CN109118025A - A kind of method and apparatus of electric system prediction - Google Patents
A kind of method and apparatus of electric system prediction Download PDFInfo
- Publication number
- CN109118025A CN109118025A CN201811113675.1A CN201811113675A CN109118025A CN 109118025 A CN109118025 A CN 109118025A CN 201811113675 A CN201811113675 A CN 201811113675A CN 109118025 A CN109118025 A CN 109118025A
- Authority
- CN
- China
- Prior art keywords
- individual
- population
- fitness
- module
- individuals
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012549 training Methods 0.000 claims abstract description 48
- 238000012706 support-vector machine Methods 0.000 claims abstract description 42
- 239000013598 vector Substances 0.000 claims abstract description 11
- 230000008569 process Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000009401 outcrossing Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 abstract description 3
- 230000006978 adaptation Effects 0.000 abstract 1
- 238000012843 least square support vector machine Methods 0.000 description 6
- 230000002068 genetic effect Effects 0.000 description 4
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 3
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Biophysics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Physiology (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Physics (AREA)
- Genetics & Genomics (AREA)
- Entrepreneurship & Innovation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
Abstract
The invention discloses a kind of method and apparatus of electric system prediction, this method comprises: S1: initialization population parameter generates population at individual；S2: by the parameter assignment in S1 to support vector machines, the fitness of each individual in population is calculated；S3: according to individual adaptation degree, the select probability of population at individual is calculated, and individual choice is carried out with the select probability；S4: the population at individual of S3 selection is intersected, is made a variation；S5: judging whether current population reaches trained termination condition, if so, the support vector machines after being optimized, and execute S6；Otherwise, S3 is executed；S6: the support vector machines that training sample is input to after optimization is trained acquisition prediction model.The present invention effectively improves support vector machines and carries out the efficiency of load prediction, and improves the accuracy of the parameter of selection, to reduce prediction error of the supporting vector machine model in actual prediction, improves the precision of load prediction.
Description
Technical Field
The invention relates to the technical field of computer algorithms, in particular to a method and a device for predicting a power system.
Background
Scientific prediction is the basis and guarantee of correct decision making. The load prediction is a traditional research problem in the field of power systems, and means that load development is estimated and speculated in advance by analyzing and researching historical data and exploring the internal connection and development change rules among things from known conditions of power systems, economy, society, weather and the like. Load forecasting is the basic work of departments such as power system planning, power utilization, scheduling and the like, and the importance of the load forecasting is known for a long time.
The load prediction is essentially to fit and regress a power curve, and since the real-time power curve is influenced by various factors such as a power system, economy, society and weather, the real-time power curve generally has the characteristic of complex nonlinearity, a prediction model with strong learning capacity on the complex nonlinearity is preferably adopted.
Currently, a relatively mature prediction method is mainly used, such as a Support Vector Machine (SVM). The SVM considers the minimum empirical risk and the minimum structural risk simultaneously, so that the model has strong popularization, great advantages in the aspect of small sample recognition are achieved, the SVM has a strict mathematical theory basis, and the decision is global optimum.
At present, a uniform method does not exist in an SVM parameter selection strategy, and the quality of SVM parameter selection directly influences the fitting and regression capacity of a model. In the prior art, the SVM parameter optimization algorithms which are commonly used include a grid search algorithm, a particle swarm algorithm and the like. Although SVM parameters can be selected by using the algorithms, particularly suitable parameter values cannot be obtained, the speed of searching an optimal solution or a satisfactory solution is too slow, and the efficiency of load prediction according to the selected parameters is low.
Disclosure of Invention
The embodiment of the invention provides a method and a device for predicting a power system, which can obtain more optimized parameter values and solve the problem of low load prediction efficiency caused by too low speed of searching an optimal solution or a satisfactory solution by a support vector machine.
In a first aspect, an embodiment of the present invention provides a method for predicting a power system, where the method includes:
s1: initializing population parameters and generating population individuals;
s2: assigning the parameters in the S1 to a support vector machine, and calculating the fitness of each individual in the population;
s3: calculating the selection probability of population individuals according to the individual fitness, and carrying out individual selection according to the selection probability;
s4: crossing and mutating the population individuals selected in the S3;
s5: judging whether the current population reaches a training termination condition, if so, obtaining an optimized support vector machine, and executing S6; otherwise, go to S3;
s6: and inputting the training samples into the optimized support vector machine for training to obtain a prediction model.
Preferably, in step S2, the parameters in S1 are assigned to a least square machine support vector machine to obtain a prediction value, and the fitness of each individual is calculated, where the calculation formula is:
wherein f isiThe fitness of the ith individual;is a predicted value; y isiAre true values.
Preferably, the formula for calculating the selection probability of the population individuals in step S3 is:
wherein, PiThe selection probability for the ith individual.
Preferably, the probability of crossing the population individuals in step S4 is:
wherein f iscCarrying out crossing on the individuals with high fitness in the two individuals of the parent generation; f. ofmaxCarrying out maximum fitness in parent population before crossing for the individual;carrying out average fitness of all individuals in the parent population before crossing for the individuals; k is a radical of1And k2Is a constant.
Preferably, the probability of mutation of the population individuals in step S4 is:
wherein f ismThe fitness of the individual needing variation; k is a radical of3And k4Is a constant.
Preferably, the training samples in step S6 are screened training samples, and the screening process includes:
m1: determining the time of the day and obtaining a feature vector of the day;
m2: respectively calculating whether the similarity between the feature vector of the historical day meeting the preset condition and the feature vector of the current day meets a preset threshold value, and if so, selecting the current historical day as a training sample; otherwise, the current history date is excluded.
In a second aspect, an embodiment of the present invention provides an apparatus for predicting a power system, where the apparatus includes: an initial module, a value assignment module, a selection module, an alternation module, a judgment module and a training module, wherein,
the initial module is used for initializing population parameters and generating population individuals;
the assignment module is used for assigning the parameters initialized by the initial module to a support vector machine and calculating the fitness of each individual in the population;
the selection module is used for calculating the selection probability of the population individuals according to the individual fitness obtained by the assignment module and carrying out individual selection according to the selection probability;
the alternating module is used for crossing and varying the population individuals selected by the selection module;
the judging module is used for judging whether the current population reaches a training termination condition, if so, obtaining an optimized support vector machine and triggering the training module; otherwise, triggering the selection module;
and the training module is used for inputting the training samples into the optimized support vector machine for training to obtain a prediction model.
Preferably, the assigning module is specifically configured to assign the parameters initialized by the initial module to a least square machine support vector machine, obtain a predicted value, calculate the fitness of each individual, and modify the fitness of the individual, wherein,
the fitness calculation formula of the individual is as follows:
wherein,is a predicted value; y isiIs the true value; f. ofiThe fitness of the ith individual; c is a constant.
Preferably, the formula for calculating the selection probability of the population individuals by the selection module is as follows:
wherein, PiThe selection probability for the ith individual.
Preferably, the probability of the alternating module performing population individual crossing is as follows:
wherein f iscCarrying out crossing on the individuals with high fitness in the two individuals of the parent generation; f. ofmaxCarrying out maximum fitness in parent population before crossing for the individual;carrying out average fitness of all individuals in the parent population before crossing for the individuals; k is a radical of1And k2Is a constant.
Preferably, the probability of the alternating module performing individual variation of the population is as follows:
wherein f ismThe fitness of the individual needing variation; k is a radical of3And k4Is a constant.
Compared with the prior art, the invention has at least the following beneficial effects:
the invention effectively improves the speed of searching the optimal solution or the satisfactory solution, thereby improving the efficiency of load prediction of the support vector machine configured by the selected parameters, obtains the load data of similar days similar to the load data of the prediction day in the parameter selection process, optimizes the algorithm, and minimizes the error value after optimization, thereby improving the accuracy of the selected parameters, reducing the prediction error of the support vector machine model in actual prediction and improving the accuracy of the load prediction.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of a method of power system prediction provided by one embodiment of the present invention;
fig. 2 is a flowchart of a screened training sample according to an embodiment of the present invention.
Fig. 3 is a block diagram of an apparatus for power system prediction according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer and more complete, the technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention, and based on the embodiments of the present invention, all other embodiments obtained by a person of ordinary skill in the art without creative efforts belong to the scope of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a method for predicting a power system, which may include the following steps:
s1: initializing population parameters and generating population individuals;
s2: assigning the parameters in the S1 to a support vector machine, and calculating the fitness of each individual in the population;
s3: calculating the selection probability of population individuals according to the individual fitness, and carrying out individual selection according to the selection probability;
s4: crossing and mutating the population individuals selected in the S3;
s5: judging whether the current population reaches a training termination condition, if so, obtaining an optimized support vector machine, and executing S6; otherwise, go to S3;
s6: and inputting the training samples into the optimized support vector machine for training to obtain a prediction model.
In this embodiment, step S1 may generate parameters such as an initial kernel function parameter and a penalty factor, encode the parameters by using a binary code with a length of 8 bits, and set a population size and an iteration number, where each individual in the population is in a parameter encoding form, and randomly generate an initial value of the individual. The training termination condition is whether the iteration number reaches a preset iteration number or whether the error is smaller than a preset threshold, and if the iteration number is smaller than the preset iteration number, the step S3 is executed; alternatively, if the error is greater than the preset threshold, step S3 is executed.
In an embodiment of the present invention, step S2 specifically includes assigning the parameters in S1 to a least square machine support vector machine to obtain predicted values, and calculating the fitness of each individual, where the calculation formula is:
wherein f isiThe fitness of the ith individual;is a predicted value; y isiAre true values.
In this embodiment, the least squares support vector machine is assigned with the parameters of step S1, and then fitness values of the respective individuals are calculated according to the prediction results.
In an embodiment of the present invention, the formula for the selection module to calculate the selection probability of the population individuals is as follows:
wherein, PiThe selection probability for the ith individual.
In this embodiment, after the fitness of the individual is obtained, the individual is selected with the individual selection probability when calculating the selection probability of the population of individuals based on the obtained fitness value.
In an embodiment of the present invention, the probability of population-individual crossing performed by the alternating module is:
wherein f iscCarrying out crossing on the individuals with high fitness in the two individuals of the parent generation; f. ofmaxCarrying out maximum fitness in parent population before crossing for the individual;carrying out average fitness of all individuals in the parent population before crossing for the individuals; k is a radical of1And k2Is a constant.
The probability of variation of the population individuals in step S4 is:
wherein f ismThe fitness of the individual needing variation; k is a radical of3And k4Is a constant.
In the embodiment, the individual parameters are converted into binary codes, the cross operation adopts single-point cross, and the variation strategy adopts multi-point variation. Generally take k1And k2Is 1, k3And k4Is 0.5. Through the adjustment of the method, the cross probability P of the genetic algorithm to the high-quality individuals (namely the fitness is higher than the average fitness value of the population) in the searching processcProbability of variation PmTaking the smaller, promoting the rapid convergence of the genetic algorithm, and for individuals with fitness values lower than the population average fitness value, the cross probability PcProbability of variation PmThe larger the value, the early convergence phenomenon caused by the trapping of the local extreme point of the genetic algorithm is avoided.
As shown in fig. 2, in an embodiment of the present invention, the training samples in step S6 are screened training samples, and the screening process includes:
m1: determining the time of the day and obtaining a feature vector of the day;
m2: respectively calculating whether the similarity between the feature vector of the historical day meeting the preset condition and the feature vector of the current day meets a preset threshold value, and if so, selecting the current historical day as a training sample; otherwise, the current history date is excluded.
