind speed of a virtual-real interactive wind power plant based on digital twinning as claimed in claim 2, wherein the wind speed prediction and setting model comprises a wind speed prediction model and a wind speed setting model which are sequentially cascaded, the wind speed prediction model predicts the wind speed of an abnormal entity sensor based on the measured wind speed of an upwind entity sensor with optimal correlation with the wind speed of the abnormal entity sensor in the fault mutual diagnosis community, and the wind speed setting model sets the predicted wind speed based on the measured wind speed of the abnormal entity sensor to determine the final virtual wind speed.
8. A digital twin based virtual-real interaction wind farm wind speed monitoring system according to claim 7, wherein the wind speed prediction model comprises a two-way long-short term memory neural network model.
9. The system for monitoring wind speed of a virtual-real interaction wind power plant based on digital twins as recited in claim 7, wherein the mathematical expression of the wind speed setting model is as follows:
wherein v ist virIndicating the set wind speed, v, at time tt mapIndicating the predicted wind speed at time t, vtActual wind speed, ξ, of an anomalous entity sensor representing the prediction at time tMAPEThe MAPE error of the wind speed prediction model is represented, and n represents the total number of moments in the prediction time series.
10. An application of the virtual-real interactive wind power plant wind speed monitoring system based on the digital twin as claimed in any one of claims 1 to 9 is characterized in that the virtual-real interactive wind power plant wind speed monitoring system based on the digital twin is used for monitoring wind speed and controlling the operation of a wind power plant, when an entity sensor in an entity sensor array is abnormal, the virtual wind speed of a virtual sensor corresponding to the abnormal entity sensor in the virtual sensor array is used for controlling the operation of the wind power plant, and when the abnormal entity sensor is recovered to be normal, the wind speed measured by the entity sensor is used for controlling the operation of the wind power plant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110987657.1A CN113780375B (en) | 2021-08-26 | 2021-08-26 | Digital twinning-based virtual-real interaction wind power plant wind power monitoring system and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110987657.1A CN113780375B (en) | 2021-08-26 | 2021-08-26 | Digital twinning-based virtual-real interaction wind power plant wind power monitoring system and application |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113780375A true CN113780375A (en) | 2021-12-10 |
| CN113780375B CN113780375B (en) | 2024-03-01 |
Family
ID=78839511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110987657.1A Active CN113780375B (en) | 2021-08-26 | 2021-08-26 | Digital twinning-based virtual-real interaction wind power plant wind power monitoring system and application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113780375B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115373278A (en) * | 2022-10-25 | 2022-11-22 | 北京安帝科技有限公司 | Industrial control service abnormity prediction response system and method based on digital twin drive |
| CN116014901A (en) * | 2023-03-24 | 2023-04-25 | 四川航洋电力工程设计有限公司 | Method for monitoring working state of power transmission and transformation equipment based on digital twin technology |
| CN117272704A (en) * | 2023-11-23 | 2023-12-22 | 湖南华自卓创智能技术有限责任公司 | Digital twin-drive data processing system for multi-source heterogeneous data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3515038A1 (en) * | 2018-01-19 | 2019-07-24 | General Electric Company | Autonomous reconfigurable virtual sensing system for cyber-attack neutralization |
| US20210110262A1 (en) * | 2019-10-14 | 2021-04-15 | Honda Research Institute Europe Gmbh | Method and system for semi-supervised deep anomaly detection for large-scale industrial monitoring systems based on time-series data utilizing digital twin simulation data |
| CN112684379A (en) * | 2020-11-25 | 2021-04-20 | 江苏科技大学 | Transformer fault diagnosis system and method based on digital twinning |
-
2021
- 2021-08-26 CN CN202110987657.1A patent/CN113780375B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3515038A1 (en) * | 2018-01-19 | 2019-07-24 | General Electric Company | Autonomous reconfigurable virtual sensing system for cyber-attack neutralization |
