CN114723252A - Wind power plant sample board machine selection method - Google Patents
Wind power plant sample board machine selection method Download PDFInfo
- Publication number
- CN114723252A CN114723252A CN202210300189.0A CN202210300189A CN114723252A CN 114723252 A CN114723252 A CN 114723252A CN 202210300189 A CN202210300189 A CN 202210300189A CN 114723252 A CN114723252 A CN 114723252A
- Authority
- CN
- China
- Prior art keywords
- wind
- wind speed
- wind power
- power plant
- grouping
- 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
- 238000010187 selection method Methods 0.000 title claims description 12
- 238000000034 method Methods 0.000 claims abstract description 39
- 238000004364 calculation method Methods 0.000 claims abstract description 22
- 239000011159 matrix material Substances 0.000 claims abstract description 21
- 238000005070 sampling Methods 0.000 claims abstract description 20
- 238000010219 correlation analysis Methods 0.000 claims abstract description 10
- 238000010606 normalization Methods 0.000 claims description 10
- 238000010248 power generation Methods 0.000 claims description 7
- 238000005315 distribution function Methods 0.000 claims description 6
- 238000003064 k means clustering Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/76—Power conversion electric or electronic aspects
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Wind Motors (AREA)
Abstract
The invention provides a method for selecting a sample board computer of a wind power plant, and belongs to the field of calculation of theoretical generated energy of the wind power plant. The method comprises the following steps: acquiring wind speed data of each wind turbine in a wind power plant; calculating a Weibull distribution curve parameter of the wind speed of each wind turbine; normalizing the wind speed data of each wind turbine generator after dimensionality reduction, normalizing the wind speed Weibull distribution curve parameters of each wind turbine generator, and constructing an input variable matrix of a wind power plant grouping model by utilizing the normalized result; clustering the input variable matrix of the wind power field grouping model according to the determined optimal grouping number of the wind power field units to obtain the grouping result of each wind power unit in the wind power field; and selecting the board sampling machine of each group by using a correlation analysis method to obtain a board sampling machine selection result of the wind power plant. According to the method, the wind power plant sample board machine is selected based on the clustering algorithm, the rationality and the representativeness of the selection of the sample board machine can be improved, the calculation efficiency is ensured, and meanwhile, the accuracy of the generated energy calculation can be improved.
Description
Technical Field
The invention belongs to the field of calculation of theoretical generated energy of a wind power plant, and particularly relates to a method for selecting a sample board machine of the wind power plant.
Background
Aiming at the problem of wind abandonment and electricity limitation, the theoretical generating capacity and the wind abandonment electric quantity level of the whole wind farm need to be scientifically and efficiently evaluated, so that the running condition of the wind farm is accurately mastered, and corresponding technical means are adopted on the basis to reduce the wind abandonment phenomenon, so that the waste of wind energy resources is reduced.
In 2012, the national electric power supervision committee released a wind farm abandoned wind power amount calculation method, and a prototype board method was popularized in China to calculate abandoned wind power amount. The sample board machine method can accurately count the abandoned wind power, but whether the sample board machine can be reasonably selected is a key factor influencing the calculation accuracy of the sample board machine method, the sample board machine selection method in a research field is used for improving the rationality and the representativeness of the sample board machine selection, and the sample board machine method has important significance for the theoretical power generation evaluation of a wind power field and the operation and the scheduling of a power system.
The conventional sampling machine selection method provides that the number of sampling machines does not exceed 10% of the total number of the wind power plant units in principle, and the conventional sampling machine method does not have a fixed sampling machine selection method and is often selected according to the experience of workers and the related data of the wind power plant units. And when the theoretical power generation amount of the wind power plant is calculated to obtain the abandoned wind power amount, the distribution coefficient tends to be evenly distributed. The theoretical generating capacity of the traditional sample plate computer method is large in calculation relative error and high in uncertainty, and calculation accuracy is also remarkably reduced when the theoretical generating capacity of the wind power plant is calculated.
