CN108646560A - Power of fan model parameter optimization method based on grey wolf algorithm improvement - Google Patents
Power of fan model parameter optimization method based on grey wolf algorithm improvement Download PDFInfo
- Publication number
- CN108646560A CN108646560A CN201810448294.2A CN201810448294A CN108646560A CN 108646560 A CN108646560 A CN 108646560A CN 201810448294 A CN201810448294 A CN 201810448294A CN 108646560 A CN108646560 A CN 108646560A
- Authority
- CN
- China
- Prior art keywords
- grey wolf
- cluster centre
- power
- fan
- grey
- 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
- 241000282461 Canis lupus Species 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000005457 optimization Methods 0.000 title claims abstract description 21
- 230000006872 improvement Effects 0.000 title claims abstract description 19
- 230000008569 process Effects 0.000 claims description 20
- 239000013598 vector Substances 0.000 claims description 14
- 241000282421 Canidae Species 0.000 claims description 7
- 230000006978 adaptation Effects 0.000 claims description 3
- 230000007423 decrease Effects 0.000 claims description 3
- 230000002068 genetic effect Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012098 association analyses Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007850 degeneration Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05F—SYSTEMS FOR REGULATING ELECTRIC OR MAGNETIC VARIABLES
- G05F1/00—Automatic systems in which deviations of an electric quantity from one or more predetermined values are detected at the output of the system and fed back to a device within the system to restore the detected quantity to its predetermined value or values, i.e. retroactive systems
- G05F1/66—Regulating electric power
- G05F1/67—Regulating electric power to the maximum power available from a generator, e.g. from solar cell
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Artificial Intelligence (AREA)
- Electromagnetism (AREA)
- Sustainable Energy (AREA)
- Life Sciences & Earth Sciences (AREA)
- Sustainable Development (AREA)
- Health & Medical Sciences (AREA)
- Power Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Structures Of Non-Positive Displacement Pumps (AREA)
- Control Of Positive-Displacement Air Blowers (AREA)
Abstract
The present invention relates to the power of fan model parameter optimization methods based on grey wolf algorithm improvement, and this approach includes the following steps:S1 determines the data area of wind speed data area and corresponding active power in fan operation data, determines the total amount of fan operation data;S2 determines cluster centre point by the fan operation data in K means++ algorithm process steps S1;S3, the cluster centre point acquired according to wind turbine operation data in step S1 and step S2 set object module;S4 carries out solving-optimizing, and acquire its optimal solution using grey wolf algorithm to the parameter of the object module of setting.This method is used for power of fan identification of Model Parameters according to the characteristics of fan operation data and the demand of power of fan identification of Model Parameters, improving grey wolf algorithm, acquires optimal solution.The power of fan model parameter optimization method has the characteristics that high precision, good operating stability and reliability are high.
Description
Technical field
The invention belongs to technical field of wind power generation, and in particular to a kind of optimization method of power of fan model parameter, needle
The feature low to power of fan identification of Model Parameters precision improves grey wolf algorithm.
Background technology
Wind turbine power generation amount is mainly influenced by geographical location wind-resources residing for its power out-put characteristic and wind turbine.Fan operation
Data are one group, and using wind speed as wind turbine, input, corresponding active power, can be largely as the two-dimensional array of power output
The relationship of reflection between the two.But under operating mode, influenced by factors such as atmospheric density and turbulence intensities, wind turbine actual operating data from
It dissipates and is distributed near power of fan model, the wind turbine of different size has different discrete features;It is influenced by bulk state, wind turbine work(
Rate model parameter also has degeneration.
Power of fan model is generally established according to fan operation data, and modeling process uses algebraic equation, the differential equation, biography
The parameter model of the forms such as delivery function expression, it is more more convenient than nonparametric model simple, it can also reflect that wind turbine is provided using wind substantially
The performance in source.The identification speed of power of fan model parameter, which is speeded, to be improved with identification precision for there is the stabilization of power grids of wind power integration
Property has important meaning.
Fan operation data volume is big, discrete, therefore power of fan identification of Model Parameters needs to improve identification speed and precision.Allusion quotation
The power of fan identification Method of type includes based on normalization " interval method " of data group, Revised genetic algorithum etc.." section
To standardize treated data group of method " in wind speed is horizontal axis, power be the longitudinal axis rectangular axes on represent and be used together line
Connect the power curve as wind turbine;Revised genetic algorithum by treated, by genetic algorithm solution established by data
The optimized parameter of model.Application No. is the Chinese patents of CN105069192A to disclose one kind based on genetic algorithm solution power of fan
The improved method of parameter of curve model, this application recognize power of fan model parameter using Revised genetic algorithum,
Running of wind generating set data are first taken by Grid Clustering and distribute weight by proposition, determine parameter model using genetic algorithm later
In parameters.The method increase solving speeds, but solving precision has much room for improvement.
Invention content
The purpose of the present invention is to provide a kind of power of fan model parameter optimization methods, and this method is according to fan operation number
According to the characteristics of and power of fan identification of Model Parameters demand, improve grey wolf algorithm, and distinguish for power of fan model parameter
Know, acquires optimal solution.The power of fan model parameter optimization method has high precision, good operating stability and reliability height etc.
Feature.
To achieve the above object, the technical scheme is that:
A kind of power of fan model parameter optimization method based on grey wolf algorithm improvement, this approach includes the following steps:
S1 determines the data area of wind speed data area and corresponding active power in fan operation data, determines that wind turbine is transported
The total amount of row data;
S2 determines cluster centre point by the fan operation data in K-means++ algorithm process steps S1;
S3, the cluster centre point acquired according to wind turbine operation data in step S1 and step S2 set object module;
S4 carries out solving-optimizing, and acquire its optimal solution using grey wolf algorithm to the parameter of the object module of setting.
The above-mentioned power of fan model parameter optimization method based on grey wolf algorithm improvement, by K-means+ in the step S2
Fan operation data in+algorithm process step S1 determine that the process of cluster centre point is:
S21 gives cluster class number K, clustering convergence precision and cluster maximum iteration, and in institute's fan operation data
In arbitrarily choose an object as first cluster centre;
S22 calculates all fan operation data in addition to first cluster centre at a distance from first cluster centre
D, and therefrom select a new data point as second cluster centre, the larger point of D values is selected as cluster centre
Probability is larger, and probability size is directly proportional to D value sizes;Then all fan operation data in addition to corresponding cluster centre are calculated
The larger point of distance D, the D value of corresponding each cluster centre selected is arrived respectively, is selected the probability as cluster centre
Larger, probability size is directly proportional to D value sizes;K cluster centre is chosen altogether;
S23, calculate separately each fan operation data object to all cluster centre points distance, then according to similitude
Criterion is divided into it apart from nearest class;
S24, using the mean value of all objects in each class after repartitioning as new cluster centre point;
S25 judges K-means++ convergences, when new cluster centre point and the cluster centre of last moment point
Between error reach clustering convergence precision or when K-means++ algorithms reach cluster maximum iteration, terminate the cluster
Process, and using last cluster centre point as optimal solution, otherwise repeatedly step S23-S25 processes.
The above-mentioned power of fan model parameter optimization method based on grey wolf algorithm improvement, the object module in the step S3
For parameter logistic equation model, number of parameters is m in object module.
The above-mentioned power of fan model parameter optimization method based on grey wolf algorithm improvement is calculated using grey wolf in the step S4
Method carries out solving-optimizing to the parameter of the object module of setting, and the process for acquiring its optimal solution is:
S41 initializes wolf pack X, and grey wolf quantity is n in wolf pack, wherein any grey wolf XjFor m dimensional vectors, 1≤j≤n;If
Determine the convergence precision of grey wolf algorithm maximum iteration T and grey wolf algorithm;
S42, calculates the fitness value of every grey wolf, and retains first three maximum grey wolf of fitness value:Xα、XβAnd Xδ,
Middle XαFor the maximum grey wolf of fitness value;
S43, according to current Xα、XβAnd XδThe position of three grey wolves and location updating equation, update the position of each grey wolf;
S44, judges whether the standard deviation between each grey wolf position is less than the convergence precision of grey wolf algorithm, or judges current
Whether iterations t is more than the grey wolf algorithm maximum iteration T of setting;If so, by XαGrey wolf exports as optimal solution, defeated
Go out the corresponding parameter vector of the grey wolf, otherwise, repeats step S42-S44 processes.
The above-mentioned power of fan model parameter optimization method based on grey wolf algorithm improvement, the fitness described in step S42
Value is calculated according to fitness function, and fitness function meets following criterion:Mean absolute error MAE or root-mean-square error
RMSE is bigger, and the ideal adaptation angle value of return is smaller;The definition of described MAE, RMSE is:
Wherein (vi,Pa(vi)) it is ith cluster point, viFor wind speed, Pa(vi) it is the active power for corresponding to wind speed, Pε(Xj,
vi) indicate grey wolf XjWhen wind speed viThe active power value that lower object module returns, N are the number of cluster point.
The above-mentioned power of fan model parameter optimization method based on grey wolf algorithm improvement, in step S43 the position of grey wolf and
Location updating equation is:
Wherein, parameterFor three grey wolves distance vector away from prey respectively, parameter A, C is and random number
r1、r2Related parameter, A=a (2r1- 1), C=2r2, a is from 2 linear decreases to 0, r1、r2Be between [0,1] it is random to
Amount.
Compared with prior art, the beneficial effects of the invention are as follows:
In the methods of the invention, with K-means++ algorithm improvement grey wolf algorithms, avoid that wolf pack is absorbed in local optimum can
Energy property, and improved grey wolf algorithm is applied to power of fan model parameter optimizing, i.e. fan operation data pass through K-means++
After algorithm process, the calculation amount of fitness function value is only and cluster class number (cluster centre number is identical as cluster class number) is related,
It is unrelated with former data so that the space complexity of data is effectively controlled, and required model solution can be improved using grey wolf algorithm
Accuracy.Request for utilization number is:Mean absolute error (MAE) required by method described in the Chinese patent of CN105069192A is
158.1307, root-mean-square error (RMSE) is 214.2984, and optimal to being acquired after original data processing using the method for the present invention
MAE is 138.2689, RMSE 203.1065.The former is compared, MAE and RMSE smallers obtained by the method that the application proposes, i.e.,
Required model precision higher.
Description of the drawings
It is specifically described the present invention by reference to attached drawing and in conjunction with example, advantages of the present invention and realization method will be more
Obviously, wherein content shown in attached drawing is only used for explanation of the present invention, without constitute to the present invention in all senses on
Limitation, in the accompanying drawings:
Fig. 1 is that the present invention is based on the algorithm main flow charts of the power of fan model parameter optimization method of grey wolf algorithm improvement.
Fig. 2 is the particular flow sheet of step S2 in the method for the present invention.
Fig. 3 is the particular flow sheet of step S4 in the method for the present invention.
Fig. 4 is wind speed-power scatter plot of certain wind turbine in embodiment 1.
Fig. 5 is the power of fan illustraton of model of certain wind turbine in embodiment 1 obtained by the application method.
Specific implementation mode
The invention will be further described with reference to the accompanying drawings and examples, but does not protect model in this, as to the application
The restriction enclosed.
A kind of power of fan model parameter optimization method based on grey wolf algorithm improvement of the present invention, specifically includes following step
Suddenly:
S1 determines the data area of wind speed data area and corresponding active power in fan operation data, determines that wind turbine is transported
The total amount of row data;
S2 determines cluster centre point by the fan operation data in K-means++ algorithm process steps S1;
S3, the cluster centre point acquired according to wind turbine operation data in step S1 and step S2 set object module;
S4 carries out solving-optimizing, and acquire its optimal solution using grey wolf algorithm to the parameter of the object module of setting.
Fan operation data described in S1 is to be inputted in the wind speed average value and corresponding period of wind turbine in certain a period of time is disconnected
The output wattful power of the average value of the active power of wind turbine output or the wind speed and synchronization wind turbine of input of a certain moment wind turbine
Rate.
Fan operation data in the steps of K-means++ algorithm process described in S2 S1 determine the specific of cluster centre point
Include the following steps:
S21 gives cluster class number K, clustering convergence precision and cluster maximum iteration, and in institute's fan operation data
In arbitrarily choose an object as first cluster centre;
S22 calculates all fan operation data in addition to first cluster centre and (refers to and selected with first cluster centre
The cluster centre selected) distance D, and therefrom select a new data point as second cluster centre, the larger point of D values,
It is selected larger as the probability of cluster centre, probability size is directly proportional to D value sizes;Then calculate except corresponding cluster centre with
Outer all fan operation data arrive the larger point of distance D, D value of the corresponding each cluster centre select respectively, selected
It is taken as larger for the probability of cluster centre, probability size is directly proportional to D value sizes;K cluster centre is chosen altogether;
S23, calculate separately each fan operation data object to all cluster centre points distance, then according to similitude
Criterion is divided into it apart from nearest class;
S24, using the mean value of all objects in each class after repartitioning as new cluster centre point;
S25 judges convergence, the error between new cluster centre point and the cluster centre of last moment point
Reach clustering convergence precision or when algorithm reaches cluster maximum iteration, terminate the cluster process, and by last cluster
Central point is as optimal solution, and otherwise (both i.e. error does not reach or iterations are inadequate either case) repeats step S23-
S25 processes.
During determining cluster centre point there are many selection modes of first cluster centre, it can randomly select,
Specified point can also be specified.
Object module described in S3 is parameter logistic equation model, and the process of establishing of the object module is the prior art,
Number of parameters is m wherein in object module;
Grey wolf algorithm is utilized described in S4, solving-optimizing is carried out to the parameter of the object module of setting, and it is optimal to acquire its
Solution, specifically includes following steps:
S41, initialization wolf pack X (X1, X2..., Xn), grey wolf quantity is n in wolf pack, wherein any grey wolf Xj(1≤j≤
N) it is m dimensional vectors;Set the convergence precision of grey wolf algorithm maximum iteration T and grey wolf algorithm;
S42, calculates the fitness value of every grey wolf, and retains first three maximum grey wolf of fitness value:Xα、XβAnd Xδ,
Middle XαFor the maximum grey wolf of fitness value;
S43, according to current Xα、XβAnd XδThe position of three grey wolves and location updating equation, update the position of each grey wolf;
S44, judges whether the standard deviation between each grey wolf position is less than the convergence precision of grey wolf algorithm, or judges current
Whether iterations t is more than the grey wolf algorithm maximum iteration T of setting;If so, by XαGrey wolf exports as optimal solution, defeated
Go out the corresponding parameter vector of the grey wolf, otherwise (both i.e. standard deviation does not reach or iterations are inadequate either case), repeats
Step S42-S44 processes.
Fitness value described in S42 is calculated according to fitness function, and fitness function should meet criterion:It is average exhausted
Bigger to error MAE or root-mean-square error RMSE, the ideal adaptation angle value of return is smaller.Described MAE, RMSE are defined as follows:
Wherein (vi,Pa(vi)) it is ith cluster point, viFor wind speed, Pa(vi) it is the active power for corresponding to wind speed, Pε(Xj,
vi) indicate grey wolf Xj(XjIndicate some grey wolf in above-mentioned wolf pack X.Grey wolf XjFor m dimensional vectors, as m required parameter number
Value) when wind speed viThe active power value that lower object module returns, N are the number of cluster point.
The position of grey wolf described in S43 and location updating equation are as follows:
Wherein, parameterFor three grey wolves distance vector away from prey respectively, parameter A, C is and random number
r1、r2Related parameter, A=a (2r1- 1), C=2r2, a is from 2 linear decreases to 0, r1、r2Be between [0,1] it is random to
Amount.
Embodiment 1
The operation data of certain wind turbine such as table 1, drafting obtain two-dimensional coordinate figure, as shown in Figure 4.
Certain the fan operation data of table 1
S1:
Ranging from (3m/s, the 25m/s) of wind speed when can determine the fan operation by table 1
Ranging from (0kW, the 2100kW) of active power of output when can determine the fan operation by table 1
It can determine that the fan operation total amount of data is 20000 by table 1;
S2:
S21 determines that cluster class number K is 200, determines that convergence precision is 1.0E-6, iterations are 200 times;
S22, according to 200 data of algorithm picks as initial K cluster centre point in 20000 data given;
S23, calculate separately each data to this 200 cluster centre points distance, then according to similarity criterion by its
It is divided into it apart from nearest class, (v can be expressed as by belonging to the data of z (z=1,2,3 ..., 200) number classyz,Pa(vyz)), z
The total N of data in number classz(y=1,2,3 ..., Nz);
S24, new cluster centre point (vz *,Pa *(vz *)), it is represented by:
S25 judges K-means++ convergences, and when algorithm reaches convergence precision or algorithm reaches iteration time
When number, terminate the cluster process, and using last cluster centre point as optimal solution, otherwise repeatedly S23-S25 processes.
S3:
Power of fan model uses typical five parameter objectives model:
Wherein v is wind speed, and P is power, and X=(a, b, c, d, g) is the special parameter vector of power of fan model, and a is pre-
The peak response of phase, b are slope factor, and c is crossover position parameter, and d is minimum response, and g is asymmetry parameter.
S4:
S41, initialization wolf pack X (X1,X2,…,Xn), any of which grey wolf (Xj, 1≤j≤n) and it is 5 dimensional vectors.Setting is most
Big iterations 200, convergence precision 1.0E-4;
S42, calculates the fitness value of every grey wolf, and retains first three maximum grey wolf of fitness value:Xα、XβAnd Xδ;
S43, according to current Xα、XβAnd XδThe position of three grey wolves and location updating equation, update the position of each grey wolf;
S44, judges whether the standard deviation between each grey wolf position is less than setting grey wolf convergence precision, or judges current change
Whether generation number t is more than the maximum iteration T of setting;If so, by XαGrey wolf exports as optimal solution, exports the grey wolf pair
Otherwise the parameter vector answered repeats S42-S44 processes.
From figure 5 it can be seen that required power module agrees with well with data and the trend of known power of fan model
It is consistent.With document《The association analysis of wind power plant operation data》It compares, model is closer to power of fan theory mould obtained by this patent
Type plays an important role to improving power of fan precision of prediction and improving operation of power networks reliability.
Unaccomplished matter of the present invention is known technology.
Claims (6)
1. a kind of power of fan model parameter optimization method based on grey wolf algorithm improvement, this approach includes the following steps:
S1 determines the data area of wind speed data area and corresponding active power in fan operation data, determines fan operation number
According to total amount;
S2 determines cluster centre point by the fan operation data in K-means++ algorithm process steps S1;
S3, the cluster centre point acquired according to wind turbine operation data in step S1 and step S2 set object module;
S4 carries out solving-optimizing, and acquire its optimal solution using grey wolf algorithm to the parameter of the object module of setting.
2. the power of fan model parameter optimization method according to claim 1 based on grey wolf algorithm improvement, feature exist
By the fan operation data in K-means++ algorithm process steps S1 in the step S2, the process of cluster centre point is determined
It is:
S21 gives cluster class number K, clustering convergence precision and cluster maximum iteration, and appoints in institute's fan operation data
Meaning chooses an object as first cluster centre;
S22 calculates all fan operation data and first cluster centre distance D in addition to first cluster centre, and
Therefrom select a new data point as second cluster centre, the larger point of D values is selected the probability as cluster centre
Larger, probability size is directly proportional to D value sizes;Then all fan operation data difference in addition to corresponding cluster centre is calculated
The point larger to distance D, the D value of the corresponding each cluster centre selected, be selected it is larger as the probability of cluster centre,
Probability size is directly proportional to D value sizes;K cluster centre is chosen altogether;
S23, calculate separately each fan operation data object to all cluster centre points distance, then according to similarity criterion
It is divided into it apart from nearest class;
S24, using the mean value of all objects in each class after repartitioning as new cluster centre point;
S25 judges K-means++ convergences, when between new cluster centre point and the cluster centre of last moment point
Error reach clustering convergence precision or when K-means++ algorithms reach cluster maximum iteration, terminate the cluster process,
And using last cluster centre point as optimal solution, otherwise repeatedly step S23-S25 processes.
