CN110489852A - Improve the method and device of the wind power system quality of data - Google Patents
Improve the method and device of the wind power system quality of data Download PDFInfo
- Publication number
- CN110489852A CN110489852A CN201910750791.2A CN201910750791A CN110489852A CN 110489852 A CN110489852 A CN 110489852A CN 201910750791 A CN201910750791 A CN 201910750791A CN 110489852 A CN110489852 A CN 110489852A
- Authority
- CN
- China
- Prior art keywords
- data
- parameter
- error
- time
- type
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Wind Motors (AREA)
Abstract
The invention discloses a kind of method and devices for improving the wind power system quality of data, which comprises determines the wrong data and its type of error in the system data of acquisition；The system data includes: the data that each time point is corresponded in certain period of time, and the data of each time point include the data segment of one or more corresponding different sensors；The wrong data in the system data is handled according to the type of error, the system data that obtains that treated；Based on mechanism model, treated that system data is modified to described, obtains revised system data.Using the present invention, the quality of the system data of wind power system acquisition can effectively improve.
Description
Technical field
The present invention relates to data processing fields, and in particular to a kind of method and device for improving the wind power system quality of data.
Background technique
Wind-powered electricity generation domain prediction maintenance system is built from terminal data acquisition, algorithm model, again to predictive maintenance, is formed
Lifecycle management system, is wind-power electricity generation fault pre-alarming, wind-power electricity generation stable operation, safety grid-connection, save O&M at
This grade direction is each provided with technological guidance.
However running of wind generating set operating condition is complicated and changeable, set state monitoring data amount is big, due to compressor emergency shutdown, off-load,
The factors such as communication noise and equipment fault, can generate a large amount of abnormal datas, such as collected source data duplicate for a long time value,
The invalid datas such as zero or collected data do not meet physics law etc., seriously affect wind-powered electricity generation predictive maintenance model essence
Degree, is reported by mistake so as to cause model pre-warning result, the situations such as fails to report.Simultaneously as wind-powered electricity generation domain-specialist knowledge and algorithm are built
The knowledge of mould personnel mismatches, and abnormal data when model modeling is caused to screen not exclusively, increases Early-warning Model modeling difficulty.
To sum up, in the prior art to the collection of running of wind generating set data, management, analysis and excavation there are still it is many not
Foot, is unable to the quality difference of the acquired data of accurate recognition, so cannot effectively support the correct screening of asperity data with it is reasonable
Optimization, so that the quality of data cannot ensure.If these data are directly used without processing, can wind-power electricity generation be counted
Characteristic is distorted, and then will affect the operating status of Wind turbines and the predictive maintenance result of operation characteristic.
Summary of the invention
The embodiment of the present invention provides a kind of method and device for improving the wind power system quality of data, can be to the system of acquisition
Data are effectively treated, and improve the quality of source data.
For this purpose, the invention provides the following technical scheme:
A method of improving the wind power system quality of data, which comprises
Determine the wrong data and its type of error in the system data of acquisition；The system data includes: certain time
The data at each time point are corresponded in section, the data of each time point include the data segment of one or more corresponding different sensors；
The wrong data in the system data is handled according to the type of error, the system number that obtains that treated
According to；
Based on mechanism model, treated that system data is modified to described, obtains revised system data.
Optionally, the type of error includes any of the following or a variety of: missing errors, type error, numerical fault,
Rule errors, repetitive error；
It is described according to the type of error to the wrong data in the system carry out processing include any of the following or
It is a variety of:
Data filling is carried out to the data segment that type of error is missing errors；
Data type conversion or deletion are carried out to the data segment that type of error is type error；
The data segment that type of error is numerical fault is deleted；
Data conversion or deletion are carried out to the data segment that type of error is rule errors；
It is the data segment of repetitive error for type of error, if all data segments of corresponding different sensors continue weight
It is more than again setting time, then deletes duplicate data in the setting time；If type of error is the data segment of repetitive error
For wind speed parameter or wind direction parameter or temperature parameter, then duplicate data segment is deleted.
Optionally, described to include: to the data segment progress data filling that type of error is missing errors
If type of error is that the data segment of missing errors is wind speed parameter or power parameter, according to wind power module into
Row data filling；
Otherwise, data filling is carried out by interpolation.
It is optionally, described that based on mechanism model, treated that system data is modified includes: to described
Wind speed parameter abnormal in treated the system data is determined using ambient wind model, and deletes the exception
Wind speed parameter；
The power parameter in treated the system data is modified using wind power module；
Temperature parameter abnormal in treated the system data is determined using modeling of energy conservation, and is deleted described different
Normal temperature parameter.
Optionally, described that packet is modified to the power parameter in treated the system data using wind power module
It includes:
Whether the data that each time point is examined successively meet wind power module；
If the data of current point in time do not meet wind power module, joined according to the wind speed in the data of current point in time
Power parameter in several and fan condition code parameters revision current point in time data.
Optionally, when wind speed parameter and current fan condition code parameters revision in the data according to current point in time
Between power parameter in the data put include:
If the wind speed parameter in the data of current point in time is less than incision wind speed, will be in the data of current point in time
Power parameter is revised as 0；
If the wind speed parameter in the data of current point in time is greater than incision wind speed and is less than rated wind speed, will be current
Power parameter in the data at time point is deleted；
If the wind speed parameter in the data of current point in time is greater than rated wind speed, check in the data of current point in time
Fan condition code whether be power limitation code；
If it is, the power parameter in the data of current point in time is revised as power limitation；
Otherwise, the power parameter in the data of current point in time is revised as full hair power.
Optionally, the method also includes:
The abnormal data in the revised system data is determined based on mathematical model；
Remove the abnormal data, the system data after being optimized.
Optionally, described to determine that the abnormal data in the revised system data includes: based on mathematical model
The abnormal data in the revised system data is determined based on any one or more following mathematical model: poly-
Class model, Remanent Model and cluster models.
Optionally, the Clustering Model includes any of the following or a variety of: k-means model, DBSCAN model；
The input of the Clustering Model is the revised system data, is exported as the corresponding data class of different time points
Other and its quantity.