In recent years, a new method for selecting a sample plate machine can identify bad data and eliminate outliers by applying a double-step K-means clustering algorithm, and provides a basis for estimating the theoretical power generation of the whole field by subsequently dividing a cluster and selecting the sample plate machine. The method can not only embody single-machine representativeness, but also avoid the complexity of calculation one by one, but the method can not make the difference between the classified clusters minimum and the difference between the clusters in different classes maximum, so that a more representative sample board machine can not be more effectively selected to calculate the theoretical generating capacity and the abandoned wind electric quantity of the wind power plant.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a method for selecting a wind power plant sample board machine. According to the method, the wind power plant sample board selecting machine is selected based on the clustering algorithm, the rationality and the representativeness of the sample board selecting machine can be improved, the calculation efficiency is guaranteed, and meanwhile the accuracy of the generated energy calculation can be improved.
The embodiment of the invention provides a method for selecting a sample board machine of a wind power plant, which comprises the following steps:
acquiring wind speed data of each wind turbine in a wind power plant;
calculating a Weibull distribution curve parameter of the wind speed of each wind turbine generator according to the wind speed data;
normalizing the wind speed data of each wind turbine after dimensionality reduction, normalizing the wind speed Weibull distribution curve parameters of each wind turbine, and constructing an input variable matrix of a wind power plant grouping model by using the normalized result;
determining the optimal grouping number of the wind power plant units, and clustering the input variable matrix of the wind power plant grouping model according to the optimal grouping number to obtain the grouping result of each wind power plant unit in the wind power plant;
and selecting the panel sampling machine of each group by using a correlation analysis method according to the grouping result so as to obtain a panel sampling machine selection result of the wind power plant.
In a specific embodiment of the invention, the wind speed weibull distribution curve parameters comprise shape parameters and scale parameters.
In a specific embodiment of the present invention, the sampling board selecting method for each group is performed by using a correlation analysis method according to the grouping result to obtain the sampling board selecting result of the wind farm, and the specific method is as follows:
1) for the ith unit in any group, calculating the mean value of the correlation coefficients of the unit and other units in the group:
in the formula, rm(i) The mean value of the correlation coefficients of the wind speed of the ith unit in any group and other units in the group is i 1, 2. q is the number of the grouped units; r (ij) is the wind speed related coefficient of the ith unit and the jth unit in the group;
2) will be r in each packetm(i) And the highest unit is used as the grouped sample board machine, and all the grouped sample board machines form the sample board machine selection result of the wind power plant.
In a specific embodiment of the present invention, the shape parameter and scale parameter calculation method is as follows:
1) uniformly dividing the wind speed data of the ith unit into n intervals, wherein the upper limit value of the wind speed of each interval is recorded asj=1,2,...,n,;
2) Counting the occurrence frequency of the corresponding wind speed data of each wind speed interval of the ith unit and recording the occurrence frequency asj=1,2,...,n;
3) Using the upper limit value of wind speed of each section of the ith unitAnd the frequency of the intervalPerforming curve fitting to obtain a wind speed Weibull distribution curve corresponding to the ith unit, wherein the expression is as follows:
f(x)=a(i)b(i)x(b(i)-1)exp(-a(i)xb(i))
respectively calculating the shape parameter k and the scale parameter c of the Weibull distribution corresponding to the curve:
k(i)=b(i))
wherein, a (i) is a constant term obtained by converting the logarithm of the mathematical form of the two-parameter Weibull distribution function of the ith unit into a linear form, and b (i) is a slope obtained by converting the logarithm of the mathematical form of the two-parameter Weibull distribution function of the ith unit into the linear form;
k (i) is the shape parameter of the wind speed Weibull distribution curve of the ith unit, and c (i) is the scale parameter of the wind speed Weibull distribution curve of the ith unit.