| US20210110262A1 (en) * | 2019-10-14 | 2021-04-15 | Honda Research Institute Europe Gmbh | Method and system for semi-supervised deep anomaly detection for large-scale industrial monitoring systems based on time-series data utilizing digital twin simulation data |
| CN112684379A (en) * | 2020-11-25 | 2021-04-20 | 江苏科技大学 | Transformer fault diagnosis system and method based on digital twinning |
Non-Patent Citations (1)
| Title |
|---|
| 胡瑾秋;郝笑笑;张来斌;: "基于虚拟传感技术的工业数据错误诊断方法", 仪器仪表学报, no. 03 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115373278A (en) * | 2022-10-25 | 2022-11-22 | 北京安帝科技有限公司 | Industrial control service abnormity prediction response system and method based on digital twin drive |
| CN115373278B (en) * | 2022-10-25 | 2023-01-06 | 北京安帝科技有限公司 | Industrial control service abnormity prediction response system and method based on digital twin drive |
| CN116014901A (en) * | 2023-03-24 | 2023-04-25 | 四川航洋电力工程设计有限公司 | Method for monitoring working state of power transmission and transformation equipment based on digital twin technology |
| CN117272704A (en) * | 2023-11-23 | 2023-12-22 | 湖南华自卓创智能技术有限责任公司 | Digital twin-drive data processing system for multi-source heterogeneous data |
| CN117272704B (en) * | 2023-11-23 | 2024-01-26 | 湖南华自卓创智能技术有限责任公司 | Digital twin-drive data processing system for multi-source heterogeneous data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113780375B (en) | 2024-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113780375B (en) | Digital twinning-based virtual-real interaction wind power plant wind power monitoring system and application | |
| CN105205569B (en) | State of fan gear box online evaluation method for establishing model and online evaluation method | |
| CN112462736B (en) | Wind turbine generator fault diagnosis method based on data analysis | |
| CN112834211A (en) | Fault early warning method for transmission system of wind turbine generator | |
| CN111946559A (en) | Method for detecting structures of wind turbine foundation and tower | |
| CN115329812A (en) | Road infrastructure abnormity monitoring method based on artificial intelligence | |
| CN112065668A (en) | Wind turbine generator state abnormity assessment method and system | |
| CN114519923A (en) | Intelligent diagnosis and early warning method and system for power plant | |
| CN110033102A (en) | A kind of huge hydroelectric power plant has the intelligent diagnosing method and expert system of learning functionality | |
| CN113777351B (en) | Fault diagnosis method and device for wind speed sensor of wind power plant | |
| CN116609055A (en) | Method for diagnosing wind power gear box fault by using graph convolution neural network | |
| CN117728587B (en) | Real-time monitoring system and method for operation data of new energy power generation equipment | |
| CN117365869A (en) | Self-adaptive early warning strategy design method for wind turbine blade tower sweeping faults | |
| CN117649044A (en) | Multi-dimensional quantitative state evaluation method for offshore wind turbine generator | |
| CN117674249A (en) | Distributed photovoltaic-containing power distribution network fault self-healing control and evaluation method | |
| CN111192163A (en) | Generator reliability medium-short term prediction method based on wind turbine generator operating data | |
| CN110794683A (en) | Wind power gear box state evaluation method based on deep neural network and kurtosis characteristics | |
| CN114320773B (en) | Wind turbine generator system fault early warning method based on power curve analysis and neural network | |
| CN114167837B (en) | Intelligent fault diagnosis method and system for railway signal system | |
| CN114442543A (en) | Computer monitoring method suitable for early warning of hydropower station fault | |
| CN114154567A (en) | Wind power plant station operation data anomaly identification method based on machine learning | |
| CN113036917A (en) | Power distribution network monitoring information monitoring system and method based on machine learning | |
| CN112884170A (en) | Predictive intelligent operation and maintenance system and method for comprehensive pipe gallery | |
| CN114151291B (en) | Early fault monitoring method for wind turbine generator | |
| Anderson et al. | On the sensitivity of wind turbine failure rate estimates to failure definitions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |