CN107274038A - A kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion - Google Patents
A kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion Download PDFInfo
- Publication number
- CN107274038A CN107274038A CN201710641743.0A CN201710641743A CN107274038A CN 107274038 A CN107274038 A CN 107274038A CN 201710641743 A CN201710641743 A CN 201710641743A CN 107274038 A CN107274038 A CN 107274038A
- Authority
- CN
- China
- Prior art keywords
- ant
- lion
- ant lion
- value
- lssvm
- 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
- 241001206881 Myrmeleon inconspicuus Species 0.000 title claims abstract description 118
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000005611 electricity Effects 0.000 title claims abstract description 32
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 241000258923 Neuroptera Species 0.000 claims abstract description 7
- 230000000717 retained effect Effects 0.000 claims abstract description 5
- 238000005295 random walk Methods 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000012843 least square support vector machine Methods 0.000 abstract description 3
- 238000012549 training Methods 0.000 description 9
- 238000010219 correlation analysis Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 241000282320 Panthera leo Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000019637 foraging behavior Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000002147 killing effect Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000002829 reductive effect Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Operations Research (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion, Prediction of annual electricity consumption method comprises the following steps：Determine the input variable of least square method supporting vector machine (Least Square Support Vector Machines, LSSVM) forecast model；Initialize ant lion optimized algorithm；The fitness value of initial ant lion is calculated, initial elite ant lion is obtained；The position of ant is updated, the fitness value of current ant is calculated, and the fitness value of ant lion corresponding to its is compared, and judges whether to update ant lion position；By the fitness value of the ant lion after more new position, the fitness value with previous generation elite ant lions is compared one by one, is retained the corresponding ant lion of smaller fitness value, is obtained the elite ant lion of current iteration；Judge whether to reach maximum iteration, if it has, then position and the corresponding Prediction of annual electricity consumption value of output elite ant lion, if it has not, continuing iteration.Compared with prior art, the present invention has the advantages that precision of prediction is higher and forecasting efficiency is higher.
Description
Technical field
The present invention relates to Prediction of annual electricity consumption technical field, used more particularly, to a kind of based on the LSSVM that ant lion optimizes
Power predicating method.
Background technology
Planning of the Prediction of annual electricity consumption to power system, runs and safeguards most important, can also be anti-to a certain extent
Reflect the economic development of a country.Accurate Prediction of annual electricity consumption can be that power system operator and economic manager are carried
For valuable reference.Therefore it is one of most important groundwork of power system electric load to be predicted, and it is to the energy
Planning, the operation of power system and control and Strategy for economic development research important in inhibiting.The method of prediction has：Regression analysis
Method, analogy method, elastic coefficient method, time series method, neural network etc..
In modern economy society, all existed between power consumption and economy, society, population and ecological environment abnormal close
Relation, i.e. electricity system is a complication system influenceed by many factors, and prediction is set up with conventional mathematical method
Model, not only workload is big, and precision of prediction is also difficult to ensure that.
The content of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind is based on ant lion optimization
LSSVM Prediction of annual electricity consumption methods.
The purpose of the present invention can be achieved through the following technical solutions：
A kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion, described Prediction of annual electricity consumption method include with
Lower step：
S1, the input variable for determining LSSVM forecast models；
S2, initialization ant lion optimized algorithm, LSSVM moulds are substituted into using initial ant lion position as nuclear parameter and regularization parameter
Type, obtains corresponding Prediction of annual electricity consumption value；
S3, set up fitness function, calculate the fitness value of initial ant lion position, obtain initial fitness value, retain most
The corresponding ant lion of small initial fitness value is used as initial elite ant lion；
S4, ant position is updated, calculate the fitness value of contemporary ant, and ant lion position corresponding with contemporary ant is suitable
Answer angle value to be compared, judge whether to update ant lion position；
S5, the fitness value by the fitness value of ant lion position obtained in the previous step one by one with previous generation elite ant lions position
Compare, retain the corresponding ant lion position of smaller fitness value, obtain current iteration elite ant lion position；
S6, judge whether to reach maximum iteration, if it has, then output elite ant lion position and its corresponding year electricity consumption
Predicted value is measured, if it has not, returning to S4.
Step S1 is specially：Using Grey Incidence Analysis, obtain between year power consumption influence factor and year power consumption
Association angle value, according to association angle value, select corresponding year power consumption influence factor to become as the input of LSSVM forecast models
Amount.
Ant lion optimized algorithm is initialized in step S2 to comprise the following steps：
S201, parameter initialization, including population scale Agents_no, the dimension d=2 of solution, the upper bound b of solution spaceup, solution
The lower bound b in spacelow, maximum iteration Max_iter；
S202, position initialization, randomly generate the location matrix M of antAntWith the location matrix M of ant lionAntlion。
The upper bound b of solution space described in S201up=[1000,1000], lower bound blow=[0.1,0.01].
The location matrix M of ant in S202AntWith the location matrix M of ant lionAntlionFor：
Wherein MAntAnd MAntlionIn value by formula A*jOr
AL*j=rand* (bupj-blowj)+blowjObtain, A*jAnd AL*jAnt location matrix and ant lion location matrix jth row are represented respectively
Value, rand be 0 to 1 between a random number, n=Agents_no, bupjAnd blowjRespectively the upper bound of jth row is with
Boundary, MAntAnd MAntlionEvery a line correspond to LSSVM one group of nuclear parameter and regularization parameter, i.e. (σ, γ).
Fitness function in step S3 is：Wherein,WithRespectively s groups are examined
The actual value and predicted value of sample are tested, N is number of samples.
The more new formula of ant position is in step S4Wherein,Expression is swum at random around ant lion
The selected ant position of the t times iteration roulette walked,Represent the t times iteration of the random walk around elite ant lion
Ant position,The position of i-th of ant during for the t times iteration.
Wherein, aiRepresent the position of random walk of i-th of ant most
Small value, hiThe position maximum of the random walk of i-th of ant is represented,Represent the actual bit of i-th of ant during the t times iteration
Put minimum value,Represent the physical location maximum of i-th of ant during the t times iteration, Wi tRepresent i-th of ant during the t times iteration
The random walk position of ant, Wi t=[0, cumsum (2r1(t) -1) ..., cumsum (2rMax_iter(t) -1)], cumsum is meter
The cumulative value function of the group that counts, t represents current iteration number of times, and Max_iter represents maximum iteration, r1(t) ..., rMax_iter
(t) it is random function and separate.
Described random function calculation formula is：
R (t) is r1(t) ..., rMax_iter(t), rand (t) be 0 to 1 between
Random number.
The fitness value of the corresponding ant lion position of the fitness value of contemporary ant is compared in step S4, judged
Ant lion position whether is updated, if the fitness value of contemporary ant is less than the fitness value of ant lion, is substituted with contemporary ant position
Ant lion position, otherwise retains former ant lion position.
Compared with prior art, the present invention has advantages below：
1st, compared to existing Forecasting Methodology, precision of prediction is higher：LSSVM Forecasting Methodologies are carried out using ant lion optimized algorithm
Optimization, it predicts the outcome closer to actual power consumption, reduces the relative error of prediction；
2nd, forecasting efficiency is improved：Automatic cycle iteration, forecasting efficiency is high, and iteration run time is short.
Brief description of the drawings
Fig. 1 is the LSSVM optimized based on ant lion of present invention Prediction of annual electricity consumption method flow diagram；
Fig. 2 is the detail flowchart that ALO optimizes LSSVM parameters；
Fig. 3 is the comparative result figure being predicted using ALO-LSSVM and GCA-ALO-LSSVM methods；
Fig. 4 is the predicted value comparison diagram with being obtained using the inventive method using other method.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is a part of embodiment of the present invention, rather than whole embodiments.Based on this hair
Embodiment in bright, the every other reality that those of ordinary skill in the art are obtained on the premise of creative work is not made
Example is applied, should all belong to the scope of protection of the invention.
The present invention provides a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion, main to include two parts：It is first
First power consumption influence factor and power consumption are carried out using grey correlation analysis (Grey Correlation Analysis, GCA)
Correlation analysis, according to grey correlation angle value, determines LSSVM input variables, simplifies LSSVM structures, optimize secondly by ant lion
Algorithm (Ant Lion Optimizer, ALO), obtains least square method supporting vector machine (Least Square Support
Vector Machines, LSSVM) optimized parameter.