In a specific embodiment of the present invention, the normalizing the wind speed data of each wind turbine after dimensionality reduction, normalizing the wind speed weibull distribution curve parameters of each wind turbine, and constructing an input variable matrix of a wind farm grouping model by using the normalized result includes:
1) reducing the dimension of the wind speed data of each wind turbine;
for the ith set of machines, the computational expression is as follows:
in the formula, vm(i) The average value of the wind speed of the ith unit is obtained; v. ofsd(i) Is the ith tableThe unit wind speed mean square difference value; p (i) represents the total wind speed data of the ith unit;
2) respectively normalizing the wind speed data of each wind turbine generator after dimensionality reduction and the shape parameters and the scale parameters of the wind speed Weibull distribution curve of the wind turbine generator;
for the ith set of machines, the computational expression is as follows:
in the formula, vm1(i) Is the average value of the wind speed v of the ith unit after normalizationsd2(i) The wind speed is the wind speed mean square difference value of the ith unit after normalization;
k3(i) is the shape parameter of the Weibull distribution curve of the wind speed of the ith unit after normalization, c4(i) The normalized wind speed Weibull distribution curve of the ith unit is the scale parameter;
vm(max) and vm(min) is respectively the maximum value and the minimum value of the average value of the wind speeds of all the units in the wind power plant, vsd(max) and vsd(min) are respectively the maximum value and the minimum value of the mean square difference value of the wind speeds of all the units in the wind power plant, k (max) and k (min) are respectively the maximum value and the minimum value of the Weibull distribution curve shape parameters of the wind speeds of all the units in the wind power plant, and c (max) and c (min) are respectively the maximum value and the minimum value of the Weibull distribution curve scale parameters of the wind speeds of all the units in the wind power plant;
3) the input variable matrix for establishing the wind power field grouping model is as follows:
wherein A is the total number of wind turbine generators in the wind power plant.
In a specific embodiment of the present invention, the frequency calculation method includes:
frequency available frequency/(group distance total)
The available frequency is the number of the wind speed data in any interval, and the group distance is the upper limit difference value of the wind speed of two adjacent intervals.
In a specific embodiment of the present invention, the determining the optimal grouping number of the wind farm units, and clustering the input variable matrix of the wind farm grouping model according to the optimal grouping number to obtain the grouping result of each wind farm unit in the wind farm includes:
1) setting a plurality of initial grouping numbers;
2) clustering the input variable matrix of the wind power field grouping model under each initial grouping number, and calculating CH index values corresponding to the grouping results of the wind power generation sets in the wind power field under the initial grouping number according to the clustering results;
3) taking the initial grouping number corresponding to the maximum value of the CH index value as the optimal grouping number of the wind power plant set; and the clustering result under the optimal grouping number is the grouping result of each wind turbine in the wind power plant.
In a specific embodiment of the present invention, the clustering method is K-means clustering.
The invention has the characteristics and beneficial effects that:
the method utilizes the optimal clustering number determined by the CH index, and then carries out wind power plant grouping based on a clustering algorithm, thereby providing a reliable basis for selection of a sample machine;
according to the sampling board machine selection result obtained by the method, a more accurate theoretical power generation amount calculation result of the wind power plant can be obtained, meanwhile, the calculation efficiency is guaranteed, and important information is provided for the evaluation of the running condition of the wind power plant.
Drawings
Fig. 1 is an overall flowchart of a method for selecting a sample board computer in a wind farm according to an embodiment of the present invention.
FIG. 2 is a Weibull distribution plot of wind speed after fitting in the embodiment of the present invention.
Fig. 3 is a schematic diagram of CH indexes for different numbers of packets according to the embodiment of the present invention.
Detailed Description
The invention provides a method for selecting a wind power plant sample board machine, which is further described in detail in the following by combining the attached drawings and specific embodiments
The embodiment of the invention provides a method for selecting a sample board machine of a wind power plant, the overall flow is shown as figure 1, and the method comprises the following steps:
1) acquiring measured wind speed data of all wind turbines in a wind power plant; in the embodiment, the longer the time is under the condition of permission, the more the data is, and the better the calculation effect is; in one embodiment of the invention, the time resolution of the data is 2h, and the time length of the data is not less than half a year.
