CN108763584A - A kind of method and its system of the filtering of wind power curve scatterplot - Google Patents
A kind of method and its system of the filtering of wind power curve scatterplot Download PDFInfo
- Publication number
- CN108763584A CN108763584A CN201810596993.1A CN201810596993A CN108763584A CN 108763584 A CN108763584 A CN 108763584A CN 201810596993 A CN201810596993 A CN 201810596993A CN 108763584 A CN108763584 A CN 108763584A
- Authority
- CN
- China
- Prior art keywords
- data
- power
- window
- value
- tracking point
- 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
Landscapes
- Wind Motors (AREA)
- Control Of Eletrric Generators (AREA)
Abstract
The invention belongs to wind-driven generator, data analysis, technical field of data processing is disclosed a kind of method of wind power curve scatterplot filtering, including is realized using following steps：Step S001 acquires the SCADA data set M01 data packets for obtaining target fan by using SCADA system；The M01 data packets input data screening washer subpackage is obtained data packet M011 and data packet M012 by step S002；Step S003 obtains updated data package M011a and updated data package M012b；Step S004 merges and obtains updated power data packet M02；The updated power data packet M02 obtained in step S004 is carried out data correction by step S005, is rejected distortion data and is obtained filtered power data set M03.
Description
Technical field
The invention belongs to wind-driven generator, data analysis, technical field of data processing, more particularly to wind power collecting
The discrete data method that either distortion data is handled or system regions in SCADA data, and in particular to a kind of wind power
The method and its system of curve scatterplot filtering.
Background technology
The wind power curve that wind power generating set SCADA system obtains in actual production be usually present a large amount of scatterplot or
Person is paramophia, and exceptional value causes in mostly different control strategies, different operating modes and data.In the data analysis of wind turbine
It is middle that the scatterplot to wind power curve is needed to clean, with obtain a smooth power curve come carry out it is subsequent analysis and build
Mould.
The existing technology being filtered to wind power curve scatterplot includes：
It is filtered using the operating condition marking variable in SCADA data, this method needs operating mode mark and becomes
Amount, but in most cases, SCADA system does not have this variable or record missing, while even if only screening normal operation
The data of operating mode, the scatterplot in wind power curve may be still widely present；
Operating mode filtering is carried out using the variable in the SCADA systems such as blade angle, rotating speed, this method is in 1 institute of method
State operating mode mark lack as when be a kind of alternative, but based on its dependent variable carry out operating mode's switch be it is limited, no
All operating modes can be covered；
Data filtering based on density, this method by power curve by two variable of wind speed-power by being divided into grid
System falls into a trap the density of evidence of counting in each grid, determines whether sparse scatterplot to reject.The defect of this method exists
In in the case where scatterplot is extremely disperseed, significant effect variation Chinese Patent Application No. CN201110432282.9 is using power point
The method that digit carries out data rejecting equally exists becoming in power curve scatterplot over-dispersion effect described in method 3
The problem of difference is to solve the defect of method as above, rejects a kind of method and system of the scatterplot filtering of power curve tracking.
Invention content
In order to solve the above problem of the existing technology, present invention aims at provide a kind of wind power curve scatterplot mistake
The method and its system of filter were carried out using the operating condition marking variable in SCADA data in the prior art for solving
Filter can still there are problems that being distorted discrete point, lead to that expected smooth power curve cannot be obtained；And use blade angle,
Variable in the SCADA systems such as rotating speed can have applicable range limitation to carry out operating mode filtering, cannot cover asking for full working scope
Topic.The present invention will be divided by the M01 data packets that will be acquired from SCADA system using screening washer according to given threshold,
Realize the subpackage of data；A point window is being carried out to the successful data of subpackage by power curve tracker, after will finally dividing window
Each window in data be iterated traversal and calculate, reject all scatterplot data for being unsatisfactory for condition, final retention data
Set carries out the purposes of subsequent analysis and modeling to obtain a smooth power curve.
The technical solution adopted in the present invention is：
A kind of method of wind power curve scatterplot filtering, including realized using following steps：
Step S001 acquires the SCADA data set M01 data packets for obtaining target fan by using SCADA system；
Step S002 sets power threshold by the M01 data packets input data screening washer, and by the screening washer
H=0.1 carries out comparison screening to all power datas in the M01 data packets, and screening obtains power and is less than or equal to 0.1
Data packet M011 and power are more than 0.1 data packet M012；
Data packet M011 described in step S002 and data packet M012 is imported parameter initialization by step S003 simultaneously
Power curve tracker in the first filter element G1 and the second filter element G2 in carry out data cleansing filtering, obtain respectively
Updated data package M011a and updated data package M012b；
The updated data package M011a obtained in step S003 and updated data package M012b are merged and are obtained more by step S004
Power data packet M02 after new；
The updated power data packet M02 obtained in step S004 is carried out data correction, rejects distortion by step S005
Data obtain filtered power data set M03.