(1)GCA
GCA is one of main contents of gray system theory, and its general principle is by statistical series set relations
Compare to distinguish the tightness degree of relation multifactor in system, the geometries of sequence curve are closer to then between them
The degree of association is bigger, otherwise smaller.Its detailed process is as follows：
1) power consumption sequence is set as X0=(x0(1),x0(2),…,x0(k),…,x0(p)), power consumption influence factor is Xg
=(xg(1),xg(2),…,xg(k),…xg(p)), (g=1,2 ..., m), p represents total year of selected power consumption, and m represents to use
Electricity influence factor number；
2) using first value operator to X0And XiInitialized, obtain initial value picture respectively Y0=(y0(1),y0(2),…,
y0(k),…,y0) and Y (p)g=(yg(1),yg(2),…,yg(k),…,yg(p)), (g=1,2 ..., m), m represents power consumption
Influence factor number；
3) Y is obtained0With YgBetween incidence coefficient γ0g(k),In formula：y0kThat is y0(k), ygkThat is yg(k), ξ ∈
(0,1] it is resolution ratio, typically it is taken as 0.5；
4) X is calculated0With XgDegree of association γ0g,In formula：γ0gkThat is γ0g(k), γ0g∈ (0,1],
γ0gIt is bigger, show XgTo X0Effect it is bigger, if γ0g≥γ0q, then factor XgBetter than factor Xq, XqRepresent q-th of power consumption
Influence factor, q=1,2 ..., m, γ0qFor X0With XqThe degree of association.
(2) the LSSVM forecast models based on ALO
ALO algorithms are a kind of new Intelligent evolution algorithms proposed by S.Mirjalili in 2015, and its main inspiration is come
Come from the foraging behavior of ant lion larva.The algorithm acts on following criterion in optimization process:
Criterion 1：Ant is using different paths in search space random walk, and random walk is applied to the ant of all dimensions
Ant, and ant random walk influenceed by ant lion trap.
Criterion 2：Ant lion can set up the trap matched with fitness value, and the ant lion capture ant of larger trap probability compared with
It is high.
Criterion 3：Random walk scope is reduced with ant to the moveable self-adaption of ant lion.
Criterion 4：If the fitness value of an ant is smaller than the fitness value of an ant lion, it means that it can be by ant lion
Capture.
Criterion 5：Ant lion repositions new position near capture ant, and rebuilds trap to adapt to capture the change after ant
Change.
The nuclear parameter and regularization parameter solved based on ALO in LSSVM is comprised the following steps that：
1) input variable, output variable to LSSVM is normalized, and normalizes to after [0,1], by 25 groups of data
It is divided into two parts：Preceding 20 groups of data are used as checking sample set Z as training sample set L, rear 5 groups of data.Then, then from training sample
L groups data formation training sample subset U is selected in this collection L (common N=20 groups data), sets up and trains surplus in LSSVM models, L
Remainder examines the model after training according to as test samples subset V.
2) parameter initialization of ALO algorithms：Including population scale Agents_no, the position of every ant lion is a solution,
That is two parameters in LSSVM：Nuclear parameter σ and regularization parameter γ, therefore the dimension d=2, the upper bound b of solution space of solutionup=
[bup1,bup2], lower bound blow=[blow1,blow2], maximum iteration Max_iter.Fitness function F is set up simultaneously,Wherein,WithThe respectively actual value and predicted value of s groups test samples, N is sample
Number；
3) ALO algorithms position initialization：The initial position of antThe initial position of ant lionWherein, n=Agents_no, MAntAnd MAntlionIn value by formula A*jOr AL*j=rand*
(bupj-blowj)+blowjObtain, A*jAnd AL*jThe value of location matrix jth row is represented, rand is a random number between 0 to 1,
bupjAnd blowjThe respectively upper bound of jth row variable and lower bound, MAntAnd MAntlionEvery a line correspond to one group of LSSVM core ginseng
Number and regularization parameter (σ, γ)；
4) it is that nuclear parameter and regularization parameter substitute into LSSVM forecast models by ant lion position, obtains corresponding predicted value, so
Afterwards according to fitness function F, the fitness value of initial ant lion is calculated, the minimum ant lion of initial fitness value is initial elite ant
Lion elite(0)；
5) position of previous generation elite ant lions is retained, according to ant lion position and elite ant lion location updating ant position, meter
The fitness value when former generation ant is calculated, and is compared with the fitness value of corresponding ant lion, when the fitness value of ant is less than ant
During the fitness value of lion, then ant is caught；
6) update ant lion position to the position for catching and killing ant, calculating more new position after ant lion fitness value, successively with
The fitness value of previous generation elite ant lions is compared, and is retained the ant lion of smaller fitness value, is obtained current iteration elite ant lion
elite(t)；
If 7) not up to maximum iteration, return to step 5), otherwise, output elite ant lion position is tried to achieve
LSSVM nuclear parameter and regularization parameter (σ, γ).
The LSSVM optimized based on ant lion Prediction of annual electricity consumption method, is comprised the following steps, as shown in Figure 2：
1) least square method supporting vector machine (Least Square Support Vector Machines, LSSVM) is determined
Input variable；
2) initialization ant lion optimized algorithm (Ant Lion Optimizer, ALO)；
3) fitness function is set up, initial fitness value is calculated, and obtain initial elite ant lion position；
4) ant position is updated, the fitness value of current ant position, and the fitness value of corresponding ant lion is calculated
It is compared, judges whether to update ant lion position；
5) the corresponding fitness value in ant lion position after updating is calculated, one by one with the fitness value ratio of previous generation elite ant lions
Compared with the ant lion of the smaller fitness value of reservation obtains current iteration elite ant lion；
6) judge whether to reach maximum iteration, if so, then going to next step, gone to after otherwise adding 1 by iterations
4)；
7) iterative process is completed, required Prediction of annual electricity consumption value is obtained.
The step 1) in LSSVM input variable determine that method is：Using Grey Incidence Analysis to power consumption shadow
The factor of sound carries out grey correlation analysis with power consumption, obtains and associates angle value between power consumption influence factor and power consumption, according to
Associate angle value, selection year power consumption influence factor as LSSVM input variable.
The step 2) be specially：
201) ALO parameter initializations, including population scale Agents_no, the position of every ant lion is a solution, i.e.,
Two parameters in LSSVM：Nuclear parameter σ and regularization parameter γ, therefore the dimension d=2, the upper bound b of solution space of solutionup=
[bup1,bup2], lower bound blow=[blow1,blow2], maximum iteration Max_iter；
202) ALO position initializations, randomly generate the location matrix M of antAntWith the location matrix M of ant lionAntlion：Wherein, n=Agents_no, MAntAnd MAntlionIn value by
Formula A*jOr AL*j=rand* (bupj-blowj)+blowjObtain, A*jAnd AL*jThe value of location matrix jth row is represented, rand is 0 to 1
Between a random number, bupjAnd blowjThe upper bound and lower bound that respectively jth is arranged, MAntAnd MAntlionEvery a line correspond to one
Group LSSVM nuclear parameter and regularization parameter (σ, γ).
The upper bound b of the solution spaceup=[1000,1000], lower bound blow=[0.1,0.01].
The step 3) be specially：
301) input variable, output variable to LSSVM is normalized, and normalizes to after [0,1], by 25 groups of numbers
According to being divided into two parts：Preceding 20 groups of data are used as checking sample set Z as training sample set L, rear 5 groups of data.Then, then from training
L groups data formation training sample subset U is selected in sample set L (common N=20 groups data), sets up and trains in LSSVM models, L
Remaining data examines the model after training as test samples subset V；
302) by step 2) in MAntlionLSSVM is substituted into, and it is trained, is carried out in advance with the LSSVM trained
Survey, obtain predicted value；
303) according to fitness function F,Wherein,WithRespectively s groups are examined
The actual value and predicted value of sample, N is number of samples, the initial fitness value of each ant lion is calculated, by all initial fitness
Value compares one by one, obtains and records initial elite ant lion position and its fitness value, elite ant lion position be required LSSVM most
Excellent (σ, γ).
The step 4) in the more new formula of ant position be：Wherein,Represent that the t times iteration exists
The ant position of random walk around the selected ant lion of roulette,Represent random around elite ant lion during the t times iteration
The ant position of migration,The position of i-th of ant during for the t times iteration.
Wherein, aiRepresent the position of random walk of i-th of ant most
Small value, hiThe position maximum of the random walk of i-th of ant is represented,Represent the actual bit of i-th of ant during the t times iteration
Put minimum value,Represent the physical location maximum of i-th of ant during the t times iteration, Wi tRepresent i-th of ant during the t times iteration
The random walk position of ant.Specifically, ant itself random walk formula is：Wi t=[0, cumsum (2r1(t)-1),…,
cumsum(2rMax_iter(t) -1)], cumsum is to calculate the cumulative value function of array, and t represents current iteration number of times, Max_iter tables
Show maximum iteration, r1(t) ..., rMax_iter(t) it is random function and separate.
Work as calculatingWhen, i.e. ant random walk around ant lion,
Represent the physical location minimum value of i-th of ant during the t times iteration, ctThe minimum value of all ant positions when being the t times iteration,Represent the physical location maximum of i-th of ant during the t times iteration, dtThe maximum of all ant positions when being the t times iteration
Value,Represent i-th of the ant lion position selected during the t times iteration.
Work as calculatingWhen, i.e. ant random walk around elite ant lion,
Represent the physical location minimum value of i-th of ant during the t times iteration, ctThe minimum value of all ant positions when being the t times iteration,Represent the physical location maximum of i-th of ant during the t times iteration, dtThe maximum of all ant positions when being the t times iteration
Value, ElitetRepresent elite ant lion position during the t times iteration.
ct, dtCalculation formula is：In formula, t represents current iteration number of times, and I is ratio, ct' represent the t times repeatedly
For when last moment all ant positions minimum value, dt' represent the t times iteration when last moment all ant positions maximum
Value.
Ratio I calculation formula are：
In formula, t represents current iteration number of times, Max_
Iter represents maximum iteration.
Random function r1(t) ..., rMax_iter(t) calculation formula is：
R (t) is r1(t) ..., rMax_iter(t), rand be 0 to 1 between it is random
Number.
Then according to F, the fitness value of current ant position is calculated, and is compared with the fitness value of corresponding ant lion,
Judge whether to update ant lion position.
When the fitness value of ant is less than the fitness value of ant lion, then ant lion position is updated.
The step 5) be specially：
501) ant lion location updating formula is：In formula,The position of i-th of ant lion selected during the t times iteration is represented,Represent the position of i-th of ant during the t times iteration
Put；
502) according to F, the ant lion fitness value after more new position, and the fitness with previous generation elite ant lions one by one are calculated
Value is compared, and is retained the ant lion of smaller fitness value, is obtained current iteration elite ant lion position.
Embodiment
The present invention uses certain city year power consumption measured data (unit：Hundred million kilowatt hours) it is tested.According to Fig. 1, use
The power consumption and power consumption influence factor sequence training pattern of nineteen ninety~2009 year, with the model prediction test set trained
The power consumption of 2010~2014.Wherein each parameter setting is as follows in ALO algorithms：Population scale Agents_no=30, variable
Dimension d=2, maximum iteration Max_iter=200, the upper bound b of solution spaceup=[1000,1000], lower bound blow=
[0.1,0.01].The LSSVM optimized parameters asked for according to ALO algorithms are [1000,1000].Respectively with ALO-LSSVM and GCA-
ALO-LSSVM represents to determine the forward and backward model of input variable using GCA, and the predicted value that two methods are obtained is shown in Table 1, Fig. 3, table 1
Compare to determine that the forward and backward ALO-LSSVM of input variable predicts the outcome.
Table 1
Simultaneously using forecast model is set up based on GCA-Bayes BP neural networks, see with the predicted value comparison diagram of the present invention
Fig. 4.
The prediction effect of several method is shown in Table 2, and table 2 represents the prediction effect evaluation of three kinds of Forecasting Methodologies used.
Table 2
According to the 1st edition that in October, 1998, China Electric Power Publishing House published by Niu Dongxiao, Cao Shuhua, Zhao Lei, Zhang Wenwen etc.
Write《Techniques for Prediction of Electric Loads and its application》The forecast model evaluation criterion that page 15 is provided in one book is understood, several
Forecasting Methodology has all reached good prediction effect.But it can be seen that what the present invention was optimized based on ant lion from specific evaluation criterion value
LSSVM Prediction of annual electricity consumption method has higher accuracy rate when predicting year power consumption, can meet requirement.
The foregoing is only a specific embodiment of the invention, but protection scope of the present invention is not limited thereto, any
Those familiar with the art the invention discloses technical scope in, various equivalent modifications can be readily occurred in or replaced
Change, these modifications or substitutions should be all included within the scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection domain be defined.
Claims (10)
1. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion, it is characterised in that described Prediction of annual electricity consumption side
Method comprises the following steps：
S1, the input variable for determining LSSVM forecast models；
S2, initialization ant lion optimized algorithm, LSSVM models are substituted into using initial ant lion position as nuclear parameter and regularization parameter,
Obtain corresponding Prediction of annual electricity consumption value；
S3, set up fitness function, calculate the fitness value of initial ant lion position, obtain initial fitness value, retain it is minimum just
The corresponding ant lion of beginning fitness value is used as initial elite ant lion；
S4, renewal ant position, calculate the fitness value of contemporary ant, and the fitness of ant lion position corresponding with contemporary ant
Value is compared, and judges whether to update ant lion position；
S5, the fitness value ratio by the fitness value of ant lion position obtained in the previous step one by one with previous generation elite ant lions position
Compared with the corresponding ant lion position of the smaller fitness value of reservation obtains current iteration elite ant lion position；
S6, judge whether to reach maximum iteration, if it has, then output elite ant lion position and its corresponding year power consumption are pre-
Measured value, if it has not, returning to S4.
2. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 1, it is characterised in that
Step S1 is specially：Using Grey Incidence Analysis, the degree of association between year power consumption influence factor and year power consumption is obtained
Value, according to association angle value, selects corresponding year power consumption influence factor as the input variable of LSSVM forecast models.
3. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 1, it is characterised in that
Ant lion optimized algorithm is initialized in step S2 to comprise the following steps：
S201, parameter initialization, including population scale Agents_no, the dimension d=2 of solution, the upper bound b of solution spaceup, solution space
Lower bound blow, maximum iteration Max_iter；
S202, position initialization, randomly generate the location matrix M of antAntWith the location matrix M of ant lionAntlion。
4. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 3, it is characterised in that
The upper bound b of solution space described in S201up=[1000,1000], lower bound blow=[0.1,0.01].
5. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 3, it is characterised in that
The location matrix M of ant in S202AntWith the location matrix M of ant lionAntlionFor：
Wherein MAntAnd MAntlionIn value by formula A*jOr AL*j=
rand*(bupj-blowj)+blowjObtain, A*jAnd AL*jThe value of ant location matrix and ant lion location matrix jth row is represented respectively,
Rand is a random number between 0 to 1, n=Agents_no, bupjAnd blowjThe upper bound and lower bound that respectively jth is arranged, MAnt
And MAntlionEvery a line correspond to LSSVM one group of nuclear parameter and regularization parameter, i.e. (σ, γ).
6. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 1, it is characterised in that
Fitness function in step S3 is：Wherein,WithRespectively s groups test samples
Actual value and predicted value, N are number of samples.
7. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 1, it is characterised in that
The more new formula of ant position is in step S4Wherein,Represent the t of the random walk around ant lion
The secondary selected ant position of iteration roulette,Represent the ant position of the t times iteration of random walk around elite ant lion
Put,The position of i-th of ant during for the t times iteration.
8. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 7, it is characterised in thatWherein, aiRepresent the position minimum value of the random walk of i-th of ant, hiTable
Show the position maximum of the random walk of i-th of ant,The physical location minimum value of i-th of ant during the t times iteration is represented,The physical location maximum of i-th of ant during the t times iteration is represented,Represent the random trip of i-th of ant during the t times iteration
Walk position,Cumsum is to calculate array accumulated value letter
Number, t represents current iteration number of times, and Max_iter represents maximum iteration, r1(t) ..., rMax_iter(t) for random function and
It is separate.
9. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 8, it is characterised in that
Described random function calculation formula is：
R (t) is r1(t) ..., rMax_iter(t), rand (t) be 0 to 1 between it is random
Number.
10. a kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion according to claim 1, it is characterised in that
The fitness value of the corresponding ant lion position of the fitness value of contemporary ant is compared in step S4, judges whether to update
Ant lion position, if the fitness value of contemporary ant is less than the fitness value of ant lion, ant lion position is substituted with contemporary ant position,
Otherwise former ant lion position is retained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710641743.0A CN107274038A (en) | 2017-07-31 | 2017-07-31 | A kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710641743.0A CN107274038A (en) | 2017-07-31 | 2017-07-31 | A kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107274038A true CN107274038A (en) | 2017-10-20 |
Family
ID=60076052
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710641743.0A Pending CN107274038A (en) | 2017-07-31 | 2017-07-31 | A kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107274038A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107798199A (en) * | 2017-11-09 | 2018-03-13 | 华中科技大学 | A kind of Hydropower Unit parameter closed-loop identification method |
| CN108614242A (en) * | 2018-03-25 | 2018-10-02 | 哈尔滨工程大学 | A kind of radar-communication integration waveform design method based on the optimization of multiple target ant lion |