2) And (2) calculating a Weibull distribution double parameter value of each unit of the wind power plant according to the wind speed data in the step 1).
The method for acquiring the Weibull distribution double-parameter value of the ith unit comprises the following steps: 2-1) grouping the measured wind speed data of the ith unit obtained in the step 1), wherein in the embodiment, the wind speed interval of the unit is uniformly divided into n intervals from the wind speed of 0, and the upper limit value of the wind speed of each interval is recorded asAn upper wind speed limit value of j, which is 1,2, …, n, of the jth interval of the ith unit;
it should be noted that, the value of the number n of the intervals has no special requirement, the more the intervals are divided, the finer the intervals are, the intervals can be divided into 50 or 100 according to the size of the wind speed data set, and the number n of the wind speed intervals of each unit is the same;
2-2) counting the number of steps according to the result of the step 2-1)The frequency of occurrence of wind speed data corresponding to each wind speed interval of the i sets is recorded asIndicating the frequency of the wind speed data of the ith unit appearing in the jth wind speed interval;
in this embodiment, the frequency may be calculated by frequency/(group pitch total), in an embodiment of the present invention, the wind speed of 8760 hours per hour in a year is counted, the total wind speed range is 0m/s to 20m/s, the wind speed range is uniformly divided into 40 intervals according to the group pitch of 0.5m/s, wherein 3 wind speed values occur in the wind speed interval of 0 to 0.5m/s, and then the corresponding frequency of the wind speed interval is 3/(0.5 × 40);
2-3) using the wind speed upper limit value speed of each section of the ith unitAnd the frequency of the intervalPerforming curve fitting by using a built-in function library of MATLAB, and minimizing the value of the error sum of squares to obtain a wind speed Weibull distribution curve corresponding to the ith unit;
the weibull distribution curve of wind speed after fitting according to a specific embodiment of the present invention is shown in fig. 2, wherein the wind speed range is uniformly divided into 40 intervals by taking the actually measured wind speed range of 0-20m/s as the group pitch according to 0.5m/s in the embodiment, the abscissa of each point in fig. 2 represents the upper limit value of the wind speed of the corresponding wind speed interval, the ordinate of each point represents the frequency value of the wind speed of the corresponding wind speed interval, and the curve shown in fig. 2 is obtained by fitting, i.e., the curve is the curve according to the scattered points Fitting a Weibull distribution curve.
Wherein, the expression of the curve fitting model is as follows:
f(x)=a(i)b(i)x(b(i)-1)exp(-a(i)xb(i)) (1)
calculating the shape parameter k and the scale parameter c of the Weibull distribution corresponding to the curve:
k(i)=b(i) (2)
wherein a (i) is a constant term obtained by converting the logarithm of the two-parameter Weibull distribution function of the i-th unit into a linear form, and b (i) is a slope obtained by converting the logarithm of the two-parameter Weibull distribution function of the i-th unit into a linear form.
Wherein k (i) is the shape parameter of the wind speed Weibull distribution curve of the ith unit, and c (i) is the scale parameter of the wind speed Weibull distribution curve of the ith unit.
3) Normalizing the wind speed data of each wind turbine generator after dimensionality reduction, normalizing the wind speed Weibull distribution curve parameters of each wind turbine generator, and constructing an input variable matrix of a wind power plant grouping model by utilizing the normalized result;
in the embodiment of the invention, the specific steps are as follows:
3-1) performing dimensionality reduction on the wind speed data of each unit of the wind power plant acquired according to the step 1), so that the main attribute beneficial to convergence of a clustering objective function is fast grasped to improve the calculation efficiency.