3. the power of fan model parameter optimization method according to claim 1 based on grey wolf algorithm improvement, feature exist
Object module in the step S3 is parameter logistic equation model, and number of parameters is m in object module.
4. the power of fan model parameter optimization method according to claim 3 based on grey wolf algorithm improvement, feature exist
Grey wolf algorithm is utilized in the step S4, solving-optimizing is carried out to the parameter of the object module of setting, and acquire its optimal solution
Process be:
S41 initializes wolf pack X, and grey wolf quantity is n in wolf pack, wherein any grey wolf XjFor m dimensional vectors, 1≤j≤n;Setting ash
The convergence precision of wolf algorithm maximum iteration T and grey wolf algorithm;
S42, calculates the fitness value of every grey wolf, and retains first three maximum grey wolf of fitness value:Xα、XβAnd Xδ, wherein Xα
For the maximum grey wolf of fitness value;
S43, according to current Xα、XβAnd XδThe position of three grey wolves and location updating equation, update the position of each grey wolf;
S44, judges whether the standard deviation between each grey wolf position is less than the convergence precision of grey wolf algorithm, or judges current iteration
Whether number t is more than the grey wolf algorithm maximum iteration T of setting;If so, by XαGrey wolf exports as optimal solution, and output should
Otherwise the corresponding parameter vector of grey wolf repeats step S42-S44 processes.
5. the power of fan model parameter optimization method according to claim 4 based on grey wolf algorithm improvement, feature exist
Fitness value described in step S42 is calculated according to fitness function, and fitness function meets following criterion:It is average exhausted
Bigger to error MAE or root-mean-square error RMSE, the ideal adaptation angle value of return is smaller;The definition of described MAE, RMSE is:
Wherein (vi,Pa(vi)) it is ith cluster point, viFor wind speed, Pa(vi) it is the active power for corresponding to wind speed, Pε(Xj,vi) table
Show grey wolf XjWhen wind speed viThe active power value that lower object module returns, N are the number of cluster point.
6. the power of fan model parameter optimization method according to claim 4 based on grey wolf algorithm improvement, feature exist
The position of grey wolf and location updating equation are in step S43:
Wherein, parameterFor three grey wolves, the distance vector away from prey, parameter A, C are and random number r respectively1、
r2Related parameter, A=a (2r1- 1), C=2r2, a is from 2 linear decreases to 0, r1、r2It is the random vector between [0,1].
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810448294.2A CN108646560A (en) | 2018-05-11 | 2018-05-11 | Power of fan model parameter optimization method based on grey wolf algorithm improvement |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810448294.2A CN108646560A (en) | 2018-05-11 | 2018-05-11 | Power of fan model parameter optimization method based on grey wolf algorithm improvement |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108646560A true CN108646560A (en) | 2018-10-12 |
Family
ID=63754628
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810448294.2A Pending CN108646560A (en) | 2018-05-11 | 2018-05-11 | Power of fan model parameter optimization method based on grey wolf algorithm improvement |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108646560A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110174919A (en) * | 2019-05-07 | 2019-08-27 | 广州水沐青华科技有限公司 | Photovoltaic system maximum power tracking method, computer readable storage medium under the conditions of part masking based on depth grey wolf algorithm |
CN110210170A (en) * | 2019-06-19 | 2019-09-06 | 河北工业大学 | A kind of large-scale wind power group of planes equivalence small-signal model modeling method |
CN110414734A (en) * | 2019-07-29 | 2019-11-05 | 河北工业大学 | A method of meter and the assessment of wind-resources usage forecast |
CN111539508A (en) * | 2020-04-01 | 2020-08-14 | 国家电网公司华东分部 | Generator excitation system parameter identification algorithm based on improved wolf algorithm |