| CN109062664A (en) * | 2018-07-25 | 2018-12-21 | 南京邮电大学 | Cloud computing method for scheduling task based on ant lion optimization algorithm |
| CN109145464A (en) * | 2018-08-28 | 2019-01-04 | 暨南大学 | Merge the Structural Damage Identification of multiple target ant lion optimization and the sparse regularization of mark |
| CN109212465A (en) * | 2018-09-01 | 2019-01-15 | 哈尔滨工程大学 | A kind of particular array dynamic direction-finding method based on cultural ant lion mechanism |
| CN109284543A (en) * | 2018-09-04 | 2019-01-29 | 河北工业大学 | IGBT method for predicting residual useful life based on optimal scale Gaussian process model |
| CN109767036A (en) * | 2018-12-28 | 2019-05-17 | 北京航空航天大学 | Support vector machines failure prediction method based on the optimization of adaptive ant lion |
| CN109829420A (en) * | 2019-01-18 | 2019-05-31 | 湖北工业大学 | A kind of feature selection approach based on the high spectrum image for improving ant lion optimization algorithm |
| CN109858816A (en) * | 2019-02-02 | 2019-06-07 | 湖北工业大学 | A method of production scheduling is carried out using ant lion algorithm |
| CN110006649A (en) * | 2018-12-24 | 2019-07-12 | 湖南科技大学 | A kind of Method for Bearing Fault Diagnosis based on improvement ant lion algorithm and support vector machines |
| CN110487519A (en) * | 2019-06-28 | 2019-11-22 | 暨南大学 | Structural Damage Identification based on ALO-INM and weighting trace norm |
| CN111598255A (en) * | 2020-04-27 | 2020-08-28 | 同济大学 | Method for compensating cogging effect of suspension gap sensor of maglev train |
| CN111985678A (en) * | 2020-07-06 | 2020-11-24 | 上海交通大学 | Photovoltaic power short-term prediction method |
| CN113253709A (en) * | 2021-06-07 | 2021-08-13 | 江苏中车数字科技有限公司 | Health diagnosis method and device suitable for rail transit vehicle |
| CN113449462A (en) * | 2021-06-09 | 2021-09-28 | 淮阴工学院 | Ultra-short-term wind speed prediction method and system based on MALO-BiGRU |
| CN116205554A (en) * | 2023-04-26 | 2023-06-02 | 浙江天柜科技有限公司 | Mobile self-service vending equipment and vending control method |
| CN117233540A (en) * | 2023-11-15 | 2023-12-15 | 广东电网有限责任公司江门供电局 | Metering pipeline fault detection method and system based on deep learning |
-
2017
- 2017-07-31 CN CN201710641743.0A patent/CN107274038A/en active Pending
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107798199B (en) * | 2017-11-09 | 2020-07-07 | 华中科技大学 | Hydroelectric generating set parameter closed-loop identification method |
| CN107798199A (en) * | 2017-11-09 | 2018-03-13 | 华中科技大学 | A kind of Hydropower Unit parameter closed-loop identification method |
| CN108614242A (en) * | 2018-03-25 | 2018-10-02 | 哈尔滨工程大学 | A kind of radar-communication integration waveform design method based on the optimization of multiple target ant lion |
| CN108614242B (en) * | 2018-03-25 | 2022-04-05 | 哈尔滨工程大学 | Radar communication integrated waveform design method based on multi-objective ant lion optimization |
| CN109062664A (en) * | 2018-07-25 | 2018-12-21 | 南京邮电大学 | Cloud computing method for scheduling task based on ant lion optimization algorithm |
| CN109145464A (en) * | 2018-08-28 | 2019-01-04 | 暨南大学 | Merge the Structural Damage Identification of multiple target ant lion optimization and the sparse regularization of mark |
| CN109145464B (en) * | 2018-08-28 | 2022-11-01 | 暨南大学 | Structural damage identification method integrating multi-target ant lion optimization and trace sparse regularization |
| CN109212465A (en) * | 2018-09-01 | 2019-01-15 | 哈尔滨工程大学 | A kind of particular array dynamic direction-finding method based on cultural ant lion mechanism |
| CN109212465B (en) * | 2018-09-01 | 2024-01-30 | 哈尔滨工程大学 | Special array dynamic direction finding method based on cultural ant lion mechanism |
| CN109284543A (en) * | 2018-09-04 | 2019-01-29 | 河北工业大学 | IGBT method for predicting residual useful life based on optimal scale Gaussian process model |
| CN110006649A (en) * | 2018-12-24 | 2019-07-12 | 湖南科技大学 | A kind of Method for Bearing Fault Diagnosis based on improvement ant lion algorithm and support vector machines |
| CN109767036A (en) * | 2018-12-28 | 2019-05-17 | 北京航空航天大学 | Support vector machines failure prediction method based on the optimization of adaptive ant lion |
| CN109829420A (en) * | 2019-01-18 | 2019-05-31 | 湖北工业大学 | A kind of feature selection approach based on the high spectrum image for improving ant lion optimization algorithm |
| CN109829420B (en) * | 2019-01-18 | 2022-12-02 | 湖北工业大学 | Hyperspectral image feature selection method based on improved ant lion optimization algorithm |