For the ith set of machines, the computational expression is as follows:
in the formula, vm(i) The average value of the wind speed of the ith unit is obtained; v. ofsd(i)And the wind speed mean square difference value of the ith unit is obtained. And p (i) represents the total wind speed data of the ith unit.
3-2) respectively normalizing the wind speed data (wind speed average value and wind speed mean square difference) of each unit subjected to dimensionality reduction in the step 3-1) and the shape parameters and scale parameters (namely 4 data characteristic values of each wind turbine generator) of the wind speed Weibull distribution curve of each unit obtained in the step 2), so that the wind speed data and the scale parameters are respectively converted into data in a [0, 1] interval, and the clustering difficulty is reduced.
For the ith set of machines, the computational expression is as follows:
in the formula, vm1(i) Is the average value of the wind speed v of the ith unit after normalizationsd2(i) The wind speed is the wind speed mean square difference value of the ith unit after normalization;
k3(i) is the shape parameter of the Weibull distribution curve of the wind speed of the ith unit after normalization, c4(i) The normalized wind speed Weibull distribution curve of the ith unit is the scale parameter;
vm(max) and vm(min) is respectively the maximum value and the minimum value of the average value of the wind speeds of all the units in the wind power plant, vsd(max) and vsd(min) are respectively the maximum value and the minimum value of the mean square difference value of the wind speeds of all the units in the wind power plant, and k (max) and k (min) are respectively the Weibull distribution curve shapes of the wind speeds of all the units in the wind power plantThe maximum value and the minimum value of the shape parameter, c (max) and c (min) are respectively the maximum value and the minimum value of the wind speed Weibull distribution curve scale parameter of all the units of the wind power plant. In an embodiment of the present invention, the normalization calculation process may be implemented by using a minmaxscale function in a Python language-based machine learning tool sklern.
3-3) utilizing the result normalized in the step 3-2), and establishing an input matrix of the wind power plant grouping model as follows:
wherein A is the total number of wind turbine generators in the wind power plant.
In a specific embodiment of the present invention, the wind farm has 33 wind turbines in total, and the input variable matrix of the grouping model is:
4) determining the optimal grouping number of the wind power plant units, and clustering the input variable matrix of the wind power plant grouping model according to the optimal grouping number to obtain the grouping result of each wind power plant unit in the wind power plant; the method comprises the following specific steps:
4-1) setting a plurality of initial grouping numbers M (the grouping numbers of the wind turbines can be 2, 3 and 4 of course) according to the number of the wind turbines, but the grouping numbers need to be determined according to the number of the wind turbines, so that the turbines with different operation characteristics are distributed to different clusters to the maximum extent, and meanwhile, the high efficiency of grouping calculation is also guaranteed. ) In the embodiment, when the number of groups is 7, the situation of an isolated unit occurs, so that the initial number of groups only considers the situation that M is 2-6;
4-2) clustering the input variable matrix of the wind power field grouping model under each initial grouping number, and calculating CH index values corresponding to the grouping results of the wind power generation sets in the wind power field under the initial grouping number according to the clustering result;
in this embodiment, for each initial grouping number, the clustering effect corresponding to the initial grouping number is evaluated by using a Calinski-Harabasz index (i.e., a CH index), and C in the CH index corresponding to the clustering result corresponding to each initial grouping number is calculatedkThe function value determines the optimal number of groups.
In a specific embodiment of the invention, under each initial grouping number, computing the K-means clustering by using Python software on the input variable matrix of the wind power plant grouping model obtained in the step 3), wherein the specific process is as follows:
reading an input variable matrix by using a read statement in a Pandas library;
normalizing the Sklearn library by using a MinMaxScaler function in the Sklearn library;
thirdly, performing K-means clustering calculation by using a fit method in a sklern library;
and obtaining a grouping result of the wind power plant. 4-3) taking the initial grouping number corresponding to the maximum value of the CH index value as the optimal grouping number of the wind power plant unit; and the clustering result under the optimal grouping number is the grouping result of each wind turbine in the wind power plant.
