CN110188914A - A kind of intelligent Forecasting for grid power transmission route ice covering thickness - Google Patents
A kind of intelligent Forecasting for grid power transmission route ice covering thickness Download PDFInfo
- Publication number
- CN110188914A CN110188914A CN201910225585.XA CN201910225585A CN110188914A CN 110188914 A CN110188914 A CN 110188914A CN 201910225585 A CN201910225585 A CN 201910225585A CN 110188914 A CN110188914 A CN 110188914A
- Authority
- CN
- China
- Prior art keywords
- support vector
- vector machines
- covering thickness
- ice covering
- optimization 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
- 230000005540 biological transmission Effects 0.000 title claims abstract description 12
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 54
- 239000002245 particle Substances 0.000 claims abstract description 37
- 238000005457 optimization Methods 0.000 claims abstract description 32
- 238000012706 support-vector machine Methods 0.000 claims abstract description 17
- 238000000034 method Methods 0.000 claims description 22
- 238000012549 training Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 4
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000005611 electricity Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000012804 iterative process Methods 0.000 description 3
- 238000012417 linear regression Methods 0.000 description 3
- 239000003016 pheromone Substances 0.000 description 3
- 244000144992 flock Species 0.000 description 2
- 238000003916 acid precipitation Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach 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
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000010429 evolutionary process Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000003595 mist Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000001373 regressive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000004781 supercooling Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000007704 transition Effects 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Biophysics (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (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 invention belongs to electric power hazard prediction technical fields, more particularly to a kind of intelligent Forecasting for grid power transmission route ice covering thickness, comprising: choose the input factor of prediction mean daily temperature, relative humidity and wind speed, 6 the previous day ice covering thickness, relative humidity and temperature variables as support vector machines;Quick global search is carried out using kernel parameter and penalty factor of the particle swarm optimization algorithm to support vector machines;The relatively figure of merit that global search is obtained is converted into the initial information element distribution of ant colony optimization algorithm, carries out route searching using ant group algorithm;The length, runing time and the number of iterations of the optimal solution that route searching generates are substituted into particle swarm optimization algorithm to speed and the position for updating each particle, until obtaining optimal solution, substitute into support vector machines;Power network line ice covering thickness is predicted after being weighted to the penalty factor of support vector machines.The present invention effectively overcomes the shortcomings that two kinds of algorithms, significantly improves computational efficiency.
Description
Technical field
The invention belongs to electric power hazard prediction technical fields more particularly to a kind of for grid power transmission route ice covering thickness
Intelligent Forecasting.
Background technique
Normal, the safe operation of transmission line of electricity are to avoid the important guarantee of power grid major accident, and the icing of transmission line of electricity
It will lead to transmission line of electricity overtension, cause the accidents such as conductor galloping, line tripping, broken string, interrupt power supply, serious shadow
Ring the stability and safety of Operation of Electric Systems.Transmission line of electricity passes through filth, High aititude, ice and snow, acid rain, strong mist etc. and extremely dislikes
When the area of bad complexity, it will a possibility that increasing the generation of transmission line icing.Powerline ice-covering, which has become, influences countries in the world
One of an important factor for electric power netting safe running.Accurately prediction and prevention and treatment are carried out to the icing of power grid, to prevent and control ice damage,
Raising power grid security, which is reliably run, to have important practical significance.
Swarm intelligence optimization algorithm is individual and mutually exchange between individual and conjunction in simulation practical biocenose life
Make, with simple, limited individual behavior and intelligence, entire group whole capability difficult to the appraisal is formed by interaction, such as
Genetic algorithm, differential evolution, ant group algorithm, particle swarm algorithm, evolutional programming scheduling algorithm.Existing ant group algorithm is due to lacking just
Beginning pheromones, it is slower in search initial stage arithmetic speed, and the setting of the parameter in algorithm is continuously, mainly to pass through experiment side
What method determined, accuracy, calculating speed and performance of this method etc. and the experience of laboratory technician are closely related, are difficult so that should
Algorithm performance is optimal;Existing particle swarm algorithm can show excellent performance in the optimization problem in continuous space,
But after obtaining locally optimal solution by n times iterative operation, then with the increase of the number of iterations, the speed of particle will be increasingly
It is small, and gradually tend to 0, easily fall into local convergence.