| CN109858816A (en) * | 2019-02-02 | 2019-06-07 | 湖北工业大学 | A method of production scheduling is carried out using ant lion algorithm |
| CN110487519A (en) * | 2019-06-28 | 2019-11-22 | 暨南大学 | Structural Damage Identification based on ALO-INM and weighting trace norm |
| CN111598255A (en) * | 2020-04-27 | 2020-08-28 | 同济大学 | Method for compensating cogging effect of suspension gap sensor of maglev train |
| CN111985678A (en) * | 2020-07-06 | 2020-11-24 | 上海交通大学 | Photovoltaic power short-term prediction method |
| CN113253709B (en) * | 2021-06-07 | 2021-09-21 | 江苏中车数字科技有限公司 | Health diagnosis method and device suitable for rail transit vehicle |
| CN113253709A (en) * | 2021-06-07 | 2021-08-13 | 江苏中车数字科技有限公司 | Health diagnosis method and device suitable for rail transit vehicle |
| CN113449462A (en) * | 2021-06-09 | 2021-09-28 | 淮阴工学院 | Ultra-short-term wind speed prediction method and system based on MALO-BiGRU |
| CN113449462B (en) * | 2021-06-09 | 2023-06-20 | 淮阴工学院 | Ultra-short-term wind speed prediction method and system based on MALO-BiGRU |
| CN116205554A (en) * | 2023-04-26 | 2023-06-02 | 浙江天柜科技有限公司 | Mobile self-service vending equipment and vending control method |
| CN116205554B (en) * | 2023-04-26 | 2024-02-09 | 浙江天柜科技有限公司 | Mobile self-service vending equipment and vending control method |
| CN117233540A (en) * | 2023-11-15 | 2023-12-15 | 广东电网有限责任公司江门供电局 | Metering pipeline fault detection method and system based on deep learning |
| CN117233540B (en) * | 2023-11-15 | 2024-02-20 | 广东电网有限责任公司江门供电局 | Metering pipeline fault detection method and system based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107274038A (en) | A kind of LSSVM Prediction of annual electricity consumption methods optimized based on ant lion | |
| CN103942457B (en) | Water quality parameter time series prediction method based on relevance vector machine regression | |
| CN109324291B (en) | Prediction method for predicting service life of proton exchange membrane fuel cell | |
| CN101251579B (en) | Analog circuit failure diagnosis method based on supporting vector machine | |
| CN103488869A (en) | Wind power generation short-term load forecast method of least squares support vector machine | |
| CN110659779B (en) | Power distribution system network loss prediction method based on long-term and short-term memory network | |
| CN112733997B (en) | Hydrological time series prediction optimization method based on WOA-LSTM-MC | |
| CN110826774B (en) | Bus load prediction method and device, computer equipment and storage medium | |
| CN106446571A (en) | Error evaluation method of wind power forecasting | |
| CN103336771B (en) | Data similarity detection method based on sliding window | |
| CN109214708A (en) | Study of Risk Evaluation Analysis for Power System method based on cross entropy theoretical optimization support vector machines | |
| CN106327357A (en) | Load identification method based on improved probabilistic neural network | |
| CN103106331B (en) | Based on the lithographic line width Intelligent Forecasting of dimensionality reduction and increment type extreme learning machine | |
| CN112990500A (en) | Transformer area line loss analysis method and system based on improved weighted gray correlation analysis | |
| CN109858665A (en) | Photovoltaic short term power prediction technique based on Feature Selection and ANFIS-PSO | |
| TW201635224A (en) | Method of short-term wind power generation forecasting | |
| CN106296434A (en) | A kind of Grain Crop Yield Prediction method based on PSO LSSVM algorithm | |
| CN106295877B (en) | Method for predicting electric energy consumption of smart power grid | |
| Dengen | Comparison of SARIMA, NARX and BPNN models in forecasting time series data of network traffic | |
| CN103489037B (en) | A kind of Forecasting Methodology that can power generating wind resource | |
| CN106257506B (en) | Three layers of joint dynamic select optimal models method of big data quantity prediction | |
| CN109961160A (en) | A kind of power grid future operation trend predictor method and system based on trend parameter | |
| CN107204616A (en) | Power system probabilistic state estimation method based on the pseudo- spectrometry of adaptive sparse | |
| Coelho et al. | A heuristic fuzzy algorithm bio-inspired by evolution strategies for energy forecasting problems | |
| CN105743088B (en) | A kind of horizontal uncertain influence elastic network(s) regression analysis of distribution network voltage |
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: 20171020 |