For the further accurate processing for realizing data, it is preferable that SCADA data set in the step S001
Data type is corresponding with the initiation parameter type in power curve tracker described in step S003 in M01 data packets.
Preferably, the initiation parameter includes that wind speed coboundary positional value, wind speed lower boundary positional value, power window are big
Small value, wind speed deviation threshold value, up-and-down boundary width threshold value, front and back window tracking point difference allow the upper bound and lower bound, window to chase after
Track point deviant, rated power starting air speed value and rated power float value.
It is worth noting that：Since SCADA system is computer-based DCS and power automation monitoring system；
Its application field is very wide, can be applied to the data acquisition and monitoring in the fields such as electric power, metallurgy, oil, chemical industry, combustion gas, railway
The numerous areas such as control and process control.Therefore, in the data acquisition for carrying out corresponding field or in the institute of different purposes
The target data type that need to be acquired is in the presence of partly or completely different, the needs of the follow-up calculating based on the present invention,
By gathered data type in SCADA system and the follow-up initiation parameter class realized data and divide the power curve tracker of window processing
The corresponding purpose of type is exactly the unified docking and acquisition in order to realize data.Such as：If power curve tracker is in calculating process
Middle meeting carry out cleaning or filter base calculating dependent on wind speed deviation threshold value as basic data, but in SCADA data collection
Calculating interruption/suspension will so be directly resulted in by closing not corresponding acquisition in M01 data packets.
In order to which with further refining the present invention, the first filter element G1 described in above-mentioned steps S003 passes through data packet
The process that M011 obtains updated data package M011a includes the following steps：
Step S0031 determines power curve starting point：Choose power all wind speed average values in the sections 1-10；
Step S0032 filters invalid data：Choose and delete power less than 5 and wind speed be more than power curve starting point it is complete
Portion's data obtain updated data package M011a.The first filter element G1 is operation one of in power curve tracker
Module, operation rule of the burning in the computing module can be real by way of including but not limited to custom logic operation
It is existing.This for current existing power curve tracker carry out setting realize from technological layer to those of ordinary skill in the art and
Speech is the prior art, is just not detailed herein.Really, although it should be strongly noted that the realization of technological layer problem surely belongs to
It is existing, but for carry out setting and calculating logic program in including but not limited to operation/computational methods be it is not existing, i.e.,
Content of the present invention.
Based on the above method, it is further preferred that the second filter element G2 described in step S003 passes through data
The process that packet M012 obtains updated data package M012b includes the following steps：
Step S0033, by all data in data packet M012, fixed width carries out a point window, obtains window W1、W2、
W3…….Wn, and extract data in each institute's split window successively by the way of cycle；
Step S0034 judges current window WnIt is interior to whether there is data；If data are not present, it is recycled to next window
Mouth Wn+1It is middle to carry out above-mentioned judgement；If there are data, window W is calculatedn+1Description value；The description value includes air speed data
Up-and-down boundary value, wind speed deviation and up-and-down boundary width；
The air speed data up-and-down boundary value：Using air speed data up-and-down boundary position as quantile calculate, below with
Upper, lower are indicated；
Wind speed deviation：It is indicated below with ws_sd；Up-and-down boundary width：It is calculated as maximum value up and down and subtracts tracking point difference
Absolute value, indicated below with upper_width, lower_width；Calculation window Wn+1The median of interior all retention datas,
Compared with the final tracking points of upper window Wn, determines the tracking point of current window, indicated below with tracker；The tracking point
Update using cross-window tracking point update method realize；
The description value calculation formula is as follows：
1) wind speed deviation
Wherein ws indicates that air speed data value, μ are the arithmetic average of air speed data, and N indicates data volume
2) up-and-down boundary width
Upper_width=| upper-tracker |
Lower_width=| lower-tracker |
Such as no data, wind speed deviation, up-and-down boundary width are set to 0；Such as the window wind speed deviation, up-and-down boundary width
More than corresponding threshold value, and current iteration has all retention datas in data and the searching loop window in the window；
Step S0035, traversal, which is deleted to current window wind speed deviation, up-and-down boundary width, meets setup parameter：
3) as above border width is more than lower boundary width, and the absolute value of the difference of two values is more than 1, then deletes；
4) following border width is more than coboundary width, and the absolute value of the difference of two values is more than 1, then deletes；
5) such as absolute value of the difference of up-and-down boundary width is less than 1, and such as current air speed data point is more than upper boundary values, or works as
Preceding air speed data point is less than lower border value, then deletes；
6) the be described value of update current window and current window tracking point, the tracking point are calculated as current window
Wn+1The wind speed median of retention data；Until the data filtering for completing to be cyclically updated all windows obtains updated data package
M012b。
In order to preferably realize the present invention, updated power data packet M02 described in step S005 carries out data correction
Method proceed as follows：
7) judge the pass between the power data power corresponding with rated power starting wind speed in the power data packet M02
System；
8) judge the power data in the power data packet M02 whether in rated power plus-minus maximum power float value model
In enclosing；
Be satisfied by 7) and 8), retained if described, otherwise, then delete it is corresponding not and meanwhile meet it is described 7) with work(8)
Rate data obtain the power data set M03 after final filtration.