Optionally, the Remanent Model includes any of the following or a variety of: linear regression, support vector machines, decision tree,
Neural network；
The input of the Remanent Model is the revised system data, is exported as the corresponding residual error of different time points.
Optionally, determine that the abnormal data in the revised system includes: based on group system
Using blower fan system as a cluster models, each column for all blowers for including in each model in cluster models are judged
Whether the difference of data mean parameter corresponding with wind field is more than given threshold, if it exceeds given threshold, it is determined that the columns
Data according to affiliated time point are abnormal data.
A kind of device improving the wind power system quality of data, described device include:
Wrong data detection module, the wrong data and its type of error in system data for determining acquisition；It is described
System data includes: the data that each time point is corresponded in certain period of time, and the data of each time point include one or more right
Answer the data segment of different sensors；
Data processing module, for being handled according to the type of error the wrong data in the system data,
The system data that obtains that treated；
Data correction module is corrected for based on mechanism model, treated that system data is modified to described
System data afterwards.
Optionally, the type of error includes any of the following or a variety of: missing errors, type error, numerical fault,
Rule errors, repetitive error；
The data processing module, specifically for the wrong data in the system carry out it is following any one or more
Processing:
Data filling is carried out to the data segment that type of error is missing errors；
Data type conversion or deletion are carried out to the data segment that type of error is type error；
The data segment that type of error is numerical fault is deleted；
Data conversion or deletion are carried out to the data segment that type of error is rule errors；
If all data segments of corresponding different sensors are persistently repeated more than setting time, when deleting the setting
Interior duplicate data；
If the data segment that type of error is repetitive error is wind speed parameter or wind direction parameter or temperature parameter, delete
Duplicate data segment.
Optionally, the data processing module carries out data to the data segment that type of error is missing errors in the following manner
It fills up:
If type of error is that the data segment of missing errors is wind speed parameter or power parameter, according to wind power module into
Row data filling；
Otherwise, data filling is carried out by interpolation.
Optionally, the data correction module includes:
Wind speed parameter amending unit, for determining wind abnormal in treated the system data using ambient wind model
Fast parameter, and delete the abnormal wind speed parameter.
Power parameter amending unit, for using wind power module to the power parameter in treated the system data
It is modified；
Temperature parameter amending unit, it is abnormal in treated the system data for being determined using modeling of energy conservation
Temperature parameter, and delete the abnormal temperature parameter；
Optionally, the power parameter amending unit includes:
Check subelement, whether the data for each time point to be examined successively meet wind power module；
Revise subelemen, for determining that the data of current point in time do not meet wind power module in the inspection subelement
Afterwards, according to the function in the data of wind speed parameter and fan condition code parameters revision current point in time in the data of current point in time
Rate parameter.
Optionally, the revise subelemen is less than incision specifically for the wind speed parameter in the data of current point in time
When wind speed, the power parameter in the data of current point in time is revised as 0；Wind speed parameter in the data of current point in time is big
When cutting wind speed and being less than rated wind speed, the power parameter in the data of current point in time is deleted；In current point in time
Data in wind speed parameter be greater than rated wind speed when, check whether the fan condition code in the data of current point in time is restriction
Power code；If it is, the power parameter in the data of current point in time is revised as power limitation；Otherwise, by current time
Power parameter in the data of point is revised as full hair power.
Optionally, described device further include:
Anomaly data detection module, for determining the abnormal number in the revised system data based on mathematical model
According to；
Abnormal data cleaning modul, the system data for removing the abnormal data, after being optimized.
Optionally, the anomaly data detection module includes any of the following or multiple module:
First detection module, for detecting the abnormal data in the revised system data based on Clustering Model；
Second detection module, for detecting the abnormal data in the revised system data based on Remanent Model；
Third detection module, for detecting the abnormal data in the revised system data based on cluster models.
Optionally, the Clustering Model includes any of the following or a variety of: k-means model, DBSCAN model；
The input of the Clustering Model is the revised system data, is exported as the corresponding data class of different time points
Other and its quantity.
Optionally, the Remanent Model includes any of the following or a variety of: linear regression, support vector machines, decision tree,
Neural network；
The input of the Remanent Model is the revised system data, is exported as the corresponding residual error of different time points.
Optionally, the third detection module is specifically used for judging cluster mould using blower fan system as a cluster models
Whether the difference of each column data for all blowers for including in each model in type mean parameter corresponding with wind field is more than setting threshold
Value, if it exceeds given threshold, it is determined that the data at the column data affiliated time point are abnormal data.
A kind of electronic equipment, comprising: one or more processors, memory；
For the memory for storing computer executable instructions, the processor is executable for executing the computer
Instruction, to realize mentioned-above method.
A kind of readable storage medium storing program for executing, is stored thereon with instruction, and described instruction is performed to realize mentioned-above method.
The method and device provided in an embodiment of the present invention for improving the wind power system quality of data, for the wind power system of acquisition
Data determine wrong data therein and type of error, are carried out first according to different type of errors to wrong data therein
Processing, the system data that obtains that treated；Then data are carried out based on mechanism model and are repaired for treated system data again
Just, revised system data is obtained.The scheme of the embodiment of the present invention is not only to caused mistake in data acquisition and transmit process
Accidentally data are handled, and the characteristics of according to wind power system data, are modified based on mechanism model to system data, thus
Make finally obtained system data that there is higher quality, efficiently avoids low-quality source data to follow-up system O&M pipe
The work such as reason generate interference, guarantee the normal operation of Wind turbines.
Further, also abnormal data inspection can be carried out to revised system data based on one or more mathematical models
It surveys, and removes the abnormal data detected, advanced optimize system data, can better ensure that the matter of system data
Amount.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only one recorded in the present invention
A little embodiments are also possible to obtain other drawings based on these drawings for those of ordinary skill in the art.