In this embodiment, the optimal grouping number is determined to be m according to the maximum value of the CH index value, the wind turbine generator is divided into m clusters, effective clustering is basically achieved, the number of the wind turbine generator in each cluster is relatively uniform, the goals of intra-class similarity and inter-class exclusivity are basically achieved, and even if the operating characteristic difference of the wind turbine generator in each cluster is minimum, the operating characteristic difference of the wind turbine generator in different clusters is maximum.
In an embodiment of the present invention, the CH index results corresponding to the clustering results with different grouping numbers are shown in fig. 3, wherein the abscissa of each point represents different initial grouping numbers, and the ordinate represents the CH index (i.e. C) corresponding to the clustering result with the grouping numberkFunction value), the CH index in fig. 3 is maximized when the number of packets is 5, so that it can be determined that the clustering effect is the best when the number of wind farm packets is 5, i.e. the optimal number of packets m is 5 in this embodiment.
5) And 4) selecting the panel sampling machine which can represent the grouping output characteristic most in each group by using a correlation analysis method according to the grouping result of the step 4), and obtaining a panel sampling machine selection result of the wind power plant.
In this embodiment, the Correlation degree between two or more random variables is studied by using a typical Correlation coefficient r in a Correlation Analysis (Correlation Analysis), so as to obtain an average value of the wind speed Correlation coefficients of each unit in m groups and other units in the same group.
For the ith machine set in any machine group, the calculation expression is as follows:
in the formula, rm(i) The mean value of the correlation coefficients of the wind speed of the ith unit in any group and other units in the group is i 1, 2. q is the number of units of the group. r (ij) is a wind speed Correlation coefficient of the ith unit and the jth unit in the group, and is obtained by Correlation Analysis (Correlation Analysis).
For each wind turbine generator group which is grouped according to the optimal grouping number, the mean value r of the wind speed correlation coefficient in the groupm(i) The highest unit is the unit which can represent the grouped output characteristic most, namely one of the sampling boards of the wind power plant, and m sampling boards in m groups are selected accordingly (one sampling board is selected in each group).
In one embodiment of the present invention, after 33 wind turbine generator sets are grouped, the correlation coefficient between the sets in a certain group is shown in table 1:
TABLE 1 correlation coefficient between sets of modules
In table 1, the wind speed correlation coefficients of the six wind turbine generators (3#, 15#, 16#, 17#, 19#, 20#) in the same group are shown, and as can be seen, the average value of the wind speed correlation coefficients of the 16# generator set is highest (0.967) compared with those of the other generator sets, so that the generator set is selected as one of the spline machines of the wind farm.
Claims (8)
1. A method for selecting a wind power plant sample board machine is characterized by comprising the following steps:
acquiring wind speed data of each wind turbine in a wind power plant;
calculating a Weibull distribution curve parameter of the wind speed of each wind turbine generator according to the wind speed data;
normalizing the wind speed data of each wind turbine generator after dimensionality reduction, normalizing the wind speed Weibull distribution curve parameters of each wind turbine generator, and constructing an input variable matrix of a wind power plant grouping model by using the normalized result;
determining the optimal grouping number of the wind power plant units, and clustering the input variable matrix of the wind power plant grouping model according to the optimal grouping number to obtain the grouping result of each wind power plant unit in the wind power plant;
and selecting the panel sampling machine of each group by using a correlation analysis method according to the grouping result so as to obtain a panel sampling machine selection result of the wind power plant.
2. Selection method according to claim 1, wherein the wind speed weibull profile parameters comprise shape parameters and scale parameters.
