CN117390592B - Method and system for constructing characteristic landscape forecast model - Google Patents
Method and system for constructing characteristic landscape forecast model Download PDFInfo
- Publication number
- CN117390592B CN117390592B CN202311686847.5A CN202311686847A CN117390592B CN 117390592 B CN117390592 B CN 117390592B CN 202311686847 A CN202311686847 A CN 202311686847A CN 117390592 B CN117390592 B CN 117390592B
- Authority
- CN
- China
- Prior art keywords
- landscape
- preset
- meteorological
- factors
- contained
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 230000000737 periodic effect Effects 0.000 claims abstract description 39
- 238000004458 analytical method Methods 0.000 claims abstract description 32
- 230000008859 change Effects 0.000 claims abstract description 29
- 238000013528 artificial neural network Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims description 46
- 238000012544 monitoring process Methods 0.000 claims description 44
- 238000006243 chemical reaction Methods 0.000 claims description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 16
- 238000012795 verification Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 7
- 238000010586 diagram Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 3
- 238000003491 array Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2131—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on a transform domain processing, e.g. wavelet transform
-
- 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
- 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
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Environmental & Geological Engineering (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for constructing a characteristic landscape forecast model, wherein the method comprises the following steps: acquiring meteorological data generated in a plurality of continuous years, and detecting a plurality of original independent variables contained in the meteorological data through a preset regression model; extracting a plurality of target independent variables from a plurality of original independent variables based on a preset rule, and setting the plurality of target independent variables as corresponding meteorological factors; detecting the change period of the meteorological factors in real time by a preset wavelet analysis method, and extracting the periodic factors contained in the change period; and inputting a plurality of meteorological factors and periodic factors into a preset neural network to train a corresponding landscape forecasting model, and forecasting the occurrence date of each landscape through the landscape forecasting model. The invention can accurately predict the occurrence date of each landscape and improves the user experience.
Description
Technical Field
The invention relates to the technical field of weather, in particular to a method and a system for constructing a characteristic landscape forecast model.
Background
With the development of the age and the progress of technology, a weather observation site is set in a plurality of areas in China, and weather data of each area are collected in real time through the weather observation site, so that future weather can be predicted in real time according to the collected weather data.
The special landscapes such as rime and cloud sea can attract a large number of tourists, so that how to accurately predict the occurrence date of the special landscapes such as rime and cloud sea has higher economic benefit.
However, in the process of predicting the characteristic landscape in the prior art, most of considered meteorological factors are single, so that a predicted result is single-sided, a certain error is generated, and the characteristic landscape is unchanged for people.
Therefore, in order to overcome the shortcomings of the prior art, it is necessary to provide a landscape forecast model capable of accurately and comprehensively predicting the occurrence date of a characteristic landscape.
Disclosure of Invention
Based on the above, the invention aims to provide a method and a system for constructing a characteristic landscape forecast model, so as to solve the problem that a predicted result in the prior art is relatively one-sided, and a certain error is generated.
The first aspect of the embodiment of the invention provides:
a method for constructing a distinctive landscape forecast model, wherein the method comprises the following steps:
acquiring meteorological data generated within a plurality of continuous years, and detecting a plurality of original independent variables contained in the meteorological data through a preset regression model;
extracting a plurality of target independent variables from a plurality of original independent variables based on a preset rule, and setting the plurality of target independent variables as corresponding meteorological factors;
detecting the change period of the meteorological factors in real time by a preset wavelet analysis method, and extracting the periodic factors contained in the change period;
and inputting a plurality of meteorological factors and the periodic factors into a preset neural network to train a corresponding landscape forecasting model, and forecasting the occurrence date of each landscape through the landscape forecasting model.
The beneficial effects of the invention are as follows: by collecting weather data in real time, further detecting original independent variables capable of affecting weather, further determining corresponding weather factors according to the current original independent variables, synchronously extracting corresponding periodic factors according to the change period of the current weather factors, and based on the corresponding periodic factors, only further inputting the acquired weather factors and the periodic factors into a preset neural network at the same time, a required landscape forecast model can be finally trained, based on the model, the occurrence date of each landscape can be simply and rapidly predicted, and the use experience of a user is correspondingly improved.
Further, the step of detecting a plurality of original independent variables contained in the meteorological data through a preset regression model includes:
when the meteorological data are acquired, detecting a plurality of meteorological elements contained in the meteorological data in real time, and correspondingly acquiring a plurality of meteorological values corresponding to the meteorological elements respectively;
generating a corresponding weather monitoring table according to a plurality of weather elements and a plurality of weather values, and generating a corresponding weather monitoring diagram according to the weather monitoring table, wherein the weather elements are horizontal coordinates, and the weather values are vertical coordinates;
inputting the weather monitoring graph into the preset regression model in real time, so that the preset regression model correspondingly outputs a plurality of original independent variables, and the weather monitoring graph has uniqueness.