In the present invention, the cross-window tracking point update method preferably uses following proposal to realize：
Such as current window Wn+1Data volume is more than 2, previous window WnFinal tracking point and current median absolute difference are small
In feasible value lower bound, then it is current window W to update tracking pointn+1Median；Such as previous window WnFinal tracking point and current middle position
Number absolute difference is more than feasible value lower bound, and the difference absolute value of two values is less than the feasible value upper bound, then it is two values to update tracking point
Mean value；Such as previous window WnFinal tracking point is more than the feasible value upper bound with current median absolute difference, updates tracking point
For previous window WnFinal tracking point adds current window Wn+1Tracking point deviant；
Such as current window Wn+1Data volume is less than or equal to 2, and update tracking point is previous window WnFinal tracking point adds window
Tracking point deviant；
Such as current window Wn+1The window W for having data for first1, current window Wn+1Tracking point and previous window WnTracking
Point, which synchronizes, is set as window Wn+1Air speed data median.
A kind of system of wind power curve scatterplot filtering, including SCADA system：SCADA numbers for acquiring target fan
According to set M01 data packets, the SCADA data set M01 data packets include wind speed coboundary position data, wind speed lower boundary position
Data, power window size Value Data, wind speed deviation threshold data, up-and-down boundary width threshold value data, front and back window is set to chase after
Track point difference allows the upper bound and lower bound data, window tracking point shift value data, rated power starting wind speed Value Data and specified
Power floating Value Data；
Screening washer：The SCADA data set M01 data packets are divided into different targets by setting power threshold h
Data acquisition system；
Power curve tracker：For realizing the different target data acquisition system that will pass through after screening washer screening according to this hair
The method of any wind power curve scatterplot filtering, which calculates, in bright obtains smooth power curve data.
Beneficial effects of the present invention are：
In the method for the invention by the way of dividing window to iterate over calculating filtering screening, it can be come from various
SCADA system includes the work condition abnormality because record lacks, and the caused scatterplot problem of operating mode mark missing carries out comprehensive filtering and picks
Except the smooth power curve of acquisition；Meanwhile the present invention carries out logical process by using the wrong differentiation of data successively, then to data
Obtain the data for meeting sets requirement and carry out window data update, can avoid so in the prior art the scope of application it is limited,
The problem of garbled data is distorted or mixes excessive scatterplot.
Description of the drawings
Fig. 1 is the flow chart of the embodiment of the present invention 1；
Fig. 2 is preferred embodiment of the present invention flow chart；
Fig. 3 is the schematic diagram before the filtering of wind power curve scatterplot；
Fig. 4 is the filtered schematic diagram of wind power curve scatterplot；
Specific implementation mode
Below in conjunction with the accompanying drawings and specific embodiment the present invention is further elaborated.It is being illustrated to the embodiment of the present invention
Before, it is more advantageous to understands that scheme of the present invention first explains following term/nouns for convenience：
1, SCADA (Supervisory Control And Data Acquisition) system
That is data acquisition and supervisor control.SCADA system is that computer-based DCS is supervised with power automation
Control system；Its application field is very wide, can be applied to the data acquisition in the fields such as electric power, metallurgy, oil, chemical industry, combustion gas, railway
With monitoring control and the numerous areas such as process control.
2, wind speed deviation threshold value
Each wind speed xi deviates the average of the distance of mean wind speed, is the root after sum of squares of deviations is average.
3, median
Median (also known as intermediate value, English：Median), the proper noun in statistics represents a sample, population or general
Numerical value set, can be divided into equal two parts up and down by a numerical value in rate distribution.For limited manifold, Ke Yitong
One that middle is found out after all observed values height is sorted is crossed as median.If observed value has even number, usually take
The average of most intermediate two values is as median.
Embodiment 1：
As shown in Fig. 1,3 and 4, a kind of method of wind power curve scatterplot filtering is present embodiments provided, including using such as
Lower step is realized：
Step S001 acquires the SCADA data set M01 data packets for obtaining target fan by using SCADA system.
Step S002 sets power threshold by the M01 data packets input data screening washer, and by the screening washer
H=0.1 carries out comparison screening to all power datas in the M01 data packets, and screening obtains power and is less than or equal to 0.1
Data packet M011 and power are more than 0.1 data packet M012；Herein, it should be noted that data were carried out according to this method
Filter, for different wind turbines or because of the difference of realistic objective demand data, it is believed that set the tool of the power threshold h
Body value, and one or more can be set, M01 data packets are divided into multiple data packets divided by the power threshold h
M011、M012……M01n。
Data packet M011 described in step S002 and data packet M012 is imported parameter initialization by step S003 simultaneously
Power curve tracker in the first filter element G1 and the second filter element G2 in carry out data cleansing filtering, obtain respectively
Updated data package M011a and updated data package M012b.
The updated data package M011a obtained in step S003 and updated data package M012b are merged and are obtained more by step S004
Power data packet M02 after new.
The updated power data packet M02 obtained in step S004 is carried out data correction, rejects distortion by step S005
Data obtain filtered power data set M03.
Embodiment 2：
For the further accurate processing for realizing data, on the basis of embodiment 1, in the step S001
Data type and the initiation parameter in power curve tracker described in step S003 in SCADA data set M01 data packets
Type is corresponding；The initiation parameter includes wind speed coboundary positional value, wind speed lower boundary positional value, power window size
Value, wind speed deviation threshold value, up-and-down boundary width threshold value, front and back window tracking point difference allow the upper bound and lower bound, window tracking
Point deviant, rated power starting air speed value and rated power float value.It is worth noting that：Since SCADA system is in terms of
DCS based on calculation machine and power automation monitoring system；Its application field is very wide, can be applied to electric power, metallurgy, oil,
The numerous areas such as the data acquisition in the fields such as chemical industry, combustion gas, railway and monitoring control and process control.Therefore, it is carrying out pair
It is in the presence of partly or complete in the required target data type being acquired of different purposes to answer the data in field to acquire either
Complete different, gathered data type in SCADA system is realized data point by the needs of the follow-up calculating based on the present invention with follow-up
Window processing power curve tracker the corresponding purpose of initiation parameter type be exactly in order to realize the unified docking of data with
Acquisition.Such as：If power curve tracker can depend on wind speed deviation threshold value to be carried out as basic data in calculating process
Cleaning or filter base calculate, but not corresponding acquisition will so directly result in SCADA data set M01 data packets
Calculate interruption/suspension.
Embodiment 3：
In order to further refining the present invention, on the basis of above-mentioned any embodiment, shown in Fig. 2,
In the present embodiment, the first filter element G1 described in above-mentioned steps S003 obtains updated data package by data packet M011
The process of M011a includes the following steps：
Step S0031 determines power curve starting point：Choose power all wind speed average values in the sections 1-10；
Step S0032 filters invalid data：Choose and delete power less than 5 and wind speed be more than power curve starting point it is complete
Portion's data obtain updated data package M011a.The first filter element G1 is operation one of in power curve tracker
Module, operation rule of the burning in the computing module can be real by way of including but not limited to custom logic operation
It is existing.This for current existing power curve tracker carry out setting realize from technological layer to those of ordinary skill in the art and
Speech is the prior art, is just not detailed herein.Really, although it should be strongly noted that the realization of technological layer problem surely belongs to
It is existing, but for carry out setting and calculating logic program in including but not limited to operation/computational methods be it is not existing, i.e.,
Content of the present invention.
Based on the above method, it is further preferred that the second filter element G2 described in step S003 passes through data
The process that packet M012 obtains updated data package M012b includes the following steps：
Step S0033, by all data in data packet M012, fixed width carries out a point window, obtains window W1、W2、
W3…….Wn, and extract data in each institute's split window successively by the way of cycle；
Step S0034 judges current window WnIt is interior to whether there is data；If data are not present, it is recycled to next window
Mouth Wn+1It is middle to carry out above-mentioned judgement；If there are data, window W is calculatedn+1Description value；The description value includes air speed data
Up-and-down boundary value, wind speed deviation and up-and-down boundary width；
The air speed data up-and-down boundary value：Using air speed data up-and-down boundary position as quantile calculate, below with
Upper, lower are indicated；
Wind speed deviation：It is indicated below with ws_sd；Up-and-down boundary width：It is calculated as maximum value up and down and subtracts tracking point difference
Absolute value, indicated below with upper_width, lower_width；Calculation window Wn+1The median of interior all retention datas,
With upper window WnFinal tracking point compares, and determines the tracking point of current window, is indicated below with tracker；The tracking point
Update using cross-window tracking point update method realize；
The description value calculation formula is as follows：
1) wind speed deviation
Wherein ws indicates that air speed data value, μ are the arithmetic average of air speed data, and N indicates data volume
2) up-and-down boundary width