3. The selection method according to claim 1, wherein the sampling board machine of each group is selected by using a correlation analysis method according to the grouping result to obtain the sampling board machine selection result of the wind farm, and the specific method is as follows:
1) for the ith unit in any group, calculating the mean value of the correlation coefficients of the unit and other units in the group:
in the formula, rm(i) The mean value of the correlation coefficients of the wind speed of the ith unit in any group and other units in the group is i 1, 2. q is the number of the grouped units; r (ij) is the wind speed related coefficient of the ith unit and the jth unit in the group;
2) will be r in each packetm(i) And the highest unit is used as the grouped sample board machine, and all the grouped sample board machines form the sample board machine selection result of the wind power plant.
4. The selection method according to claim 2, wherein the shape parameter and scale parameter calculation method is as follows:
1) uniformly dividing the wind speed data of the ith unit into n intervals, wherein the upper limit value of the wind speed of each interval is recorded as
2) Counting the occurrence frequency of the corresponding wind speed data of each wind speed interval of the ith unit and recording the frequency as
3) Using the upper limit value of the wind speed of each section of the ith unitAnd the frequency of the intervalPerforming curve fitting to obtain a wind speed Weibull distribution curve corresponding to the ith unit, wherein the expression is as follows:
f(x)=a(i)b(i)x(b(i)-1)exp(-a(i)xb(i))
respectively calculating the shape parameter k and the scale parameter c of the Weibull distribution corresponding to the curve:
k(i)=b(i))
wherein, a (i) is a constant term obtained by converting the logarithm of the mathematical form of the two-parameter Weibull distribution function of the ith unit into a linear form, and b (i) is a slope obtained by converting the logarithm of the mathematical form of the two-parameter Weibull distribution function of the ith unit into the linear form;
k (i) is a shape parameter of a Weibull distribution curve of the wind speed of the ith unit, and c (i) is a scale parameter of the Weibull distribution curve of the wind speed of the ith unit.
5. The selection method according to claim 4, wherein the normalizing the wind speed data of each wind turbine after dimensionality reduction, the normalizing the wind speed Weibull distribution curve parameters of each wind turbine, and the constructing an input variable matrix of the wind farm grouping model using the normalized result comprises:
1) reducing the dimension of the wind speed data of each wind turbine;
for the ith set of machines, the computational expression is as follows:
in the formula, vm(i) The average value of the wind speed of the ith unit is obtained; v. ofsd(i) The wind speed mean square difference value of the ith unit is obtained; p (i) represents the total wind speed data of the ith unit;
2) respectively normalizing the wind speed data of each wind turbine generator after dimensionality reduction and the shape parameters and the scale parameters of the wind speed Weibull distribution curve of the wind turbine generator;
for the ith set of machines, the computational expression is as follows:
in the formula, vm1(i) Is the average value of the wind speed v of the ith unit after normalizationsd2(i) The wind speed mean square difference value of the ith unit after normalization;
k3(i) is the shape parameter of the Weibull distribution curve of the wind speed of the ith unit after normalization, c4(i) The normalized wind speed Weibull distribution curve of the ith unit is the scale parameter;
vm(max) and vm(min) is respectively the maximum value and the minimum value of the average value of the wind speeds of all the units in the wind power plant, vsd(max) and vsd(min) are respectively the maximum value and the minimum value of the mean square difference value of the wind speeds of all the units in the wind power plant, k (max) and k (min) are respectively the maximum value and the minimum value of the Weibull distribution curve shape parameters of the wind speeds of all the units in the wind power plant, and c (max) and c (min) are respectively the maximum value and the minimum value of the Weibull distribution curve scale parameters of the wind speeds of all the units in the wind power plant;
3) the input variable matrix for establishing the wind power field grouping model is as follows:
wherein A is the total number of wind turbine generators in the wind power plant.
6. Selection method according to claim 4, characterized in that the frequency is calculated by:
frequency available frequency/(group distance total)
The available frequency is the number of the wind speed data in any interval, and the group distance is the upper limit difference of the wind speeds in two adjacent intervals.