Further, the step of inputting the weather monitoring graph into the preset regression model in real time so that the preset regression model correspondingly outputs a plurality of original independent variables includes:
when the weather monitoring graph is obtained, extracting a plurality of change curves contained in the weather monitoring graph one by one, and preprocessing each change curve to generate a plurality of corresponding target curves;
analyzing each target curve through the preset regression model to correspondingly detect variable elements contained in each target curve;
and setting a plurality of variable elements as a plurality of original independent variables respectively, wherein each variable element has uniqueness.
Further, the expression of the preset regression model is:
wherein alpha is j Represents the j-th meteorological element, alpha 0 Representing regression coefficient, x i，j Representing the weather value of coordinates (i, j), Y i Represents the i-th variable element, i=1, …, n.
Further, the step of inputting the meteorological factors and the periodic factors into a preset neural network to train a corresponding landscape forecasting model includes:
when the meteorological factors and the periodic factors are respectively acquired, carrying out mixed processing on the meteorological factors and the periodic factors based on preset weights so as to generate corresponding target data sets, and splitting the target data sets into corresponding training sets and verification sets according to preset proportions;
and detecting a conversion layer, an analysis layer and a learning layer which are respectively contained in the preset neural network, and correspondingly training the landscape forecast model through the conversion layer, the analysis layer and the learning layer based on the training set and the verification set.
Further, the training the landscape forecast model based on the training set and the verification set through the conversion layer, the analysis layer and the learning layer correspondingly includes:
when the training set is acquired, carrying out normalization processing on the training set through a first preset algorithm in the conversion layer so as to output a plurality of corresponding characteristic values;
analyzing the characteristic values through a second preset algorithm in the analysis layer to analyze the characteristic values into corresponding characteristic sequences;
inputting a plurality of feature sequences into the learning layer to correspondingly train the landscape forecast model, wherein the expression of the first preset algorithm is as follows:
wherein Y represents the characteristic value, X represents the meteorological factor, min i Sum max i Representing the minimum value and the maximum value of the periodic factors respectively, wherein each characteristic sequence has uniqueness.
Further, the step of inputting the feature sequences into the learning layer to train the landscape forecast model correspondingly includes:
when a plurality of feature sequences are obtained, feature factors contained in each feature sequence are detected one by one, and learning networks contained in the learning layer are extracted;
detecting a plurality of network nodes contained in the learning network, and adding corresponding identifiers to each network node according to the connection sequence among the network nodes;
extracting initial network parameters respectively contained in each network node, and respectively and correspondingly replacing each initial network parameter with each characteristic factor according to the identification.
A second aspect of an embodiment of the present invention proposes:
a featured landscape forecast model construction system, wherein the system comprises:
the detection module is used for acquiring meteorological data generated in a plurality of continuous years and detecting a plurality of original independent variables contained in the meteorological data through a preset regression model;
the extraction module is used for extracting a plurality of target independent variables from a plurality of original independent variables based on a preset rule and setting the plurality of target independent variables as corresponding meteorological factors;
the analysis module is used for detecting the change period of the meteorological factors in real time through a preset wavelet analysis method and extracting the periodic factors contained in the change period;
the training module is used for inputting a plurality of meteorological factors and periodic factors into a preset neural network to train a corresponding landscape forecasting model, and forecasting the occurrence date of each landscape through the landscape forecasting model.
Further, the detection module is specifically configured to:
when the meteorological data are acquired, detecting a plurality of meteorological elements contained in the meteorological data in real time, and correspondingly acquiring a plurality of meteorological values corresponding to the meteorological elements respectively;
generating a corresponding weather monitoring table according to a plurality of weather elements and a plurality of weather values, and generating a corresponding weather monitoring diagram according to the weather monitoring table, wherein the weather elements are horizontal coordinates, and the weather values are vertical coordinates;
inputting the weather monitoring graph into the preset regression model in real time, so that the preset regression model correspondingly outputs a plurality of original independent variables, and the weather monitoring graph has uniqueness.
Further, the detection module is specifically further configured to:
when the weather monitoring graph is obtained, extracting a plurality of change curves contained in the weather monitoring graph one by one, and preprocessing each change curve to generate a plurality of corresponding target curves;
analyzing each target curve through the preset regression model to correspondingly detect variable elements contained in each target curve;
and setting a plurality of variable elements as a plurality of original independent variables respectively, wherein each variable element has uniqueness.
Further, the expression of the preset regression model is:
wherein alpha is j Represents the j-th meteorological element, alpha 0 Representing regression coefficient, x i，j Representing the weather value of coordinates (i, j), Y i Represents the i-th variable element, i=1, …, n.
Further, the training module is specifically configured to:
when the meteorological factors and the periodic factors are respectively acquired, carrying out mixed processing on the meteorological factors and the periodic factors based on preset weights so as to generate corresponding target data sets, and splitting the target data sets into corresponding training sets and verification sets according to preset proportions;
and detecting a conversion layer, an analysis layer and a learning layer which are respectively contained in the preset neural network, and correspondingly training the landscape forecast model through the conversion layer, the analysis layer and the learning layer based on the training set and the verification set.
Further, the training module is specifically configured to:
when the training set is acquired, carrying out normalization processing on the training set through a first preset algorithm in the conversion layer so as to output a plurality of corresponding characteristic values;
analyzing the characteristic values through a second preset algorithm in the analysis layer to analyze the characteristic values into corresponding characteristic sequences;
inputting a plurality of feature sequences into the learning layer to correspondingly train the landscape forecast model, wherein the expression of the first preset algorithm is as follows:
wherein Y represents the characteristic value, X represents the meteorological factor, min i Sum max i Representing the minimum value and the maximum value of the periodic factors respectively, wherein each characteristic sequence has uniqueness.
Further, the training module is specifically configured to:
when a plurality of feature sequences are obtained, feature factors contained in each feature sequence are detected one by one, and learning networks contained in the learning layer are extracted;
detecting a plurality of network nodes contained in the learning network, and adding corresponding identifiers to each network node according to the connection sequence among the network nodes;
extracting initial network parameters respectively contained in each network node, and respectively and correspondingly replacing each initial network parameter with each characteristic factor according to the identification.
A third aspect of an embodiment of the present invention proposes:
a computer comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the distinctive landscape forecast model construction method as described above when executing the computer program.
A fourth aspect of the embodiment of the present invention proposes:
a readable storage medium having stored thereon a computer program, wherein the program when executed by a processor implements the distinctive landscape forecast model construction method as described above.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
FIG. 1 is a flowchart of a method for constructing a model for forecasting a characteristic landscape according to a first embodiment of the present invention;
fig. 2 is a block diagram of a feature landscape forecast model building system according to a sixth embodiment of the present invention.
The invention will be further described in the following detailed description in conjunction with the above-described figures.
Detailed Description
In order that the invention may be readily understood, a more complete description of the invention will be rendered by reference to the appended drawings. Several embodiments of the invention are presented in the figures. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
It will be understood that when an element is referred to as being "mounted" on another element, it can be directly on the other element or intervening elements may also be present. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. The terms "vertical," "horizontal," "left," "right," and the like are used herein for illustrative purposes only.
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 invention belongs. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
Referring to fig. 1, a method for constructing a distinctive landscape forecast model according to a first embodiment of the present invention is shown, where the method for constructing a distinctive landscape forecast model according to the present invention can simply and rapidly predict the occurrence date of each landscape, and correspondingly improves the use experience of a user.
Specifically, the method for constructing the characteristic landscape forecast model provided by the embodiment specifically includes the following steps:
step S10, acquiring meteorological data generated in a plurality of continuous years, and detecting a plurality of original independent variables contained in the meteorological data through a preset regression model;
step S20, extracting a plurality of target independent variables from a plurality of original independent variables based on a preset rule, and setting the plurality of target independent variables as corresponding meteorological factors;
step S30, detecting the change period of the meteorological factors in real time through a preset wavelet analysis method, and extracting the periodic factors contained in the change period;
and S40, inputting a plurality of meteorological factors and periodic factors into a preset neural network to train a corresponding landscape forecasting model, and forecasting the occurrence date of each landscape through the landscape forecasting model.
In particular, in this embodiment, it is first required to be described that, in order to accurately predict the occurrence date of each feature landscape, a large amount of weather data generated in the past year needs to be obtained as samples, so as to cover various possible situations, and correspondingly improve the prediction accuracy. Based on the above, firstly, the weather data generated by the weather station in a plurality of consecutive years are acquired, and further, in order to clearly acquire the factors influencing the characteristic landscape, a plurality of original independent variables contained in the current weather data, namely, influence factors, need to be further detected in real time through a preset regression model.
Furthermore, in order to avoid the interference of external factors, a plurality of target independent variables are randomly extracted from a plurality of current original independent variables according to a certain proportion. Preferably, the above ratio may be 10%. Based on this, the current several target independent variables can be further set directly to the required meteorological factors. Furthermore, as each meteorological factor is fluctuating, the change period among a plurality of current meteorological factors can be further detected in real time through a preset wavelet analysis method, and the period factors contained in the current change period can be correspondingly and directly extracted. Based on this, data that can be used for subsequent training, namely the weather factor and the period factor, are obtained in the above manner. Simultaneously, the current meteorological factors and the periodic factors are synchronously input into a preset neural network to finally train a required landscape forecasting model, and the appearance date of each special landscape is accurately predicted through the landscape forecasting model. Preferably, the neural network may be a CNN neural network.
Second embodiment
Specifically, in this embodiment, it should be noted that the step of detecting, by a preset regression model, a plurality of original independent variables included in the meteorological data includes:
when the meteorological data are acquired, detecting a plurality of meteorological elements contained in the meteorological data in real time, and correspondingly acquiring a plurality of meteorological values corresponding to the meteorological elements respectively;
generating a corresponding weather monitoring table according to a plurality of weather elements and a plurality of weather values, and generating a corresponding weather monitoring diagram according to the weather monitoring table, wherein the weather elements are horizontal coordinates, and the weather values are vertical coordinates;
inputting the weather monitoring graph into the preset regression model in real time, so that the preset regression model correspondingly outputs a plurality of original independent variables, and the weather monitoring graph has uniqueness.
Specifically, in this embodiment, after a large amount of weather data is obtained through the above steps, in order to accurately screen out an original independent variable that can be conveniently predicted later, it is necessary to first obtain a plurality of weather elements included in the current weather data, specifically, detect weather elements such as visibility, temperature, precipitation, and the like, and obtain a plurality of weather values corresponding to each weather element. Based on the method, a needed weather monitoring table can be simply constructed correspondingly, namely, a table with complete data can be counted.
Further, a corresponding weather monitoring chart is generated in real time according to the data in the current weather monitoring table, and preferably, weather elements are horizontal coordinates and weather values are vertical coordinates. Meanwhile, the generated meteorological monitoring graph is further input into a preset regression model, so that the current preset regression model can output required original independent variables, and subsequent processing is facilitated.
Specifically, in this embodiment, it should be further noted that the step of inputting the weather monitoring graph into the preset regression model in real time so that the preset regression model correspondingly outputs a plurality of original independent variables includes:
when the weather monitoring graph is obtained, extracting a plurality of change curves contained in the weather monitoring graph one by one, and preprocessing each change curve to generate a plurality of corresponding target curves;
analyzing each target curve through the preset regression model to correspondingly detect variable elements contained in each target curve;
and setting a plurality of variable elements as a plurality of original independent variables respectively, wherein each variable element has uniqueness.
Specifically, in this embodiment, it should be further noted that, after the weather monitoring chart is obtained through the steps described above, a plurality of change curves included in the current weather monitoring chart may be intuitively observed at this time, and further, in order to avoid interference of redundant data, filtering and noise reduction may be sequentially performed on the current plurality of change curves at this time to further generate a required target curve. Based on the above, the corresponding input of each current standard curve is performed in the preset regression model, so as to perform corresponding analysis processing, and further analyze the variable elements contained in each current standard curve, namely the elements affecting the feature landscape appearance date. Furthermore, the corresponding setting of the current variable elements to the original independent variables can be directly performed, so that the subsequent processing is facilitated.
Third embodiment
In addition, in this embodiment, it should be noted that the expression of the preset regression model is:
wherein alpha is j Represents the j-th meteorological element, alpha 0 Representing regression coefficient, x i，j Representing the weather value of coordinates (i, j), Y i Represents the i-th variable element, i=1, …, n.
In addition, in this embodiment, in order to accurately obtain the required original argument, the regression model is preset in this embodiment, and specifically, the required original argument can be output by simply inputting the obtained data into the regression model.
Further, the original independent variable output in real time is subjected to screening processing, and a target independent variable which can be used for subsequent calculation is further extracted, so that subsequent processing is facilitated.
In addition, in this embodiment, it should be further noted that the step of inputting the meteorological factors and the periodic factors into the preset neural network to train the corresponding landscape forecast model includes:
when the meteorological factors and the periodic factors are respectively acquired, carrying out mixed processing on the meteorological factors and the periodic factors based on preset weights so as to generate corresponding target data sets, and splitting the target data sets into corresponding training sets and verification sets according to preset proportions;
and detecting a conversion layer, an analysis layer and a learning layer which are respectively contained in the preset neural network, and correspondingly training the landscape forecast model through the conversion layer, the analysis layer and the learning layer based on the training set and the verification set.
In addition, in this embodiment, it should be further noted that, after the required meteorological factors and the periodic factors are obtained through the above steps, at this time, the current meteorological factors and the periodic factors may be mixed according to a preset weight of 1:1, and a target data set for subsequent training may be generated. Further, the current target data set is split into a required training set and a verification set according to the ratio of 7:3. Further, the conversion layer, the analysis layer and the learning layer contained in the preset neural network are synchronously detected, so that a required landscape forecasting model is finally trained.
Fourth embodiment
In this embodiment, it should be noted that the step of training the landscape forecast model based on the training set and the verification set through the conversion layer, the analysis layer and the learning layer includes:
when the training set is acquired, carrying out normalization processing on the training set through a first preset algorithm in the conversion layer so as to output a plurality of corresponding characteristic values;
analyzing the characteristic values through a second preset algorithm in the analysis layer to analyze the characteristic values into corresponding characteristic sequences;
inputting a plurality of feature sequences into the learning layer to correspondingly train the landscape forecast model, wherein the expression of the first preset algorithm is as follows:
wherein Y represents the characteristic value, X represents the meteorological factor, min i Sum max i Representing the minimum value and the maximum value of the periodic factors respectively, wherein each characteristic sequence has uniqueness.
In this embodiment, it should be noted that, after the required training set and verification set are obtained through the above steps, the training set needs to be processed first at this time, specifically, the weather factors and the periodic factors contained in the current training set are normalized through the first preset algorithm in the conversion layer, and a plurality of corresponding feature values are output, where the feature values are specific numerical values. Further, the DTW algorithm pre-stored in the analysis layer is used for analyzing the current characteristic values and analyzing corresponding characteristic sequences, wherein the characteristic sequences are computer codes which are convenient for computer processing. On the basis, a needed landscape forecasting model can be trained correspondingly through a plurality of current feature sequences and the learning layers.
Fifth embodiment
In this embodiment, it should be noted that, the step of inputting the plurality of feature sequences into the learning layer to correspondingly train the landscape forecast model includes:
when a plurality of feature sequences are obtained, feature factors contained in each feature sequence are detected one by one, and learning networks contained in the learning layer are extracted;
detecting a plurality of network nodes contained in the learning network, and adding corresponding identifiers to each network node according to the connection sequence among the network nodes;
extracting initial network parameters respectively contained in each network node, and respectively and correspondingly replacing each initial network parameter with each characteristic factor according to the identification.
In this embodiment, it should be noted that, after the required feature sequences are obtained through the above steps, since each sequence has a certain feature factor composition, based on this, the feature factors respectively included in each feature sequence can be correspondingly detected, and at the same time, the learning grids included in the learning layer are synchronously extracted, and specifically, the learning grids are constructed by horizontally and longitudinally distributed learning chains, and each learning chain includes a plurality of network nodes.
Further, all network nodes included in the current learning network are detected in real time, corresponding identifiers are further added to each current network node according to the connection sequence among the network nodes, and preferably, the representation can be marked by using numbers. Based on the above, the initial network parameters respectively contained in each current network node are finally extracted in real time, and meanwhile, each current initial network parameter is replaced by each characteristic factor one by one according to the identification sequence, so that the network parameters in the preset neural network can be changed in real time, and a required landscape preparation model can be correspondingly trained.
Referring to fig. 2, a sixth embodiment of the present invention provides:
a featured landscape forecast model construction system, wherein the system comprises:
the detection module is used for acquiring meteorological data generated in a plurality of continuous years and detecting a plurality of original independent variables contained in the meteorological data through a preset regression model;
the extraction module is used for extracting a plurality of target independent variables from a plurality of original independent variables based on a preset rule and setting the plurality of target independent variables as corresponding meteorological factors;
the analysis module is used for detecting the change period of the meteorological factors in real time through a preset wavelet analysis method and extracting the periodic factors contained in the change period;
the training module is used for inputting a plurality of meteorological factors and periodic factors into a preset neural network to train a corresponding landscape forecasting model, and forecasting the occurrence date of each landscape through the landscape forecasting model.
In the above feature landscape forecast model building system, the detection module is specifically configured to:
when the meteorological data are acquired, detecting a plurality of meteorological elements contained in the meteorological data in real time, and correspondingly acquiring a plurality of meteorological values corresponding to the meteorological elements respectively;
generating a corresponding weather monitoring table according to a plurality of weather elements and a plurality of weather values, and generating a corresponding weather monitoring diagram according to the weather monitoring table, wherein the weather elements are horizontal coordinates, and the weather values are vertical coordinates;
inputting the weather monitoring graph into the preset regression model in real time, so that the preset regression model correspondingly outputs a plurality of original independent variables, and the weather monitoring graph has uniqueness.
In the above feature landscape forecast model building system, the detection module is further specifically configured to:
when the weather monitoring graph is obtained, extracting a plurality of change curves contained in the weather monitoring graph one by one, and preprocessing each change curve to generate a plurality of corresponding target curves;
analyzing each target curve through the preset regression model to correspondingly detect variable elements contained in each target curve;
and setting a plurality of variable elements as a plurality of original independent variables respectively, wherein each variable element has uniqueness.
In the feature landscape forecast model construction system, the expression of the preset regression model is as follows:
wherein alpha is j Represents the j-th meteorological element, alpha 0 Representing regression coefficient, x i，j Representing the weather value of coordinates (i, j), Y i Represents the i-th variable element, i=1, …, n.
In the above feature landscape forecast model building system, the training module is specifically configured to:
when the meteorological factors and the periodic factors are respectively acquired, carrying out mixed processing on the meteorological factors and the periodic factors based on preset weights so as to generate corresponding target data sets, and splitting the target data sets into corresponding training sets and verification sets according to preset proportions;
and detecting a conversion layer, an analysis layer and a learning layer which are respectively contained in the preset neural network, and correspondingly training the landscape forecast model through the conversion layer, the analysis layer and the learning layer based on the training set and the verification set.
In the above feature landscape forecast model building system, the training module is further specifically configured to:
when the training set is acquired, carrying out normalization processing on the training set through a first preset algorithm in the conversion layer so as to output a plurality of corresponding characteristic values;
analyzing the characteristic values through a second preset algorithm in the analysis layer to analyze the characteristic values into corresponding characteristic sequences;
inputting a plurality of feature sequences into the learning layer to correspondingly train the landscape forecast model, wherein the expression of the first preset algorithm is as follows:
wherein Y represents the characteristic value, X represents the meteorological factor, min i Sum max i Representing the minimum value and the maximum value of the periodic factors respectively, wherein each characteristic sequence has uniqueness.
In the above feature landscape forecast model building system, the training module is further specifically configured to:
when a plurality of feature sequences are obtained, feature factors contained in each feature sequence are detected one by one, and learning networks contained in the learning layer are extracted;
detecting a plurality of network nodes contained in the learning network, and adding corresponding identifiers to each network node according to the connection sequence among the network nodes;
extracting initial network parameters respectively contained in each network node, and respectively and correspondingly replacing each initial network parameter with each characteristic factor according to the identification.
A seventh embodiment of the present invention provides a computer, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the method for building a feature landscape forecast model provided in the foregoing embodiment when executing the computer program.
An eighth embodiment of the present invention provides a readable storage medium having stored thereon a computer program, wherein the program when executed by a processor implements the distinctive landscape forecast model construction method provided by the above embodiment.
In summary, the method and the system for constructing the characteristic landscape forecast model provided by the embodiment of the invention can simply and rapidly forecast the occurrence date of each landscape, and correspondingly improve the use experience of users.
The above-described respective modules may be functional modules or program modules, and may be implemented by software or hardware. For modules implemented in hardware, the various modules described above may be located in the same processor; or the above modules may be located in different processors in any combination.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The foregoing examples illustrate only a few embodiments of the invention and are described in detail herein without thereby limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (7)
1. The method for constructing the characteristic landscape forecast model is characterized by comprising the following steps:
acquiring meteorological data generated within a plurality of continuous years, and detecting a plurality of original independent variables contained in the meteorological data through a preset regression model;
extracting a plurality of target independent variables from a plurality of original independent variables based on a preset rule, and setting the plurality of target independent variables as corresponding meteorological factors;
detecting the change period of the meteorological factors in real time by a preset wavelet analysis method, and extracting the periodic factors contained in the change period;
inputting a plurality of meteorological factors and periodic factors into a preset neural network to train a corresponding landscape forecasting model, and forecasting the occurrence date of each landscape through the landscape forecasting model;
the step of inputting the meteorological factors and the periodic factors into a preset neural network to train a corresponding landscape forecasting model comprises the following steps of:
when the meteorological factors and the periodic factors are respectively acquired, carrying out mixed processing on the meteorological factors and the periodic factors based on preset weights so as to generate corresponding target data sets, and splitting the target data sets into corresponding training sets and verification sets according to preset proportions;
detecting a conversion layer, an analysis layer and a learning layer which are respectively contained in the preset neural network, and correspondingly training the landscape forecast model through the conversion layer, the analysis layer and the learning layer based on the training set and the verification set;
the step of training the landscape forecast model based on the training set and the verification set through the conversion layer, the analysis layer and the learning layer correspondingly comprises the following steps:
when the training set is acquired, carrying out normalization processing on the training set through a first preset algorithm in the conversion layer so as to output a plurality of corresponding characteristic values;
analyzing the characteristic values through a second preset algorithm in the analysis layer to analyze the characteristic values into corresponding characteristic sequences;
inputting a plurality of feature sequences into the learning layer to correspondingly train the landscape forecast model, wherein the expression of the first preset algorithm is as follows:
wherein Y represents the characteristic value, X represents the meteorological factor, min i Sum max i Respectively representing the minimum value and the maximum value of the periodic factors, wherein each characteristic sequence has uniqueness;
the step of inputting the feature sequences into the learning layer to correspondingly train the landscape forecast model comprises the following steps:
when a plurality of feature sequences are obtained, feature factors contained in each feature sequence are detected one by one, and learning networks contained in the learning layer are extracted;
detecting a plurality of network nodes contained in the learning network, and adding corresponding identifiers to each network node according to the connection sequence among the network nodes;
extracting initial network parameters respectively contained in each network node, and respectively and correspondingly replacing each initial network parameter with each characteristic factor according to the identification.
2. The method for constructing the distinctive landscape forecast model according to claim 1, wherein: the step of detecting a plurality of original independent variables contained in the meteorological data through a preset regression model comprises the following steps:
when the meteorological data are acquired, detecting a plurality of meteorological elements contained in the meteorological data in real time, and correspondingly acquiring a plurality of meteorological values corresponding to the meteorological elements respectively;
generating a corresponding weather monitoring table according to a plurality of weather elements and a plurality of weather values, and generating a corresponding weather monitoring diagram according to the weather monitoring table, wherein the weather elements are horizontal coordinates, and the weather values are vertical coordinates;
inputting the weather monitoring graph into the preset regression model in real time, so that the preset regression model correspondingly outputs a plurality of original independent variables, and the weather monitoring graph has uniqueness.
3. The method for constructing the distinctive landscape forecast model according to claim 2, wherein: the step of inputting the weather monitoring graph into the preset regression model in real time so that the preset regression model correspondingly outputs a plurality of original independent variables comprises the following steps:
when the weather monitoring graph is obtained, extracting a plurality of change curves contained in the weather monitoring graph one by one, and preprocessing each change curve to generate a plurality of corresponding target curves;
analyzing each target curve through the preset regression model to correspondingly detect variable elements contained in each target curve;
and setting a plurality of variable elements as a plurality of original independent variables respectively, wherein each variable element has uniqueness.
4. The method for constructing a distinctive landscape forecast model according to claim 3, wherein: the expression of the preset regression model is as follows:
wherein alpha is j Represents the j-th meteorological element, alpha 0 Representing regression coefficient, x i，j Representing the weather value of coordinates (i, j), Y i Represents the i-th variable element, i=1, …, n.
5. A distinctive landscape forecast model construction system for implementing the distinctive landscape forecast model construction method according to any one of claims 1 to 4, the system comprising:
the detection module is used for acquiring meteorological data generated in a plurality of continuous years and detecting a plurality of original independent variables contained in the meteorological data through a preset regression model;
the extraction module is used for extracting a plurality of target independent variables from a plurality of original independent variables based on a preset rule and setting the plurality of target independent variables as corresponding meteorological factors;
the analysis module is used for detecting the change period of the meteorological factors in real time through a preset wavelet analysis method and extracting the periodic factors contained in the change period;
the training module is used for inputting a plurality of meteorological factors and periodic factors into a preset neural network to train a corresponding landscape forecasting model, and forecasting the occurrence date of each landscape through the landscape forecasting model.
6. A computer comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of constructing the featured landscape forecast model of any of claims 1 to 4 when the computer program is executed by the processor.
7. A readable storage medium having stored thereon a computer program, which when executed by a processor implements the distinctive landscape forecast model construction method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311686847.5A CN117390592B (en) | 2023-12-11 | 2023-12-11 | Method and system for constructing characteristic landscape forecast model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311686847.5A CN117390592B (en) | 2023-12-11 | 2023-12-11 | Method and system for constructing characteristic landscape forecast model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117390592A CN117390592A (en) | 2024-01-12 |
| CN117390592B true CN117390592B (en) | 2024-03-22 |
Family
ID=89441338
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311686847.5A Active CN117390592B (en) | 2023-12-11 | 2023-12-11 | Method and system for constructing characteristic landscape forecast model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117390592B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109543906A (en) * | 2018-11-23 | 2019-03-29 | 长三角环境气象预报预警中心（上海市环境气象中心） | A kind of method and apparatus of atmospheric visibility prediction |
| WO2021052140A1 (en) * | 2019-09-17 | 2021-03-25 | 中国科学院分子细胞科学卓越创新中心 | Anticipatory learning method and system oriented towards short-term time series prediction |
| CN114280696A (en) * | 2021-12-23 | 2022-04-05 | 无锡九方科技有限公司 | Intelligent sea fog level forecasting method and system |
| CN114676822A (en) * | 2022-03-25 | 2022-06-28 | 东南大学 | Multi-attribute fusion air quality forecasting method based on deep learning |
| CN115079309A (en) * | 2022-08-08 | 2022-09-20 | 广东省气象公共服务中心(广东气象影视宣传中心) | Method, device, equipment and medium for constructing prediction model of multi-type sea fog |
| CN115237896A (en) * | 2022-07-12 | 2022-10-25 | 四川大学 | Data preprocessing method and system for forecasting air quality based on deep learning |
| CN115730684A (en) * | 2022-12-09 | 2023-03-03 | 安徽大学 | Air quality detection system based on LSTM-CNN model |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10871594B2 (en) * | 2019-04-30 | 2020-12-22 | ClimateAI, Inc. | Methods and systems for climate forecasting using artificial neural networks |
| CN114547017B (en) * | 2022-04-27 | 2022-08-05 | 南京信息工程大学 | Meteorological big data fusion method based on deep learning |
| US20230351155A1 (en) * | 2022-04-27 | 2023-11-02 | Humana Inc. | Ensemble Time Series Model for Forecasting |
| CN114626512B (en) * | 2022-05-17 | 2022-09-06 | 南京信息工程大学 | High-temperature disaster forecasting method based on directed graph neural network |
-
2023
- 2023-12-11 CN CN202311686847.5A patent/CN117390592B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109543906A (en) * | 2018-11-23 | 2019-03-29 | 长三角环境气象预报预警中心（上海市环境气象中心） | A kind of method and apparatus of atmospheric visibility prediction |
| WO2021052140A1 (en) * | 2019-09-17 | 2021-03-25 | 中国科学院分子细胞科学卓越创新中心 | Anticipatory learning method and system oriented towards short-term time series prediction |
| CN114280696A (en) * | 2021-12-23 | 2022-04-05 | 无锡九方科技有限公司 | Intelligent sea fog level forecasting method and system |
| CN114676822A (en) * | 2022-03-25 | 2022-06-28 | 东南大学 | Multi-attribute fusion air quality forecasting method based on deep learning |
| CN115237896A (en) * | 2022-07-12 | 2022-10-25 | 四川大学 | Data preprocessing method and system for forecasting air quality based on deep learning |
| CN115079309A (en) * | 2022-08-08 | 2022-09-20 | 广东省气象公共服务中心(广东气象影视宣传中心) | Method, device, equipment and medium for constructing prediction model of multi-type sea fog |
| CN115730684A (en) * | 2022-12-09 | 2023-03-03 | 安徽大学 | Air quality detection system based on LSTM-CNN model |
Non-Patent Citations (2)
| Title |
|---|
| Deep Learning and Time Series-to-Image Encoding for Financial Forecasting;Silvio Barra等;IEEE;20200515(第03期);全文 * |
| 人工神经网络在短期降水预测方面的应用研究;张继学;王鹏;张琳;王一;;科技风;20160915(第17期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117390592A (en) | 2024-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112685950B (en) | Method, system and equipment for detecting abnormality of ocean time sequence observation data | |
| CN111008640A (en) | Image recognition model training and image recognition method, device, terminal and medium | |
| CN108760305B (en) | Bearing fault detection method, device and equipment | |
| CN108629326A (en) | The action behavior recognition methods of objective body and device | |
| CN112288021A (en) | Medical wastewater monitoring data quality control method, device and system | |
| CN108491931B (en) | Method for improving nondestructive testing precision based on machine learning | |
| CN109978870A (en) | Method and apparatus for output information | |
| CN113761259A (en) | Image processing method and device and computer equipment | |
| CN114519508A (en) | Credit risk assessment method based on time sequence deep learning and legal document information | |
| CN111176872B (en) | Monitoring data processing method, system, device and storage medium for IT operation and maintenance | |
| CN117473275B (en) | Energy consumption detection method for data center | |
| CN115147353A (en) | Defect detection model training method, device, equipment, medium and program product | |
| CN110555384A (en) | Beef marbling automatic grading system and method based on image data | |
| Adha et al. | Automated building classification framework using convolutional neural network | |
| CN114154622A (en) | Algorithm model for traffic operation system flow data acquisition missing completion | |
| CN117390592B (en) | Method and system for constructing characteristic landscape forecast model | |
| CN111158918B (en) | Supporting point parallel enumeration load balancing method, device, equipment and medium | |
| CN112487819A (en) | Method, system, electronic device and storage medium for identifying homonyms among enterprises | |
| DE112014006799T5 (en) | Method for measuring a user behavior consistency level based on a complex correspondence system | |
| CN117315355A (en) | Image classification method, system, electronic equipment and storage medium | |
| CN116884435A (en) | Voice event detection method and device based on audio prompt learning | |
| CN111680572B (en) | Dynamic judgment method and system for power grid operation scene | |
| CN115511341A (en) | Landslide time-varying failure probability assessment method and device | |
| CN109361539A (en) | A kind of performance alarm method and apparatus based on development network | |
| Kotenok et al. | The e-government’s influence on the country’s economy (at the example of Ukraine and Estonia) |
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 |