CN112068033A (en) * | 2020-09-02 | 2020-12-11 | 河北工业大学 | 1/6 periodic current-based inverter power tube open-circuit fault online identification method |
CN112249905A (en) * | 2020-11-02 | 2021-01-22 | 上海海事大学 | Dynamic weighing method of crane based on K-means clustering and linear regression |
CN112289329A (en) * | 2020-10-22 | 2021-01-29 | 国网青海省电力公司海西供电公司 | High-voltage circuit breaker fault diagnosis method based on GWO-KFCM |
CN112800103A (en) * | 2020-12-09 | 2021-05-14 | 华能陕西发电有限公司 | System and method for acquiring power curve of wind turbine generator based on SCADA data mining |
CN113051769A (en) * | 2021-04-09 | 2021-06-29 | 中南大学 | Power curve modeling method based on asymmetric loss and hybrid intelligent optimization algorithm |
CN113239503A (en) * | 2021-05-10 | 2021-08-10 | 上海电气工程设计有限公司 | New energy output scene analysis method and system based on improved k-means clustering algorithm |
CN113798918A (en) * | 2021-11-03 | 2021-12-17 | 天津工业大学 | Numerical control machine tool geometric error identification method based on wolf algorithm |
CN116203845A (en) * | 2023-03-03 | 2023-06-02 | 中国华能集团清洁能源技术研究院有限公司 | Method and system for configuring fan frequency response controller based on wolf algorithm |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105069192A (en) * | 2015-07-20 | 2015-11-18 | 河北工业大学 | Improved method for solving fan power curve parameter model based on genetic algorithm |
CN107067121A (en) * | 2017-06-13 | 2017-08-18 | 广东工业大学 | A kind of improvement grey wolf optimized algorithm based on multiple target |
CN107133695A (en) * | 2017-04-28 | 2017-09-05 | 北京天诚同创电气有限公司 | A kind of wind power forecasting method and system |
-
2018
- 2018-05-11 CN CN201810448294.2A patent/CN108646560A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105069192A (en) * | 2015-07-20 | 2015-11-18 | 河北工业大学 | Improved method for solving fan power curve parameter model based on genetic algorithm |
CN107133695A (en) * | 2017-04-28 | 2017-09-05 | 北京天诚同创电气有限公司 | A kind of wind power forecasting method and system |
CN107067121A (en) * | 2017-06-13 | 2017-08-18 | 广东工业大学 | A kind of improvement grey wolf optimized algorithm based on multiple target |
Non-Patent Citations (3)
Title |
---|
G.M.KOMAKI等: "Grey Wolf Optimizer algorithm for the two-stage assembly flow shop scheduling problem with release time", 《JOURNAL OF COMPUTATIONAL SCIENCE》 * |
ZENGWH513: "K-means++", 《HTTPS://BLOG.CSDN.NET/ZWHLXL/ARTICLE/DETAILS/41415585?UTM_MEDIUM=DISTRIBUTE.WAP_RELEVANT.NONE-TASK-BLOG-BLOGCOMMENDFROMMACHINELEARNPAI2-4.NONECASE&DEPTH_1-UTM_SOURCE=DISTRIBUTE.WAP_RELEVANT.NONE-TASK-BLOG-BLOGCOMMENDFROMMACHINELEARNPAI2-4.NONECASE》 * |
杨红光等: "一种结合灰狼优化和K-均值的混合聚类算法", 《江西理工大学学报》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110174919A (en) * | 2019-05-07 | 2019-08-27 | 广州水沐青华科技有限公司 | Photovoltaic system maximum power tracking method, computer readable storage medium under the conditions of part masking based on depth grey wolf algorithm |
CN110210170A (en) * | 2019-06-19 | 2019-09-06 | 河北工业大学 | A kind of large-scale wind power group of planes equivalence small-signal model modeling method |
CN110414734B (en) * | 2019-07-29 | 2022-03-22 | 河北工业大学 | Method for forecasting and evaluating wind resource utilization rate |
CN110414734A (en) * | 2019-07-29 | 2019-11-05 | 河北工业大学 | A method of meter and the assessment of wind-resources usage forecast |
CN111539508A (en) * | 2020-04-01 | 2020-08-14 | 国家电网公司华东分部 | Generator excitation system parameter identification algorithm based on improved wolf algorithm |
CN111539508B (en) * | 2020-04-01 | 2023-09-01 | 国家电网公司华东分部 | Generator excitation system parameter identification algorithm based on improved gray wolf algorithm |
CN112068033A (en) * | 2020-09-02 | 2020-12-11 | 河北工业大学 | 1/6 periodic current-based inverter power tube open-circuit fault online identification method |
CN112068033B (en) * | 2020-09-02 | 2024-03-26 | 河北工业大学 | On-line identification method for open-circuit faults of inverter power tube based on 1/6 period current |
CN112289329A (en) * | 2020-10-22 | 2021-01-29 | 国网青海省电力公司海西供电公司 | High-voltage circuit breaker fault diagnosis method based on GWO-KFCM |