Fig. 1 is a kind of flow chart for the method that the embodiment of the present invention improves the wind power system quality of data；
Fig. 2 is the process being modified using wind power module to the power parameter in system data in the embodiment of the present invention
Figure；
Fig. 3 is another flow chart for the method that the embodiment of the present invention improves the wind power system quality of data；
Fig. 4 is that the present invention implements to improve a kind of structural block diagram of the device of the wind power system quality of data；
Fig. 5 is a kind of structural block diagram of data correction module in the embodiment of the present invention；
Fig. 6 is another structural block diagram for the device that the embodiment of the present invention improves the wind power system quality of data.
Specific embodiment
The scheme of embodiment in order to enable those skilled in the art to better understand the present invention with reference to the accompanying drawing and is implemented
Mode is described in further detail the embodiment of the present invention.
The embodiment of the present invention provides a kind of method and device for improving the wind power system quality of data, for the wind-powered electricity generation system of acquisition
Unite data, determine wrong data therein and type of error, first according to different type of errors to wrong data therein into
Row processing, the system data that obtains that treated；Then data are carried out based on mechanism model and are repaired for treated system data again
Just, revised system data is obtained.
As shown in Figure 1, be a kind of flow chart for the method that the embodiment of the present invention improves the wind power system quality of data, including with
Lower step:
Step 101, the wrong data and its type of error in the system data of acquisition are determined.
The system data includes: the data that each time point is corresponded in certain period of time, and the data of each time point include
The data segment of the corresponding different sensors of one or more.
Since wind power system unit is numerous, operating status is different, it is therefore desirable to which the data volume for monitoring acquisition is very big；In addition exist
It inevitably will receive the influence of the factors such as noise jamming in data transfer procedure, so that some data hair in the system data of acquisition
Raw mistake, generates wrong data.Due to generating wrong data, the differences such as object, the mistake of different wrong data
Type would also vary from.
In practical applications, the wrong data and its wrong class in the system data can be determined using the prior art
Type.Wherein, the type of error mainly include the following types:
Missing errors, the i.e. data of some data segment are sky；
Type error, the i.e. data type of some data segment are not consistent with the actual type of the data segment, such as certain data
The numerical value of section should be 12.5, but be registered as character string ' 12.5 '；
Numerical fault, the i.e. numerical value of some data segment not within the specified scope, if ambient temperature value is registered as 100 DEG C,
Obviously convention is not met；
Rule errors, i.e. some data segment do not meet setting rule, should be integer, the wind between 0-n such as fan condition code
Machine start and stop state should be the rules such as 0 or 1；
Repetitive error, i.e. some data generate repetition within a certain period of time.
Certainly, according to the actual application, there can also be other type of errors, without limitation to this embodiment of the present invention.
In addition, in order to it is more acurrate, comprehensively determine wrong data therein, the embodiment of the invention also provides a kind of determinations
The method of wrong data and its type of error, this method not only carry out it for the system data acquired in certain period of time
Laterally detection is detected for the data at each time point, but also carries out longitudinal detection to it, that is, be directed to certain period of time
Interior each data segment is detected.This method specifically includes: laterally detection and longitudinal detection；Wherein:
Laterally detection referred to as unit of time point, successively detects to the data at each time point, marks the data
The position of middle abnormal data section, and misregistration type；
Longitudinal detection refers to as unit of data segment, successively detects to each data segment in the certain period of time,
And recording exceptional data section and type of error.
Wherein, detection is carried out to the data at each time point in the laterally detection to specifically include that
Whether detect has missing data section in the data；If so, then marking the position of missing data section, and record mistake
Accidentally type are as follows: missing errors；
Whether the type for detecting each data segment in the data is correct；If incorrect, the position of the data segment is marked
It sets, and misregistration type are as follows: type error；
Whether within the specified scope to detect the numerical value of each data segment in the data；If it is not, then marking the data
The position of section, and misregistration type are as follows: numerical fault；
Detect whether each data segment in the data meets setting rule；If do not met, the position of flag data section,
And misregistration type are as follows: rule errors.
Carrying out detection to each data segment in the certain period of time in longitudinal detection includes:
Repeatability of the data segment in the certain period of time is detected, and marks repeated data section, misregistration class
Type are as follows: repetitive error；
Continuity of the data segment in the certain period of time is detected, and marks the data segment of missing, misregistration
Type are as follows: missing errors.
Step 102, the wrong data in the system data is handled according to the type of error, after obtaining processing
System data.
For the wrong data of different type of errors, different processing modes can be used.In embodiments of the present invention, needle
To above-mentioned several type of errors, following processing mode can be respectively adopted:
(1) data filling is carried out to the data segment that type of error is missing errors, specifically there are following several situations:
If certain row data (the i.e. corresponding data sometime put) or certain column data (i.e. corresponding one in the system data
Some data segment fixed time in section) all lack, then the row is directly deleted without filling up to the row data or the column data
Or the column.
If the individual data segments missing in certain row or certain column, needs the system according to representated by the data segment of missing to join
Number difference takes different measures of filling up, such as: if the data segment that type of error is missing errors is wind speed parameter or power
Parameter then can carry out data filling according to wind power module, that is to say, that corresponding according to time point belonging to missing data section
Other data segments and wind power module, are calculated the numerical value of the missing data section, which is filled up corresponding position；Institute
Stating wind power module is that a kind of Wind turbines wind speed and the functional relation of active power and the factory design parameter of blower itself have
It closes.
If type of error is not wind speed parameter for the data segment of missing errors, nor power parameter, due to wind-powered electricity generation system
Data of uniting would generally carry out data acquisition using high sampling rate, therefore can carry out data filling by interpolation method, specifically may be used
It is filled up with using the forward and backward corresponding data segment of time point belonging to missing data section to carry out interpolation.If when belonging to missing data section
Between put forward and backward corresponding data segment and also lack, then can use random number and carry out interpolation and fill up, wherein random number comes from the data
Within the scope of section normal distribution 3sigma.
(2) data type conversion or deletion are carried out to the data segment that type of error is type error.
(3) data segment that type of error is numerical fault is deleted.
(4) data conversion or deletion are carried out to the data segment that type of error is rule errors.
For example, fan condition code column should be integer, if corresponding data segment is not integer, its value is carried out at rounding
Reason, if corresponding data segment is character string, can not carry out rounding processing, then delete the data segment.