Upper_width=| upper-tracker |
Lower_width=| lower-tracker |
Such as no data, wind speed deviation, up-and-down boundary width are set to 0；Such as the window wind speed deviation, up-and-down boundary width
More than corresponding threshold value, and current iteration has all retention datas in data and the searching loop window in the window；
Step S0035, traversal, which is deleted to current window wind speed deviation, up-and-down boundary width, meets setup parameter：
3) as above border width is more than lower boundary width, and the absolute value of the difference of two values is more than 1, then deletes；
4) following border width is more than coboundary width, and the absolute value of the difference of two values is more than 1, then deletes；
5) such as absolute value of the difference of up-and-down boundary width is less than 1, and such as current air speed data point is more than upper boundary values, or works as
Preceding air speed data point is less than lower border value, then deletes；
6) the be described value of update current window and current window tracking point, the tracking point are calculated as current window
Wn+1The wind speed median of retention data；Until the data filtering for completing to be cyclically updated all windows obtains updated data package
M012b。
In the present embodiment, updated power data packet M02 described in step S005 carry out the method for data correction according to
It is following to carry out：
7) judge the pass between the power data power corresponding with rated power starting wind speed in the power data packet M02
System；
8) judge the power data in the power data packet M02 whether in rated power plus-minus maximum power float value model
In enclosing；
Be satisfied by 7) and 8), retained if described, otherwise, then delete it is corresponding not and meanwhile meet it is described 7) with work(8)
Rate data obtain the power data set M03 after final filtration.As shown in Figure 3 and Figure 4, it is certain area H1-01F wind turbine 1-8 months
Part is filtered front and back power scatter plot.Certainly, the work(acquired within the scope of different wind turbine models and specified period
Rate scatter plot is different, but can obtain smooth a, curve with one fixed width by the above method,
To achieve the purpose that filter discrete point.
Embodiment 4：
In the present embodiment, the cross-window tracking point update method preferably uses following proposal to realize：
Such as current window Wn+1Data volume is more than 2, previous window WnFinal tracking point and current median absolute difference are small
In feasible value lower bound, then it is current window W to update tracking pointn+1Median；Such as previous window WnFinal tracking point and current middle position
Number absolute difference is more than feasible value lower bound, and the difference absolute value of two values is less than the feasible value upper bound, then it is two values to update tracking point
Mean value；Such as previous window WnFinal tracking point is more than the feasible value upper bound with current median absolute difference, updates tracking point
For previous window WnFinal tracking point adds current window Wn+1Tracking point deviant；
Such as current window Wn+1Data volume is less than or equal to 2, and update tracking point is previous window WnFinal tracking point adds window
Tracking point deviant；
Such as current window Wn+1The window W for having data for first1, current window Wn+1Tracking point and previous window WnTracking
Point, which synchronizes, is set as window Wn+1Air speed data median.
Embodiment 5：
A kind of system of wind power curve scatterplot filtering, including SCADA system：SCADA numbers for acquiring target fan
According to set M01 data packets, the SCADA data set M01 data packets include wind speed coboundary position data, wind speed lower boundary position
Data, power window size Value Data, wind speed deviation threshold data, up-and-down boundary width threshold value data, front and back window is set to chase after
Track point difference allows the upper bound and lower bound data, window tracking point shift value data, rated power starting wind speed Value Data and specified
Power floating Value Data；
Screening washer：The SCADA data set M01 data packets are divided into different targets by setting power threshold h
Data acquisition system；
Power curve tracker：For realizing the different target data acquisition system that will pass through after screening washer screening according to this hair
The method of any wind power curve scatterplot filtering, which calculates, in bright obtains smooth power curve data.
The present invention is not limited to above-mentioned optional embodiment, anyone can show that other are various under the inspiration of the present invention
The product of form, however, make any variation in its shape or structure, it is every to fall into the claims in the present invention confining spectrum
Technical solution, be within the scope of the present invention.
Claims (9)
1. a kind of method of wind power curve scatterplot filtering, it is characterised in that：Including being realized using following steps：
Step S001 acquires the SCADA data set M01 data packets for obtaining target fan by using SCADA system；
Step S002 sets power threshold h=by the M01 data packets input data screening washer, and by the screening washer
0.1, comparison screening is carried out to all power datas in the M01 data packets, screening obtains the data that power is less than or equal to 0.1
Wrap the data packet M012 of M011 and power more than 0.1；
Data packet M011 described in step S002 and data packet M012 is imported the work(of parameter initialization by step S003 simultaneously
Data cleansing filtering is carried out in the first filter element G1 and the second filter element G2 in rate curve tracing device, is updated respectively
Data packet M011a and updated data package M012b；
Step S004 merges the updated data package M011a obtained in step S003 and updated data package M012b after obtaining update
Power data packet M02；
The updated power data packet M02 obtained in step S004 is carried out data correction, rejects distortion data by step S005
Obtain filtered power data set M03.
