CN110942182A - Method for establishing typhoon prediction model based on support vector regression - Google Patents
Method for establishing typhoon prediction model based on support vector regression Download PDFInfo
- Publication number
- CN110942182A CN110942182A CN201911109552.5A CN201911109552A CN110942182A CN 110942182 A CN110942182 A CN 110942182A CN 201911109552 A CN201911109552 A CN 201911109552A CN 110942182 A CN110942182 A CN 110942182A
- Authority
- CN
- China
- Prior art keywords
- typhoon
- model
- support vector
- data
- vector regression
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 31
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000010219 correlation analysis Methods 0.000 claims abstract description 6
- 238000010606 normalization Methods 0.000 claims abstract description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000000611 regression analysis Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 abstract description 9
- 238000010801 machine learning Methods 0.000 abstract description 5
- 238000012795 verification Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 27
- 238000011156 evaluation Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000012443 analytical study Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
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
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Abstract
The invention relates to a method for establishing a typhoon prediction model based on support vector regression, in particular to a typhoon prediction method based on support vector regression, which comprises the following steps: step S1: acquiring typhoon data, namely training samples, of the years, and preprocessing the training samples; step S2: performing correlation analysis and standard normalization processing on the preprocessed typhoon data; step S3: establishing a support vector regression model; step S4: and inputting the typhoon parameters into the support vector regression model to predict the typhoon maximum wind speed. The method utilizes a large amount of previous typhoon data to train the support vector regression model by a machine learning method, and the verification of the test set shows that the model has good precision and can be well used for predicting the typhoon maximum wind speed.
Description
Technical Field
The invention relates to the field of typhoon prediction, in particular to a method for establishing a typhoon prediction model based on support vector regression.
Background
As the economic loss caused by typhoons is increasing, more scholars are beginning to study typhoons. At present, typhoon is generally researched by a meteorological typhoon model for meteorological forecasting and an engineering wind field model considering engineering practical application. The meteorological typhoon model mainly simulates some basic characteristics of typhoon, such as atmospheric evolution process, temperature field, flow field, energy balance, typhoon eye, typhoon rainfall and the like. The meteorological typhoon model has a complex structure, is simulated by adopting a Monte Carlo method, and is not suitable for engineering application. In order to construct a typhoon model suitable for engineering application, a large amount of research is conducted by scholars at home and abroad.
Russel proposed to combine a hurricane generation model with a hurricane wind field distribution model in 1971, and found that the generation of hurricanes followed periodic poisson's law. Since then, people began conducting analytical studies on typhoons using numerical simulations. The Batts wind field model was proposed by l.r.russel and m.e.batts in 1980. The Batts wind field model is a numerical model based on a typhoon gradient equilibrium equation. This is followed by the Shapiro typhoon wind field model, the Georgiou model and now more commonly improved CE wind field numerical models. The YanMeng wind field model is a model proposed by YanMeng et al in 1995 that takes into account boundary layer friction correction. YanMeng typhoon models are subjected to parameter analysis by Li Tao, Racek and the like at the university of mansion, a value taking method is provided, and then numerical simulation is carried out on typhoon DAN through the models. The YanMeng wind field model is used as a typhoon model by the zhang of the power university in north China, model parameters are determined by inverting the wilson typhoon, and the power response of the power transmission line under the action of the typhoon is researched. With the development of neural networks, people gradually use the neural networks to predict typhoons. A new typhoon maximum wind speed prediction model is established by combining Li Hongli and Wangxin with a propagation clustering method and a sparse Bayesian regression model. The typhoon is influenced by a plurality of factors, so the accuracy of the traditional numerical simulation method is very high. With the development of machine learning and the improvement of computer computing power, machine learning is also gradually applied to numerical simulation of typhoon.
Disclosure of Invention
In view of this, the present invention provides a method for building a typhoon prediction model based on support vector regression, which can well predict typhoons.
The invention is realized by adopting the following scheme: a method for building a typhoon prediction model based on support vector regression comprises the following steps:
step S1: acquiring typhoon data, namely training samples, of the years, and preprocessing the training samples;
step S2: performing correlation analysis and standard normalization processing on the preprocessed typhoon data;
step S3: establishing a support vector regression model, namely a typhoon prediction model;
step S4: and inputting parameters including air temperature, typhoon central air pressure, two-minute average wind speed, two-minute average wind direction and typhoon central dimension data when typhoon arrives into the support vector regression model for predicting typhoon maximum wind speed.
Further, the specific content of the preprocessing of the training samples in step S1 is as follows: coding the non-digital parameters in the training sample by using a LabeleEncoder function in a skleern library of Python, and converting the non-digital parameter codes into digital parameters; if the number of null values existing in a certain data set of the training sample is less than 10% of that of the data set, averaging the data before and after the position of the null value, and taking the average value as the null value; and if the number of null values existing in the training sample is more than 10% of the data set, rejecting the data set where the group of null values are located.
Further, the specific content of step S2 is:
the correlation coefficient is represented by r; taking a parameter with the absolute value of r greater than 0.2 of the maximum wind speed to perform regression analysis;
the normalized formula is expressed as:
where s is the standard deviation of the samples and μ is the mean of the samples.
Further, the specific content of step S3 is:
for the collected historical typhoon data sample set D { (x)1,y1),(x2,y2),…,(xm,ym)}, yiE.g. R, establishing a regression model such as the formula (3) to make f (x) and y as close as possible;
f(x)=ωTx+b (3)
wherein ω ═ ω (ω)1;ω2;...;ωd) And b is a model parameter.
Compared with the prior art, the invention has the following beneficial effects:
the method trains the support vector regression model through a large amount of data, and the verification of the test set shows that the model has good precision and can be well used for predicting typhoon maximum wind speed.
Drawings
FIG. 1 is a diagram illustrating a distribution of real values and output values of four kernel function models according to an embodiment of the present invention.
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiment provides a method for establishing a typhoon prediction model based on support vector regression, which comprises the following steps:
step S1: acquiring typhoon data, namely training samples, of the years, and preprocessing the training samples;
step S2: performing correlation analysis and standard normalization processing on the preprocessed typhoon data;
step S3: establishing a support vector regression model, namely a typhoon prediction model;
step S4: and inputting parameters including air temperature, typhoon central air pressure, two-minute average wind speed, two-minute average wind direction and typhoon central dimension data when typhoon arrives into the support vector regression model for predicting typhoon maximum wind speed.
In this embodiment, the specific content of the preprocessing of the training samples in step S1 is as follows: coding the non-digital parameters in the training sample by using a LabeleEncoder function in a skleern library of Python, and converting the non-digital parameter codes into digital parameters; if the number of null values existing in a certain data set of the training sample is less than 10% of that of the data set, averaging the data before and after the position of the null value, and taking the average value as the null value; and if the number of null values existing in the training sample is more than 10% of the data set, rejecting the data set where the group of null values are located.
In this embodiment, the specific content of step S2 is:
the correlation coefficient is represented by r; taking a parameter with the absolute value of r greater than 0.2 of the maximum wind speed to perform regression analysis; the parameters are shown in table 1;
TABLE 1 correlation coefficient table of different parameters and maximum wind speed
Parameter(s) | Two minute average wind speed | Two minute average wind direction | Maximum wind speed | Maximum wind direction | Maximum wind direction | Air temperature |
Correlation coefficient | 0.891889 | -0.30062 | 0.937323 | -0.30345 | -0.2812 | 0.360267 |
Parameter(s) | Maximum air temperature | Lowest air temperature | Typhoon center longitude | Central air pressure | Maximum wind speed | Maximum wind speed |
Correlation coefficient | 0.349691 | 0.378095 | 0.249058 | 0.256066 | -0.26054 | 1 |
The normalized formula is expressed as:
where s is the standard deviation of the samples and μ is the mean of the samples.
In this embodiment, the specific content of step S3 is:
for a given sample set D { (x)1,y1),(x2,y2),…,(xm,ym)},ytE { -1, +1}, divide a hyperplane in sample space for sample set D, separate different categories; in the sample space, the hyperplane is divided as described by the equation of equation (3):
ωTx+b=0 (3)
wherein ω ═ ω (ω)1;ω2;...;ωd) Determining the direction of the hyperplane for the normal vector; b is a displacement term, and determines the distance between the hyperplane and the origin; therefore, a hyperplane can be determined by ω and b; the classification model of the support vector machine is described by the model of equation (4):
f(x)=ωTx+b (4)
by usingRepresents a vector that maps x into a high-dimensional space, so the partition hyperplane model in space is represented by equation (5):
where ω and b are model parameters; due to the presence of the kernel function κ (x) in the low-dimensional spacei,xj) Of a value equal to that of the high dimensional spaceSo that a large number of computations are simplified by using the kernel function; the kernel function adopts a Gaussian kernel function;
as to the problem of the classification there are,
for the regression problem, it is for a given sample D { (x)1,y1),(x2,y2),…,(xm,ym)}, yiE.g., R, it is desirable to train a model form such as (4) so that the difference between f (x) and y is as small as possible. For conventional regression models, the loss is usually calculated by calculating the difference between the output value of the model and the true value, and is zero when and only when the difference is zero. The support vector regression has a tolerance epsilon, and when the absolute value of the difference between the model output value and the true value is less than the tolerance epsilon, the loss is considered to be zero; the loss is only calculated when the absolute value of the difference between the model output value and the true value is greater than the tolerance epsilon, i.e. the loss is calculatedThe support vector regression takes f (x) as a center to establish an interval with the width of 2 epsilon; if the training sample is in the interval, the prediction is considered to be correct; the problem of support vector regression translates to equation (7):
where C is a regularization constant, leIs an epsilon-insensitive loss function;
in this embodiment, the basic task of typhoon prediction is to relate a variable x at a certain moment in time to the typhoon maximum wind speed1,x2,...,xmAs an input, the maximum wind speed y of the typhoon is predicted, and a model f (x) is established, so that f (x) approaches y. When the model is built and the testing precision meets the requirements, the typhoon maximum wind speed can be predicted through typhoon related factors.
In this example, the data was derived from 27273 data collected by 2085 automated weather stations in Fujian province during the 18 th month 18 th Molandi typhoon 2016 and the Molandi typhoon path data provided by the Chinese typhoon net. The training samples are processed before the support vector regression model is established, and the effect of machine learning has a great relationship with the processing of sample data. If the parameters in the training samples are not of the numeric type, the parameters are encoded. For example, typhoon moves to, the data provided by the Chinese typhoon network is in the Chinese direction, and needs to be coded. In this embodiment, a LabelEncoder function in a sklern library of Python is used to perform encoding processing, so as to encode a non-digital parameter into a digital parameter. Of the 27273 sample data, some have null values (null) for various reasons and must be processed. This embodiment performs the following processing on the null value: if the data collected in an automatic monitoring station only has a few null values, for example, the rainfall at a certain moment is missing, the values of the rainfall before and after the moment are averaged, and the average value is taken as the value at the moment to replace the null value. And if a large amount of data of one automatic monitoring station is lost, the data of the automatic monitoring station is rejected.
In this embodiment, the selection of the kernel function is specifically as follows:
training samples if there is not a hyperplane in the original sample space that can be divided into different classes, the samples need to be mapped from the original space to a higher-dimensional feature space, so that the samples are linearly separable in the high-dimensional feature space. By usingRepresenting a vector that maps x into a high-dimensional space, the partition hyperplane model in space can be expressed as:
where ω and b are model parameters. Due to the presence of the kernel function κ (x) in the low-dimensional spacei,xj) Of a value equal to that of the high dimensional spaceSo that a large number of computations can be simplified using the kernel function. The selection of the kernel function has a great influence on the accuracy of the support vector regression prediction, and four kernel functions are selected for modeling comparison in the embodiment, as shown in table 2.
TABLE 2 Kernel function
The processed samples were divided into a training set and a test set, wherein the training set accounts for 70% of the total samples. Four kernel functions were selected and the model was trained with training samples. Inputting data of a test group into the trained model, calculating the difference between the output value and the true value of the model, and randomly acquiring a group of data to obtain a scatter diagram of the fitting curve of the output value and the true value of each kernel function model shown in fig. 1. The black dispersion point is the true value, and is not shown because the Sigmoid kernel function has a large error. The four kernel function model evaluation scores and errors are shown in table 3:
TABLE 3 evaluation scores and errors of four kernel function models
The evaluation is a system default evaluation rule to evaluate the model, the closer to 1, the better the model. The mean square error represents the degree of dispersion between the output value and the true value of the model, and the smaller the mean square error, the better the model. The R2 decision coefficient represents the goodness of fit of the model, with R2 being the better the closer to the model. It can be seen that the gaussian kernel function is the optimal model.
Without correlation analysis, the obtained parameters were all used to train the regression model, and the results as in table 4 were obtained.
Table 4 all parameters are involved in training, and the evaluation scores and errors of four kernel function models
It can be seen that except for the sigmoid kernel function, the model with parameter selection is scored better than the model without parameter selection, and the training time of the model is shorter than that of the model without parameter selection, so that parameter correlation analysis is generally carried out before model pre-training. Meanwhile, the optimal kernel function is also a Gaussian kernel function, and a support vector regression model of the Gaussian kernel function has two parameters, namely a kernel coefficient gamma of the Gaussian kernel function, and the other parameter is a penalty parameter C. And solving the two parameters through a greedy algorithm to obtain an optimal solution model. The solved kernel coefficient gamma is 0.05, and the penalty parameter C is 40. And (3) scoring of a training set: 0.9428 test set score: 0.9267 test set mean square error: 0.0731, test set R2 points: 0.9268.
preferably, the typhoon great wind speed of the embodiment is influenced by a plurality of factors, and the simple typhoon model is difficult to accurately perform numerical simulation. According to the method, the support vector regression model is trained through a large amount of data and a machine learning method, and the verification of the test set shows that the model has good precision and can be well used for predicting the typhoon maximum wind speed.
The above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.
Claims (4)
1. A method for building a typhoon prediction model based on support vector regression is characterized in that: the method comprises the following steps:
step S1: acquiring typhoon data, namely training samples, of the years, and preprocessing the training samples;
step S2: performing correlation analysis and standard normalization processing on the preprocessed typhoon data;
step S3: establishing a support vector regression model, namely a typhoon prediction model;
step S4: and inputting parameters including air temperature, typhoon central air pressure, two-minute average wind speed, two-minute average wind direction and typhoon central dimension data when typhoon arrives into the support vector regression model for predicting typhoon maximum wind speed.
2. The method of claim 1, wherein the method comprises the following steps: the specific content of the training sample preprocessing in step S1 is as follows: coding the non-digital parameters in the training sample by using a LabeleEncoder function in a skleern library of Python, and converting the non-digital parameter codes into digital parameters; if the number of null values existing in a certain data set of the training sample is less than 10% of that of the data set, averaging the data before and after the position of the null value, and taking the average value as the null value; and if the number of null values existing in the training sample is more than 10% of the data set, rejecting the data set where the group of null values are located.
3. The method of claim 1, wherein the method comprises the following steps: the specific content of step S2 is:
the correlation coefficient is represented by r; taking a parameter with the absolute value of r greater than 0.2 of the maximum wind speed to perform regression analysis;
the normalized formula is expressed as:
where s is the standard deviation of the samples and μ is the mean of the samples.
4. The method of claim 1, wherein the method comprises the following steps: the specific content of step S3 is:
for the collected historical typhoon data sample set D { (x)1,y1),(x2,y2),…,(xm,ym)},yiE.g. R, establishing a regression model such as the formula (3) to make f (x) and y as close as possible;
f(x)=ωTx+b (3)
wherein ω ═ ω (ω)1;ω2;...;ωd) And b is a model parameter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911109552.5A CN110942182A (en) | 2019-11-14 | 2019-11-14 | Method for establishing typhoon prediction model based on support vector regression |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911109552.5A CN110942182A (en) | 2019-11-14 | 2019-11-14 | Method for establishing typhoon prediction model based on support vector regression |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110942182A true CN110942182A (en) | 2020-03-31 |
Family
ID=69907508
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911109552.5A Pending CN110942182A (en) | 2019-11-14 | 2019-11-14 | Method for establishing typhoon prediction model based on support vector regression |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110942182A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111523087A (en) * | 2020-04-10 | 2020-08-11 | 北京航空航天大学 | Typhoon intensity long-term change trend analysis method |
CN111967918A (en) * | 2020-09-01 | 2020-11-20 | 上海积成能源科技有限公司 | System model for predicting electricity price based on support vector regression algorithm |
CN112686254A (en) * | 2020-12-31 | 2021-04-20 | 山西三友和智慧信息技术股份有限公司 | Typhoon center positioning method based on infrared satellite cloud picture |
CN112748255A (en) * | 2020-12-29 | 2021-05-04 | 国网河北省电力有限公司保定供电分公司 | Typhoon model-based transmission tower typhoon wind speed and wind direction calculation method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100830518B1 (en) * | 2006-12-19 | 2008-05-21 | 영남대학교 산학협력단 | Wind speed estimating method of wind generation system using svr algorithm |
CN106529706A (en) * | 2016-10-25 | 2017-03-22 | 国家电网公司 | Support-vector-machine-regression-based method for predicting wind speed of wind power plant |
CN109102101A (en) * | 2017-06-21 | 2018-12-28 | 北京金风科创风电设备有限公司 | Wind speed prediction method and system for wind power plant |
CN110210641A (en) * | 2018-02-28 | 2019-09-06 | 北京金风科创风电设备有限公司 | Wind direction prediction method and device for wind power plant |
-
2019
- 2019-11-14 CN CN201911109552.5A patent/CN110942182A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100830518B1 (en) * | 2006-12-19 | 2008-05-21 | 영남대학교 산학협력단 | Wind speed estimating method of wind generation system using svr algorithm |
CN106529706A (en) * | 2016-10-25 | 2017-03-22 | 国家电网公司 | Support-vector-machine-regression-based method for predicting wind speed of wind power plant |
CN109102101A (en) * | 2017-06-21 | 2018-12-28 | 北京金风科创风电设备有限公司 | Wind speed prediction method and system for wind power plant |
CN110210641A (en) * | 2018-02-28 | 2019-09-06 | 北京金风科创风电设备有限公司 | Wind direction prediction method and device for wind power plant |
Non-Patent Citations (2)
Title |
---|
RACHEL_NANA: "支持向量机SVM原理及推导", 《HTTPS://BLOG.CSDN.NET/ABCDRACHEL/ARTICLE/DETAILS/95842833》 * |
林涛等: "基于支持向量回归的风电场短期功率预测", 《中南民族大学学报(自然科学版)》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111523087A (en) * | 2020-04-10 | 2020-08-11 | 北京航空航天大学 | Typhoon intensity long-term change trend analysis method |
CN111967918A (en) * | 2020-09-01 | 2020-11-20 | 上海积成能源科技有限公司 | System model for predicting electricity price based on support vector regression algorithm |
CN112748255A (en) * | 2020-12-29 | 2021-05-04 | 国网河北省电力有限公司保定供电分公司 | Typhoon model-based transmission tower typhoon wind speed and wind direction calculation method |
CN112686254A (en) * | 2020-12-31 | 2021-04-20 | 山西三友和智慧信息技术股份有限公司 | Typhoon center positioning method based on infrared satellite cloud picture |
CN112686254B (en) * | 2020-12-31 | 2022-08-09 | 山西三友和智慧信息技术股份有限公司 | Typhoon center positioning method based on infrared satellite cloud picture |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110942182A (en) | Method for establishing typhoon prediction model based on support vector regression | |
CN108009674A (en) | Air PM2.5 concentration prediction methods based on CNN and LSTM fused neural networks | |
CN108197744B (en) | Method and system for determining photovoltaic power generation power | |
CN110321361B (en) | Test question recommendation and judgment method based on improved LSTM neural network model | |
CN109657881A (en) | A kind of neural network photovoltaic power generation prediction technique and system suitable for small sample | |
CN111563706A (en) | Multivariable logistics freight volume prediction method based on LSTM network | |
CN111126704B (en) | Multi-region precipitation prediction model construction method based on multi-graph convolution and memory network | |
CN110942194A (en) | Wind power prediction error interval evaluation method based on TCN | |
CN109635245A (en) | A kind of robust width learning system | |
CN106022954A (en) | Multiple BP neural network load prediction method based on grey correlation degree | |
CN106778838A (en) | A kind of method for predicting air quality | |
CN113344288B (en) | Cascade hydropower station group water level prediction method and device and computer readable storage medium | |
CN107798426A (en) | Wind power interval Forecasting Methodology based on Atomic Decomposition and interactive fuzzy satisfying method | |
CN112232561A (en) | Power load probability prediction method based on constrained parallel LSTM quantile regression | |
CN117035201B (en) | Multi-target scheduling rule making method and system for plain river network water engineering cluster | |
CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
CN106570592A (en) | Artificial-neural-network-based intelligent numerical value forecasting correction system | |
CN113516304A (en) | Space-time joint prediction method and device for regional pollutants based on space-time graph network | |
CN113393057A (en) | Wheat yield integrated prediction method based on deep fusion machine learning model | |
CN110260914B (en) | Engineering safety monitoring system region division method based on measuring point space-time characteristics | |
CN110648023A (en) | Method for establishing data prediction model based on quadratic exponential smoothing improved GM (1,1) | |
CN112307536A (en) | Dam seepage parameter inversion method | |
CN111242266A (en) | Operation data management system | |
CN113673777B (en) | Desert succession prediction method under climate change condition | |
CN115877483A (en) | Typhoon path forecasting method based on random forest and GRU |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200331 |
|
RJ01 | Rejection of invention patent application after publication |