CN114970719A - Internet of things operation index prediction method based on improved SVR model - Google Patents
Internet of things operation index prediction method based on improved SVR model Download PDFInfo
- Publication number
- CN114970719A CN114970719A CN202210584561.5A CN202210584561A CN114970719A CN 114970719 A CN114970719 A CN 114970719A CN 202210584561 A CN202210584561 A CN 202210584561A CN 114970719 A CN114970719 A CN 114970719A
- Authority
- CN
- China
- Prior art keywords
- svr model
- internet
- model
- operation index
- data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 29
- 238000005457 optimization Methods 0.000 claims abstract description 11
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims description 21
- 239000002245 particle Substances 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 18
- 238000002790 cross-validation Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000010200 validation analysis Methods 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000007637 random forest analysis Methods 0.000 description 19
- 230000000694 effects Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000010845 search algorithm Methods 0.000 description 3
- 238000010219 correlation analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an Internet of things operation index prediction method based on an improved SVR model, which comprises the following steps: preprocessing the climate data and the Internet of things operation index data; performing feature selection on the preprocessed climate data, and eliminating redundant features; establishing an RF-SVR model, and optimizing parameters of the RF-SVR model by using an intelligent optimization algorithm PSO to obtain the RF-PSO-SVR model; and outputting an Internet of things operation index prediction result through an RF-PSO-SVR model. The method improves the reliability of feature selection, is beneficial to improving the prediction precision of a subsequent prediction model, and optimizes the SVR model after feature selection by using a PSO algorithm aiming at the problem that the accurate prediction of local Internet of things operation index data cannot be realized by using the SVR only, thereby improving the model fitting goodness, reducing the mean square error and effectively improving the prediction precision of the Internet of things operation index data.
Description
Technical Field
The invention belongs to the field of prediction of an operation index of an electric power system Internet of things, and particularly relates to an Internet of things operation index prediction method based on an improved SVR model.
Background
The internet technology in China has gradually penetrated from the initial combination with the consumption service field to the combination with the industrial field, and the traditional industrial field is in the important window period of digital transformation and upgrading. In the face of mass data of a user side of a power system, the prior art lacks deep analysis on the influence factors and the influence strength of the local Internet of things operation indexes, and an effective prediction method is difficult to form.
A prediction model between the Internet of things operation index and the climate factor can be analyzed and established by developing the research on the machine learning algorithm. The machine learning prediction method comprises two important research directions of data preprocessing and model building. The research of the data preprocessing direction focuses on a feature selection part, and common feature selection methods include a gray correlation analysis method, a genetic algorithm, a Random Forest (RF) and the like. The grey correlation analysis method is suitable for occasions with strong subjectivity and can influence the specialty of the feature selection result to a certain extent. While genetic algorithms require many iterations and are far higher in computational complexity than RF. The RF algorithm, while more efficient, is used alone, ignoring the contribution of features to model fitness. Therefore, researchers have proposed a method for determining feature dimensions by calculating feature importance using RF and using a Sequence Forward Selection (SFS) algorithm based on classification accuracy. In the aspect of further enhancing the reliability of the training set feature selection result, the K-fold cross validation and search algorithm can play a good role. The operation index of the Internet of things belongs to a continuous variable, so a regression model needs to be established. Among the regression models, linear regression is most widely used, but has a large limitation in dealing with non-linear problems. In the electric power field, many variables are in nonlinear relation. Therefore, a model needs to be established by an algorithm with higher generalization capability, and Support Vector machine Regression (SVR) is used to map data to a high-dimensional feature space, so as to convert an optimization problem into a convex quadratic programming problem, and realize modeling of nonlinear data. But accurate prediction of local internet of things operation index data cannot be realized by using the SVR only.
Therefore, a new technical solution is needed to solve this problem.
Disclosure of Invention
The purpose of the invention is as follows: in order to overcome the defects in the prior art, the method for predicting the Internet of things operation index based on the improved SVR model is provided, is simple to use, and can effectively improve the prediction accuracy.
The technical scheme is as follows: in order to achieve the purpose, the invention provides an Internet of things operation index prediction method based on an improved SVR model, which comprises the following steps:
s1: preprocessing the climate data and the Internet of things operation index data;
s2: performing feature selection on the preprocessed climate data, and eliminating redundant features;
s3: establishing a climate and internet of things operation index prediction (RF-SVR) model according to the preprocessed internet of things operation index data and the climate data after feature selection, and optimizing parameters of the RF-SVR model by using an intelligent optimization algorithm PSO to obtain an RF-PSO-SVR model;
s4: and outputting the prediction result of the operation index of the Internet of things through the RF-PSO-SVR model.
Further, the preprocessing method in step S1 includes:
a1: randomly dividing a climate and thing-internet operation index data set into a training set and a testing set according to a proportion;
a2: and respectively carrying out standardization processing on the training set and the test set so as to eliminate the influence of dimension.
Further, the step S2 is specifically:
b1: selecting training set data for feature selection;
b2: traversing the set number of RF trees using a trellis search;
b3: determining the number of RF trees by using K-fold cross validation with the minimum validation error as a criterion;
b4: sorting the feature importance by using the RF after the parameters are optimized;
b5: for the feature with the sorted importance degree, the dimension Selection is carried out by using a Sequence Forward Selection (SFS) algorithm based on the goodness of fit. The purpose of dimension selection is to eliminate redundant and interference features, improve the accuracy of the SVR model and reduce the computational complexity of the SVR model.
Further, the step B3 is specifically:
dividing a data set into K groups by L-fold cross validation, circulating for K times, selecting a group of data which is not repeated with the previous data as a validation set in each circulation, and using the rest K-1 groups as a training set; after the cycle is finished, K parameter values and K verification errors corresponding to the K parameter values can be obtained, the average error of the K errors is calculated and used as a new verification error, and finally, the parameter value corresponding to the minimum verification error is selected, namely the number of the RF trees is introduced into the RF feature selection model.
Further, the method for performing dimension selection in step B5 by using the sequence forward selection algorithm based on goodness of fit is:
and adding the features into the SVR model one by one from the empty set according to the importance, and selecting the feature dimension which enables the fitting goodness of the SVR model to be highest.
Further, the process of optimizing the parameters of the climate and physical link operation index prediction model by using the intelligent optimization algorithm PSO in step S3 includes:
c1: parameter information such as iteration times, population number and the like of the particle swarm algorithm is set;
c2: initializing penalty parameters (C), loss functions (epsilon) and kernel coefficients (gamma) of the SVR model, initializing the position and the speed of particles in the particle swarm optimization, and then calculating the goodness of fit (R) of the SVR model in the training set 2 ) As fitness;
c3: continuously iterating, updating the position and the speed of the particles, calculating corresponding fitness, and recording C, epsilon and gamma corresponding to the global optimal fitness;
c4: and (4) bringing the optimal parameters C, epsilon and gamma obtained in the step C3 into the test set SVR model for prediction.
Further, R is used in the step S3 2 And Mean Square Error (MSE) to test the SVR model performance.
Further, the position and velocity of the particles in the steps C2 and C3 are realized by the following formulas:
for particle i 1,2, 3.., N, the velocity calculation formula is:
v i (t+1)＝v i (t)+c 1 *rand()*(pbest i (t)-x i (t))+c 2 *rand()*(gbest i (t)-x i (t))
wherein, c 1 And c 2 As a learning factor, pbest i For optimal position obtained by a single search, gbest i For the optimal position obtained by global search, rand () represents the number between randomly selected intervals (0, 1);
the position calculation formula is:
x i (t+1)＝x i (t)+v i (t)。
further, the SVR model in step C4 is implemented by the following formula:
wherein, alpha * Is a Lagrange multiplier, k (x) i ,x j ) Representing the kernel function, b is the displacement.
The method aims at predicting the local Internet of things operation index of the metering equipment at the client side of the smart grid under different climatic conditions. Firstly, the characteristics of the climate data are selected so as to improve the model performance and reduce the calculation complexity, and then the optimization algorithm is used for further optimizing the model performance. In feature selection, a grid search algorithm and K-fold cross validation are used for Random Forest (RF) parameter selection. For the calculated RF feature importance, dimension Selection is performed using a goodness-of-fit based Sequence Forward Selection (SFS) algorithm. After the characteristics are selected, a Particle Swarm Optimization (PSO) algorithm is used for optimizing a Support Vector Regression (SVR) model.
Has the beneficial effects that: compared with the prior art, the invention has the following advantages:
1. according to the method, the grid search algorithm is utilized to traverse the RF parameters, the K-fold cross validation is utilized to determine the RF parameters, and for the calculated feature importance, the SFS algorithm based on the goodness of fit is utilized to select the feature dimension, so that the reliability of feature selection is improved, and the prediction accuracy of a subsequent prediction model is improved.
2. Aiming at the problem that accurate prediction of local Internet of things operation index data cannot be achieved only by using the SVR, the method optimizes the SVR model after feature selection by using the PSO algorithm, thereby improving the goodness of fit of the model, reducing the mean square error and effectively improving the prediction accuracy of the Internet of things operation index data.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a diagram showing the goodness of fit of SVR models corresponding to different dimensional features;
FIG. 3 is a diagram of the prediction effect of different algorithm models.
Detailed Description
The present invention is further illustrated by the following figures and specific examples, which are to be understood as illustrative only and not as limiting the scope of the invention, which is to be given the full breadth of the appended claims and any and all equivalent modifications thereof which may occur to those skilled in the art upon reading the present specification.
The embodiment provides an internet of things operation index prediction method based on an improved SVR model, as shown in fig. 1, including the following steps:
s1: preprocessing the climate data and the Internet of things operation index data;
the climate variable data selected in the embodiment is the highest temperature, the lowest temperature, the average temperature, the relative humidity, the rainfall, the wind speed and the air pressure data of a certain area of Jiangsu province for three continuous years; the index variable of the internet of things is success rate data acquired for three continuous years.
The pretreatment process in this example is:
s1-1: dividing the climate data and the IoT operation index data set randomly according to the proportion of 80% of the training set and 20% of the testing set;
s1-2: and respectively carrying out standardization processing on the training set and the test set to eliminate the influence of dimensions.
S2: the method specifically includes the following steps:
s2-1: selecting training set data for feature selection;
s2-2: the set number of RF trees is traversed using a trellis search.
S2-3: the number of the preset trees traversed by the gridding search is as follows: 20. 40, 60, 80, 100, 120, 140, 160, 180, 200, 300, 500.
S2-4: determining the number of RF trees by using K-fold cross validation with the minimum validation error as a criterion;
in this example, 5-fold cross-validation was used to divide the data set into 5 groups, and 5 cycles were performed. Each cycle picks a set of data that is not repeated before as a validation set and uses the remaining 4 sets as training sets. After the loop is over, 5 parameter values and 5 verification errors corresponding to the parameter values can be obtained. The average error of these 5 errors is calculated and used as the new validation error. Finally, the parameter value 160 corresponding to the minimum verification error is selected and substituted into the RF characteristic selection model.
S2-5: sorting the feature importance by using the RF after the parameters are optimized;
the feature importance ranking results in this example are shown in table 1:
TABLE 1 training set feature importance
The importance of the maximum temperature, the minimum temperature, the average temperature, the relative humidity, the air pressure, the rainfall and the average wind speed are respectively as follows: 0.45573, 0.25744, 0.21525, 0.02632, 0.02465, 0.01339, 0.00722.
S2-6: and (4) utilizing an SFS algorithm based on goodness of fit to perform dimension selection.
The results of the descending order of the importance of the features in table 1 are combined, the features are added to the SVR model one by one and the corresponding goodness-of-fit is calculated, and the results are shown in fig. 2. And eliminating the average wind speed characteristic according to the model goodness of fit and the characteristic sorting result shown in the figure 2 and the table 1.
S3: according to the preprocessed internet of things operation index data and the climate data after feature selection, a climate and internet of things operation index prediction (RF-SVR) model is established, parameters of the RF-SVR model are optimized by an intelligent optimization algorithm PSO, and the RF-PSO-SVR model is obtained;
the steps in this embodiment specifically include the following processes:
s3-1: parameter information such as iteration times, population number and the like of the particle swarm algorithm is set;
the PSO algorithm parameter setting conditions are as follows: iteration is carried out for 20 times, the population size is 20, the weight factor is 0.4, and the learning factor is 2.
S3-2: initializing punishment parameters (C), loss functions (epsilon) and kernel coefficients (gamma) of the SVR model, initializing the position and the speed of particles in the particle swarm optimization, and then calculating R of the SVR model in a training set 2 As fitness;
the initialized C is 2.08346, epsilon is 0.23507, and gamma is 0.31808, the initial velocity of the particle is 0.8, and the initial position is a random number. Finally calculated R of training set SVR model 2 Is 0.28.
S3-3: continuously iterating, updating the position and the speed of the particles, calculating corresponding fitness, and recording C, epsilon and gamma corresponding to the global optimal fitness;
for particle i 1,2, 3.., N, the velocity calculation formula is:
v i (t+1)＝v i (t)+c 1 *rand()*(pbest i (t)-x i (t))+c 2 *rand()*(gbest i (t)-x i (t))
wherein, c 1 And c 2 As a learning factor, pbest i For optimal position obtained by a single search, gbest i The optimal position obtained for the global search. And rand () represents a number between randomly selected intervals (0, 1).
The position calculation formula is:
x i (t+1)＝x i (t)+v i (t)
after 20 iterations, the optimal parameters are C-7.27886351 and epsilon-0.29907942 gamma-0.13297306.
S3-4: bringing the optimal parameters into a test set SVR model for prediction;
the SVR model is implemented by the following formula:
wherein, alpha * Is a Lagrange multiplier, k (x) i ,x j ) Represents the kernel function, b is the displacement.
S3-5: using R 2 And Mean Square Error (MSE) to test the model performance. SVR model test set R corresponding to optimal parameters 2 At 0.79849, the test set MSE was 0.20061.
S4: and outputting an Internet of things operation index prediction result through the SVR model after the parameters are optimized.
In order to verify the prediction effect of the RF-PSO-SVR model provided by the invention, the RF-PSO-SVR model, the SVR model before feature selection and the RF-SVR model after feature selection are compared with the same data in terms of performance, and specific comparison data are shown in Table 2 and FIG. 3.
TABLE 2 comparison of different model Performance
As can be seen from Table 2, the SVR model test set R after feature selection is compared with the SVR model test set R before feature selection 2 The increase was about 4.2% and the MSE in the test set was about 4.2% lower. Illustrating the explainability enhancement of the weather factor characteristics to the success rate of daily collection. Interference features are effectively eliminated after feature selection, and prediction accuracy is improved while models are simplified. The prediction accuracy of the SVR model after the feature selection optimized by the PSO algorithm is further improved, and a test set R 2 The increase is about 9.9%, and the MSE in the test set is about 9.9%.
FIG. 3 further illustrates the predicted effects of the SVR, RF-PSO-SVR models by plotting a scatter plot with true values as abscissa and predicted values as ordinate. FIG. 3(a) shows the sameThe regression model data points are most dispersed, the data point distribution converges after the feature selection in fig. 3(b), and the model data points optimized by PSO in fig. 3(c) are more concentrated near y-x, so that the fitting effect is better. The three graphs (a), (b) and (c) of FIG. 3 verify the test set R under different models 2 The change in MSE in the test set. The superiority of the RF-PSO-SVR algorithm in the aspect of improving the model performance is further proved.