2. a kind of method of wind power curve scatterplot filtering according to claim 1, it is characterised in that：The step S001
Data type is joined with the initialization in power curve tracker described in step S003 in middle SCADA data set M01 data packets
Several classes of types are corresponding.
3. a kind of method of wind power curve scatterplot filtering according to claim 2, it is characterised in that：The initialization ginseng
Number includes wind speed coboundary positional value, wind speed lower boundary positional value, power window size value, wind speed deviation threshold value, upper following
Boundary's width threshold value, front and back window tracking point difference allow the upper bound and lower bound, window tracking point deviant, rated power starting wind speed
Value and rated power float value.
4. according to a kind of method of wind power curve scatterplot filtering of claim 1-3 any one of them, it is characterised in that：Step
The process that first filter element G1 described in S003 obtains updated data package M011a by data packet M011 includes the following steps：
Step S0031 determines power curve starting point：Choose power all wind speed average values in the sections 1-10；
Step S0032 filters invalid data：It chooses and deletes whole numbers that power is less than 5 and wind speed is more than power curve starting point
According to acquisition updated data package M011a.
5. a kind of method of wind power curve scatterplot filtering according to claim 4, it is characterised in that：Institute in step S003
The second filter element G2 is stated by the process of data packet M012 acquisition updated data packages M012b to include the following steps：
Step S0033, by all data in data packet M012, fixed width carries out a point window, obtains window W1、W2、W3……
.Wn, and extract data in each institute's split window successively by the way of cycle；
Step S0034 judges current window WnIt is interior to whether there is data；If data are not present, it is recycled to next window Wn+1
It is middle to carry out above-mentioned judgement；If there are data, window W is calculatedn+1Description value；The description value includes air speed data or more
Boundary value, wind speed deviation and up-and-down boundary width；
The air speed data up-and-down boundary value：Using air speed data up-and-down boundary position as quantile calculate, below with upper,
Lower is indicated；
Wind speed deviation：It is indicated below with ws_sd；Up-and-down boundary width：It is calculated as upper and lower maximum value and subtracts the exhausted of tracking point difference
To value, indicated below with upper_width, lower_width；Calculation window Wn+1The median of interior all retention datas, and it is upper
The one final tracking points of window Wn compare, and determine the tracking point of current window, are indicated below with tracker；The tracking point is more
Newly realized using cross-window tracking point update method；
The description value calculation formula is as follows：
1) wind speed deviation
Wherein ws indicates that air speed data value, μ are the arithmetic average of air speed data, and N indicates data volume
2) up-and-down boundary width
Upper_width=| upper-tracker |
Lower_width=| lower-tracker |
Such as no data, wind speed deviation, up-and-down boundary width are set to 0；Such as the window wind speed deviation, up-and-down boundary width are more than
Corresponding threshold value, and current iteration has all retention datas in data and the searching loop window in the window；
Step S0035, traversal, which is deleted to current window wind speed deviation, up-and-down boundary width, meets setup parameter：
3) as above border width is more than lower boundary width, and the absolute value of the difference of two values is more than 1, then deletes；
4) following border width is more than coboundary width, and the absolute value of the difference of two values is more than 1, then deletes；
5) such as absolute value of the difference of up-and-down boundary width is less than 1, and such as current air speed data point is more than upper boundary values or current wind
Fast data point is less than lower border value, then deletes；
6) the be described value of update current window and current window tracking point, the tracking point are calculated as current window Wn+1It protects
The wind speed median of residual evidence；Until the data filtering for completing to be cyclically updated all windows obtains updated data package M012b.
6. according to a kind of method of wind power curve scatterplot filtering of claim 1-3 any one of them, it is characterised in that：Step
The method that updated power data packet M02 carries out data correction described in S005 proceeds as follows：
7) judge the relationship between the power data power corresponding with rated power starting wind speed in the power data packet M02；
8) judge the power data in the power data packet M02 whether in rated power plus-minus maximum power float value range
It is interior；
Be satisfied by 7) and 8), retained if described, otherwise, then delete it is corresponding not and meanwhile meet it is described 7) with power number 8)
According to the power data set M03 after acquisition final filtration.
7. a kind of method of wind power curve scatterplot filtering according to claim 5, it is characterised in that：Institute in step S005
The method that updated power data packet M02 carries out data correction is stated to proceed as follows：
7) judge the relationship between the power data power corresponding with rated power starting wind speed in the power data packet M02；
8) judge the power data in the power data packet M02 whether in rated power plus-minus maximum power float value range
It is interior；
Be satisfied by 7) and 8), retained if described, otherwise, then delete it is corresponding not and meanwhile meet it is described 7) with power number 8)
According to the power data set M03 after acquisition final filtration.