CN112249905A (en) * | 2020-11-02 | 2021-01-22 | 上海海事大学 | Dynamic weighing method of crane based on K-means clustering and linear regression |
CN112800103A (en) * | 2020-12-09 | 2021-05-14 | 华能陕西发电有限公司 | System and method for acquiring power curve of wind turbine generator based on SCADA data mining |
CN113051769A (en) * | 2021-04-09 | 2021-06-29 | 中南大学 | Power curve modeling method based on asymmetric loss and hybrid intelligent optimization algorithm |
CN113239503A (en) * | 2021-05-10 | 2021-08-10 | 上海电气工程设计有限公司 | New energy output scene analysis method and system based on improved k-means clustering algorithm |
CN113798918A (en) * | 2021-11-03 | 2021-12-17 | 天津工业大学 | Numerical control machine tool geometric error identification method based on wolf algorithm |
CN116203845A (en) * | 2023-03-03 | 2023-06-02 | 中国华能集团清洁能源技术研究院有限公司 | Method and system for configuring fan frequency response controller based on wolf algorithm |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108646560A (en) | Power of fan model parameter optimization method based on grey wolf algorithm improvement | |
CN108846472A (en) | A kind of optimization method of Adaptive Genetic Particle Swarm Mixed Algorithm | |
CN105426954A (en) | Particle swarm optimization method based on multi-strategy synergistic function | |
CN106960244A (en) | A kind of genetic algorithm and the evolution algorithm of particle cluster algorithm Parallel Fusion | |
CN111178440B (en) | Path optimization method based on efficient reduction of total transport cost of AGV | |
CN108062720A (en) | A kind of load forecasting method based on similar day selection and random forests algorithm | |
CN111811111A (en) | Central air conditioner energy consumption control method based on improved particle swarm algorithm | |
CN108564592A (en) | Based on a variety of image partition methods for being clustered to differential evolution algorithm of dynamic | |
CN113783224A (en) | Power distribution network double-layer optimization planning method considering operation of various distributed energy sources | |
CN110929399A (en) | Wind power output typical scene generation method based on BIRCH clustering and Wasserstein distance | |
CN108830408A (en) | A kind of equipment support task multi-objective planning method that oriented mission is resource matched | |
CN109831790A (en) | Cooperation caching method based on brainstorming optimization algorithm in mist wireless access network | |
CN115577864A (en) | Distribution network operation optimization scheduling method based on multi-model combined operation | |
CN115940292A (en) | Wind-containing power storage system optimal scheduling method and system based on circle search algorithm | |
CN107229998A (en) | A kind of autonomous pathfinding strategy process of unmanned plane | |
CN111160654B (en) | Transportation path optimization method for reducing total cost based on fuzzy C-means-simulated annealing algorithm | |
CN109635999B (en) | Hydropower station scheduling method and system based on particle swarm-bacterial foraging | |
CN116207797B (en) | Coordinated optimization control method for multiple series-parallel equipment of power distribution network | |
CN108805463A (en) | A kind of production scheduling method for supporting peak clipping type electricity needs to respond | |
CN105894034A (en) | Commercial load characteristic clustering analysis method based on quadratic clustering | |
CN113937808B (en) | Distributed power source location and volume-fixing optimization method based on improved sparrow search algorithm | |
CN115526267A (en) | Power distribution network operation scene extraction method and device | |
CN110932334B (en) | Wind power plant power control method with constraint multi-objective optimization | |
CN110970936B (en) | Method for calculating primary frequency modulation performance of deep peak shaving unit | |
CN113947321A (en) | Distributed information energy network topological structure construction method and system |
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: 20181012 |
|
RJ01 | Rejection of invention patent application after publication |