7. The selection method according to claim 5, wherein the determining an optimal grouping number of the wind farm units, and clustering an input variable matrix of the wind farm grouping model according to the optimal grouping number to obtain a grouping result of each wind farm unit in the wind farm comprises:
1) setting a plurality of initial grouping numbers;
2) clustering the input variable matrix of the wind power field grouping model under each initial grouping number, and calculating CH index values corresponding to the grouping results of the wind power generation sets in the wind power field under the initial grouping number according to clustering results;
3) taking the initial grouping number corresponding to the maximum value of the CH index value as the optimal grouping number of the wind power plant set; and the clustering result under the optimal grouping number is the grouping result of each wind turbine in the wind power plant.
8. Selection method according to claim 1 or 7, characterized in that the clustering method is K-means clustering.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210300189.0A CN114723252A (en) | 2022-03-25 | 2022-03-25 | Wind power plant sample board machine selection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210300189.0A CN114723252A (en) | 2022-03-25 | 2022-03-25 | Wind power plant sample board machine selection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114723252A true CN114723252A (en) | 2022-07-08 |
Family
ID=82239724
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210300189.0A Pending CN114723252A (en) | 2022-03-25 | 2022-03-25 | Wind power plant sample board machine selection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114723252A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116993026A (en) * | 2023-09-26 | 2023-11-03 | 无锡九方科技有限公司 | Large-scale wind power plant unit operation parameter optimization method |
-
2022
- 2022-03-25 CN CN202210300189.0A patent/CN114723252A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116993026A (en) * | 2023-09-26 | 2023-11-03 | 无锡九方科技有限公司 | Large-scale wind power plant unit operation parameter optimization method |
CN116993026B (en) * | 2023-09-26 | 2023-12-19 | 无锡九方科技有限公司 | Large-scale wind power plant unit operation parameter optimization method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104808587B (en) | A kind of mobility statistical method based on machining apparatus running status | |
CN108376262B (en) | Analytical model construction method for typical characteristics of wind power output | |
CN103887815A (en) | Wind power plant parameter identification and dynamic equivalence method based on operation data | |
CN112186761B (en) | Wind power scene generation method and system based on probability distribution | |
CN103489046A (en) | Method for predicting wind power plant short-term power | |
CN103279803A (en) | Load modeling method and system based on comprehensive information theory and modern interior point theory | |
CN114723252A (en) | Wind power plant sample board machine selection method | |
CN112651576A (en) | Long-term wind power prediction method and device | |
CN110263834A (en) | A kind of detection method of new energy power quality exceptional value | |
CN117559443A (en) | Ordered power utilization control method for large industrial user cluster under peak load | |
CN115660232A (en) | Ultra-short-term prediction method, device and system for wind power | |
CN113505909A (en) | Error compensation method for short-term wind power trend prediction | |
CN115898787A (en) | Method and device for dynamically identifying static yaw error of wind turbine generator | |
CN110943473A (en) | Generator coherence identification method based on wide area measurement system and clustering theory | |
CN115271423A (en) | Wind generating set running state evaluation method based on hybrid similarity | |
CN113554203B (en) | Wind power prediction method and device based on high-dimensional meshing and LightGBM | |
CN115344996A (en) | Wind power typical scene construction method and system based on multi-characteristic quantity indexes of improved K-means algorithm | |
Qiao et al. | Research on SCADA data preprocessing method of Wind Turbine | |
CN112365081A (en) | Photovoltaic power station power generation capacity prediction method and device | |
CN109066791B (en) | Method and device for determining wind power simulation abandoned wind sequence | |
CN111898871A (en) | Power grid power end data quality evaluation method, device and system | |
Gu et al. | Online monitoring of wind turbine operation efficiency and optimization based on benchmark values | |
Miao et al. | A Transformer District Line Loss Calculation Method Based on Data Mining and Machine Learning | |
Xie et al. | Energy System Time Series Data Quality Maintenance System Based on Data Mining Technology | |
CN118194607B (en) | Dynamic modeling method and system for boiler unit data |
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 |