8. a kind of method of wind power curve scatterplot filtering according to claim 5, it is characterised in that：The cross-window chases after
Track point update method includes following content：
Such as current window Wn+1Data volume is more than 2, previous window WnFinal tracking point is less than with current median absolute difference to be held
Perhaps it is worth lower bound, then it is current window W to update tracking pointn+1Median；Such as previous window WnFinal tracking point and current median are poor
It is worth absolute value and is more than feasible value lower bound, and the difference absolute value of two values is less than the feasible value upper bound, then it is the equal of two values to update tracking point
Value；Such as previous window WnFinal tracking point is more than the feasible value upper bound with current median absolute difference, before update tracking point is
One window WnFinal tracking point adds current window Wn+1Tracking point deviant；
Such as current window Wn+1Data volume is less than or equal to 2, and update tracking point is previous window WnFinal tracking point is tracked plus window
Point deviant；
Such as current window Wn+1The window W for having data for first1, current window Wn+1Tracking point and previous window WnTracking point is same
Step is set as window Wn+1Air speed data median.
9. a kind of system of wind power curve scatterplot filtering, it is characterised in that：
Including SCADA system：SCADA data set M01 data packets for acquiring target fan, the SCADA data set
M01 data packets include wind speed coboundary position data, wind speed lower boundary position data, power window size Value Data, wind speed mark
Quasi- difference threshold data, up-and-down boundary width threshold value data, front and back window tracking point difference allow the upper bound and lower bound data, window to chase after
Track point shift value data, rated power starting wind speed Value Data and rated power floating Value Data；
Screening washer：The SCADA data set M01 data packets are divided into different target datas by setting power threshold h
Set；
Power curve tracker：For realizing the different target data acquisition system that will pass through after screening washer screening according to claim
The method of wind power curve scatterplot filtering described in 8, which calculates, obtains smooth power curve data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810596993.1A CN108763584B (en) | 2018-06-11 | 2018-06-11 | Method and system for filtering scattered points of wind power curve |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810596993.1A CN108763584B (en) | 2018-06-11 | 2018-06-11 | Method and system for filtering scattered points of wind power curve |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108763584A true CN108763584A (en) | 2018-11-06 |
| CN108763584B CN108763584B (en) | 2021-11-02 |
Family
ID=64021355
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810596993.1A Active CN108763584B (en) | 2018-06-11 | 2018-06-11 | Method and system for filtering scattered points of wind power curve |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108763584B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110334310A (en) * | 2019-07-01 | 2019-10-15 | 北京源清慧虹信息科技有限公司 | A kind of the determination method and determining device of power of fan curve |
| CN114091354A (en) * | 2022-01-07 | 2022-02-25 | 国能日新科技股份有限公司 | Method and device for acquiring wind turbine generator power prediction model sample set |
| CN114579827A (en) * | 2022-05-06 | 2022-06-03 | 昆仑智汇数据科技(北京)有限公司 | Method, device and equipment for processing data performance curve of industrial equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102588210A (en) * | 2011-12-21 | 2012-07-18 | 中能电力科技开发有限公司 | Filtering method for preprocessing fitting data of power curve |
| CN103291544A (en) * | 2013-06-21 | 2013-09-11 | 华北电力大学 | Method for drawing digital wind power generating set power curve |
| CN103955521A (en) * | 2014-05-08 | 2014-07-30 | 华北电力大学 | Cluster classification method for wind power plant |
| CN105719029A (en) * | 2016-03-11 | 2016-06-29 | 中国农业大学 | Combined wind power prediction method based on wind speed fluctuation characteristic extraction |
| US20160224391A1 (en) * | 2012-01-30 | 2016-08-04 | International Business Machines Corporation | Full exploitation of parallel processors for data processing |
| CN105930933A (en) * | 2016-04-26 | 2016-09-07 | 华北电力科学研究院有限责任公司 | Wind power plant theoretical power curve determination method and device |
| CN106897941A (en) * | 2017-01-03 | 2017-06-27 | 北京国能日新***控制技术有限公司 | A kind of blower fan method for processing abnormal data and device based on quartile box traction substation |
| CN106997391A (en) * | 2017-04-10 | 2017-08-01 | 华北电力大学（保定） | A kind of method of steady state condition data in quick screening large scale process data |
| CN107330183A (en) * | 2017-06-29 | 2017-11-07 | 华北电力大学 | A kind of wind power utilization computational methods based on service data |
-
2018
- 2018-06-11 CN CN201810596993.1A patent/CN108763584B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102588210A (en) * | 2011-12-21 | 2012-07-18 | 中能电力科技开发有限公司 | Filtering method for preprocessing fitting data of power curve |
| US20160224391A1 (en) * | 2012-01-30 | 2016-08-04 | International Business Machines Corporation | Full exploitation of parallel processors for data processing |
| CN103291544A (en) * | 2013-06-21 | 2013-09-11 | 华北电力大学 | Method for drawing digital wind power generating set power curve |
| CN103955521A (en) * | 2014-05-08 | 2014-07-30 | 华北电力大学 | Cluster classification method for wind power plant |
| CN105719029A (en) * | 2016-03-11 | 2016-06-29 | 中国农业大学 | Combined wind power prediction method based on wind speed fluctuation characteristic extraction |
| CN105930933A (en) * | 2016-04-26 | 2016-09-07 | 华北电力科学研究院有限责任公司 | Wind power plant theoretical power curve determination method and device |
| CN106897941A (en) * | 2017-01-03 | 2017-06-27 | 北京国能日新***控制技术有限公司 | A kind of blower fan method for processing abnormal data and device based on quartile box traction substation |
| CN106997391A (en) * | 2017-04-10 | 2017-08-01 | 华北电力大学（保定） | A kind of method of steady state condition data in quick screening large scale process data |
| CN107330183A (en) * | 2017-06-29 | 2017-11-07 | 华北电力大学 | A kind of wind power utilization computational methods based on service data |
Non-Patent Citations (2)
| Title |
|---|
| HANGTAO LI等: "《Wind turbine power curve drawing based on method of grid》", 《IEEE XPLORE》 * |
| 朱倩雯等: "《风电场输出功率异常数据识别与重构方法研究》", 《电力***保护与控制》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110334310A (en) * | 2019-07-01 | 2019-10-15 | 北京源清慧虹信息科技有限公司 | A kind of the determination method and determining device of power of fan curve |
| CN110334310B (en) * | 2019-07-01 | 2023-04-07 | 北京源清慧虹信息科技有限公司 | Method and device for determining power curve of fan |
| CN114091354A (en) * | 2022-01-07 | 2022-02-25 | 国能日新科技股份有限公司 | Method and device for acquiring wind turbine generator power prediction model sample set |
| CN114091354B (en) * | 2022-01-07 | 2022-05-17 | 国能日新科技股份有限公司 | Method and device for acquiring wind turbine generator power prediction model sample set |
| CN114579827A (en) * | 2022-05-06 | 2022-06-03 | 昆仑智汇数据科技(北京)有限公司 | Method, device and equipment for processing data performance curve of industrial equipment |
| CN114579827B (en) * | 2022-05-06 | 2022-08-05 | 昆仑智汇数据科技(北京)有限公司 | Method, device and equipment for processing data performance curve of industrial equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108763584B (en) | 2021-11-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108763584A (en) | A kind of method and its system of the filtering of wind power curve scatterplot | |
| CN109674080B (en) | Tobacco leaf conditioning water adding amount prediction method, storage medium and terminal equipment | |
| CN104283737B (en) | The treating method and apparatus of data flow | |
| CN109472769A (en) | A kind of bad image defect detection method and system | |
| CN109343344A (en) | Cigarette machine operating parameter optimization method | |
| CN106991508A (en) | A kind of running of wind generating set state identification method based on DBSCAN | |
| CN104933483A (en) | Wind power forecasting method dividing based on weather process | |
| CN107220449A (en) | A kind of power of fan curve modeling method based on piecewise nonlinear function | |
| CN104199971B (en) | Visualization intelligence analysis method and system based on standard knowledge framework | |
| CN110224862A (en) | Multi-agent system network appearance based on multilayer perceptron invades capability assessment method | |
| CN110348488A (en) | A kind of modal identification method based on local density's peak value cluster | |
| CN107092755A (en) | A kind of parameter for product design relies on the update method of model | |
| CN108205727B (en) | Digital workshop process decision method based on decision tree and expert system | |
| CN109614394A (en) | A kind of data processing method based on BIM | |
| CN108198268A (en) | A kind of production equipment data scaling method | |
| CN105240058B (en) | Steam turbine flow curve identifying and optimizing method based on spray nozzle flow calculation | |
| CN104680017B (en) | Time-varying stability of power system analysis system and method | |
| CN107679536A (en) | A kind of artificial tooth bead line extraction method based on DBSCAN clusters | |
| CN108647483A (en) | A kind of SCR inlet NO based on fuzzy tree modeling methodXThe flexible measurement method of concentration | |
| CN105740521A (en) | Small grid removing method and apparatus in oil reservoir numerical simulation system solution process | |
| CN101515371B (en) | Human body movement data fragment extracting method | |
| CN107587982A (en) | A kind of running of wind generating set state demarcation method and device | |
| CN103955743A (en) | Ultrahigh-pressure water jet road mark line cleaning effect forecasting method and device | |
| CN109670640A (en) | The efficiency evaluation method of DEA ethylene unit based on AP algorithm | |
| CN107315845A (en) | Sulphur unit tail gas method up to standard |
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 |