Claims (9)
1. An Internet of things operation index prediction method based on an improved SVR model is characterized by comprising the following steps:
s1: preprocessing the climate data and the Internet of things operation index data;
s2: selecting characteristics of the preprocessed climate data;
s3: establishing an RF-SVR model according to the preprocessed Internet of things operation index data and the climate data after feature selection, and optimizing parameters of the RF-SVR model by using an intelligent optimization algorithm PSO to obtain the RF-PSO-SVR model;
s4: and outputting an Internet of things operation index prediction result through an RF-PSO-SVR model.
2. The method for predicting the operation index of the internet of things based on the improved SVR model as claimed in claim 1, wherein said preprocessing method of step S1 is:
a1: randomly dividing a climate and thing-internet operation index data set into a training set and a testing set according to a proportion;
a2: and respectively carrying out standardization processing on the training set and the test set so as to eliminate the influence of dimension.
3. The method for predicting the operation index of the internet of things based on the improved SVR model as claimed in claim 2, wherein said step S2 specifically comprises:
b1: selecting training set data for feature selection;
b2: traversing the set number of RF trees using a trellis search;
b3: determining the number of RF trees by using K-fold cross validation with the minimum validation error as a criterion;
b4: sorting the feature importance by using the RF after optimizing the parameters;
b5: and for the features with the sorted importance degrees, carrying out dimension selection by using a sequence forward selection algorithm based on goodness-of-fit.
4. The method for predicting the IoT operation index based on the improved SVR model as claimed in claim 3, wherein said step B3 is specifically:
dividing a data set into K groups by K-fold cross validation, circulating for K times, selecting a group of data which is not repeated with the previous data as a validation set in each circulation, and using the rest K-1 groups as a training set; after the cycle is finished, K parameter values and K verification errors corresponding to the K parameter values can be obtained, the average error of the K errors is calculated and used as a new verification error, and finally, the parameter value corresponding to the minimum verification error is selected, namely the number of the RF trees is introduced into the RF feature selection model.
5. The method for predicting the operation index of the internet of things based on the improved SVR model as claimed in claim 3, wherein said step B5 of using the sequence forward selection algorithm based on goodness of fit for dimension selection comprises:
and adding the features into the SVR model one by one from the empty set according to the importance, and selecting the feature dimension which enables the fitting goodness of the SVR model to be highest.
6. The method for predicting the performance index of the internet of things based on the improved SVR model as claimed in claim 1, wherein the step S3 of optimizing the parameters of the climate and performance index prediction model by using the PSO is performed by:
c1: setting parameter information of a particle swarm algorithm;
c2: initializing penalty parameters (C), loss functions (epsilon) and kernel coefficients (gamma) of the SVR model, initializing the position and the speed of particles in the particle swarm optimization, and then calculating the goodness of fit (R) of the SVR model in the training set 2 ) As fitness;
c3: continuously iterating, updating the position and the speed of the particles, calculating corresponding fitness, and recording C, epsilon and gamma corresponding to the global optimal fitness;
c4: and (4) bringing the optimal parameters C, epsilon and gamma obtained in the step C3 into the test set SVR model for prediction.
7. The method as claimed in claim 6, wherein R is used in step S3 2 And the two indexes of the mean square error are used for testing the performance of the SVR model.
