CN117057475A - Charging station load prediction method, charging station load prediction device, storage medium and computer equipment - Google Patents

Charging station load prediction method, charging station load prediction device, storage medium and computer equipment Download PDF

Info

Publication number
CN117057475A
CN117057475A CN202311092251.2A CN202311092251A CN117057475A CN 117057475 A CN117057475 A CN 117057475A CN 202311092251 A CN202311092251 A CN 202311092251A CN 117057475 A CN117057475 A CN 117057475A
Authority
CN
China
Prior art keywords
load
charging station
function
training
support vector
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
Application number
CN202311092251.2A
Other languages
Chinese (zh)
Inventor
龙羿
高辉
李炜卓
徐婷婷
陈良亮
龙虹毓
张谦
胡晓锐
胡文
黄会
高芸
王松
池磊
李顺
曹登焜
李涛永
蒋林洳
李培军
杨烨
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Chongqing Electric Power Co Marketing Service Center
State Grid Corp of China SGCC
Original Assignee
State Grid Chongqing Electric Power Co Marketing Service Center
State Grid Corp of China SGCC
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by State Grid Chongqing Electric Power Co Marketing Service Center, State Grid Corp of China SGCC filed Critical State Grid Chongqing Electric Power Co Marketing Service Center
Priority to CN202311092251.2A priority Critical patent/CN117057475A/en
Publication of CN117057475A publication Critical patent/CN117057475A/en
Pending legal-status Critical Current

Links

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
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • 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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0202—Market predictions or forecasting for commercial activities
    • 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
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Resources & Organizations (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Tourism & Hospitality (AREA)
  • Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Operations Research (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Primary Health Care (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Quality & Reliability (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Power Engineering (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a charging station load prediction method, a charging station load prediction device, a storage medium and computer equipment. Comprising the following steps: acquiring historical load information of a charging station; processing the historical load information and determining a historical load value corresponding to the load influence factor; determining a characteristic vector of a load influence factor; training a support vector regression model based on the sample data; and inputting the feature vector of the target data into a support vector regression model conforming to a preset condition to obtain a predicted load value of the charging station. According to the method, the influence factors of various different dimensions are fully considered, the load change of the charging station in different time and states can be accurately predicted, the defect that the load change is difficult to accurately predict in the prior art in new energy charging is overcome, model training and prediction can be performed more quickly even in a charging station load prediction scene with a smaller data set, and effective prediction service is provided for safe and stable operation of the electric vehicle charging station.

Description

Charging station load prediction method, charging station load prediction device, storage medium and computer equipment
Technical Field
The application relates to the technical field of new energy, in particular to a charging station load prediction method, a charging station load prediction device, a storage medium and computer equipment.
Background
As the amount of electric vehicles kept increases year by year, the demand for charging stations is also increasing. However, charging stations can place a significant load on the grid during charging, especially during peak hours. Therefore, load overload can be avoided by accurately predicting the load of the charging station, and the charging efficiency and the user experience are improved.
In the related art, the number of charging piles and the charging rate are used as supports in most cases, and a regression analysis method, a time sequence method, an exponential smoothing method and a gray prediction method are adopted to predict the load of the charging station, but factors influencing the load of the charging station are various and complex, and the relevance between the load of the charging station and other influencing factors cannot be fully considered, so that the prediction accuracy is not high. And the model accuracy is further reduced due to the problems of excessive network parameters, slow training and the like when the neural network is trained by using a large-scale sample.
Disclosure of Invention
In view of the above, the present application provides a charging station load prediction method, apparatus, storage medium and computer device, which solve the technical problem that it is difficult to rapidly and accurately predict the charging station load change.
According to an embodiment of one aspect of the present application, there is provided a charging station load prediction method including:
Acquiring historical load information of a charging station;
processing the historical load information, and determining a historical load value corresponding to a load influence factor, wherein the load influence factor comprises a user charging habit factor, a time type factor, a weather factor and a season factor;
determining a characteristic vector of a load influence factor;
training a support vector regression model based on sample data, wherein the sample data comprises a feature vector and a historical load value corresponding to the feature vector;
and inputting the feature vector of the target data into a support vector regression model conforming to a preset condition to obtain a predicted load value of the charging station.
Further, determining a feature vector of the load influencing factor includes:
and carrying out dummy coding on the load influence factors to obtain feature vectors corresponding to the load influence factors.
Further, before training the support vector regression model based on the sample data, the charging station load prediction method further includes:
and carrying out normalization processing on the sample data.
Further, the sample data includes training data; training a support vector regression model based on the sample data, comprising:
mapping training data to a high-dimensional feature space based on a kernel function of a support vector regression machine, and determining a regression function of the training data;
And constructing a support vector regression model based on the regression function.
Further, mapping the training data to a high-dimensional feature space based on a kernel function of a support vector regression machine, determining a regression function of the training data, comprising:
constructing an objective function based on the kernel function and the relaxation variable;
solving an objective function to obtain a hyperplane parameter of a support vector regression machine;
a regression function is determined based on the hyperplane parameters.
Further, before mapping the training data to the high-dimensional feature space based on the kernel function of the support vector regression machine, the charging station load prediction method further comprises:
constructing a kernel function based on the Gaussian radial basis function;
determining a parameter range of a kernel function, wherein the parameter range comprises a plurality of groups of function parameters, and the function parameters comprise kernel width and penalty coefficients;
dividing training data into a plurality of training subsets and verification subsets by adopting a cross algorithm;
model training is carried out on a plurality of training subsets based on each group of function parameters, and a plurality of intermediate models corresponding to each group of function parameters are constructed;
sample verification is carried out on the verification subset based on the plurality of intermediate models, and performance indexes of function parameters corresponding to the plurality of intermediate models are determined, wherein the performance indexes comprise at least one of the following: average relative error, root mean square error and determining coefficient;
Calculating a score for each set of function parameters based on an average of the performance metrics of the plurality of intermediate models;
and determining the function parameter with the largest score as the target parameter of the kernel function.
Further, the sample data includes test data, and after training the support vector regression model based on the sample data, the charging station load prediction method further includes:
inputting the feature vector in the test data into a support vector regression model to obtain a test load value;
and if the difference value between the test load value and the historical load value in the test data is in a preset range, determining that the support vector regression model meets preset conditions.
Further, after the historical load information of the charging station is obtained, the charging station load prediction method further comprises the following steps:
and filtering and complementing the historical load information.
According to an embodiment of another aspect of the present application, there is provided a charging station load prediction apparatus including:
the acquisition module is used for acquiring historical load information of the charging station;
the processing module is used for processing the historical load information and determining a historical load value corresponding to a load influence factor, wherein the load influence factor comprises a user charging habit factor, a time type factor, a weather factor and a season factor; the method comprises the steps of,
Determining a characteristic vector of a load influence factor;
the construction module is used for training a support vector regression model based on sample data, wherein the sample data comprises a characteristic vector and a historical load value corresponding to the characteristic vector;
and the prediction module is used for inputting the feature vector of the target data into a support vector regression model conforming to a preset condition to obtain a predicted load value of the charging station.
Further, the processing module is specifically configured to perform dummy coding on the load influencing factor, so as to obtain a feature vector corresponding to the load influencing factor.
Further, the processing module is further used for carrying out normalization processing on the sample data.
Further, the sample data includes training data; the construction module is specifically used for mapping the training data to a high-dimensional feature space based on a kernel function of the support vector regression machine and determining a regression function of the training data; and constructing a support vector regression model based on the regression function.
Further, a construction module is specifically configured to construct an objective function based on the kernel function and the relaxation variable; solving an objective function to obtain a hyperplane parameter of a support vector regression machine; a regression function is determined based on the hyperplane parameters.
Further, a construction module is specifically used for constructing a kernel function based on the Gaussian radial basis function; determining a parameter range of a kernel function, wherein the parameter range comprises a plurality of groups of function parameters, and the function parameters comprise kernel width and penalty coefficients; dividing training data into a plurality of training subsets and verification subsets by adopting a cross algorithm; model training is carried out on a plurality of training subsets based on each group of function parameters, and a plurality of intermediate models corresponding to each group of function parameters are constructed; sample verification is carried out on the verification subset based on the plurality of intermediate models, and performance indexes of function parameters corresponding to the plurality of intermediate models are determined, wherein the performance indexes comprise at least one of the following: average relative error, root mean square error and determining coefficient; calculating a score for each set of function parameters based on an average of the performance metrics of the plurality of intermediate models; and determining the function parameter with the largest score as the target parameter of the kernel function.
Further, the sample data includes test data, and the charging station load prediction apparatus further includes:
the test module is used for inputting the feature vector in the test data into a support vector regression model to obtain a test load value; and if the difference value between the test load value and the historical load value in the test data is in a preset range, determining that the support vector regression model meets preset conditions.
Further, the processing module is also used for filtering and complementing the historical load information.
According to still another aspect of the present application, there is provided a readable storage medium having stored thereon a program or instructions which when executed by a processor, implement the steps of the charging station load prediction method described above.
According to still another aspect of the present application, there is provided a computer device including a storage medium, a processor and a computer program stored on the storage medium and executable on the processor, the processor implementing the steps of the charging station load prediction method described above when executing the program.
By means of the technical scheme of the embodiment, historical load data provided by the charging station is collected first. By processing the historical load information, the historical load value of the charging station under the condition of various different load influencing factors is determined. And converting the load influence factors with different formats into feature vectors suitable for model identification due to format differences of the different load influence factors. And training a support vector regression (Support Vector Regression, SVR) model for predicting the load of the charging station by taking the characteristic vector and the historical load value corresponding to the characteristic vector as sample data. And finally, inputting the feature vector of the target data into a support vector regression model which meets the preset condition, namely, can reach the prediction accuracy requirement, and obtaining the prediction load value of the charging station. On the one hand, the historical load values under the user charging habit factors, the time type factors, the meteorological factors and the seasonal factors are used as training samples, and the support vector regression model is trained, so that the influence factors of various different dimensions are fully considered, the load change of the charging station in different time and states can be accurately predicted, and the defect that the load change is difficult to accurately predict in the prior art in the new energy charging process is overcome. On the other hand, the load influence factors which do not meet the machine learning modeling requirement are converted into the numerical characteristic vectors, so that the accurate description of the load influence factors through a physical model is realized, the characteristics of data can be better understood, and the prediction efficiency and accuracy are improved. On the other hand, the complex nonlinear relation in the data can be better captured by adopting the support vector regression model, and the method has better robustness on noise and abnormal values in the data, and particularly under the scene of smaller data volume such as the predicted charging station load, the accuracy and efficiency of the charging station load prediction are effectively improved.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
fig. 1 is a schematic flow chart of a charging station load prediction method according to an embodiment of the present application;
FIG. 2 shows a summer load graph provided by an embodiment of the present application;
FIG. 3 shows a winter load graph provided by an embodiment of the present application;
FIG. 4 shows a graph comparing test load values and historical load values provided by an embodiment of the present application;
fig. 5 shows a schematic structural diagram of a charging station load prediction device according to an embodiment of the present application.
Detailed Description
The application will be described in detail hereinafter with reference to the drawings in conjunction with embodiments. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the application.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly fused. The term "and/or" as used herein includes all or any element and all combination of one or more of the associated listed items.
Exemplary embodiments according to the present application will now be described in more detail with reference to the accompanying drawings. These exemplary embodiments may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. It should be appreciated that these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of these exemplary embodiments to those skilled in the art.
In this embodiment, a charging station load prediction method is provided, as shown in fig. 1, and the method includes:
step 110, acquiring historical load information of a charging station;
the historical load information comprises load values of different charging guns of different charging piles of the charging station under different load influence factors. Load influencing factors include user charging habit factors (e.g., charging preferences at different times, etc.), time type factors (e.g., whether it is a weekday, whether it is a specified holiday, etc.), weather factors (e.g., sunny day, rainy day, temperature, humidity, etc.), and seasonal factors (e.g., spring, summer, etc.).
Specifically, the historical load information can be obtained from a charging station background database, and the historical load information can be obtained through the output electric quantity of the timing device, the weather recording device and the charging pile.
It should be noted that, in an embodiment, after step 110, the charging station load prediction method further includes: and filtering and complementing the historical load information.
In this embodiment, statistical methods (e.g., mean, standard deviation) are used to check the historical load information for the presence of outliers, such as extremely high or extremely low values, that are outside of the threshold range, and to remove outliers from the historical load information. Meanwhile, for the missing load value or the deleted abnormal value, interpolation methods (such as linear interpolation and polynomial interpolation) can be used to fill in the gap. Therefore, errors or data with low precision can be eliminated, fine screening of the data is realized, the quality and reliability of the processed data are ensured, and the subsequent data processing speed is improved.
Step 120, processing the historical load information, and determining a historical load value corresponding to the load influence factor;
in this embodiment, all the historical load values are associated with load influencing factors such as user charging habit factors, time type factors, weather factors, season factors, and the like by processing the historical load information. Therefore, the correlation between the load influence factors and the load values is assisted to analyze, the influence factors of various different dimensions are fully considered, reliable and comprehensive data support is provided for subsequent model training, the accurate prediction of the future load trend and fluctuation of the charging station in different time and states is facilitated, and a corresponding load management strategy is formulated.
Further, as a refinement and extension of the foregoing embodiment, to fully describe the implementation process of this embodiment, after step 120, the charging station load prediction method further includes: performing mutual information calculation on the historical load value and different load influence factors respectively, and determining the mutual information value of the load influence factors; sorting the load influence factors corresponding to the historical load values based on the mutual information values, and determining a load influence factor list; and outputting a load influence factor list.
The mutual information (Mutual Information, MI) is an index for measuring the degree of association between two variables, and can be used for evaluating the influence degree of load influence factors on the historical load value.
In this embodiment, the charging station load may be used as one random variable, and the load influencing factors (such as weather, season, working day, etc.) may be used as another random variable, and the mutual information value between each historical load value and the different load influencing factors may be calculated respectively using a calculation formula of the mutual information. The larger the mutual information value is, the higher the correlation between the load influencing factors and the historical load value is. And then sorting the load influence factors corresponding to the historical load values according to the mutual information values, so that a load influence factor list with correlation as a sorting basis can be obtained. Therefore, the correlation between the historical load value and the load influence factors is measured by utilizing the mutual information, so that a user can intuitively know the influence degree of each load influence factor on the load value of the charging station through the load influence factor list output by the system.
In addition, the load influence factors with mutual information values larger than a preset value and the historical load values can be used as sample data, available data in the historical load information can be further filtered, the data space density is reduced, the data redundancy is effectively reduced, the sample data for training the model can be more attached to the actual charging condition, and the prediction accuracy of the model is higher.
For example, as shown in fig. 2 and 3, since the air conditioning load of the electric vehicle is more in summer (6 months to 9 months) and winter (11 months to 2 months), the electric vehicle is overall more energy-efficient, but the load profile in summer is significantly higher than in winter, which results in more charging time required for the electric vehicle in summer. Therefore, when predicting charging station load, it is necessary to use seasonal factors as training data.
Further, as a refinement and extension of the foregoing embodiment, to fully describe the implementation process of this embodiment, after step 120, the charging station load prediction method further includes: calculating correlation coefficients among different load influence factors according to the historical load values; a load influencing factor thermodynamic diagram is generated and output based on the correlation coefficient. Therefore, the influence of the visual characteristic factors on the load prediction is realized, so that a user is assisted in understanding the interrelationship among the load influence factor variables, and guidance and decision basis are provided for the aspects of load prediction, energy management, system optimization and the like.