In this embodiment, since the characteristics of the daily data include weather (sunny, cloudy, rainy), highest temperature, lowest temperature, average temperature, humidity, and the like, similar historical daily data may be selected to improve the accuracy of the selected parameters. The process of selecting similar historical daily data, that is, the process of screening training samples, may be: 1) the day to be predicted (i.e., the current day) differs from the month of the current day by no more than 2 months or the same month of the previous year; 2) the working characteristics of the date to be predicted are the same, namely the working date, weekend or holiday; 3) giving a similarity threshold; 4) calculating the similarity of the feature vectors of the days to be predicted and the historical days meeting the conditions 1) and 2) (the similarity calculation can adopt Euclidean distance); 5) when the similarity value is smaller than a given threshold, the day is taken as a training sample. Inputting the screened training samples to a least square support vector machine, obtaining a prediction model after training is completed, and calling the model to obtain a predicted value of 24 hours in the future. When the method is applied to a power system, a predicted heat load value of the future 24 hours can be obtained. In addition, the conditions of the selected area historical days can be properly modified according to the weather characteristics of a certain future day given by the weather forecast, and training samples can be screened for the certain future day.
The superiority of the invention is verified by using experiments. The predicted heat load value (24 hours, one heat load value per hour) for 30 days was selected for comparison. The test data was one day, 24 points. Three algorithms are compared respectively:
(1) searching a least square support vector machine algorithm of the selected parameters by using a grid;
(2) training a least square support vector machine algorithm (parameters of the least square support vector machine are obtained by grid search) by using a training set selected on a similar day;
(3) the algorithm of the invention is as follows: training a least square support vector machine algorithm by using a training set selected on a similar day, wherein parameters of the least square support vector machine are obtained by using an improved genetic algorithm;
the root mean square error RMSE and the mean relative error MAPE indicators for the three methods are compared and the data is as follows:
mean relative error MAPE:
root mean square error RMSE:
the calculation results are shown in table 1 below:
TABLE 1
| Index (I) | SVM algorithm | Similar day + SVM algorithm | Algorithm of the invention |
| RMSE | 0.96 | 0.72 | 0.49 |
| MAPE | 8.2％ | 6.7％ | 5.9％ |
Through comparison of experimental data, it can be seen that the method provided herein can achieve better effects on prediction of thermal load.
It should be noted that, in an embodiment of the present invention, the individual fitness calculated in step S2 may be modified, and the specific process of the modification is as follows:
n1: obtaining the average fitness of the current populationMaximum fitness in a current population
fmaxAnd the minimum fitness f in the current populationmin；
N2: if it isThen N3 is executed; otherwise, perform N4;
N3：
N4：
n5: obtaining the corrected individual fitness f'i＝afi+b，i＝1，2，3…，n；
Wherein, f'iThe corrected individual fitness is the ith; f. ofiThe ith uncorrected individual fitness; c is a constant.
In this embodiment, the individual fitness after correction can be effectively prevented from being smaller than 0 by correcting the individual fitness. In the following steps, the corrected individual fitness can be adopted for calculation when the selection probability of the population individuals is calculated, the probability of crossing the population individuals is calculated, and the probability of variation of the population individuals is calculated.
As shown in fig. 3, an embodiment of the present invention provides an apparatus for predicting a power system, where the apparatus includes: an initial module, a value assignment module, a selection module, an alternation module, a judgment module and a training module, wherein,
the initial module is used for initializing population parameters and generating population individuals;
the assignment module is used for assigning the parameters initialized by the initial module to a support vector machine and calculating the fitness of each individual in the population;
the selection module is used for calculating the selection probability of the population individuals according to the individual fitness obtained by the assignment module and carrying out individual selection according to the selection probability;
the alternating module is used for crossing and mutating the population individuals selected by the selection module;
the judging module is used for judging whether the current population reaches a training termination condition, if so, obtaining an optimized support vector machine and triggering the training module; otherwise, triggering the selection module;
and the training module is used for inputting the training samples into the optimized support vector machine for training to obtain the prediction model.