8. The method of claim 6, wherein the positions and velocities of the particles in steps C2 and C3 are determined by the following equations:
for particle i ═ 1,2, 3.., N, the velocity calculation formula is:
v i (t+1)＝v i (t)+c 1 *rand()*(pbest i (t)-x i (t))+c 2 *rand()*(gbest i (t)-x i (t))
wherein, c 1 And c 2 As a learning factor, pbest i For optimal position obtained by a single search, gbest i For the optimal position obtained by global search, rand () represents the number between randomly selected intervals (0, 1);
the position calculation formula is:
x i (t+1)＝x i (t)+v i (t)。
9. the method as claimed in claim 6, wherein the SVR model in step C4 is implemented by the following formula:
wherein, alpha,α * Is a Lagrange multiplier, k (x) i ,x j ) Representing the kernel function, b is the displacement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210584561.5A CN114970719B (en) | 2022-05-27 | 2022-05-27 | Internet of things operation index prediction method based on improved SVR model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210584561.5A CN114970719B (en) | 2022-05-27 | 2022-05-27 | Internet of things operation index prediction method based on improved SVR model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114970719A true CN114970719A (en) | 2022-08-30 |
| CN114970719B CN114970719B (en) | 2024-04-26 |
Family
ID=82956187
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210584561.5A Active CN114970719B (en) | 2022-05-27 | 2022-05-27 | Internet of things operation index prediction method based on improved SVR model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114970719B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102419584A (en) * | 2011-11-30 | 2012-04-18 | 锦州华冠环境科技实业公司 | Method of estimating and evaluating emission law of pollution source by using internet of things and internet-of-things controller |
| CN109190979A (en) * | 2018-09-03 | 2019-01-11 | 深圳市智物联网络有限公司 | A kind of industry internet of things data analysis method, system and relevant device |
| CN113887571A (en) * | 2021-09-10 | 2022-01-04 | 上海工业自动化仪表研究院有限公司 | Electronic equipment fault prediction method for improving SVR algorithm |
-
2022
- 2022-05-27 CN CN202210584561.5A patent/CN114970719B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102419584A (en) * | 2011-11-30 | 2012-04-18 | 锦州华冠环境科技实业公司 | Method of estimating and evaluating emission law of pollution source by using internet of things and internet-of-things controller |
| CN109190979A (en) * | 2018-09-03 | 2019-01-11 | 深圳市智物联网络有限公司 | A kind of industry internet of things data analysis method, system and relevant device |
| CN113887571A (en) * | 2021-09-10 | 2022-01-04 | 上海工业自动化仪表研究院有限公司 | Electronic equipment fault prediction method for improving SVR algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114970719B (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111738512B (en) | Short-term power load prediction method based on CNN-IPSO-GRU hybrid model | |
| CN106920006B (en) | Subway station air conditioning system energy consumption prediction method based on ISOA-LSSVM | |
| CN111199016A (en) | DTW-based improved K-means daily load curve clustering method | |
| CN110263998B (en) | Double-layer correction method for multisource numerical weather forecast set | |
| CN109492748B (en) | Method for establishing medium-and-long-term load prediction model of power system based on convolutional neural network | |
| CN110598854A (en) | GRU model-based transformer area line loss rate prediction method | |
| CN116937579B (en) | Wind power interval prediction considering space-time correlation and interpretable method thereof | |
| CN111832839B (en) | Energy consumption prediction method based on sufficient incremental learning | |
| CN113762387B (en) | Multi-element load prediction method for data center station based on hybrid model prediction | |
| CN112884236B (en) | Short-term load prediction method and system based on VDM decomposition and LSTM improvement | |
| CN112288157A (en) | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning | |
| CN114021483A (en) | Ultra-short-term wind power prediction method based on time domain characteristics and XGboost | |
| CN116526473A (en) | Particle swarm optimization LSTM-based electrothermal load prediction method | |
| CN115995810A (en) | Wind power prediction method and system considering weather fluctuation self-adaptive matching | |
| CN115115090A (en) | Wind power short-term prediction method based on improved LSTM-CNN | |
| CN114119273A (en) | Park comprehensive energy system non-invasive load decomposition method and system | |
| CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
| CN115374710A (en) | Wind speed prediction method based on improved gull optimization algorithm and multi-core extreme learning machine | |
| CN116757057A (en) | Air quality prediction method based on PSO-GA-LSTM model | |
| CN113627674A (en) | Distributed photovoltaic power station output prediction method and device and storage medium | |
| CN116826745B (en) | Layered and partitioned short-term load prediction method and system in power system background | |
| CN114970719A (en) | Internet of things operation index prediction method based on improved SVR model | |
| CN116307250A (en) | Short-term load prediction method and system based on typical daily feature selection | |
| CN113449466B (en) | Solar radiation prediction method and system for optimizing RELM based on PCA and chaos GWO | |
| CN115796327A (en) | Wind power interval prediction method based on VMD (vertical vector decomposition) and IWOA-F-GRU (empirical mode decomposition) -based models |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |