CN108205698A - A kind of cloud resource load predicting method based on the double string whale optimization algorithms of just remaining chaos - Google Patents
A kind of cloud resource load predicting method based on the double string whale optimization algorithms of just remaining chaos Download PDFInfo
- Publication number
- CN108205698A CN108205698A CN201711164036.3A CN201711164036A CN108205698A CN 108205698 A CN108205698 A CN 108205698A CN 201711164036 A CN201711164036 A CN 201711164036A CN 108205698 A CN108205698 A CN 108205698A
- Authority
- CN
- China
- Prior art keywords
- whale
- individual
- chaos
- formula
- algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Probability & Statistics with Applications (AREA)
- Pure & Applied Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Feedback Control In General (AREA)
Abstract
In order to solve the disadvantage that traditional cloud computing resources load predicting method is not high to load sequence high fdrequency component precision of prediction and generalization ability is weak, the present invention proposes a kind of hybrid wavelet packet transform and the short-term cloud computing resources load predicting method of double string whale optimization (CSCWOA) the algorithm optimization multilayer perceptron neural networks (MLP) of just remaining chaos.Multiband pretreatment is carried out to load sequence by wavelet package transforms to decompose, then the load subsequence of single branch reconstruct gained is predicted using the MLP neural networks of CSCWOA algorithm optimizations, is finally superimposed the predicted value of each subsequence to obtain actual prediction result.The program can grasp the changing rule of each frequency range impact burr of load sequence, have preferable precision of prediction and generalization ability.
Description
Technical field
Resource load the invention belongs to cloud computing predicts field, especially a kind of based on the double string whale optimizations of just remaining chaos
The cloud resource load predicting method of algorithm.
Background technology
Cloud computing resources load estimation be cloud computing system platform planning important component, the direct shadow of precision of prediction
Ring economy, safety and the service quality to cloud computing system.It according to historic load when axis on establish now with
Quantitative relationship between the load of following cloud resource come the short-term cloud resource load state information that obtains, be cloud resource planning distribution,
The performance optimization and the cost minimization of cloud operator of cloud computing platform provide rational foundation.
With the multiplication of cloud computing data magnanimity and complicated and changeable, the autoregressive moving average (ARMA) of conventional linear statistic
Model, difference autoregressive moving average (ARIMA) model and difference autoregression summation sliding average (FARIMA) model without
Method ensures the accuracy that the cloud computing resources of chaotic sea are predicted in switch-time load.Due to neural network and support vector machines
SVM is realized simply, suitable for Nonlinear Time Series, therefore is gradually applied in load estimation field.However, nerve net
The threshold value and weights of network are adjusted using gradient descent method, lead to that algorithm is easily trapped into local optimum and convergence rate is slow.Vector machine
The selection of SVM parameters decides the quality of prediction result, causes prediction model generalization ability weak.Although scholars introduce manually
Ant group algorithm[7], particle cluster algorithm, the swarm intelligence algorithms such as grey wolf algorithm neural network and vector machine SVM are optimized, still
Since this kind of swarm intelligence algorithm itself is existing insufficient, its precision of forecasting model optimized is caused also to there is the place promoted.Together
When, load sequence uses the relevant preprocess methods such as wavelet transformation before prediction,
Lack the decomposition to sequence high fdrequency component, affect precision of prediction indirectly.
Invention content
The characteristics of changing and existing optimization algorithm and the deficiency of data prediction are loaded for cloud computing resources, invention is a kind of
Hybrid wavelet packet transform (Wavelet Packet Decomposition, WPD) and the double string whale optimization (Whale of just remaining chaos
Optimization algorithm based on chaotic sine cosine operator, CSCWOA) algorithm optimization
The short-term cloud computing resources load predicting method of multilayer perceptron (MultiLayer Perceptron, MLP) neural network
(WPD-CSCWOA-MLP), the prediction of 100h in advance and to cloud computing platform has been carried out.The experimental results showed that wavelet package transforms are more
Effectively optimization neural network is searched in dimensional analysis load sequence and the double string optimizing of the just remaining chaos of CSCWOA, improves precision of prediction
And generalization ability.
The purpose of the present invention is achieved through the following technical solutions:
1. wavelet package transforms pre-process the time series of cloud computing resources load data
WAVELET PACKET DECOMPOSITION (WPD) is that only parsing original signal is low in wavelet decomposition (Wavelet decomposition, WD)
On the basis of frequency part, increase a kind of frequency-domain analysis method parsed to high frequency section.Relative to conventionally employed wavelet decomposition
It predicts reason method, increases the parsing to high fdrequency component, the variation rule that burr is impacted in load sequence are caught in can be finer and smoother catch
Rule helps to improve the precision of prediction and generalization ability of prediction model.
2. the double string whale optimization algorithms of just remaining chaos
The double string whale optimization algorithms (CSCWOA) of just remaining chaos are in whale optimization algorithm (Whale Optimization
Algorithm, WOA) on the basis of a kind of innovatory algorithm for proposing.CSCWOA algorithms add in the predation of artificial whale
The double string mechanism of just remaining chaos make up the defects of WOA algorithms are easily trapped into locally optimal solution, and accelerate its convergence rate, and raising is searched
The accuracy of rope optimizing
(1) basic whale algorithm
The Optimum search process of WOA algorithms is broadly divided into following three links, and wherein migration, which is looked for food and surrounded, shrinks link
Switching, by section [- 2,2] random parameter A determine.
1) migration is looked for food.As -1≤A≤1, artificial whale utilizes the random individual coordinate X of populationrandCarry out location navigation
Search of food, as shown in formula (1).
Xt+1=Xrand-A·D (1)
Wherein, Xt+1It is the coordinate after t generations individual location updating, D=| CXrand-Xt| represent current individual X with it is random
Individual XrandDistance, and C is the random number on section [0,2], controls XrandThe influence of distance X how fars.
2) it surrounds and shrinks.As A ＜ -1 or A ＞ 1, if artificial whale searches out the position X of global optimum's individualbest, open
Begin to surround food, predation range is shunk, as shown in formula (2).
Xt+1=Xbest-A·|C·Xbest-Xt| (2)
3) helical is preyed on.Artificial whale is surrounding optimum individual XbestWhile, it can also follow the track of log spiral
Movement, which is caught, grabs food, such as shown in (3).
Xt+1=Dbest·ebl·cos(2πl)+Xt (3)
Wherein, Dbest=| Xbest-Xt| represent individual X and optimum individual XbestDistance, b is to mould spiral trajectory
Constant, l are the random numbers on section [- 1,1], and as l=-1, artificial whale is nearest apart from food, as l=1, artificial whale
It is farthest apart from food.Because formula (3) allows artificial whale to capture the arbitrary food for being distributed in search space, ensure artificial whale
Global search of the fish in optimizing is developed with part, reduces optimizing blind spot
(2) the double string mechanism of just remaining chaos
Artificial whale is during helical is preyed on optimum individual XbestCoordinate carry out location navigation, accelerate the later stage receive
While holding back speed, also it is easy to cause population at individual and flocks together rapidly in solution space, accelerate the decline of population diversity,
And increase the probability of algorithm precocity.In order to reduce the possibility that individual is gathered to local minimum region, just remaining double strings are incorporated herein
Mechanism and chaos operator control the moving region of population at individual, improve the ability that algorithm jumps out local optimum, calculate difference
As shown in formula (4) and (5).
In the sinusoidal chaos helical predation movement of formula (4), artificial whale is with random population individual XrandFor navigation coordinate,
Logarithmic sine screw motion is done, global search capture food maintains the diversity of population, individual is avoided to gather optimizing, is absorbed in office
Portion's optimal solution.Movement is preyed on by the cosine chaos helical of formula (5) simultaneously, with optimum individual XbestFor navigation coordinate, raising is sought
The speed of excellent positioning.Wherein, parameter r3For the random number on [0,2] section, control random individual XrandWith optimum individual Xbest
With the influence of current individual X distance degree.And parameter r1Effect be that the sinusoidal global search of control and cosine are locally developed
Regional extent, calculate as shown in formula (6).In formula (6), a is constant, and t is current iteration number, and T is total iterations, r1
It is adaptively reduced with the increase of iterations, reduces the optimizing regional extent of the just remaining double strings of artificial whale, finally make algorithm
It converges in same optimal solution, ensure that its convergence.
Parameter r2The Optimizing operator based on cube chaotic maps, relative to general chaos operator sequence, it have compared with
Good balanced ergodic and convergence efficiency[13], calculate as shown in formula (7).The randomness that formula (7) passes through chaotic maps itself
And ergodic, the degree of variation of population at individual is adaptively adjusted, enhances artificial whale and jumps out part in just remaining double string optimizing
The ability of optimal solution.
Fig. 1 illustrate the double strings predations of just remaining chaos in same space to carried out in the range of different zones global search and
Local optimal searching.Sinusoidal chaos global search helps cosine chaos, and locally exploitation reduces optimizing blind spot, avoids potential optimal solution quilt
It loses.And locally exploitation compensates for the defects of sinusoidal chaos global search convergence rate is slow to cosine chaos, improves efficiency of algorithm.
And passing through the greedy mechanism of introducing, the filial generation solution that more sinusoidal and cosine generates in chaos predation preferentially retains.Just remaining chaos is double
String intersects optimizing, complements each other, and on the one hand prevents algorithm precocious, improves solving precision, on the other hand accelerates convergence rate, improves
Solution efficiency.The two promotes propagation of the individual information in population, and algorithm is enabled to quickly converge on same optimal solution.
3. improve whale algorithm optimization neural network
Multilayer perceptron (MLP) neural network is a kind of feed forward-fuzzy control.The MLP neural networks of three-decker are
Suitable for the most simple network of arbitrary non-stationary signal prediction processing, generalization ability and treatment effeciency are relative to other nerve nets
Network, it is more advantageous.However MLP neural networks are during prediction, using gradient descent method come adjust the threshold values of its network and
Weights cause MLP neural metwork trainings slow-footed while are easily trapped into local extremum, learn to be not enough, reduce pre-
Survey precision.CSCWOA algorithms have preferable convergence rate and jump out the ability of locally optimal solution, for overcome MLP neural networks with
Upper shortcoming provides possibility.
Fig. 2 illustrates the flow of CSCWOA training MLP neural networks.Using the MLP neural networks of three-decker, if its is defeated
The number of nodes for entering layer is n, and the hidden layer number of plies is 1, number of nodes h, and threshold values and weights are respectively θ, ω, then artificial whale is individual
It is encoded to WI=1 ..., N={ ω11,ω12,...,ωnn,θ1,θ2,...,θh}.Weigh in artificial whale population every it is individual suitable
Shown in the calculating such as formula (8) for answering angle value fobj.
In formula (8), m is the number of nodes of output layer, and o is the real output value of MLP neural networks, and d is then defeated for its expectation
Go out value, S is training sample sum.
4. establish prediction model
WPD-CSCWOA-MLP prediction models before being predicted, first with WAVELET PACKET DECOMPOSITION pre-process load sequence, by when
Between it is Sequence Transformed into complete frequency-domain analysis subsequence, each subsequence is then handled by CSCWOA-MLP, in Optimized model,
MLP provides the mean square error MSE of prediction for CSCWOA, and CSCWOA then trains optimizing to provide valve for MLP according to mean square error MSE
Value θ and weights ω, each prediction subsequence component superposition for finally obtaining CSCWOA-MLP training, obtain actual prediction as a result,
As shown in Figure 3
Description of the drawings
Fig. 1 is the double string optimizing figures of just remaining chaos of the present invention
Fig. 2 is the CSCWOA training MLP neural network flow charts of the present invention
Fig. 3 is the WPD-CSCWOA-MLP prediction model figures of the present invention
Specific embodiment
The present invention is described in detail With reference to embodiment.
A. the time series of cloud computing resources load data is pre-processed by wavelet package transforms
The burr of the time series load curve of cloud computing resources load data is more, but hair caused by impact load
Thorn is not bad value.If cannot be effectively addressed in data prediction, precision of prediction can be influenced.By introducing wavelet packet
The variation that burr is impacted in load sequence is caught in the time series of preconditioning cloud computing resources load data, finer and smoother catching
Rule helps to improve the precision of prediction and generalization ability of prediction model.
B. pass through the double string mechanism optimization whale algorithms of just remaining chaos
Artificial whale is during helical is preyed on optimum individual XbestCoordinate carry out location navigation, accelerate the later stage receive
While holding back speed, also it is easy to cause population at individual and flocks together rapidly in solution space, accelerate the decline of population diversity,
And increase the probability of algorithm precocity.In order to reduce the possibility that individual is gathered to local minimum region, just remaining double strings are incorporated herein
Mechanism and chaos operator control the moving region of population at individual, improve the ability that algorithm jumps out local optimum, calculate difference
As shown in formula (4) and (5).
In the sinusoidal chaos helical predation movement of formula (4), artificial whale is with random population individual XrandFor navigation coordinate,
Logarithmic sine screw motion is done, global search capture food maintains the diversity of population, individual is avoided to gather optimizing, is absorbed in office
Portion's optimal solution.Movement is preyed on by the cosine chaos helical of formula (5) simultaneously, with optimum individual XbestFor navigation coordinate, raising is sought
The speed of excellent positioning.Wherein, parameter r3For the random number on [0,2] section, control random individual XrandWith optimum individual Xbest
With the influence of current individual X distance degree.And parameter r1Effect be that the sinusoidal global search of control and cosine are locally developed
Regional extent, calculate as shown in formula (6).In formula (6), a is constant, and t is current iteration number, and T is total iterations, r1
It is adaptively reduced with the increase of iterations, reduces the optimizing regional extent of the just remaining double strings of artificial whale, finally make algorithm
It converges in same optimal solution, ensure that its convergence.
Parameter r2The Optimizing operator based on cube chaotic maps, relative to general chaos operator sequence, it have compared with
Good balanced ergodic and convergence efficiency[13], calculate as shown in formula (7).The randomness that formula (7) passes through chaotic maps itself
And ergodic, the degree of variation of population at individual is adaptively adjusted, enhances artificial whale and jumps out part in just remaining double string optimizing
The ability of optimal solution.
Fig. 1 illustrate the double strings predations of just remaining chaos in same space to carried out in the range of different zones global search and
Local optimal searching.Sinusoidal chaos global search helps cosine chaos, and locally exploitation reduces optimizing blind spot, avoids potential optimal solution quilt
It loses.And locally exploitation compensates for the defects of sinusoidal chaos global search convergence rate is slow to cosine chaos, improves efficiency of algorithm.
And pass through the greedy mechanism of introducing,
Compare the filial generation solution that sinusoidal and cosine generates in chaos predation, preferentially retain.The double strings of just remaining chaos intersect optimizing,
It complements each other, on the one hand prevents algorithm precocious, improve solving precision, on the other hand accelerate convergence rate, improve solution efficiency.Two
Person promotes propagation of the individual information in population, and algorithm is enabled to quickly converge on same optimal solution.
C. pass through improved whale algorithm optimization neural network
Multilayer perceptron (MLP) neural network is a kind of feed forward-fuzzy control.The MLP neural networks of three-decker are
Suitable for the most simple network of arbitrary non-stationary signal prediction processing, generalization ability and treatment effeciency are relative to other nerve nets
Network, it is more advantageous.
However MLP neural networks adjust the threshold values of its network and weights using gradient descent method, lead during prediction
It causes MLP neural metwork trainings slow-footed while is easily trapped into local extremum, learn to be not enough, reduce precision of prediction.
CSCWOA algorithms have preferable convergence rate and jump out the ability of locally optimal solution, to overcome MLP neural network disadvantage mentioned above
Provide possibility.
Fig. 2 illustrates the flow of CSCWOA training MLP neural networks.Using the MLP neural networks of three-decker, if its is defeated
The number of nodes for entering layer is n, and the hidden layer number of plies is 1, number of nodes h, and threshold values and weights are respectively θ, ω, then artificial whale is individual
It is encoded to WI=1 ..., N={ ω11,ω12,...,ωnn,θ1,θ2,...,θh}.Weigh in artificial whale population every it is individual suitable
Shown in the calculating such as formula (8) for answering angle value fobj.
In formula (8), m is the number of nodes of output layer, and o is the real output value of MLP neural networks, and d is then defeated for its expectation
Go out value, S is training sample sum.
D.WPD-CSCWOA-MLP prediction models are established
WPD-CSCWOA-MLP prediction models before being predicted, first with WAVELET PACKET DECOMPOSITION pre-process load sequence, by when
Between it is Sequence Transformed into complete frequency-domain analysis subsequence, each subsequence is then handled by CSCWOA-MLP, in Optimized model,
MLP provides the mean square error MSE of prediction for CSCWOA, and CSCWOA then trains optimizing to provide valve for MLP according to mean square error MSE
Value θ and weights ω, each prediction subsequence component superposition for finally obtaining CSCWOA-MLP training, obtain actual prediction as a result,
As shown in Figure 3.
Claims (4)
- A kind of 1. cloud resource load predicting method based on the double string whale optimization algorithms of just remaining chaos, which is characterized in that the side Method includes the following steps：1. the time series of cloud computing resources load data is pre-processed by WAVELET PACKET DECOMPOSITION method；2. whale algorithm is improved by sine and cosine mechanism；3. use the improvement whale algorithm optimization neural network；4. using the neural network prediction model, and the pretreated time sequence is inputted in the prediction model Column data carries out the cloud resource load estimation.
- 2. Forecasting Methodology as described in claim 1, which is characterized in that described specific by sine and cosine mechanism improvement whale algorithm Including：Just remaining double string mechanism and chaos operator are introduced in whale algorithm to control the moving region of population at individual, algorithm is improved and jumps Go out the ability of local optimum, calculate respectively as shown in formula (4) and (5)：In the sinusoidal chaos helical predation movement of formula (4), artificial whale is with random population individual XrandFor navigation coordinate, do just String logatithmic spiral moves, and global search capture food maintains the diversity of population, individual is avoided to gather optimizing, is absorbed in part most Excellent solution；Movement is preyed on by the cosine chaos helical of formula (5) simultaneously, with optimum individual XbestFor navigation coordinate, improve optimizing and determine The speed of position；Wherein, parameter r3For the random number on [0,2] section, for controlling random individual XrandWith optimum individual XbestWith The influence of current individual X distance degree；And parameter r1The sinusoidal global search of effect control and cosine locally develop Regional extent is calculated as shown in formula (6).In formula (6), a is constant, and t is current iteration number, and T is total iterations, r1With It the increase of iterations and adaptively reduces, reduce the optimizing regional extent of the just remaining double strings of artificial whale, finally receive algorithm It holds back in same optimal solution, ensure that its convergence；Parameter r2It is the Optimizing operator based on cube chaotic maps, relative to general chaos operator sequence, it has preferable equal The ergodic that weighs and convergence efficiency are calculated as shown in formula (7).Formula (7) by the randomness and ergodic of chaotic maps itself, The degree of variation of population at individual is adaptively adjusted, enhances the energy that artificial whale jumps out locally optimal solution in just remaining double string optimizing Power：
- 3. Forecasting Methodology as claimed in claim 2, which is characterized in that described to use the improvement whale algorithm optimization nerve net Network specifically includes：Using the MLP neural networks of three-decker, the number of nodes of input layer is n, and the hidden layer number of plies is 1, number of nodes h, Threshold value and weights are respectively θ, ω, and artificial whale individual UVR exposure is WI=1 ..., N={ ω11,ω12,...,ωnn,θ1,θ2,..., θh}；Shown in the calculating such as formula (8) for weighing every individual fitness value fobj in artificial whale population：Wherein, m is the number of nodes of output layer, and o is the real output value of MLP neural networks, and d is then its desired output, and S is instruction Practice total sample number.
- 4. Forecasting Methodology as claimed in claim 3, which is characterized in that the cloud resource load estimation specifically includes：Pass through institute It states pretreatment and the time series is converted to complete frequency-domain analysis subsequence, each son is then handled by the prediction model Sequence obtains actual prediction result.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710452380 | 2017-06-15 | ||
| CN2017104523806 | 2017-06-15 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108205698A true CN108205698A (en) | 2018-06-26 |
Family
ID=62603705
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711164036.3A Pending CN108205698A (en) | 2017-06-15 | 2017-11-21 | A kind of cloud resource load predicting method based on the double string whale optimization algorithms of just remaining chaos |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108205698A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830431A (en) * | 2018-08-03 | 2018-11-16 | 广东工业大学 | A kind of Electricity price forecasting solution and relevant apparatus based on whale optimization algorithm |
| CN109738519A (en) * | 2019-01-04 | 2019-05-10 | 国网四川省电力公司广安供电公司 | A kind of denoising method of transformer high-voltage bushing lead ultrasound detection |
| CN109992412A (en) * | 2019-02-19 | 2019-07-09 | 广州视源电子科技股份有限公司 | Capacity adjusting method and device of cloud server, storage medium and cloud server |
| CN110322050A (en) * | 2019-06-04 | 2019-10-11 | 西安邮电大学 | A kind of wind energy resources compensation data method |
| CN110377957A (en) * | 2019-06-20 | 2019-10-25 | 浙江大学 | A kind of overhead crane neural network modeling approach of whale search strategy grey wolf algorithm |
| CN110472046A (en) * | 2019-07-11 | 2019-11-19 | 广东奥博信息产业股份有限公司 | A kind of government and enterprises' service Text Clustering Method |
| CN110516831A (en) * | 2019-06-18 | 2019-11-29 | 国网（北京）节能设计研究院有限公司 | A kind of short-term load forecasting method based on MWOA algorithm optimization SVM |
| CN110927490A (en) * | 2019-12-02 | 2020-03-27 | 吉林松江河水力发电有限责任公司 | Transformer winding fault diagnosis method based on IVMD (integrated virtual machine direction) permutation entropy and CWAA-SVM (continuous wave operation-support vector machine) |
| CN110996333A (en) * | 2019-11-06 | 2020-04-10 | 湖北工业大学 | Wireless sensor network node positioning method based on whale algorithm |
| CN112163808A (en) * | 2020-09-23 | 2021-01-01 | 贵州工程应用技术学院 | Method for solving logistics center addressing problem by self-adaptive whale algorithm based on opponent learning |
| CN112507610A (en) * | 2020-11-20 | 2021-03-16 | 东北大学 | Interval prediction method for hot rolling coiling temperature |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184679A (en) * | 2015-09-22 | 2015-12-23 | 国网山东省电力公司电力科学研究院 | Wind power plant power multi-time scale joint prediction method |
| CN106127298A (en) * | 2016-06-16 | 2016-11-16 | 太原科技大学 | A kind of Fault Diagnosis of Hydraulic Pump method based on self adaptation PSO BP neutral net |
| CN106789214A (en) * | 2016-12-12 | 2017-05-31 | 广东工业大学 | It is a kind of based on the just remaining pair network situation awareness method and device of string algorithm |
-
2017
- 2017-11-21 CN CN201711164036.3A patent/CN108205698A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184679A (en) * | 2015-09-22 | 2015-12-23 | 国网山东省电力公司电力科学研究院 | Wind power plant power multi-time scale joint prediction method |
| CN106127298A (en) * | 2016-06-16 | 2016-11-16 | 太原科技大学 | A kind of Fault Diagnosis of Hydraulic Pump method based on self adaptation PSO BP neutral net |
| CN106789214A (en) * | 2016-12-12 | 2017-05-31 | 广东工业大学 | It is a kind of based on the just remaining pair network situation awareness method and device of string algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 刘竹松 等: "正余混沌双弦鲸鱼优化算法", 《计算机工程与应用》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830431A (en) * | 2018-08-03 | 2018-11-16 | 广东工业大学 | A kind of Electricity price forecasting solution and relevant apparatus based on whale optimization algorithm |
| CN109738519A (en) * | 2019-01-04 | 2019-05-10 | 国网四川省电力公司广安供电公司 | A kind of denoising method of transformer high-voltage bushing lead ultrasound detection |
| CN109738519B (en) * | 2019-01-04 | 2021-08-17 | 国网四川省电力公司广安供电公司 | Denoising method for ultrasonic detection of lead of high-voltage bushing of transformer |
| CN109992412B (en) * | 2019-02-19 | 2021-08-10 | 广州视源电子科技股份有限公司 | Capacity adjusting method and device of cloud server, storage medium and cloud server |
| CN109992412A (en) * | 2019-02-19 | 2019-07-09 | 广州视源电子科技股份有限公司 | Capacity adjusting method and device of cloud server, storage medium and cloud server |
| CN110322050A (en) * | 2019-06-04 | 2019-10-11 | 西安邮电大学 | A kind of wind energy resources compensation data method |
| CN110322050B (en) * | 2019-06-04 | 2023-04-07 | 西安邮电大学 | Wind energy resource data compensation method |
| CN110516831A (en) * | 2019-06-18 | 2019-11-29 | 国网（北京）节能设计研究院有限公司 | A kind of short-term load forecasting method based on MWOA algorithm optimization SVM |
| CN110377957A (en) * | 2019-06-20 | 2019-10-25 | 浙江大学 | A kind of overhead crane neural network modeling approach of whale search strategy grey wolf algorithm |
| CN110377957B (en) * | 2019-06-20 | 2020-10-20 | 浙江大学 | Bridge crane neural network modeling method of whale search strategy wolf algorithm |
| CN110472046B (en) * | 2019-07-11 | 2022-02-22 | 广东奥博信息产业股份有限公司 | Government and enterprise service text clustering method |
| CN110472046A (en) * | 2019-07-11 | 2019-11-19 | 广东奥博信息产业股份有限公司 | A kind of government and enterprises' service Text Clustering Method |
| CN110996333A (en) * | 2019-11-06 | 2020-04-10 | 湖北工业大学 | Wireless sensor network node positioning method based on whale algorithm |
| CN110927490A (en) * | 2019-12-02 | 2020-03-27 | 吉林松江河水力发电有限责任公司 | Transformer winding fault diagnosis method based on IVMD (integrated virtual machine direction) permutation entropy and CWAA-SVM (continuous wave operation-support vector machine) |
| CN112163808A (en) * | 2020-09-23 | 2021-01-01 | 贵州工程应用技术学院 | Method for solving logistics center addressing problem by self-adaptive whale algorithm based on opponent learning |
| CN112163808B (en) * | 2020-09-23 | 2022-04-01 | 贵州工程应用技术学院 | Method for solving logistics center addressing problem by self-adaptive whale algorithm based on opponent learning |
| CN112507610A (en) * | 2020-11-20 | 2021-03-16 | 东北大学 | Interval prediction method for hot rolling coiling temperature |
| CN112507610B (en) * | 2020-11-20 | 2023-09-26 | 东北大学 | Interval prediction method for hot rolling coiling temperature |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108205698A (en) | A kind of cloud resource load predicting method based on the double string whale optimization algorithms of just remaining chaos | |
| CN109344959A (en) | Neural network training method, nerve network system and computer system | |
| CN108038507A (en) | Local receptor field extreme learning machine image classification method based on particle group optimizing | |
| CN104866904A (en) | Parallelization method of BP neural network optimized by genetic algorithm based on spark | |
| CN106357456A (en) | Prediction method of network traffic and device thereof | |
| CN107798379A (en) | Improve the method for quantum particle swarm optimization and the application based on innovatory algorithm | |
| CN109803341A (en) | Adaptive path planning method in wireless sensor network | |
| CN109934422A (en) | Neural network wind speed prediction method based on time series data analysis | |
| CN108154003A (en) | Prediction of Blasting Vibration method based on the optimization of Spark gene expressions | |
| CN114741955A (en) | Multi-objective optimization task scheduling method based on security cloud | |
| Wang et al. | DeepNetQoE: Self-adaptive QoE optimization framework of deep networks | |
| CN116614394A (en) | Service function chain placement method based on multi-target deep reinforcement learning | |
| CN111445065A (en) | Energy consumption optimization method and system for refrigeration group control of data center | |
| Zhao et al. | Adaptive swarm intelligent offloading based on digital twin-assisted prediction in VEC | |
| CN108896330B (en) | Fault diagnosis method for hydroelectric generating sets | |
| CN114444755A (en) | Wind power short-term power prediction method and device and readable storage medium | |
| CN112417761A (en) | Mooring truncation design method based on multi-objective cuckoo optimization algorithm | |
| CN116451859A (en) | Bayesian optimization-based stock prediction method for generating countermeasure network | |
| CN107995027A (en) | Improved quantum particle swarm optimization and the method applied to prediction network traffics | |
| Shuang et al. | Task scheduling based on Grey Wolf optimizer algorithm for smart meter embedded operating system | |
| CN115271254A (en) | Short-term wind power prediction method for optimizing extreme learning machine based on gull algorithm | |
| Guosheng et al. | Cloud security situation prediction method based on grey wolf optimization and BP neural network | |
| Li et al. | Improved harmony search algorithm for global optimization | |
| Balevi et al. | An inherent fog network: brain-spinal cord-nerve networks | |
| Sun et al. | Optimal scheduling method of marine transportation resources based on wolf swarm algorithm |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180626 |