Preferably, the assigning module is specifically configured to assign the parameters initialized by the initial module to a least square machine support vector machine, obtain a predicted value, and calculate the fitness of each individual, wherein,
the fitness calculation formula of the individual is as follows:
wherein,is a predicted value; y isiIs the true value; f. ofiThe fitness of the ith individual; c is a constant.
In an embodiment of the present invention, the formula for the selection module to calculate the selection probability of the population individuals is as follows:
wherein, PiThe selection probability for the ith individual.
In an embodiment of the present invention, the probability of population-individual crossing performed by the alternating module is:
wherein f iscCarrying out crossing on the individuals with high fitness in the two individuals of the parent generation; f. ofmaxCarrying out maximum fitness in parent population before crossing for the individual;carrying out average fitness of all individuals in the parent population before crossing for the individuals; k is a radical of1And k2Is a constant.
In an embodiment of the present invention, the probability of the alternating module performing population individual variation is:
wherein f ismThe fitness of the individual needing variation; k is a radical of3And k4Is a constant.
It should be noted that, in an embodiment of the present invention, the individual fitness calculated by the assignment module may be modified, and the specific process of the modification is as follows:
the specific process for correcting the individual fitness comprises the following steps:
n1: obtaining the average fitness of the current populationMaximum fitness in a current population
fmaxAnd the minimum fitness f in the current populationmin；
N2: if it isThen N3 is executed; otherwise, perform N4;
N3：
N4：
n5: obtaining the corrected individual fitness f'i＝afi+b，i＝1，2，3…，n；
Wherein, f'iThe corrected individual fitness is the ith; f. ofiThe ith uncorrected individual fitness; c is a constant.
In this embodiment, the individual fitness after correction can be effectively prevented from being smaller than 0 by correcting the individual fitness. In other modules of the device, the corrected individual fitness can be adopted for calculation when the selection probability of the population individuals is calculated, the probability of crossing the population individuals is calculated, and the probability of variation of the population individuals is calculated.
Because the content of information interaction, execution process, and the like among the modules in the device is based on the same concept as the method embodiment of the present invention, specific content can be referred to the description in the method embodiment of the present invention, and is not described herein again.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a" does not exclude the presence of other similar elements in a process, method, article, or apparatus that comprises the element.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it is to be noted that: the above description is only a preferred embodiment of the present invention, and is only used to illustrate the technical solutions of the present invention, and not to limit the protection scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A method of power system prediction, the method comprising:
s1: initializing population parameters and generating population individuals;
s2: assigning the parameters in the S1 to a support vector machine, and calculating the fitness of each individual in the population;
s3: calculating the selection probability of population individuals according to the individual fitness, and carrying out individual selection according to the selection probability;
s4: crossing and mutating the population individuals selected in the S3;
s5: judging whether the current population reaches a training termination condition, if so, obtaining an optimized support vector machine, and executing S6; otherwise, go to S3;
s6: and inputting the training samples into the optimized support vector machine for training to obtain a prediction model.
2. The method for predicting the power system according to claim 1, wherein the step S2 is specifically to assign the parameters in S1 to a least square machine support vector machine, obtain the predicted values, and calculate the fitness of each individual, and the calculation formula is as follows:
wherein f isiThe fitness of the ith individual;is a predicted value; y isiAre true values.
3. The method for predicting the power system according to claim 2, wherein the formula for calculating the selection probability of the population individuals in step S3 is as follows:
wherein, PiThe selection probability for the ith individual.
4. The method of power system prediction according to claim 2,
the probability of crossing the population individuals in step S4 is:
wherein f iscIs an individualCarrying out crossing on the individuals with high fitness in the parent generation; f. ofmaxCarrying out maximum fitness in parent population before crossing for the individual;carrying out average fitness of all individuals in the parent population before crossing for the individuals; k is a radical of1And k2Is a constant.
5. The method of power system prediction according to claim 2,
the probability of variation of the population individuals in step S4 is:
wherein f ismThe fitness of the individual needing variation; k is a radical of3And k4Is a constant.
6. The method for predicting the power system according to claim 1, wherein the training samples in step S6 are screened training samples, and the screening process includes:
m1: determining the time of the day and obtaining a feature vector of the day;
m2: respectively calculating whether the similarity between the feature vector of the historical day meeting the preset condition and the feature vector of the current day meets a preset threshold value, and if so, selecting the current historical day as a training sample; otherwise, the current history date is excluded.
7. An apparatus for power system prediction, the apparatus comprising: an initial module, a value assignment module, a selection module, an alternation module, a judgment module and a training module, wherein,
the initial module is used for initializing population parameters and generating population individuals;
the assignment module is used for assigning the parameters initialized by the initial module to a support vector machine and calculating the fitness of each individual in the population;
the selection module is used for calculating the selection probability of the population individuals according to the individual fitness obtained by the assignment module and carrying out individual selection according to the selection probability;
the alternating module is used for crossing and varying the population individuals selected by the selection module;
the judging module is used for judging whether the current population reaches a training termination condition, if so, obtaining an optimized support vector machine and triggering the training module; otherwise, triggering the selection module;
and the training module is used for inputting the training samples into the optimized support vector machine for training to obtain a prediction model.
8. The power system prediction device of claim 7,
the assignment module is specifically used for assigning the parameters initialized by the initial module to a least square machine support vector machine to obtain predicted values and calculate the fitness of each individual, wherein,
the fitness calculation formula of the individual is as follows:
wherein,is a predicted value; y isiIs the true value; f. ofiThe fitness of the ith individual; c is a constant.
9. The power system prediction device of claim 8,
the formula for calculating the selection probability of the population individuals by the selection module is as follows:
wherein, PiThe selection probability for the ith individual.
10. The power system prediction device of claim 8,
the probability of the alternating module for carrying out population individual crossing is as follows:
wherein f iscCarrying out crossing on the individuals with high fitness in the two individuals of the parent generation; f. ofmaxCarrying out maximum fitness in parent population before crossing for the individual;carrying out average fitness of all individuals in the parent population before crossing for the individuals; k is a radical of1And k2Is a constant;
the probability of the alternating module for carrying out individual variation of the population is as follows:
wherein f ismThe fitness of the individual needing variation; k is a radical of3And k4Is a constant.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811113675.1A CN109118025A (en) | 2018-09-25 | 2018-09-25 | A kind of method and apparatus of electric system prediction |
| PCT/CN2019/107947 WO2020063690A1 (en) | 2018-09-25 | 2019-09-25 | Electrical power system prediction method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811113675.1A CN109118025A (en) | 2018-09-25 | 2018-09-25 | A kind of method and apparatus of electric system prediction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109118025A true CN109118025A (en) | 2019-01-01 |
Family
ID=64856590
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811113675.1A Pending CN109118025A (en) | 2018-09-25 | 2018-09-25 | A kind of method and apparatus of electric system prediction |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109118025A (en) |
| WO (1) | WO2020063690A1 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020063690A1 (en) * | 2018-09-25 | 2020-04-02 | 新智数字科技有限公司 | Electrical power system prediction method and apparatus |
| CN112348597A (en) * | 2020-11-27 | 2021-02-09 | 新智数字科技有限公司 | Power price prediction method and device based on least square support vector machine model |
| CN112598282A (en) * | 2020-12-24 | 2021-04-02 | 清华大学 | Method and device for identifying load model parameters of power system |
| CN113642640A (en) * | 2021-08-13 | 2021-11-12 | 华北电力科学研究院有限责任公司 | Regional power grid minimum output prediction method and device |
Families Citing this family (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113627639A (en) * | 2020-05-07 | 2021-11-09 | 中国石油化工股份有限公司 | Well testing productivity prediction method and system for carbonate fracture-cave reservoir |
| CN111859627B (en) * | 2020-06-29 | 2023-09-22 | 珠海格力电器股份有限公司 | Parameter optimization method and device for component model |
| CN111914900B (en) * | 2020-07-03 | 2023-10-27 | 深圳供电局有限公司 | User electricity utilization mode classification method |
| CN112036436B (en) * | 2020-07-23 | 2024-02-13 | 国网江苏省电力有限公司检修分公司 | Data noise processing method and processing system of camera oil temperature prediction system |
| CN112070529A (en) * | 2020-08-24 | 2020-12-11 | 贵州民族大学 | Passenger carrying hotspot parallel prediction method, system, terminal and computer storage medium |
| CN117238379A (en) * | 2020-09-17 | 2023-12-15 | 温州大学 | Storage medium storing gene selection method program |
| CN117238453A (en) * | 2020-09-23 | 2023-12-15 | 温州大学 | Storage medium storing medical image feature processing method program |
| CN112182968B (en) * | 2020-09-28 | 2024-01-30 | 长安大学 | Method, system and equipment for constructing equivalent circuit model of lithium ion battery |
| CN112669169B (en) * | 2020-12-15 | 2024-04-30 | 国网辽宁省电力有限公司阜新供电公司 | Short-term photovoltaic power prediction device and method |
| CN113033064B (en) * | 2021-01-04 | 2024-03-01 | 国网浙江省电力有限公司营销服务中心 | Power equipment impact load simulation method based on support vector machine |
| CN113326917B (en) * | 2021-04-29 | 2024-06-25 | 开放智能机器（上海）有限公司 | Genetic algorithm-based automatic operator optimizing method and system |
| CN113239821A (en) * | 2021-05-18 | 2021-08-10 | 北京科技大学 | Real-time automatic identification method for home behaviors |
| CN113268432B (en) * | 2021-06-24 | 2023-09-01 | 广东电网有限责任公司计量中心 | Electric energy meter driver testing method and system based on evolutionary algorithm |
| CN113780768B (en) * | 2021-08-26 | 2024-07-19 | 浙江大学常州工业技术研究院 | Chaotic variation genetic high-dimensionality multi-index comprehensive energy system planning method and system |
| CN114550283A (en) * | 2022-01-10 | 2022-05-27 | 南京邮电大学 | Wireless network signal action recognition system and method based on improved support vector machine |
| CN114565239B (en) * | 2022-02-15 | 2023-06-09 | 石河子大学 | Comprehensive low-carbon energy scheduling method and system for industrial park |
| CN116010291A (en) * | 2023-03-28 | 2023-04-25 | 江西财经大学 | Multipath coverage test method based on equalization optimization theory and gray prediction model |
| CN116108763B (en) * | 2023-04-13 | 2023-06-27 | 湖南工商大学 | Method for predicting bubble collapse critical point of foaming material based on temperature |
| CN116579253B (en) * | 2023-07-11 | 2023-09-22 | 中国海洋大学 | Method and device for determining physical properties of sealing strip, electronic equipment and storage medium |
| CN117252114B (en) * | 2023-11-17 | 2024-02-13 | 湖南华菱线缆股份有限公司 | Cable torsion resistance experiment method based on genetic algorithm |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106096771A (en) * | 2016-06-07 | 2016-11-09 | 广州供电局有限公司 | The parameter selection method of support vector machine and system and load forecasting method and system |
| CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学（保定） | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
| US20180150766A1 (en) * | 2016-11-30 | 2018-05-31 | Daegu Gyeongbuk Institute Of Science And Technology | Classification method based on support vector machine |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109118025A (en) * | 2018-09-25 | 2019-01-01 | 新智数字科技有限公司 | A kind of method and apparatus of electric system prediction |
-
2018
- 2018-09-25 CN CN201811113675.1A patent/CN109118025A/en active Pending
-
2019
- 2019-09-25 WO PCT/CN2019/107947 patent/WO2020063690A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学（保定） | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
| CN106096771A (en) * | 2016-06-07 | 2016-11-09 | 广州供电局有限公司 | The parameter selection method of support vector machine and system and load forecasting method and system |
| US20180150766A1 (en) * | 2016-11-30 | 2018-05-31 | Daegu Gyeongbuk Institute Of Science And Technology | Classification method based on support vector machine |
Non-Patent Citations (3)
| Title |
|---|
| 李顺昕 等: "基于加权粒子群优化的LSSVM的钢铁企业电力负荷预测", 《华北电力大学学报》 * |
| 邓军 等: "基于支持向量机的无线电干扰预测算法", 《电机与控制学报》 * |
| 郭晓原: "遗传算法中适应度尺度变换与操作算子的比较研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020063690A1 (en) * | 2018-09-25 | 2020-04-02 | 新智数字科技有限公司 | Electrical power system prediction method and apparatus |
| CN112348597A (en) * | 2020-11-27 | 2021-02-09 | 新智数字科技有限公司 | Power price prediction method and device based on least square support vector machine model |
| CN112598282A (en) * | 2020-12-24 | 2021-04-02 | 清华大学 | Method and device for identifying load model parameters of power system |
| CN113642640A (en) * | 2021-08-13 | 2021-11-12 | 华北电力科学研究院有限责任公司 | Regional power grid minimum output prediction method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020063690A1 (en) | 2020-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109118025A (en) | A kind of method and apparatus of electric system prediction | |
| CN111199016B (en) | Daily load curve clustering method for improving K-means based on DTW | |
| CN109345027B (en) | Micro-grid short-term load prediction method based on independent component analysis and support vector machine | |
| CN110751318B (en) | Ultra-short-term power load prediction method based on IPSO-LSTM | |
| CN109886589A (en) | A method of low-carbon Job-Shop is solved based on whale optimization algorithm is improved | |
| CN108932557A (en) | A kind of Short-term Load Forecasting Model based on temperature cumulative effect and grey relational grade | |
| CN116596044B (en) | Power generation load prediction model training method and device based on multi-source data | |
| CN108805743A (en) | A kind of power grid enterprises' sale of electricity company operation Benefit Evaluation Method | |
| CN105447509A (en) | Short-term power prediction method for photovoltaic power generation system | |
| CN105117461B (en) | A kind of enquiring and optimizing method based on Revised genetic algorithum | |
| CN109376950A (en) | A kind of polynary Load Forecasting based on BP neural network | |
| CN114792156A (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
| CN110163444A (en) | A kind of water demand prediction method based on GASA-SVR | |
| CN103345660A (en) | Cold load prediction method based on support vector machine parameters optimized with cat swarm algorithm | |
| CN115186803A (en) | Data center computing power load demand combination prediction method and system considering PUE | |
| CN110837929A (en) | Least square support vector machine electricity utilization prediction method based on adaptive genetic algorithm | |
| Guo et al. | Support vector machine model in electricity load forecasting | |
| CN112036651A (en) | Electricity price prediction method based on quantum immune optimization BP neural network algorithm | |
| CN112749211B (en) | Novel tea yield prediction method based on electric power big data | |
| CN113222234A (en) | Gas demand prediction method and system based on integrated modal decomposition | |
| CN114650321A (en) | Task scheduling method for edge computing and edge computing terminal | |
| CN109685242B (en) | Photovoltaic ultra-short term combined prediction method based on Adaboost algorithm | |
| CN116826699A (en) | Photovoltaic cluster power prediction method, device, equipment and medium | |
| CN110991748A (en) | Short-term load prediction method for urban power grid | |
| CN109934394A (en) | A kind of Demand Side Response prediction technique based on grey and Markov theory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190101 |
|
| RJ01 | Rejection of invention patent application after publication |