For example, the maximum load on the day, the minimum load on the day, the maximum air temperature on the day, the minimum air temperature on the day, the weather type, the season type, whether or not the working day is taken as the load influencing factor associated with the historical load value. The horizontal axis and the vertical axis of the thermodynamic diagram respectively represent different influencing factors, the value range of the correlation coefficient of each cell is-1 to 1, positive values are positive correlations, negative values are negative correlations, wherein-1 represents complete negative correlations, 1 represents complete positive correlations, 0 represents no correlations, and the bigger the correlation coefficient is, the stronger the correlations are. Similarly, the darker the color of each cell in the thermodynamic diagram indicates the magnitude of the correlation coefficient, and the lighter the color the weaker or no correlation.
Step 130, determining a characteristic vector of the load influencing factors;
in this embodiment, since data items such as weather and seasons are classified variants and are generally expressed in text, the machine learning modeling requirement is not satisfied. The load influence factors which do not accord with the machine learning modeling requirement are converted into the numerical characteristic vectors, so that the accurate description of the load influence factors through a physical model is realized, the characteristics of data can be better understood, and the prediction efficiency and the prediction accuracy are improved.
In an actual application scenario, step 130, namely determining a feature vector of the load influencing factor, specifically includes:
and 131, performing dummy coding on the load influence factors to obtain feature vectors corresponding to the load influence factors.
In this embodiment, for each classification variable (load influencing factor), a new binary feature vector is created. The length of the feature vector is equal to the number of possible values of the variable. For each variable sample, only one element in the feature vector is 1, which means that the sample has the value, and the other elements are all 0. And combining the eigenvectors of all the load influence factors into an eigenvector matrix, wherein each column corresponds to the eigenvector of one load influence factor. Therefore, discrete load influence factors can be converted into binary vector representations through a dummy coding technology, wherein each characteristic vector value corresponds to a unique binary code, so that the method is suitable for various machine learning algorithms and statistical analysis methods, and the load influence factors can be used for predicting, modeling and optimizing the load of the charging station conveniently.
Step 140, training a support vector regression model based on the sample data;
The sample data comprises a characteristic vector and a historical load value corresponding to the characteristic vector.
It can be understood that the sample data can be divided into training data and test data, the training data is used for training a model, and the test data is used for testing the trained model, so that the consistency of sources of the training data and the test data is ensured, and the test result is more accurate.
In the embodiment, a load influence factor with high relevance is used as an input vector, a historical load value is used as an output value to perform model training, and a support vector regression model is built. Under the condition of ensuring the prediction precision, the complex nonlinear relation in the data can be better captured, the robustness to noise and abnormal values in the data is better, and the accuracy and the efficiency of the load prediction of the charging station are further improved.
Further, as a refinement and extension of the foregoing embodiment, for a complete description of the implementation process of this embodiment, step 140, that is, before training the support vector regression model based on the sample data, the charging station load prediction method further includes: and carrying out normalization processing on the sample data.
Specifically, the normalization process may employ the following formula:
wherein X is the value of the feature vector before conversion, X norm X is the value of the converted feature vector max For the maximum value of the sample, X min Is the sample minimum.
In this embodiment, the feature space may increase in dimension due to the dummy encoding process, especially in the case of having a large number of discrete feature values, which may increase the complexity of the model and the calculation cost. For this purpose, after determining the sample data, all the sample data are normalized, and each dimension of the feature vectors of different dimensions in the sample data is normalized to be within the [0,1] interval. Therefore, the dimension influence of the feature vectors with different dimensions can be eliminated, errors caused by different data dimensions are prevented, the weight of each feature vector in the model is more balanced, the stability of data distribution is ensured, and the stability and the interpretability of the model after training are improved. In an actual application scenario, the sample data is divided into training data. Step 140, namely training a support vector regression model based on the sample data, specifically includes:
step 141, mapping training data to a high-dimensional feature space based on a kernel function of a support vector regression machine, and determining a regression function of the training data;
Specifically, the kernel functions include a linear kernel function, a polynomial kernel function, a Radial Basis Function (RBF) kernel function, and the like, and an appropriate kernel function may be selected according to characteristics of data and requirements of problems.
Step 142, constructing a support vector regression model based on the regression function;
in this embodiment, the kernel function is used to map the training data from the original feature space to the high-dimensional feature space, without explicitly calculating the coordinates of the high-dimensional feature space, saving computational resources. And then carrying out regression analysis on the training data in a high-dimensional space to find a nonlinear relation between the input variable and the output variable in the high-dimensional space, so as to train a regression function with the aim of minimizing risks for constructing the model. And finally, constructing a support vector regression model based on the regression function, so that the support vector regression model can process nonlinear relations and has stronger generalization capability and robustness. In addition, even in a scene of smaller data volume such as the predicted charging station load, the support vector regression model can be ensured to be modeled and predicted more efficiently.
In an actual application scenario, step 142, that is, mapping the training data to the high-dimensional feature space based on the kernel function of the support vector regression machine, determines a regression function of the training data, specifically includes:
Step 142-1, constructing an objective function based on the kernel function and the relaxation variables;
wherein the objective function targets risk minimization, the objective function may be expressed as:
wherein w is a hyperplane parameter, C is a penalty coefficient (regularization parameter), n is the number of training data, and ζ i And xi * i And (3) as a relaxation variable corresponding to the ith training data, wherein the relaxation variable is used for representing errors of the sample points on two sides of the hyperplane.
Wherein, xi i And xi * i The following relationship is satisfied:
wherein x is i Is the original feature vector of the ith training data, y i Is the true value of the ith training data, f (x i ) For the predicted value (x of the ith training data i Mapping to a high-dimensional feature space), epsilon is the maximum error allowed by the regression analysis.
Further, in an embodiment, prior to step 142-1, the charging station load prediction method further comprises:
step 210, constructing a kernel function based on the Gaussian radial basis function;
specifically, the gaussian Radial Basis Function (RBF) is expressed as:
wherein x is i For the center of the kernel function, σ is the width of the kernel function (kernel width), σ controls the influence range of the sample point in the feature space, a smaller σ will make the influence range larger, and a larger σ will make the influence range smaller.
Step 220, determining a parameter range of the kernel function;
the parameter range comprises a plurality of groups of function parameters, and the function parameters comprise kernel width and penalty coefficients.
It should be noted that the size of the parameter grid may be reduced appropriately according to the complexity of the problem and the limitation of the computing resources, or other more efficient parameter optimization methods may be used, such as random search, bayesian optimization, etc.
Step 230, dividing the training data into a plurality of training subsets and verification subsets by adopting a crossover algorithm;
for example, training data is equally divided into v shares, v-1 subsets as training subsets and the v-th subset as verification subset.
Step 240, respectively performing model training on the plurality of training subsets based on each group of function parameters, and constructing a plurality of intermediate models corresponding to each group of function parameters;
step 250, performing sample verification on the verification subset based on the plurality of intermediate models, and determining performance indexes of function parameters corresponding to the plurality of intermediate models;
wherein the performance index comprises at least one of the following: average relative error (Mean Absolute Percentage Error, MAPE), root mean square error (Root Mean Squared Error, RMSE), and decision coefficient (Coefficient of Determination). The closer the average relative error or root mean square error is to 0, the smaller the prediction error of the model, and the closer to positive infinity, the larger the prediction error of the model. The value range of the decision coefficient is between 0 and 1, and the closer to 1 is the better the fitting degree of the model to the data, and the closer to 0 is the worse the fitting degree of the model to the data.
The average relative error MAE can be expressed as:
the root mean square error RMSE can be expressed as:
determining the coefficient R 2 Can be expressed as:
wherein n is the number of samples, s i As a predictor of the intermediate model,to verify the measured values of the subset +.>Is the mean of the measured values.
Step 260, calculating the score of each set of function parameters based on the average value of the performance indexes of the plurality of intermediate models;
wherein a larger score indicates a smaller average relative error and root mean square error, the closer the decision coefficient is to 1.
It will be appreciated that if the performance indicator comprises a plurality of data items, for example, the average relative error and the root mean square error are used as the performance indicators, the average value of each data item may be calculated separately, and the score for each set of function parameters may be determined by a weighted calculation.
Step 270, determining the function parameter with the largest score as the target parameter of the kernel function.
In this embodiment, the parameter ranges of the kernel width and penalty coefficients may be determined by a grid search, and multiple sets of function parameters are selected from the parameter ranges, each set of function parameters including a different kernel width and penalty coefficient. The training data is equally divided into a plurality of subsets using a cross-validation algorithm, with a portion of the subsets being training subsets for training. Model training is carried out on the training subsets based on each group of function parameters, and a plurality of intermediate models corresponding to each group of function parameters are constructed. Predicting another partial subset (verification subset) through the intermediate model to obtain a plurality of performance indexes for determining the function parameters corresponding to the plurality of intermediate models, calculating the score of each group of function parameters according to the average value of the plurality of performance indexes, and grading the evaluation standard of the function parameters. And so on, a score for each set of function parameters is obtained. By comparing the scores, the performances of different parameter combinations are compared, so that the target parameter with the largest score can be conveniently screened out from multiple groups of function parameters. The reasonable kernel function parameters are configured in a calculation cross-validation mode, so that the accuracy of the support vector regression model in predicting the load value is improved, and the technical problem of fewer samples for model training is solved.
Step 142-2, solving an objective function to obtain hyperplane parameters of a support vector regression machine;
specifically, the hyperplane parameters include a weight vector for the hyperplane.
In step 142-3, a regression function is determined based on the hyperplane parameters.
Specifically, the hyperplane of the support vector regression machine (Support Vector Regression, SVR) refers to a hyperplane in the feature space, and the shape and fitting degree of the regression function can be changed by adjusting the parameters of the hyperplane. The regression function can be expressed as:
wherein w is a hyperplane parameter,to map the hidden function, x is the feature vector of the training data in the original space, b is the bias term, and f (x) is the predicted value of the training data (the result of x mapping to the high-dimensional feature space). Thus, by calculating the inner product of the input feature vector and the hyperplane parameter (weight vector) and adding the bias term, the predicted output value can be obtained.
In this embodiment, the regression problem can be converted into an optimization problem by introducing relaxation variables and regularization terms. And solving the objective function to obtain the hyperplane parameters of the support vector regression machine. And calculating a regression function according to the parameters of the hyperplane and the input feature vector, so that the regression function can carry out regression prediction based on the hyperplane. Therefore, the training data is mapped to the high-dimensional feature space, so that the nonlinear problem is converted into the linear problem, and the regression function can be fitted better.
Further, as a refinement and extension of the foregoing embodiment, for fully describing the implementation process of this embodiment, in a case where the sample data includes test data, step 140, that is, after training the support vector regression model based on the sample data, the charging station load prediction method further includes:
step 310, inputting the feature vector in the test data into a support vector regression model to obtain a test load value;
step 320, if the difference between the test load value and the historical load value in the test data is within the preset range, determining that the support vector regression model meets the preset condition.
Wherein the difference may be an absolute value or a relative value. The preset range can be reasonably set according to the model precision requirement, and the smaller the preset range is, the higher the model precision requirement is, so that the embodiment of the application is not particularly limited.
In this embodiment, the support vector regression model is tested using the test data to determine that the support vector regression model is adequate to meet the prediction accuracy requirements. If the test load value obtained by the support vector regression model has a larger difference from the historical load value in the test data, the support vector regression model is not enough in precision, and model training is needed again to adjust and optimize the support vector regression model. Otherwise, if the difference between the test load value and the historical load value in the test data is in the preset range, that is, the difference between the test load value and the historical load value is small, the support vector regression model can be judged to reach the required precision, and the support vector regression model is used for carrying out charging station load prediction. Thereby ensuring the accuracy of the support vector regression model in predicting the load value.
As shown in fig. 4, the test load value obtained by the test data substantially coincides with the fluctuation of the history load value in the test data. The candidate model has good fitting effect, and is used as a support vector regression model for predicting the load value of the charging station, so that the prediction accuracy of the support vector regression model is ensured.
And 150, inputting the feature vector of the target data into a support vector regression model meeting preset conditions to obtain a predicted load value of the charging station.
According to the charging station load prediction method provided by the embodiment, historical load data provided by a charging station is collected. By processing the historical load information, the historical load value of the charging station under the condition of various different load influencing factors is determined. And converting the load influence factors with different formats into feature vectors suitable for model identification due to format differences of the different load influence factors. And training a support vector regression (Support Vector Regression, SVR) model for predicting the load of the charging station by taking the characteristic vector and the historical load value corresponding to the characteristic vector as sample data. And finally, inputting the feature vector of the target data into a support vector regression model to obtain a predicted load value of the charging station. On the one hand, the historical load values under the user charging habit factors, the time type factors, the meteorological factors and the seasonal factors are used as training samples, and the support vector regression model is trained, so that the influence factors of various different dimensions are fully considered, the load change of the charging station in different time and states can be accurately predicted, and the defect that the load change is difficult to accurately predict in the prior art in the new energy charging process is overcome. On the other hand, the load influence factors which do not meet the machine learning modeling requirement are converted into the numerical characteristic vectors, so that the accurate description of the load influence factors through a physical model is realized, the characteristics of data can be better understood, and the prediction efficiency and accuracy are improved. On the other hand, the complex nonlinear relation in the data can be better captured by adopting the support vector regression model, and the method has better robustness on noise and abnormal values in the data, and particularly under the scene of smaller data volume such as the predicted charging station load, the accuracy and efficiency of the charging station load prediction are further improved.
It should be noted that, the sequence number of each step in the above embodiment does not mean the sequence of execution sequence, and the execution sequence of each process should be determined by its function and internal logic, and should not limit the implementation process of the embodiment of the present application in any way.
Further, as shown in fig. 5, as a specific implementation of the charging station load prediction method, an embodiment of the present application provides a charging station load prediction apparatus 500, where the charging station load prediction apparatus 500 includes: an acquisition module 501, a processing module 502, a construction module 503 and a prediction module 504.
An obtaining module 501, configured to obtain historical load information of a charging station;
the processing module 502 is configured to process the historical load information, and determine a historical load value corresponding to a load influence factor, where the load influence factor includes a user charging habit factor, a time type factor, a weather factor and a seasonal factor; and determining a feature vector of the load influencing factor;
a building module 503, configured to train a support vector regression model based on sample data, where the sample data includes a feature vector and a historical load value corresponding to the feature vector;
the prediction module 504 is configured to input the feature vector of the target data into a support vector regression model that meets a preset condition, and obtain a predicted load value of the charging station.
Further, the processing module 502 is specifically configured to perform dummy coding on the load influencing factor, so as to obtain a feature vector corresponding to the load influencing factor.
Further, the processing module 502 is further configured to normalize the sample data.
Further, the sample data includes training data, and a construction module 503 is specifically configured to map the training data to a high-dimensional feature space based on a kernel function of the support vector regression machine, and determine a regression function of the training data; and constructing a support vector regression model based on the regression function.
Further, a construction module 503 is specifically configured to construct an objective function based on the kernel function and the relaxation variable; solving an objective function to obtain a hyperplane parameter of a support vector regression machine; a regression function is determined based on the hyperplane parameters.
Further, a construction module 503 is specifically configured to construct a kernel function based on the gaussian radial basis function; determining a parameter range of a kernel function, wherein the parameter range comprises a plurality of groups of function parameters, and the function parameters comprise kernel width and penalty coefficients; dividing training data into a plurality of training subsets and verification subsets by adopting a cross algorithm; model training is carried out on a plurality of training subsets based on each group of function parameters, and a plurality of intermediate models corresponding to each group of function parameters are constructed; sample verification is carried out on the verification subset based on the plurality of intermediate models, and performance indexes of function parameters corresponding to the plurality of intermediate models are determined, wherein the performance indexes comprise at least one of the following: average relative error, root mean square error and determining coefficient; calculating a score for each set of function parameters based on an average of the performance metrics of the plurality of intermediate models; and determining the function parameter with the largest score as the target parameter of the kernel function.
Further, the sample data includes test data, and the charging station load prediction apparatus 500 further includes: the test module (not shown in the figure) is used for inputting the feature vectors in the test data into the support vector regression model to obtain a test load value; and if the difference value between the test load value and the historical load value in the test data is in a preset range, determining that the support vector regression model meets preset conditions.
Further, the processing module 502 is further configured to perform filtering processing and complement processing on the historical load information.
For specific limitations on the charging station load prediction device, reference may be made to the above limitations on the charging station load prediction method, and no further description is given here. The above-described respective modules in the charging station load prediction apparatus may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
Based on the method shown in fig. 1, correspondingly, the embodiment of the application also provides a readable storage medium, on which a computer program is stored, which when being executed by a processor, implements the charging station load prediction method shown in fig. 1.
Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.), and includes several instructions for causing a computer device (may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective implementation scenario of the present application.
Based on the method shown in fig. 1 and the virtual device embodiment shown in fig. 5, in order to achieve the above object, the embodiment of the present application further provides a computer device, which may specifically be a personal computer, a server, a network device, etc., where the computer device includes a storage medium and a processor; a storage medium storing a computer program; a processor for executing a computer program to implement the charging station load prediction method as shown in fig. 1.
Optionally, the computer device may also include a user interface, a network interface, a camera, radio Frequency (RF) circuitry, sensors, audio circuitry, WI-FI modules, and the like. The user interface may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (e.g., bluetooth interface, WI-FI interface), etc.
It will be appreciated by those skilled in the art that the architecture of a computer device provided in the present embodiment is not limited to the computer device, and may include more or fewer components, or may combine certain components, or may be arranged in different components.
The storage medium may also include an operating system, a network communication module. An operating system is a program that manages and saves computer device hardware and software resources, supporting the execution of information handling programs and other software and/or programs. The network communication module is used for realizing communication among all components in the storage medium and communication with other hardware and software in the entity equipment.
From the above description of the embodiments, it will be clear to those skilled in the art that the present application may be implemented by means of software plus necessary general hardware platform, or may be implemented by hardware to obtain historical load information of a charging station; processing the historical load information, and determining a historical load value corresponding to a load influence factor, wherein the load influence factor comprises a user charging habit factor, a time type factor, a weather factor and a season factor; determining a characteristic vector of a load influence factor; training a support vector regression model based on sample data, wherein the sample data comprises a feature vector and a historical load value corresponding to the feature vector; and inputting the feature vector of the target data into a support vector regression model conforming to a preset condition to obtain a predicted load value of the charging station. According to the embodiment of the application, on one hand, the historical load values under the charging habit factors, the time type factors, the weather factors and the seasonal factors of the user are used as training samples, and the support vector regression model is trained, so that the influence factors of various different dimensions are fully considered, the load change of the charging station in different time and states can be accurately predicted, and the defect that the load change is difficult to accurately predict in the prior art in the new energy charging is overcome. On the other hand, the load influence factors which do not meet the machine learning modeling requirement are converted into the numerical characteristic vectors, so that the accurate description of the load influence factors through a physical model is realized, the characteristics of data can be better understood, and the prediction efficiency and accuracy are improved. On the other hand, the support vector regression model is adopted to better capture the complex nonlinear relation in the data, and has better robustness on noise and abnormal values in the data, so that the accuracy and efficiency of the load prediction of the charging station are further improved.
Those skilled in the art will appreciate that the drawing is merely a schematic illustration of a preferred implementation scenario and that the modules or flows in the drawing are not necessarily required to practice the application. Those skilled in the art will appreciate that modules in an apparatus in an implementation scenario may be distributed in an apparatus in an implementation scenario according to an implementation scenario description, or that corresponding changes may be located in one or more apparatuses different from the implementation scenario. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.
The above-mentioned inventive sequence numbers are merely for description and do not represent advantages or disadvantages of the implementation scenario. The foregoing disclosure is merely illustrative of some embodiments of the application, and the application is not limited thereto, as modifications may be made by those skilled in the art without departing from the scope of the application.

Claims (10)

1. A charging station load prediction method, the method comprising:
acquiring historical load information of the charging station;
processing the historical load information, and determining a historical load value corresponding to a load influence factor, wherein the load influence factor comprises a user charging habit factor, a time type factor, a weather factor and a season factor;
Determining a feature vector of the load influencing factor;
training a support vector regression model based on sample data, wherein the sample data includes the feature vector and the historical load value corresponding to the feature vector;
and inputting the feature vector of the target data into the support vector regression model meeting preset conditions to obtain the predicted load value of the charging station.
2. The charging station load prediction method according to claim 1, wherein,
the determining the characteristic vector of the load influencing factor comprises the following steps:
performing dummy coding on the load influence factors to obtain the feature vectors corresponding to the load influence factors;
before the training of the support vector regression model based on the sample data, the method further includes:
and carrying out normalization processing on the sample data.
3. The charging station load prediction method of claim 1, wherein the sample data comprises training data, and the training support vector regression model based on the sample data comprises:
mapping the training data to a high-dimensional feature space based on a kernel function of a support vector regression machine, and determining a regression function of the training data;
And constructing the support vector regression model based on the regression function.
4. The charging station load prediction method of claim 3, wherein the mapping the training data to a high-dimensional feature space based on a kernel function of a support vector regression machine, determining a regression function of the training data, comprises:
constructing an objective function based on the kernel function and the relaxation variable;
solving the objective function to obtain a hyperplane parameter of a support vector regression machine;
the regression function is determined based on the hyperplane parameters.
5. The charging station load prediction method of claim 3, wherein before mapping the training data to the high-dimensional feature space based on a kernel function of a support vector regression machine, the method further comprises:
constructing the kernel function based on a Gaussian radial basis function;
determining a parameter range of the kernel function, wherein the parameter range comprises a plurality of groups of function parameters, and the function parameters comprise kernel width and penalty coefficients;
dividing the training data into a plurality of training subsets and verification subsets by adopting a crossover algorithm;
model training is carried out on a plurality of training subsets based on each group of function parameters, and a plurality of intermediate models corresponding to each group of function parameters are constructed;
Sample verification is carried out on the verification subset based on the plurality of intermediate models, and performance indexes of the function parameters corresponding to the plurality of intermediate models are determined, wherein the performance indexes comprise at least one of the following: average relative error, root mean square error and determining coefficient;
calculating a score for each set of function parameters based on an average of the performance metrics of the plurality of intermediate models;
and determining the function parameter with the largest score as a target parameter of the kernel function.
6. The charging station load prediction method of claim 1, wherein the sample data comprises test data, and wherein after training a support vector regression model based on the sample data, the method further comprises:
inputting the characteristic vector in the test data into the support vector regression model to obtain a test load value;
and if the difference value between the test load value and the historical load value in the test data is in a preset range, determining that the support vector regression model meets the preset condition.
7. The charging station load prediction method according to claim 1, wherein after the obtaining the historical load information of the charging station, the method further comprises:
And filtering and complementing the historical load information.
8. A charging station load prediction apparatus, the apparatus comprising:
the acquisition module is used for acquiring the historical load information of the charging station;
the processing module is used for processing the historical load information and determining a historical load value corresponding to a load influence factor, wherein the load influence factor comprises a user charging habit factor, a time type factor, a weather factor and a season factor; the method comprises the steps of,
determining a feature vector of the load influencing factor;
the construction module is used for training a support vector regression model based on sample data, wherein the sample data comprises the characteristic vector and the historical load value corresponding to the characteristic vector;
and the prediction module is used for inputting the feature vector of the target data into the support vector regression model meeting the preset condition to obtain the predicted load value of the charging station.
9. A readable storage medium having stored thereon a program or instructions, which when executed by a processor, implement the steps of the charging station load prediction method of any one of claims 1 to 7.
10. A computer device comprising a storage medium, a processor and a computer program stored on the storage medium and executable on the processor, characterized in that the processor implements the charging station load prediction method according to any one of claims 1 to 7 when executing the program.
CN202311092251.2A 2023-08-28 2023-08-28 Charging station load prediction method, charging station load prediction device, storage medium and computer equipment Pending CN117057475A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311092251.2A CN117057475A (en) 2023-08-28 2023-08-28 Charging station load prediction method, charging station load prediction device, storage medium and computer equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311092251.2A CN117057475A (en) 2023-08-28 2023-08-28 Charging station load prediction method, charging station load prediction device, storage medium and computer equipment

Publications (1)

Publication Number Publication Date
CN117057475A true CN117057475A (en) 2023-11-14

Family

ID=88669099

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311092251.2A Pending CN117057475A (en) 2023-08-28 2023-08-28 Charging station load prediction method, charging station load prediction device, storage medium and computer equipment

Country Status (1)

Country Link
CN (1) CN117057475A (en)

Similar Documents

Publication Publication Date Title
CN111327377B (en) Method, device, equipment and storage medium for field intensity prediction
CN116453438B (en) Display screen parameter detection method, device, equipment and storage medium
CN113469426A (en) Photovoltaic output power prediction method and system based on improved BP neural network
CN115936262B (en) Yield prediction method, system and medium based on big data environment interference
CN112287980B (en) Power battery screening method based on typical feature vector
CN112288137A (en) LSTM short-term load prediction method and device considering electricity price and Attention mechanism
CN113570007A (en) Method, device and equipment for optimizing construction of part defect identification model and storage medium
CN116401954A (en) Prediction method, prediction device, equipment and medium for cycle life of lithium battery
CN115423594A (en) Enterprise financial risk assessment method, device, equipment and storage medium
CN114781520A (en) Natural gas behavior abnormity detection method and system based on improved LOF model
CN115018200A (en) Power load prediction method and system based on deep learning and considering multiple influence factors
CN113376541B (en) Lithium ion battery health state prediction method based on CRJ network
CN109242141B (en) Method and device for predicting inventory quantity of commodities
CN108256693B (en) Photovoltaic power generation power prediction method, device and system
CN114169502A (en) Rainfall prediction method and device based on neural network and computer equipment
CN116960962A (en) Mid-long term area load prediction method for cross-area data fusion
CN113890833B (en) Network coverage prediction method, device, equipment and storage medium
CN116304713A (en) Wind power plant fault scene prediction model generation method and device and computer equipment
CN116662893A (en) Water quality prediction method for optimizing SVM (support vector machine) based on improved goblet sea squirt algorithm
CN116245259A (en) Photovoltaic power generation prediction method and device based on depth feature selection and electronic equipment
CN117057475A (en) Charging station load prediction method, charging station load prediction device, storage medium and computer equipment
CN116431346A (en) Compensation method for main memory capacity of electronic equipment
CN114234392B (en) Air conditioner load fine prediction method based on improved PSO-LSTM
CN115932661A (en) Method for analyzing and judging noise compliance of transformer
CN112348700B (en) Line capacity prediction method combining SOM clustering and IFOU equation

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