Summary of the invention
In view of the above technical problems, the invention proposes a kind of intelligent predicting sides for grid power transmission route ice covering thickness
Method, comprising:
Step 1: choosing prediction mean daily temperature, relative humidity and wind speed, the previous day ice covering thickness, relative humidity and temperature
Input factor of 6 variables as support vector machines;
Step 2: being carried out using kernel parameter and penalty factor of the particle swarm optimization algorithm to support vector machines quickly global
Search;
Step 3: the relatively figure of merit that global search is obtained is converted into the initial information element distribution of ant colony optimization algorithm, utilizes ant
Group's algorithm carries out route searching;
Step 4: length, runing time and the number of iterations for the optimal solution that route searching is generated substitute into Particle Swarm Optimization
Speed and the position of each particle are updated in method, until obtaining optimal solution, substitute into support vector machines;
Step 5: power network line ice covering thickness being predicted after being weighted to the penalty factor of support vector machines.
The optimal solution is using the minimum value of predicted value and the mean percent ratio error of actual value as target.
The weighting calculates the similarity between sample using grey relational grade and determines weight.
Beneficial effects of the present invention:
Comprehensively consider the influence factors such as environment temperature, relative humidity, wind speed, wind direction, height above sea level, increases in conjunction with used icing
Long Statistical Prediction Model and icing meteorologic parameter prediction model, propose weighted support vector regression algorithm (WSVR), calculate sample
This incidence coefficient thereby determines that the weight size of different samples, and passes through population-ant colony (PSO-ACO) hybrid optimization
Algorithm comes Optimal Parameters g and C, and particle swarm optimization algorithm ant colony optimization algorithm is organically combined, and joins to support vector machines
The shortcomings that number optimizes, can effectively overcome two kinds of algorithms, significantly improves computational efficiency.
Detailed description of the invention
Fig. 1 is the mobile figure of particle.
Fig. 2 is ACO-PSO mixing swarm intelligence optimization algorithm flow chart.
Fig. 3 is the parameter optimization iterative convergent process of PSO-ACO algorithm.
Fig. 4 is that PSO-ACO hybrid optimization WSVR prediction result curve is based in embodiment.
Fig. 5 is algorithms of different prediction result correlation curve in embodiment.
Fig. 6 is the error distribution situation of algorithms of different in embodiment.
Specific embodiment
With reference to the accompanying drawing, it elaborates to embodiment.
The invention proposes a kind of intelligent Forecastings for grid power transmission route ice covering thickness, utilize ant colony and particle
The shortcomings that two kinds of algorithms are organically combined, can effectively overcome two kinds of algorithms by the mixed characteristic of colony optimization algorithm, significantly mentions
Computationally efficient.
(1) ant colony optimization algorithm
Although the not too many intelligence of the single ant of the algorithm, can not also grasp neighbouring geography information, entire ant colony
But the physical feature of an optimal path between nest to food source can be found, basic ACO model is by following three
Formula description:
(if k-th of ant have passed through the path by i to j)
Wherein, m is ant number, and n is the number of iterations, and i is ant position, and j is the position that ant can reach, Λ
It can be with the set of in-position, η for antijFor enlightening information, here for by the visibility in the path of i to j, i.e.,
LkFor objective function, τijFor by the pheromones intensity in the path of i to j ,+τij kIt is ant k by the information that is left on the path of i to j
Prime number amount, α are path power, and β is the power of enlightening information, and ρ is the evaporation coefficient of pheromones quantity on path, and Q is information quality
Coefficient of discharge, Pij kThe transition probability of position j is moved to from position i for ant k.
(2) particle swarm optimization algorithm
Particle swarm optimization algorithm, derived from by the observation to Bird search of food behavior, using in flock of birds per each and every one
Body to the shared of information so that the movement of entire flock of birds generates the conversion process from disorder to order, to find food, such as Fig. 1
It is shown.The method that PSO optimizes problem are as follows: PSO initializes the random particle of a group, is found by successive ignition optimal
Solution.In an iterative process, each particle constantly updates the direction and position of oneself, first extreme value with two " extreme values " for parameter
It is the optimal solution that particle itself is found, i.e., individual extreme value pbest, another extreme value is the optimal solution that current entire population is found,
That is global extremum gbest.For particle populations when starting iteration, the position where each particle of initialization is individual extreme value,
Best position is global extremum in particle populations.After particle all in population all completes an iteration operation, to every
Position before and after a particle iteration is compared, and is updated if due to individual extreme value before, i.e., by the optimal of this iteration
Solution is set as individual extreme value;The individual extreme value of particles all in population is compared again, obtains optimal solution in all individual extreme values,
If it be better than before global extremum, it is updated to global extremum.Such circulate operation, after successive ignition, most
Obtained global extremum is obtained optimal solution eventually.
When each particle is after iterative process obtains individual extreme value and global extremum, need to update oneself by following formula
Speed and position:
Pi,j+1=Pi,j+vi,j+1 (3)
Wherein, vi,jIndicate speed of i-th of particle after j iteration, Pi,jIndicate i-th of particle by j iteration
Position afterwards, pbesti,jAnd gbesti,jRespectively represent individual extreme value and global extremum of i-th of particle after j iteration, w
It is primary updated speed to the inertia weight of speed before updating, random () is the random number between (0,1), c1
And c2Studying factors, value range be (0,2].The speed of each particle is limited in range (0, v in groupmax) in,
Particle in an iterative process, if updated speed exceed maximum value vmax, then the speed of the particle is updated to vmax。
It is this unidirectional by shared global extremum gbest to particles other in population in particle swarm algorithm optimization process
Shared information and data flowing so that entire search process can follow current optimal solution in population.Therefore, Particle Swarm Optimization
Method initial stage has quick global convergence ability.
(3) ACO_PSO hybrid optimization algorithm
Fusion improvement is carried out to above two algorithm, proposes a kind of population-ant colony algorithm, that is, utilizes population
Optimization algorithm carries out quick global search, determines the parameter of ant colony search, and converts ant group algorithm for the obtained relatively figure of merit
Initial information element distribution, using ant group algorithm carry out route searching, while by this group of parameter generate optimal solution length, fortune
Row time and the number of iterations are brought into PSO algorithm, and speed and the position of each particle are updated according to formula, and so on, until
Obtain optimal solution.The basic step of ACO-PSO hybrid algorithm is as shown in Figure 2.
(4) Weighted Support Vector
It is situation mutually indepedent between same distribution and sample that traditional algorithm of support vector machine, which is suitable for sample data,
It is to the punishment parameter C and error requirements parameter ε of different samples it is identical, model can obtain preferable effect at this time.But by
It is more complicated in real data situation, it is often found that certain sample data importance are big, it is desirable that small training error, and some samples
The importance of notebook data is relatively lower, allows a certain size training error.Therefore, when describing optimization problem, each sample
Notebook data should have different error requirements and penalty coefficient, can just obtain more accurate regression estimates.In response to this problem, originally
Invention is predicted using Weighted Support Vector regression algorithm.
If the variance of stochastic error has the property that
In σi 2In different situation, traditional support vector machine regression model hardly results in more satisfactory result.Excellent
During change, parameters ξ and ξ in model*Status be it is identical, this result in the tropic be easy to be pulled to variance it is biggish
, and the lesser item fitting effect of variance is more relatively poor.By introducing a weight λiSummarize to adjust items in model
It influences, the influence caused by control error term is different.Optimized model are as follows:
ω·φ(x)+b-yi≤ξi*+εti (6)
yi-ω·φ(x)-b≤ξi+εti (7)
ξi,ξi>=0, i=1, * ... (8) l
In formula, λi、tiWeighting coefficient of respectively i-th of the training sample to parameter C and ε.The weighting of sample punishment parameter C
System performance can be significantly improved, and the weighting of error requirements parameter ε improves less system performance, therefore, the present invention is only
To λiCarry out optimal weighting processing.
The basic thought of grey relational grade is to judge correlation degree according to similarity degree between curve.Therefore, present invention benefit
The similarity between sample is calculated with grey relational grade, so that it is determined that weight, grey relational grade calculation formula are as follows:
Grey relational grade γ (x0,xi) be
Then weight λiIt is defined as
Wherein, i=1 ... n, j are the number of arguments, and ε is a smaller number, SjIt (i) is the data after normalization.
In practical applications, each weight can be chosen according to the actual situation.
Influence power grid icing many factors in, meteorologic factor is factor the most main, as temperature, humidity, wind speed,
The external climates such as wind direction.Many documents have Primary Study to influence factor, and by summarizing and learning, the present invention summarizes to be formed
The big necessary condition of the three of wire icing are as follows: relative air humidity is 85% or more;Wind speed is greater than 1m/s;Temperature reach 0 DEG C and with
Under.Water in higher levels of humidity air is the water source for generating icing, and wind action is that the supercooling water droplet in air is made to generate fortune
It is dynamic, it is captured after colliding with conducting wire, lower temperature freezes water droplet generation.Therefore, the present invention considers mean temperature, phase
To three humidity, wind speed factors, prediction mean daily temperature, relative humidity and wind speed, the previous day ice covering thickness, relative humidity are chosen
Input factor with 6 variables of temperature as support vector machines, predicts power network line ice covering thickness.
For the validity for verifying model, the present invention chooses Hunan Province somewhere " good fortune outside line " shaft tower 20 days -4 March in 2008
The data of month icing field real-time acquisition on the 10th are calculated.Before sample training, it is necessary to first to initial data carry out screening and
Normalized, relative humidity is too small, under temperature is excessively high and the too small situation of wind speed, lead to data of the ice covering thickness close to 0
It deletes, finally retains 72 groups of valid data, wherein first 50 groups are used as training set, latter 22 groups are used as test set, verify model
Validity.
The formula of normalized is as follows:
After normalized, the value of each variable eliminates the influence of dimension between [0,1].
Since climatic factor has the characteristics that fluctuation is big, randomness is strong, SVM can multiple influences on ice covering thickness because
Element is comprehensively considered, and has preferable non-linear mapping capability and generalization ability.In SVM regressive prediction model, since core is joined
Influence of the selection of number g and penalty factor to model accuracy is very big, and the present invention will be optimized using PSO-ACO mixing swarm intelligence
Algorithm comes Optimal Parameters g and C simultaneously, finds an optimal parameter to improve the precision of prediction of model.
The present invention chooses the mean percent ratio error MAPE of predicted value and actual value as objective function, to search the target
Functional minimum value is target, and the optimal fitness function value of each ant colony corresponding to global minimum when iteration ends is
The nuclear parameter g and penalty factor of SVM model.Optimized obtained parameter is updated to SVM prediction model to (g, C), then to covering
Ice thickness is predicted.
Fig. 3 is the parameter optimization iterative convergent process of PSO-ACO hybrid optimization algorithm, is adjusted by 100 iteration, dynamic
SVM parameter, MAPE is stepped to be converged to steadily.It is found by training result, in evolutionary process, MAPE is in the 10th iteration
Convergence, minimum average B configuration absolute percent error are 1.5%, and corresponding best kernel function g and penalty factor are respectively as follows: 12.201
With 0.1, by obtained parameter to substituting into weighed SVM model, regression forecasting is carried out to ice covering thickness, as a result as shown in Figure 4.
In order to assess the reasonability of model proposed by the present invention, ant group optimization support vector machines (ACO-SVM), SVM are chosen
Model and linear regression model (LRM) model as a comparison.The prediction result of these three algorithms is shown in Fig. 4, and error condition is shown in Fig. 5.
The present invention chooses average absolute percentage error MAPE index and carries out quantitative evaluation to prediction result.
The error distribution situation of distinct methods is as shown in Figure 6.
It can be seen that four kinds of methods to actual ice covering thickness curve by the comparison of prediction curve and realized load curve
It approaches.Wherein, proposed by the present invention that (WSVR) method is returned based on PSO-ACO mixing intelligent optimizing Weighted Support Vector
Whole fitting effect is best, and the mean absolute percentage error of this method prediction result is 2.71%, and ACO-SVM, tradition
It is respectively 4.67%, 5.22%, 6.2% under SVM and linear regression method, error is much higher than method proposed by the invention.Accidentally
Poor evaluation result has shown precision of prediction of the Weighted Support Vector than single algorithm optimization SVM through hybrid algorithm optimization
It improves, hybrid optimization algorithm compensates for the defect of single algorithm, and Weighted Support Vector takes full advantage of sample information, and the two is total
With the precision and generalization ability for improving model, so that prediction effect is more preferable.It is compared with other methods, side proposed by the invention
Method has apparent advantage, can be used for electric power line ice-covering thickness prediction.
This embodiment is merely preferred embodiments of the present invention, but scope of protection of the present invention is not limited thereto,
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by anyone skilled in the art,
It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with scope of protection of the claims
Subject to.
Claims (3)
1. a kind of intelligent Forecasting for grid power transmission route ice covering thickness characterized by comprising
Step 1: choosing prediction mean daily temperature, relative humidity and wind speed, the previous day ice covering thickness, relative humidity and temperature 6
Input factor of the variable as support vector machines;
Step 2: carrying out quick global search using kernel parameter and penalty factor of the particle swarm optimization algorithm to support vector machines;
Step 3: the relatively figure of merit that global search is obtained is converted into the initial information element distribution of ant colony optimization algorithm, is calculated using ant colony
Method carries out route searching;
Step 4: length, runing time and the number of iterations for the optimal solution that route searching is generated substitute into particle swarm optimization algorithm
Speed and the position of each particle are updated, until obtaining optimal solution, substitutes into support vector machines;
Step 5: power network line ice covering thickness being predicted after being weighted to the penalty factor of support vector machines.
2. prediction technique according to claim 1, which is characterized in that the optimal solution is with average the hundred of predicted value and actual value
The minimum value for dividing ratio error is target.
3. prediction technique according to claim 1 or claim 2, which is characterized in that the weighting calculates sample using grey relational grade
Between similarity determine weight.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910225585.XA CN110188914A (en) | 2019-03-25 | 2019-03-25 | A kind of intelligent Forecasting for grid power transmission route ice covering thickness |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910225585.XA CN110188914A (en) | 2019-03-25 | 2019-03-25 | A kind of intelligent Forecasting for grid power transmission route ice covering thickness |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110188914A true CN110188914A (en) | 2019-08-30 |
Family
ID=67713756
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910225585.XA Pending CN110188914A (en) | 2019-03-25 | 2019-03-25 | A kind of intelligent Forecasting for grid power transmission route ice covering thickness |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110188914A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110866693A (en) * | 2019-11-14 | 2020-03-06 | 国网湖北省电力有限公司电力科学研究院 | Overhead transmission line icing risk assessment method based on GIS model |
CN112990678A (en) * | 2021-03-04 | 2021-06-18 | 国网安徽省电力有限公司电力科学研究院 | Icing early warning judgment method based on multi-source data fusion |
CN114353872A (en) * | 2021-12-30 | 2022-04-15 | 中国电信股份有限公司 | Prediction method and device for machine room temperature |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103778305A (en) * | 2014-02-27 | 2014-05-07 | 南京工程学院 | k-VNN- and LS-SVM-based modelling method for icing of electric transmission line |
CA2836991A1 (en) * | 2012-12-21 | 2014-06-21 | Murray W. Davis | Portable self powered line mounted conductor ice thickness measuring system for overhead electric power lines |
CN104361414A (en) * | 2014-11-24 | 2015-02-18 | 武汉大学 | Power transmission line icing prediction method based on relevance vector machine |
CN105139274A (en) * | 2015-08-16 | 2015-12-09 | 东北石油大学 | Power transmission line icing prediction method based on quantum particle swarm and wavelet nerve network |
CN109242159A (en) * | 2018-08-17 | 2019-01-18 | 华北电力大学 | A kind of novel power transmission transmission line icing prediction technique |
-
2019
- 2019-03-25 CN CN201910225585.XA patent/CN110188914A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA2836991A1 (en) * | 2012-12-21 | 2014-06-21 | Murray W. Davis | Portable self powered line mounted conductor ice thickness measuring system for overhead electric power lines |
CN103778305A (en) * | 2014-02-27 | 2014-05-07 | 南京工程学院 | k-VNN- and LS-SVM-based modelling method for icing of electric transmission line |
CN104361414A (en) * | 2014-11-24 | 2015-02-18 | 武汉大学 | Power transmission line icing prediction method based on relevance vector machine |
CN105139274A (en) * | 2015-08-16 | 2015-12-09 | 东北石油大学 | Power transmission line icing prediction method based on quantum particle swarm and wavelet nerve network |
CN109242159A (en) * | 2018-08-17 | 2019-01-18 | 华北电力大学 | A kind of novel power transmission transmission line icing prediction technique |
Non-Patent Citations (3)
Title |
---|
XIAOMIN XU,ETC: "The Weighted Support Vector Machine Based on Hybrid Swarm Intelligence Optimization for Icing Prediction of Transmission Line", 《MATHEMATICAL PROBLEMS IN ENGINEERING》 * |
侯雨伸 等: "气象过程信息挖掘与输电线路覆冰预测", 《西安交通大学学报》 * |
尹子任 等: "基于粒子群算法优化支持向量机的输电线路覆冰预测", 《电力学报》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110866693A (en) * | 2019-11-14 | 2020-03-06 | 国网湖北省电力有限公司电力科学研究院 | Overhead transmission line icing risk assessment method based on GIS model |
CN110866693B (en) * | 2019-11-14 | 2022-05-17 | 国网湖北省电力有限公司电力科学研究院 | Overhead transmission line icing risk assessment method based on GIS model |
CN112990678A (en) * | 2021-03-04 | 2021-06-18 | 国网安徽省电力有限公司电力科学研究院 | Icing early warning judgment method based on multi-source data fusion |
CN112990678B (en) * | 2021-03-04 | 2022-06-24 | 国网安徽省电力有限公司电力科学研究院 | Icing early warning judgment method based on multi-source data fusion |
CN114353872A (en) * | 2021-12-30 | 2022-04-15 | 中国电信股份有限公司 | Prediction method and device for machine room temperature |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Abou Houran et al. | COA-CNN-LSTM: Coati optimization algorithm-based hybrid deep learning model for PV/wind power forecasting in smart grid applications | |
An et al. | Short-term wind power prediction based on particle swarm optimization-extreme learning machine model combined with AdaBoost algorithm | |
Liu et al. | Short-term photovoltaic power prediction on modal reconstruction: A novel hybrid model approach | |
CN110188914A (en) | A kind of intelligent Forecasting for grid power transmission route ice covering thickness | |
CN105046374A (en) | Power interval predication method based on nucleus limit learning machine model | |
CN104484833A (en) | Photovoltaic power generation output power tracking algorithm based on genetics algorithm improved RBF-BP neural network | |
CN105701572B (en) | Photovoltaic short-term output prediction method based on improved Gaussian process regression | |
CN110837915B (en) | Low-voltage load point prediction and probability prediction method for power system based on hybrid integrated deep learning | |
CN110942205A (en) | Short-term photovoltaic power generation power prediction method based on HIMVO-SVM | |
CN108428017A (en) | Wind power interval prediction method based on core extreme learning machine quantile estimate | |
CN113468790B (en) | Wind speed characteristic simulation method and system based on improved particle swarm optimization | |
Shao et al. | An advanced weighted system based on swarm intelligence optimization for wind speed prediction | |
CN109242200A (en) | A kind of wind power interval prediction method of Bayesian network forecasting model | |
CN112819189A (en) | Wind power output prediction method based on historical predicted value | |
CN115271253A (en) | Water-wind power generation power prediction model construction method and device and storage medium | |
CN106786499B (en) | Based on the short-term wind power forecast method for improving AFSA optimization ELM | |
Zhang et al. | Short-Term Load Forecasting Based on DBO-LSTM Model | |
CN106771847B (en) | A kind of 35kV distribution network transmission line thunderbolt risk prediction technique | |
CN108694475A (en) | Short-term time scale photovoltaic cell capable of generating power amount prediction technique based on mixed model | |
CN108009684A (en) | A kind of micro-grid connection state energy management method comprising short-term load forecasting | |
CN103268416A (en) | Wind speed prediction method based on STCP-BP | |
Wu et al. | Combined IXGBoost-KELM short-term photovoltaic power prediction model based on multidimensional similar day clustering and dual decomposition | |
Chen et al. | Short-term wind power prediction based on BP neural network improved by t-tent-SSA algorithm | |
Shi et al. | Short-term photovoltaic power forecast based on long short-term memory network | |
Cheng et al. | A hybrid probabilistic estimation method for photovoltaic power generation forecasting |
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 |