(5) be the data segment of repetitive error for type of error: whether all data segments of the corresponding different sensors of inspection
Setting time (such as 30 minutes) are persistently repeated more than, if it is, deleting duplicate data in the setting time；Otherwise
Check whether the data segment that type of error is repetitive error is wind speed parameter or wind direction parameter or temperature parameter, if it is,
Delete duplicate data segment.
It should be noted that in practical applications, can according to need using any one of the above or a variety of processing modes,
And the processing sequence of different types of wrong data is not required.Certainly, the embodiment of the present invention is also not limited to above-mentioned
These processing, for other type of errors, can also there is corresponding processing mode.
Step 103, based on mechanism model, treated that system data is modified to described, obtains revised system number
According to.
For the data characteristics of wind power system acquisition, for example, wind speed, active power, gear-box oil temperature, box bearing temperature
Degree, generator bearing temperature, cabin temperature, engine room control cabinet temperature etc., should comply with certain relationship on these data theories,
Meet corresponding mechanism model, such as: wind power module, modeling of energy conservation, ambient wind model etc..Therefore, the present invention is implemented
In the method for example, above-mentioned each mechanism model can be utilized respectively, some data segments in the system data are modified, it is main
There are following three kinds of processing modes:
(1) wind speed parameter abnormal in treated the system data is determined using ambient wind model, and described in deleting
Abnormal wind speed parameter.
For example, wind speed is unchanged, but when wind vector is more than 1 degree, then corresponding wind direction parameter is abnormal data, to its into
Row is deleted.
(2) power parameter in treated the system data is modified using wind power module.
It can using the detailed process that wind power module is modified the power parameter in treated the system data
It is shown in Figure 2, comprising the following steps:
Step 201, whether the data that each time point is examined successively meet wind power module；If not, thening follow the steps
202；Otherwise terminate.
Step 202, judge whether the wind speed parameter in the data of current point in time is less than incision wind speed；If it is, holding
Row step 203；Otherwise, step 204 is executed；
Step 203, the power parameter in the data of current point in time is revised as 0；
Step 204, judge whether the wind speed parameter in the data of current point in time is less than rated wind speed；If it is, holding
Row step 205；It is no to then follow the steps 206；
Step 205, the power parameter in the data of current point in time is deleted；
Step 206, check whether the fan condition code in the data of current point in time is power limitation code；If it is,
Execute step 207；It is no to then follow the steps 208；
Step 207, the power parameter in the data of current point in time is revised as power limitation；
Step 208, the power parameter in the data of current point in time is revised as full hair power.
(3) temperature parameter abnormal in treated the system data is determined using modeling of energy conservation, and deletes institute
State abnormal temperature parameter.
For example, modeling of energy conservation is established to the approximate enclosure space such as cabin, control cabinet, gear-box in blower fan system,
When the real data of acquisition does not meet modeling of energy conservation, corresponding data is deleted.
It should be noted that in practical applications, it is above-mentioned that the suitable of processing is modified to system data using different models
Sequence is without limitation.
The method provided in an embodiment of the present invention for improving the wind power system quality of data, for the wind power system data of acquisition,
It determines wrong data therein and type of error, wrong data therein is handled according to different type of errors first,
The system data that obtains that treated；Then again for treated system data, data correction is carried out based on mechanism model, is obtained
Revised system data.The scheme of the embodiment of the present invention not only to data acquisition and transmit process in caused wrong data into
Row processing, and the characteristics of according to wind power system data, is modified system data based on mechanism model, to make final
The system data arrived has higher quality, efficiently avoids low-quality source data to work such as follow-up system operation managements
Interference is generated, guarantees the normal operation of Wind turbines.
As shown in figure 3, be another flow chart for the method that the embodiment of the present invention improves the wind power system quality of data, including
Following steps:
Step 301, the wrong data and its type of error in the system data of acquisition are determined.
Step 302, the wrong data in the system data is handled according to the type of error, after obtaining processing
System data.
Step 303, based on mechanism model, treated that system data is modified to described, obtains revised system number
According to.
Above-mentioned steps 301 are identical to step 103 as the step 101 in prior figures 1 to step 303, and details are not described herein.
Step 304, the abnormal data in the revised system data is determined based on mathematical model.
Specifically, it can be determined in the revised system data based on any one or more following mathematical model
Abnormal data: Clustering Model, Remanent Model and cluster models.
It, can be with it should be noted that if determining abnormal data based on both the above or two or more mathematical models
Using the union of the abnormal data obtained according to different models as final abnormal data, or the exception that different models are obtained
Data carry out fusion calculation (such as weighted calculation), and final abnormal data is determined according to calculated result.
The Clustering Model can specifically include it is following any one or more: k-means model, DBSCAN
(Density-Based Spatial Clustering of Applications with Noise) model；The cluster mould
The input of type is the revised system data, is exported as the corresponding data category of different time points and its quantity.
If certain categorical measure is less than setting value (such as 3), it is determined that category data are abnormal data.
The Remanent Model can specifically include it is following any one or more: linear regression, support vector machines, decision
Tree, neural network；The input of the Remanent Model be the revised system data, export for different time points it is corresponding residual
Difference.
If the residual error at certain corresponding time point is more than given threshold, it is determined that all data at the time point are abnormal number
According to.
In wind power plant, a wind field often includes tens Fans, and blower external environment is similar, therefore same wind field
The operating status of different blowers has similitude.Based on this feature, in the embodiment of the present invention, collect blower fan system as one
Group model judges each column data for all blowers for including in each model in cluster models mean parameter corresponding with wind field (ratio
Such as temperature, wind speed, revolving speed, power) difference whether be more than given threshold, if it exceeds given threshold, it is determined that the columns
Data according to affiliated time point are abnormal data.
Step 305, the abnormal data, the system data after being optimized are removed.
The method provided in an embodiment of the present invention for improving the wind power system quality of data, not only acquires data in wind power system
And caused wrong data is handled in transmit process, and the characteristics of according to wind power system data, is based on mechanism model pair
System data is modified, and carries out anomaly data detection to revised system data based on one or more mathematical models, and
The abnormal data detected is removed, system data is advanced optimized, can better ensure that the quality of system data.
Correspondingly, the embodiment of the present invention also provides a kind of device for improving the wind power system quality of data, as shown in figure 4, being
The present invention implements to improve a kind of structural block diagram of the device of the wind power system quality of data.
In this embodiment, described device includes following module:
Wrong data detection module 401, the wrong data and its type of error in system data for determining acquisition；Institute
Stating system data includes: the data that each time point is corresponded in certain period of time, and the data of each time point include one or more
The data segment of corresponding different sensors；
Data processing module 402, for according to the type of error to the wrong data in the system data at
Reason, the system data that obtains that treated；
Data correction module 403 is repaired for based on mechanism model, treated that system data is modified to described
System data after just.
In practical applications, above-mentioned wrong data detection module 401 can use the prior art to determine the system number
Wrong data and its type of error in.Wherein, the type of error mainly include the following types: missing errors, type error,
Numerical fault, rule errors, repetitive error.Certainly, according to the actual application, there can also be other type of errors, to this
Inventive embodiments are without limitation.
The meaning of above-mentioned each type of error has been described in detail in front, and details are not described herein.
In addition, in order to it is more acurrate, comprehensively determine wrong data therein, above-mentioned wrong data detection module 401 may be used also
To determine wrong data and its type of error in the following manner:
Laterally detection: as unit of time point, successively the data at each time point are detected, are marked different in the data
The position of regular data section, and misregistration type；
Longitudinal detection: as unit of data segment, successively each data segment in the certain period of time is detected, and remembers
Record abnormal data section and type of error.
The above-mentioned mode and process for laterally detecting and longitudinally detecting has had a detailed description in front, and details are not described herein.
For the wrong data of different type of errors, the data processing module 402 can use different processing modes.
For example, any one or more following processing can be carried out to the wrong data in the system:
Data filling is carried out to the data segment that type of error is missing errors, for example if type of error is missing errors
Data segment is wind speed parameter or power parameter, then carries out data filling according to wind power module；Otherwise, data are carried out by interpolation
It fills up；
Data type conversion or deletion are carried out to the data segment that type of error is numerical fault；
The data segment that type of error is rule errors is deleted；
If all data segments of corresponding different sensors are persistently repeated more than setting time, when deleting the setting
Interior duplicate data；
If the data segment that type of error is repetitive error is wind speed parameter or wind direction parameter or temperature parameter, delete
Duplicate data segment.
According to the data characteristics that wind power system acquires, above-mentioned data correction module 403 can use a variety of each mechanism models
Some data segments in the system data are modified, for example, a kind of specific structure of the data correction module 403 is such as
Shown in Fig. 5, including following each unit:
Wind speed parameter amending unit 433, it is abnormal in treated the system data for being determined using ambient wind model
Wind speed parameter, and delete the abnormal wind speed parameter；
Power parameter amending unit 431, for using wind power module to the power in treated the system data
Parameter is modified；
Temperature parameter amending unit 432, it is different in treated the system data for being determined using modeling of energy conservation
Normal temperature parameter, and delete the abnormal temperature parameter.
Wherein, the power parameter amending unit includes: inspection unit and revise subelemen to ambient wind model, in which:
It is described to check subelement for being examined successively whether the data of each time point meet wind power module；
The revise subelemen is used to determine that the data of current point in time do not meet wind power mould in the inspection subelement
After type, according in the data of wind speed parameter and fan condition code parameters revision current point in time in the data of current point in time
Power parameter.Specifically, when the wind speed parameter in the data of current point in time is less than incision wind speed, by the number of current point in time
Power parameter in is revised as 0；Wind speed parameter in the data of current point in time is greater than incision wind speed and is less than specified
When wind speed, the power parameter in the data of current point in time is deleted；Wind speed parameter in the data of current point in time is greater than
When rated wind speed, check whether the fan condition code in the data of current point in time is power limitation code；If it is, by current
Power parameter in the data at time point is revised as power limitation；Otherwise, the power parameter in the data of current point in time is repaired
It is changed to full hair power.Specific makeover process can refer to front embodiment illustrated in fig. 2.
The device provided in an embodiment of the present invention for improving the wind power system quality of data, for the wind power system data of acquisition,
It determines wrong data therein and type of error, wrong data therein is handled according to different type of errors first,
The system data that obtains that treated；Then again for treated system data, data correction is carried out based on mechanism model, is obtained
Revised system data.The scheme of the embodiment of the present invention not only to data acquisition and transmit process in caused wrong data into
Row processing, and the characteristics of according to wind power system data, is modified system data based on mechanism model, to make final
The system data arrived has higher quality, efficiently avoids low-quality source data to work such as follow-up system operation managements
Interference is generated, guarantees the normal operation of Wind turbines.
As shown in fig. 6, being another structural block diagram for the device that the embodiment of the present invention improves the wind power system quality of data.
Unlike embodiment illustrated in fig. 4, in this embodiment, described device further includes following module:
Anomaly data detection module 404, for determining the exception in the revised system data based on mathematical model
Data；
Abnormal data cleaning modul 405, the system data for removing the abnormal data, after being optimized.
In practical applications, the anomaly data detection module 404 can determine institute based on one or more mathematical models
The abnormal data in revised system data is stated, for example, a kind of specific structure of the anomaly data detection module 404 can be with
Any one or more including but not limited to following module:
First detection module, for detecting the abnormal data in the revised system data based on Clustering Model；
Second detection module, for detecting the abnormal data in the revised system data based on Remanent Model；
Third detection module, for detecting the abnormal data in the revised system data based on cluster models.
Wherein, the Clustering Model can include but is not limited to it is following any one or more: k-means model,
DBSCAN model；The input of the Clustering Model is the revised system data, is exported as the corresponding number of different time points
According to classification and its quantity.
The Remanent Model can include but is not limited to it is following any one or more: linear regression, support vector machines, certainly
Plan tree, neural network；The input of the Remanent Model be the revised system data, export for different time points it is corresponding
Residual error.
The third detection module is specifically used for judging each mould in cluster models using blower fan system as a cluster models
Whether the difference of each column data for all blowers for including in type mean parameter corresponding with wind field is more than given threshold, if super
Cross given threshold, it is determined that the data at the column data affiliated time point are abnormal data.
It should be noted that for each embodiment of device of the above-mentioned raising wind power system quality of data, due to each mould
Block, the function realization of unit are similar with corresponding method, therefore describe to compare to each embodiment of the dialogue generating means
Simply, related place can be found in the corresponding portion explanation of embodiment of the method.
The device provided in an embodiment of the present invention for improving the wind power system quality of data, not only acquires data in wind power system
And caused wrong data is handled in transmit process, and the characteristics of according to wind power system data, is based on mechanism model pair
System data is modified, and carries out anomaly data detection to revised system data based on one or more mathematical models, and
The abnormal data detected is removed, system data is advanced optimized, can better ensure that the quality of system data.
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to
Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product
Or other step or units that equipment is intrinsic.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Moreover, described above
System embodiment it is only schematical, wherein module and unit can be or can not also as illustrated by the separation member
It is to be physically separated, it can be located in a network unit, or may be distributed over multiple network units.It can root
According to actual need that some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Ordinary skill
Personnel can understand and implement without creative efforts.
Those of ordinary skill in the art will appreciate that all or part of the steps in realization above method embodiment is can
It is completed with instructing relevant hardware by program, the program can store in computer-readable storage medium,
Storage medium designated herein, such as: ROM/RAM, magnetic disk, CD.
Correspondingly, the embodiment of the present invention also provides a kind of for improving the device of the method for the wind power system quality of data, should
Device is a kind of electronic equipment, for example, can be mobile terminal, computer, tablet device, Medical Devices, body-building equipment, individual
Digital assistants etc..The electronic equipment may include one or more processors, memory；Wherein, the memory is for depositing
Computer executable instructions are stored up, the processor is for executing the computer executable instructions, to realize previous embodiments
The method.
The embodiment of the present invention has been described in detail above, and specific embodiment used herein carries out the present invention
It illustrates, method and device of the invention that the above embodiments are only used to help understand, is only the present invention one
The embodiment divided, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, should fall within the scope of the present invention, this specification
Content should not be construed as limiting the invention.Therefore, all within the spirits and principles of the present invention, it is made it is any modification,
Equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (18)
1. a kind of method for improving the wind power system quality of data, which is characterized in that the described method includes:
Determine the wrong data and its type of error in the system data of acquisition；The system data includes: in certain period of time
The data at corresponding each time point, the data of each time point include the data segment of one or more corresponding different sensors；
The wrong data in the system data is handled according to the type of error, the system data that obtains that treated；
Based on mechanism model, treated that system data is modified to described, obtains revised system data.
2. the method according to claim 1, wherein the type of error includes any of the following or a variety of:
Missing errors, type error, numerical fault, rule errors, repetitive error；
It is described processing is carried out to the wrong data in the system according to the type of error to include any of the following or a variety of:
Data filling is carried out to the data segment that type of error is missing errors；
Data type conversion or deletion are carried out to the data segment that type of error is type error；
The data segment that type of error is numerical fault is deleted；
Data conversion or deletion are carried out to the data segment that type of error is rule errors；
It is the data segment of repetitive error for type of error, if all data segments of corresponding different sensors continue to repeat to surpass
Setting time is crossed, then deletes duplicate data in the setting time；If the data segment that type of error is repetitive error is wind
Fast parameter or wind direction parameter or temperature parameter, then delete duplicate data segment.
3. according to the method described in claim 2, it is characterized in that, described carry out the data segment that type of error is missing errors
Data filling includes:
If type of error is that the data segment of missing errors is wind speed parameter or power parameter, counted according to wind power module
According to filling up；
Otherwise, data filling is carried out by interpolation.
4. the method according to claim 1, wherein it is described based on mechanism model to treated the system number
Include: according to being modified
Wind speed parameter abnormal in treated the system data is determined using ambient wind model, and deletes the abnormal wind
Fast parameter；
The power parameter in treated the system data is modified using wind power module；
Temperature parameter abnormal in treated the system data is determined using modeling of energy conservation, and is deleted described abnormal
Temperature parameter.
5. according to the method described in claim 4, it is characterized in that, it is described using wind power module to treated the system
Power parameter in data, which is modified, includes:
Whether the data that each time point is examined successively meet wind power module；
If the data of current point in time do not meet wind power module, according in the data of current point in time wind speed parameter and
Power parameter in the data of fan condition code parameters revision current point in time.
6. according to the method described in claim 5, it is characterized in that, wind speed parameter in the data according to current point in time
And the power parameter in the data of fan condition code parameters revision current point in time includes:
If the wind speed parameter in the data of current point in time is less than incision wind speed, by the power in the data of current point in time
Parameter is revised as 0；
If the wind speed parameter in the data of current point in time is greater than incision wind speed and is less than rated wind speed, by current time
Power parameter in the data of point is deleted；
If the wind speed parameter in the data of current point in time is greater than rated wind speed, the wind in the data of current point in time is checked
Whether machine status code is power limitation code；
If it is, the power parameter in the data of current point in time is revised as power limitation；
Otherwise, the power parameter in the data of current point in time is revised as full hair power.
7. method according to any one of claims 1 to 6, which is characterized in that the method also includes:
The abnormal data in the revised system data is determined based on mathematical model；
Remove the abnormal data, the system data after being optimized.
8. the method according to the description of claim 7 is characterized in that described determine the revised system based on mathematical model
Abnormal data in data includes:
The abnormal data in the revised system data is determined based on any one or more following mathematical model: cluster mould
Type, Remanent Model and cluster models.
9. according to the method described in claim 8, it is characterized in that, being determined in the revised system based on group system
Abnormal data includes:
Using blower fan system as a cluster models, each column data for all blowers for including in each model in cluster models is judged
Whether the difference of mean parameter corresponding with wind field is more than given threshold, if it exceeds given threshold, it is determined that the column data institute
The data for belonging to time point are abnormal data.
10. a kind of device for improving the wind power system quality of data, which is characterized in that described device includes:
Wrong data detection module, the wrong data and its type of error in system data for determining acquisition；The system
Data include: that the data at each time point are corresponded in certain period of time, and the data of each time point include one or more correspond to not
With the data segment of sensor；
Data processing module is obtained for being handled according to the type of error the wrong data in the system data
Treated system data；
Data correction module obtains revised for based on mechanism model, treated that system data is modified to described
System data.
11. device according to claim 10, which is characterized in that the type of error includes any of the following or more
Kind: missing errors, type error, numerical fault, rule errors, repetitive error；
The data processing module, to the wrong data progress in the system below any one or more
Reason:
Data filling is carried out to the data segment that type of error is missing errors；
Data type conversion or deletion are carried out to the data segment that type of error is type error；
The data segment that type of error is numerical fault is deleted；
Data conversion or deletion are carried out to the data segment that type of error is rule errors；
If all data segments of corresponding different sensors are persistently repeated more than setting time, delete in the setting time
Duplicate data；
If the data segment that type of error is repetitive error is wind speed parameter or wind direction parameter or temperature parameter, repetition is deleted
Data segment.
12. device according to claim 11, which is characterized in that the data processing module is in the following manner to wrong class
Type is that the data segment of missing errors carries out data filling:
If type of error is that the data segment of missing errors is wind speed parameter or power parameter, counted according to wind power module
According to filling up；
Otherwise, data filling is carried out by interpolation.
13. device according to claim 10, which is characterized in that the data correction module includes:
Wind speed parameter amending unit, for determining that wind speed abnormal in treated the system data is joined using ambient wind model
Number, and delete the abnormal wind speed parameter.
Power parameter amending unit, for being carried out using wind power module to the power parameter in treated the system data
Amendment；
Temperature parameter amending unit, for determining temperature abnormal in treated the system data using modeling of energy conservation
Parameter, and delete the abnormal temperature parameter.
14. device according to claim 13, which is characterized in that the power parameter amending unit includes:
Check subelement, whether the data for each time point to be examined successively meet wind power module；
Revise subelemen, after determining that the data of current point in time do not meet wind power module in the inspection subelement, root
According to the power ginseng in the data of wind speed parameter and fan condition code parameters revision current point in time in the data of current point in time
Number.
15. device according to claim 14, which is characterized in that
The revise subelemen will be worked as when being less than incision wind speed specifically for the wind speed parameter in the data of current point in time
Power parameter in the data at preceding time point is revised as 0；Wind speed parameter in the data of current point in time is greater than incision wind speed
And when being less than rated wind speed, the power parameter in the data of current point in time is deleted；In the data of current point in time
When wind speed parameter is greater than rated wind speed, check whether the fan condition code in the data of current point in time is power limitation code；Such as
Fruit is that the power parameter in the data of current point in time is then revised as power limitation；It otherwise, will be in the data of current point in time
Power parameter be revised as full hair power.
16. device according to any one of claims 10 to 15, which is characterized in that described device further include:
Anomaly data detection module, for determining the abnormal data in the revised system data based on mathematical model；
Abnormal data cleaning modul, the system data for removing the abnormal data, after being optimized.
17. device according to claim 16, which is characterized in that the anomaly data detection module includes following any one
Kind or multiple module:
First detection module, for detecting the abnormal data in the revised system data based on Clustering Model；
Second detection module, for detecting the abnormal data in the revised system data based on Remanent Model；
Third detection module, for detecting the abnormal data in the revised system data based on cluster models.
18. device according to claim 17, which is characterized in that
The third detection module is specifically used for judging each model in cluster models using blower fan system as a cluster models
In include the differences of each column data mean parameter corresponding with wind field of all blowers whether be more than given threshold, if it exceeds
Given threshold, it is determined that the data at the column data affiliated time point are abnormal data.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910750791.2A CN110489852A (en) | 2019-08-14 | 2019-08-14 | Improve the method and device of the wind power system quality of data |
| PCT/CN2019/106357 WO2021027011A1 (en) | 2019-08-14 | 2019-09-18 | Method and apparatus for improving data quality of wind power system |
| PCT/CN2020/082958 WO2021027294A1 (en) | 2019-08-14 | 2020-04-02 | Method and apparatus for improving wind power system data quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910750791.2A CN110489852A (en) | 2019-08-14 | 2019-08-14 | Improve the method and device of the wind power system quality of data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110489852A true CN110489852A (en) | 2019-11-22 |
Family
ID=68549935
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910750791.2A Pending CN110489852A (en) | 2019-08-14 | 2019-08-14 | Improve the method and device of the wind power system quality of data |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110489852A (en) |
| WO (2) | WO2021027011A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111323081A (en) * | 2020-03-20 | 2020-06-23 | 江南造船(集团)有限责任公司 | Ship test voyage performance index correction method, correction system, storage medium and electronic terminal |
| WO2021027294A1 (en) * | 2019-08-14 | 2021-02-18 | 北京天泽智云科技有限公司 | Method and apparatus for improving wind power system data quality |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389634A (en) * | 2015-12-01 | 2016-03-09 | 广东智造能源科技研究有限公司 | Combined short-term wind power prediction system and method |
| CN107392304A (en) * | 2017-08-04 | 2017-11-24 | 中国电力科学研究院 | A kind of Wind turbines disorder data recognition method and device |
| CN108509752A (en) * | 2018-05-02 | 2018-09-07 | 东南大学 | A kind of superheater mechanism model dynamic parameter setting method of combination field data |
| CN108549689A (en) * | 2018-04-12 | 2018-09-18 | 华北电力大学 | A kind of running of wind generating set data cleaning method |
| CN109388519A (en) * | 2017-08-07 | 2019-02-26 | 上海寒武纪信息科技有限公司 | Error recovery method and device, processor |
| CN109740175A (en) * | 2018-11-18 | 2019-05-10 | 浙江大学 | A kind of point judging method that peels off towards Wind turbines power curve data |
| US20190230119A1 (en) * | 2018-01-19 | 2019-07-25 | General Electric Company | Dynamic concurrent learning method to neutralize cyber attacks and faults for industrial asset monitoring nodes |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9354968B2 (en) * | 2011-09-30 | 2016-05-31 | Johnson Controls Technology Company | Systems and methods for data quality control and cleansing |
| US20160179599A1 (en) * | 2012-10-11 | 2016-06-23 | University Of Southern California | Data processing framework for data cleansing |
| CN103631681B (en) * | 2013-12-10 | 2016-04-20 | 国家电网公司 | A kind of method of online reparation abnormal data of wind power plant |
| CN107330183B (en) * | 2017-06-29 | 2020-04-17 | 华北电力大学 | Wind power utilization rate calculation method based on operation data |
| CN107291927B (en) * | 2017-06-29 | 2019-11-05 | 同济大学 | A kind of air speed data cleaning method for Wind turbines wind speed correlation analysis |
| CN108646261B (en) * | 2018-02-14 | 2021-03-26 | 上饶市中科院云计算中心大数据研究院 | Data processing method and device for vehicle GPS data set |
| CN109783486B (en) * | 2019-01-17 | 2020-11-24 | 华北电力大学 | Data cleaning method and device and server |
| CN110489852A (en) * | 2019-08-14 | 2019-11-22 | 北京天泽智云科技有限公司 | Improve the method and device of the wind power system quality of data |
-
2019
- 2019-08-14 CN CN201910750791.2A patent/CN110489852A/en active Pending
- 2019-09-18 WO PCT/CN2019/106357 patent/WO2021027011A1/en active Application Filing
-
2020
- 2020-04-02 WO PCT/CN2020/082958 patent/WO2021027294A1/en active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389634A (en) * | 2015-12-01 | 2016-03-09 | 广东智造能源科技研究有限公司 | Combined short-term wind power prediction system and method |
| CN107392304A (en) * | 2017-08-04 | 2017-11-24 | 中国电力科学研究院 | A kind of Wind turbines disorder data recognition method and device |
| CN109388519A (en) * | 2017-08-07 | 2019-02-26 | 上海寒武纪信息科技有限公司 | Error recovery method and device, processor |
| US20190230119A1 (en) * | 2018-01-19 | 2019-07-25 | General Electric Company | Dynamic concurrent learning method to neutralize cyber attacks and faults for industrial asset monitoring nodes |
| CN108549689A (en) * | 2018-04-12 | 2018-09-18 | 华北电力大学 | A kind of running of wind generating set data cleaning method |
| CN108509752A (en) * | 2018-05-02 | 2018-09-07 | 东南大学 | A kind of superheater mechanism model dynamic parameter setting method of combination field data |
| CN109740175A (en) * | 2018-11-18 | 2019-05-10 | 浙江大学 | A kind of point judging method that peels off towards Wind turbines power curve data |
Non-Patent Citations (4)
| Title |
|---|
| ANDREW KUSIAK: "Modles for monitoring wind farm power", 《RENEWABLE》 * |
| 张瀚超 等: "偏差修正算法的风电短期功率预测", 《新型工业化》 * |
| 戴巨川 等: "风电场SCADA数据预处理方法及评价策略", 《太阳能学报》 * |
| 温延立: "风电机组监测数据可视分析研究", 《中国优秀硕士学位论文全文数据库（工程科技Ⅱ辑）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021027294A1 (en) * | 2019-08-14 | 2021-02-18 | 北京天泽智云科技有限公司 | Method and apparatus for improving wind power system data quality |
| CN111323081A (en) * | 2020-03-20 | 2020-06-23 | 江南造船(集团)有限责任公司 | Ship test voyage performance index correction method, correction system, storage medium and electronic terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021027011A1 (en) | 2021-02-18 |
| WO2021027294A1 (en) | 2021-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109086889B (en) | Terminal fault diagnosis method, device and system based on neural network | |
| CN110569867A (en) | Decision tree algorithm-based power transmission line fault reason distinguishing method, medium and equipment | |
| CN105677791B (en) | For analyzing the method and system of the operation data of wind power generating set | |
| CN106779200A (en) | Based on the Wind turbines trend prediction method for carrying out similarity in the historical data | |
| CN111523785A (en) | Power system dynamic security assessment method based on generation countermeasure network | |
| CN109004679B (en) | Renewable energy source accessed power grid safety and stability monitoring method | |
| CN113570138B (en) | Method and device for predicting residual service life of equipment of time convolution network | |
| CN110119787B (en) | Working condition detection method and equipment for rotary mechanical equipment | |
| CN107330573B (en) | State evaluation method and device for photovoltaic system key equipment | |
| CN106919982A (en) | A kind of method for diagnosing faults towards semiconductor manufacturing equipment | |
| CN109061391B (en) | Power grid fault diagnosis method and system based on computer vision tidal current diagram | |
| CN116882790B (en) | Carbon emission equipment management method and system for mine ecological restoration area | |
| CN110489852A (en) | Improve the method and device of the wind power system quality of data | |
| CN115021679A (en) | Photovoltaic equipment fault detection method based on multi-dimensional outlier detection | |
| CN115453356A (en) | Power equipment running state monitoring and analyzing method, system, terminal and medium | |
| CN116799796A (en) | Photovoltaic power generation power prediction method, device, equipment and medium | |
| CN116842337A (en) | Transformer fault diagnosis method based on LightGBM (gallium nitride based) optimal characteristics and COA-CNN (chip on board) model | |
| CN114997566A (en) | Power grid blocking risk assessment method and system considering node connectivity loss | |
| CN111612149A (en) | Main network line state detection method, system and medium based on decision tree | |
| CN108988347B (en) | Method and system for adjusting class imbalance of transient voltage stabilization sample set of power grid | |
| CN111091141B (en) | Photovoltaic backboard fault diagnosis method based on layered Softmax | |
| CN116316629A (en) | Data-driven optimal power flow calculation method considering topology feature learning | |
| CN113807462A (en) | AI-based network equipment fault reason positioning method and system | |
| CN112699598A (en) | Intelligent diagnosis method and device for abnormal oil temperature of gear box | |
| CN113111588A (en) | NO of gas turbineXEmission concentration prediction method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191122 |
|
| RJ01 | Rejection of invention patent application after publication |