CN111860970A - River flow prediction method and device, electronic equipment and storage medium - Google Patents
River flow prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111860970A CN111860970A CN202010605978.6A CN202010605978A CN111860970A CN 111860970 A CN111860970 A CN 111860970A CN 202010605978 A CN202010605978 A CN 202010605978A CN 111860970 A CN111860970 A CN 111860970A
- Authority
- CN
- China
- Prior art keywords
- ground observation
- remote sensing
- flow
- river
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000012545 processing Methods 0.000 claims abstract description 43
- 238000012706 support-vector machine Methods 0.000 claims abstract description 42
- 238000013528 artificial neural network Methods 0.000 claims abstract description 37
- 230000009467 reduction Effects 0.000 claims abstract description 26
- 239000013598 vector Substances 0.000 claims description 33
- 238000004088 simulation Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 230000005236 sound signal Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 241000932075 Priacanthus hamrur Species 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000003068 static effect Effects 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Game Theory and Decision Science (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a river flow prediction method, a river flow prediction device, an electronic device and a computer readable storage medium, wherein the method comprises the following steps: acquiring satellite remote sensing image data and ground observation station data; carrying out dimensionality reduction processing on the satellite remote sensing image data by using an encoder to obtain remote sensing river flow characteristics; performing dimensionality enhancement processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics; inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result. Therefore, the river flow prediction method provided by the application improves the accuracy of river flow prediction by performing dimension reduction processing on satellite remote sensing image data and performing dimension increasing processing on ground observation station data.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a method and an apparatus for predicting river discharge, an electronic device, and a computer-readable storage medium.
Background
At present, a common Machine learning method Support Vector Machine (SVM) (Chinese full name: Support Vector Machine) and a Neural Network (NN) (Chinese full name: neural network model) have been used for predicting river flow, wherein the NN comprises various neural network classification models. The SVM and the NN have large prediction errors when the ground observation value time sequence contains large noise or has large correlation, and the measurement errors of the satellite remote sensing image data observation values are large during river flow prediction, so that the observation values contain a lot of noise, and the prediction power of the model is reduced.
Therefore, how to improve the accuracy of river discharge prediction is a technical problem to be solved by those skilled in the art.
Disclosure of Invention
The invention aims to provide a river flow prediction method, a river flow prediction device, electronic equipment and a computer readable storage medium, and accuracy of river flow prediction is improved.
In order to achieve the above object, the present application provides a river discharge prediction method, including:
acquiring satellite remote sensing image data and ground observation station data;
carrying out dimensionality reduction processing on the satellite remote sensing image data by using an encoder to obtain regional remote sensing river flow characteristics;
Performing dimensionality enhancement processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics;
inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
Wherein the encoder comprises a wide convolutional neural network, a Gaussian hidden Markov model, and a Bayesian flattening variational inference model;
the method for obtaining the river flow characteristics of the regional remote sensing by carrying out dimension reduction processing on the satellite remote sensing image data by using the encoder comprises the following steps:
inputting the satellite remote sensing image data into the wide convolution neural network to obtain a first area remote sensing river flow characteristic vector;
inputting the satellite remote sensing image data into the Gaussian hidden Markov model to obtain a second region remote sensing river flow characteristic vector;
inputting the satellite remote sensing image data into the Bayesian flattening variation inference model to obtain a third area remote sensing river flow characteristic vector;
and splicing the first region remote sensing river flow characteristic vector, the second region remote sensing river flow characteristic vector and the third region remote sensing river flow characteristic vector into the region remote sensing river flow characteristic.
The method for obtaining the ground observation flow and rainfall characteristics and the ground observation weather characteristics by using the decoder to perform dimension-increasing processing on the data of the ground observation station comprises the following steps:
generating analog data corresponding to the ground observation station data;
and mixing the data of the ground observation station and the simulation data into the ground observation flow, rainfall characteristic and ground observation weather characteristic by using a multiple interpolation method.
The generating of the analog data corresponding to the ground observation station data includes:
and calculating a space-time probability distribution model of the ground observation station data, and generating simulation data corresponding to the ground observation station data by using a Bayesian generative learner based on the space-time probability distribution model.
The generating of the analog data corresponding to the ground observation station data includes:
and generating analog data corresponding to the ground observation station data through an importance sampling model based on probability distribution.
After the data of the ground observation station is subjected to dimensionality reduction processing by the decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics, the method further comprises the following steps:
reducing the correlation between the ground observation flow and the rainfall characteristics and the ground observation weather characteristics based on a regularization method;
Correspondingly, inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM or a neural network to obtain a river flow prediction result, wherein the river flow prediction result comprises the following steps:
inputting the river flow characteristics of the regional remote sensing, the ground observation flow and rainfall characteristics for reducing the correlation and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
Wherein the reducing the correlation between the ground observation flow and the rainfall characteristics and the ground observation weather characteristics based on the regularization method comprises the following steps:
returning the ith ground observation flow and rainfall characteristics and the ground observation weather characteristics of the t-1 th period to the-ith ground observation flow and rainfall characteristics and the ground observation weather characteristics of the t-1 th period by utilizing penalty regression to obtain the ground observation flow and rainfall characteristics and the ground observation weather characteristics of the t-1 th period with reduced correlation;
correspondingly, inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics for reducing correlation and the ground observation weather characteristics into an SVM or a neural network to obtain a river flow prediction result, wherein the river flow prediction result comprises the following steps:
Inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics of the t-th period into the SVM or the neural network to obtain the river flow prediction result of the t-th period.
To achieve the above object, the present application provides a river discharge prediction apparatus, comprising:
the acquisition module is used for acquiring satellite remote sensing image data and ground observation station data;
the dimension reduction module is used for carrying out dimension reduction processing on the satellite remote sensing image data by utilizing an encoder to obtain the river flow characteristic of the regional remote sensing;
the dimension increasing module is used for performing dimension increasing processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics;
and the input module is used for inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
To achieve the above object, the present application provides an electronic device including:
a memory for storing a computer program;
a processor for implementing the steps of the river discharge prediction method as described above when executing the computer program.
To achieve the above object, the present application provides a computer-readable storage medium having a computer program stored thereon, which when executed by a processor, implements the steps of the river discharge prediction method as described above.
According to the scheme, the river flow prediction method provided by the application comprises the following steps: acquiring satellite remote sensing image data and ground observation station data; carrying out dimensionality reduction processing on the satellite remote sensing image data by using an encoder to obtain regional remote sensing river flow characteristics; performing dimensionality enhancement processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics; inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
According to the river flow prediction method, encoder processing is carried out on satellite remote sensing image data, low-dimensional regional remote sensing river flow characteristic vectors are calculated, and noise influence is reduced. And interpolating the low-frequency ground observation station data, improving the dimensionality of the data, forming mapping from the ground observation station data to the regional remote sensing flow characteristics, and facilitating the comprehensive calculation of the data. Therefore, the river flow prediction method provided by the application improves the accuracy of river flow prediction by performing dimension reduction processing on satellite remote sensing image data and performing dimension increasing processing on ground observation station data. The application also discloses a river flow prediction device, an electronic device and a computer readable storage medium, which can also achieve the technical effects.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow chart illustrating a river discharge prediction method according to an exemplary embodiment;
FIG. 2 is a schematic diagram illustrating a front-end data processing model architecture of a river discharge comprehensive prediction model SFAP, according to an exemplary embodiment;
FIG. 3 is a flow chart illustrating another river discharge prediction method according to an exemplary embodiment;
FIG. 4 is a diagram illustrating a back-end data processing model architecture of a river discharge comprehensive prediction model SFAP, in accordance with an exemplary embodiment;
FIG. 5 is a block diagram illustrating a river discharge prediction device in accordance with an exemplary embodiment;
FIG. 6 is a block diagram illustrating an electronic device in accordance with an exemplary embodiment.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The embodiment of the application discloses a river flow prediction method, which improves the accuracy of river flow prediction.
Referring to fig. 1, a flow chart of a river discharge prediction method according to an exemplary embodiment is shown, as shown in fig. 1, including:
s101: acquiring satellite remote sensing image data and ground observation station data;
s102: carrying out dimensionality reduction processing on the satellite remote sensing image data by using an encoder to obtain regional remote sensing river flow characteristics;
It is understood that an Encoder-Decoder method (EDL) can flexibly perform dimension reduction and dimension increase on data, and facilitates analysis by combining data from different sources.
The river flow comprehensive prediction model sfap (streamflow Aggregated prediction) includes a front-end data processing model and a back-end prediction model. The front-end data processing model structure of the river discharge comprehensive prediction model SFAP is shown in fig. 2. The method comprises the following steps of inputting satellite remote sensing image data into an encoder, processing the satellite remote sensing image data through the encoder, calculating a regional remote sensing river flow characteristic vector, reducing the dimension of the satellite remote sensing image data, extracting the regional remote sensing river flow characteristic vector, wherein the dimension reduction formula of the satellite remote sensing image data is as follows:
wherein phi iseFor the dimension-reduced parameters of the encoder,in order to remotely sense the image data for the satellite,for regional remote sensing of river discharge characteristics, nx>nh。
As a possible implementation, the encoder includes a wide convolutional neural network, a gaussian hidden markov model, and a bayesian flattened variance inference model; this step may include: inputting the satellite remote sensing image data into the wide convolution neural network to obtain a first area remote sensing river flow characteristic vector; inputting the satellite remote sensing image data into the Gaussian hidden Markov model to obtain a second region remote sensing river flow characteristic vector; inputting the satellite remote sensing image data into the Bayesian flattening variation inference model to obtain a third area remote sensing river flow characteristic vector; and splicing the first region remote sensing river flow characteristic vector, the second region remote sensing river flow characteristic vector and the third region remote sensing river flow characteristic vector into the region remote sensing river flow characteristic.
In the specific implementation, wide convolution neural networks, Gaussian hidden Markov models and Bayesian flattening variation inference models are adopted to respectively calculate the feature vectors of the regional remote sensing river flow, and then the calculated feature vectors are spliced together. The regional remote sensing river flow characteristic vectors calculated by different methods reflect different aspects of the flow time sequence, and the information contained in the input time sequence is retained to a greater extent while dimension reduction is realized by splicing the characteristic vectors.
S103: performing dimensionality enhancement processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics;
the input of the decoder is the data of the ground observation station, the frequency is low, the missing value is more, and the panel data of the ground observation station is processed through the decoder. Specifically, the step may include: generating analog data corresponding to the ground observation station data; and mixing the data of the ground observation station and the simulation data into the ground observation flow, rainfall characteristic and ground observation weather characteristic by using a multiple interpolation method.
As a possible implementation manner, for the simulation data, a space-time probability distribution model of the ground observation station data is calculated, and the simulation data corresponding to the ground observation station data is generated by using a bayesian generative learner based on the space-time probability distribution model. As another possible implementation, the simulation data corresponding to the ground observation station data is generated through an importance sampling model based on probability distribution.
And mixing the generated simulation data with the real data, interpolating the low-frequency ground observation station data by using a Multiple interpolation MI (Chinese full name: Multiple interpolation method, English full name: Multiple interpolation) method, generating high-frequency simulation data reflecting the characteristics of the real data, and improving the frequency of the ground observation station data. The rising dimension formula of the data of the ground observation station is as follows:
wherein phi isdIs an up-dimensional parameter of the decoder,the data of the ground observation station is represented by theta, the ground observation flow and rainfall characteristics and the ground observation weather characteristics are represented by theta, and the dimensionality of theta is higher than that of thetaOf (c) is calculated.
The decoder improves the dimension of input ground observation station data, generates analog samples through a Bayesian generative learner or performs importance sampling from estimated probability distribution to obtain the analog samples, performs multiple interpolation MI on the analog samples, and encrypts original low-frequency or missing ground observation station panel data. The formula for sampling is as follows:
wherein,calculated by a representative decoderProbability distribution of data of ground observation station, znFeatures representing data of ground observation stations, wnRepresenting samples taken by significant samples in the probability distribution computed by the decoder. In the decoder, the simulated samples and the real samples are consistent under the probability distribution model.
S104: inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
In the step, the characteristics of the regional remote sensing river flow, the characteristics of the ground observation flow and rainfall and the characteristics of the ground observation weather are input into the SVM or the neural network to predict the river flow.
According to the river flow prediction method provided by the embodiment of the application, the satellite remote sensing image data is processed by the encoder, the low-dimensional regional remote sensing river flow characteristic vector is calculated, and the noise influence is reduced. And the low-frequency ground observation station data is interpolated, so that the dimensionality of the data is improved, and the comprehensive calculation of the data is facilitated. Therefore, the river flow prediction method provided by the embodiment of the application improves the accuracy of river flow prediction by performing dimension reduction processing on satellite remote sensing image data and performing dimension increasing processing on ground observation station data.
The embodiment of the application discloses a river flow prediction method, and compared with the previous embodiment, the embodiment further explains and optimizes the technical scheme. Specifically, the method comprises the following steps:
referring to fig. 3, a flow chart of another river discharge prediction method is shown according to an exemplary embodiment, as shown in fig. 3, including:
S201: acquiring satellite remote sensing image data and ground observation station data;
s202: carrying out dimensionality reduction processing on the satellite remote sensing image data by using an encoder to obtain regional remote sensing river flow characteristics;
s203: performing dimensionality enhancement processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics;
s204: reducing the correlation between the ground observation flow and the rainfall characteristics and the ground observation weather characteristics based on a regularization method;
it should be noted that there is a large correlation between the ground observation flow and the rainfall and weather characteristics, and the SVM and NN perform poorly in the case of large correlation of characteristics or large data noise. Therefore, the regularization method is adopted in the embodiment to reduce the correlation among the features, and the SVM and NN are used for prediction conveniently.
Specifically, the ith ground observation flow and rainfall characteristics and the ground observation weather characteristics in the t-1 stage are regressed to the-ith ground observation flow and rainfall characteristics and the ground observation weather characteristics in the t-1 stage by punishment regression, so that the ground observation flow and rainfall characteristics and the ground observation weather characteristics in the t-1 stage with reduced correlation are obtained. The structure of the back-end data processing model of the river flow comprehensive prediction model SFAP is shown in FIG. 4, and the ground observed flow, rainfall and weather characteristics C tThe correlation between them tends to be large, in CtFor example, the rest of features can be normalized according to actual conditions and then follow-up calculation is continued, the calculation formula of the regularization method is as follows, penalty regression is used, and the ith feature C of the t-1 stage is obtainedt-1,iRegression to other characteristics Ct-1,-iThe method comprises the following steps:E(∈)=0,E(∈2)=σ2;
wherein,is a regression model for adjusting the ith characteristic Ct-1,iAnd other characteristics Ct-1,-iCorrelation between them, phiiIs a parameter, R (phi), estimated by the regression modeli) The method is a regular term and has various choices, and mainly has penalty regression options such as Lasso, adaptive Lasso, SCAD and MCP, wherein E () is an error formula, and sigma is an error parameter. Model selection is carried out while model estimation is carried out by punishment regression, the trouble of carrying out relevance adjustment on the features with low relevance in other features is saved, and only the model selection is carried outAnd performing relevance adjustment on other characteristics with larger relevance. Features i after penalizing regression and decorrelationReplaces the original ith characteristic Ct-1,iFeatures with reduced correlationAs input next in the machine learning model SVM or NN.
S205: inputting the river flow characteristics of the regional remote sensing, the ground observation flow and rainfall characteristics for reducing the correlation and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
In the step, the regional remote sensing river flow characteristics in the t stage, the ground observation flow and rainfall characteristics in the t-1 stage and the ground observation weather characteristics are input into an SVM or a neural network, and the river flow prediction result in the t stage is obtained.
Specifically, the t-1 stage ground is used for observing the flow, the rainfall and the weather characteristics Ct-1EDL processing and regularization decorrelation processing are carried out, and SVM and NN are used for carrying out the region remote sensing river flow characteristic r in the t periodtAnd (6) predicting. The EDL reduces data noise and adjusts data frequency, and the regularization method reduces correlation among features, so that the predictive power of the SVM and the NN is improved.
Order toAnd isWherein the total number of the characteristics is N,is the ground observed flow and rainfall and weather characteristics C of the t-1 periodt-1The reduced correlation characteristic obtained after the regularization,is the parameter estimation in the fitted model. The river flow prediction method comprises the following steps: will be provided withInputting the data into SVM or NN for model fitting:inputting data of t periodCarrying out the next-phase regional remote sensing river flow characteristic r in the fitted modelt+1Prediction of (2):wherein r is obtainedt+1The flow characteristics of the remote sensing river in the next period region can be reflected. Remote sensing river flow characteristic r by using next phase regiont+1As a prediction of the river section flow.
Therefore, according to the method, the relevance between the ground observed flow and the rainfall and weather features is reduced by using the penalty regression containing the regular term, and then the decorrelated features are input into the machine learning model to predict the river flow features, so that the negative influence of the relevance between the features on the model prediction capability is reduced, the stability of the model is improved, and the accuracy of river flow prediction is further improved.
In the following, a river discharge prediction apparatus provided by an embodiment of the present application is introduced, and a river discharge prediction apparatus described below and a river discharge prediction method described above may be referred to each other.
Referring to fig. 5, a block diagram of a river discharge prediction apparatus according to an exemplary embodiment is shown, as shown in fig. 5, including:
the acquisition module 501 is used for acquiring satellite remote sensing image data and ground observation station data;
the dimension reduction module 502 is used for performing dimension reduction processing on the satellite remote sensing image data by using an encoder to obtain the river flow characteristics of the regional remote sensing;
the dimension increasing module 503 is configured to perform dimension increasing processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics;
And the input module 504 is used for inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
The river flow prediction device provided by the embodiment of the application carries out encoder processing on satellite remote sensing image data, calculates low-dimensional regional remote sensing river flow characteristic vectors and reduces noise influence. And the low-frequency ground observation station data is interpolated, so that the dimensionality of the data is improved, and the comprehensive calculation of the data is facilitated. Therefore, the river flow prediction method provided by the embodiment of the application improves the accuracy of river flow prediction by performing dimension reduction processing on satellite remote sensing image data and performing dimension increasing processing on ground observation station data.
On the basis of the above embodiment, as a preferred implementation, the encoder includes a wide convolutional neural network, a gaussian hidden markov model and a bayesian flattened variance inference model; the dimension reduction module 502 comprises:
the first input unit is used for inputting the satellite remote sensing image data into the wide convolution neural network to obtain a first area remote sensing river flow characteristic vector;
The second input unit is used for inputting the satellite remote sensing image data into the Gaussian hidden Markov model to obtain a second region remote sensing river flow characteristic vector;
the third input unit is used for inputting the satellite remote sensing image data into the Bayesian flattening variation inference model to obtain a third area remote sensing river flow characteristic vector;
and the splicing unit is used for splicing the first region remote sensing river flow characteristic vector, the second region remote sensing river flow characteristic vector and the third region remote sensing river flow characteristic vector into the region remote sensing river flow characteristic.
On the basis of the above embodiment, as a preferred implementation, the dimension-increasing module 503 includes:
the generating unit is used for generating analog data corresponding to the ground observation station data;
and the mixing unit is used for mixing the ground observation station data and the simulation data into the ground observation flow, rainfall characteristic and ground observation weather characteristic by using a multiple interpolation method.
In addition to the above embodiments, as a preferred implementation manner, the generating unit is specifically a unit that calculates a spatio-temporal probability distribution model of the ground observation station data, and generates simulated data corresponding to the ground observation station data by using a bayesian generation type learner based on the spatio-temporal probability distribution model.
On the basis of the foregoing embodiment, as a preferred implementation manner, the generating unit is specifically a unit that generates analog data corresponding to the ground observation station data through an importance sampling model based on probability distribution.
On the basis of the above embodiment, as a preferred implementation, the method further includes:
the correlation reduction module is used for reducing the correlation between the ground observation flow and the rainfall characteristics and the ground observation weather characteristics based on a regularization method;
correspondingly, the input module 504 is specifically a module for inputting the regional remote sensing river flow characteristics, the correlation-reduced ground observation flow and rainfall characteristics, and the ground observation weather characteristics into an SVM or a neural network to obtain a river flow prediction result.
On the basis of the above embodiment, as a preferred implementation manner, the correlation reduction module is specifically a module that uses penalty regression to regress the ith ground observation flow and rainfall characteristics and the ground observation weather characteristics in the t-1 th period to the-ith ground observation flow and rainfall characteristics and the ground observation weather characteristics in the t-1 th period, so as to obtain the correlation-reduced ground observation flow and rainfall characteristics and the ground observation weather characteristics in the t-1 th period;
Correspondingly, the input module 504 is specifically a module for inputting the regional remote sensing river flow characteristics at the t-th stage, the ground observation flow and rainfall characteristics at the t-1 th stage, and the ground observation weather characteristics into an SVM or a neural network to obtain the river flow prediction result at the t-th stage.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
The present application further provides an electronic device, and referring to fig. 6, a structure diagram of an electronic device 600 provided in an embodiment of the present application may include a processor 11 and a memory 12, as shown in fig. 6. The electronic device 600 may also include one or more of a multimedia component 13, an input/output (I/O) interface 14, and a communication component 15.
The processor 11 is configured to control the overall operation of the electronic device 600, so as to complete all or part of the steps of the river discharge prediction method. The memory 12 is used to store various types of data to support operation at the electronic device 600, such as instructions for any application or method operating on the electronic device 600 and application-related data, such as contact data, transmitted and received messages, pictures, audio, video, and so forth. The Memory 12 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk or optical disk. The multimedia component 13 may include a screen and an audio component. Wherein the screen may be, for example, a touch screen and the audio component is used for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may further be stored in the memory 12 or transmitted via the communication component 15. The audio assembly also includes at least one speaker for outputting audio signals. The I/O interface 14 provides an interface between the processor 11 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication module 15 is used for wired or wireless communication between the electronic device 600 and other devices. Wireless communication, such as Wi-Fi, bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so that the corresponding communication component 15 may include: Wi-Fi module, bluetooth module, NFC module.
In an exemplary embodiment, the electronic Device 600 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the river flow prediction method described above.
In another exemplary embodiment, a computer readable storage medium comprising program instructions which, when executed by a processor, implement the steps of the river discharge prediction method described above is also provided. For example, the computer readable storage medium may be the memory 12 described above including program instructions that are executable by the processor 11 of the electronic device 600 to perform the river discharge prediction method described above.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and such improvements and modifications also fall within the scope of the claims of the present application.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. A river discharge prediction method, comprising:
acquiring satellite remote sensing image data and ground observation station data;
carrying out dimensionality reduction processing on the satellite remote sensing image data by using an encoder to obtain regional remote sensing river flow characteristics;
Performing dimensionality enhancement processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics;
inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
2. The river discharge prediction method of claim 1 wherein the encoder comprises a wide convolutional neural network, a gaussian hidden markov model, and a bayesian flattened variance inference model;
the method for obtaining the river flow characteristics of the regional remote sensing by carrying out dimension reduction processing on the satellite remote sensing image data by using the encoder comprises the following steps:
inputting the satellite remote sensing image data into the wide convolution neural network to obtain a first area remote sensing river flow characteristic vector;
inputting the satellite remote sensing image data into the Gaussian hidden Markov model to obtain a second region remote sensing river flow characteristic vector;
inputting the satellite remote sensing image data into the Bayesian flattening variation inference model to obtain a third area remote sensing river flow characteristic vector;
and splicing the first region remote sensing river flow characteristic vector, the second region remote sensing river flow characteristic vector and the third region remote sensing river flow characteristic vector into the region remote sensing river flow characteristic.
3. The river discharge prediction method according to claim 1, wherein the obtaining of the ground observed flow and rainfall features and the ground observed weather features by performing dimension-increasing processing on the ground observation station data by using a decoder comprises:
generating analog data corresponding to the ground observation station data;
and mixing the data of the ground observation station and the simulation data into the ground observation flow, rainfall characteristic and ground observation weather characteristic by using a multiple interpolation method.
4. The method of predicting river discharge according to claim 3, wherein the generating simulated data corresponding to the data from the ground observation station comprises:
and calculating a space-time probability distribution model of the ground observation station data, and generating simulation data corresponding to the ground observation station data by using a Bayesian generative learner based on the space-time probability distribution model.
5. The method of predicting river discharge according to claim 3, wherein the generating simulated data corresponding to the data from the ground observation station comprises:
and generating analog data corresponding to the ground observation station data through an importance sampling model based on probability distribution.
6. The river discharge prediction method according to any one of claims 1 to 5, wherein after the dimension reduction processing is performed on the data of the ground observation station by using the decoder to obtain the ground observation discharge and rainfall characteristics and the ground observation weather characteristics, the method further comprises:
Reducing the correlation between the ground observation flow and the rainfall characteristics and the ground observation weather characteristics based on a regularization method;
correspondingly, inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM or a neural network to obtain a river flow prediction result, wherein the river flow prediction result comprises the following steps:
inputting the river flow characteristics of the regional remote sensing, the ground observation flow and rainfall characteristics for reducing the correlation and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
7. The river discharge prediction method of claim 6 wherein the regularization based approach to reduce the correlation between the ground observed discharge and the rainfall and ground observed weather features comprises:
returning the ith ground observation flow and rainfall characteristics and the ground observation weather characteristics of the t-1 th period to the-ith ground observation flow and rainfall characteristics and the ground observation weather characteristics of the t-1 th period by utilizing penalty regression to obtain the ground observation flow and rainfall characteristics and the ground observation weather characteristics of the t-1 th period with reduced correlation;
correspondingly, inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics for reducing correlation and the ground observation weather characteristics into an SVM or a neural network to obtain a river flow prediction result, wherein the river flow prediction result comprises the following steps:
Inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics of the t-th period into the SVM or the neural network to obtain the river flow prediction result of the t-th period.
8. A river discharge prediction device, comprising:
the acquisition module is used for acquiring satellite remote sensing image data and ground observation station data;
the dimension reduction module is used for carrying out dimension reduction processing on the satellite remote sensing image data by utilizing an encoder to obtain the river flow characteristic of the regional remote sensing;
the dimension increasing module is used for performing dimension increasing processing on the data of the ground observation station by using a decoder to obtain ground observation flow, rainfall characteristics and ground observation weather characteristics;
and the input module is used for inputting the regional remote sensing river flow characteristics, the ground observation flow and rainfall characteristics and the ground observation weather characteristics into an SVM (support vector machine) or a neural network to obtain a river flow prediction result.
9. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the river discharge prediction method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the river discharge prediction method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010605978.6A CN111860970B (en) | 2020-06-29 | 2020-06-29 | River flow prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010605978.6A CN111860970B (en) | 2020-06-29 | 2020-06-29 | River flow prediction method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111860970A true CN111860970A (en) | 2020-10-30 |
CN111860970B CN111860970B (en) | 2022-03-22 |
Family
ID=72988123
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010605978.6A Active CN111860970B (en) | 2020-06-29 | 2020-06-29 | River flow prediction method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111860970B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114330123A (en) * | 2021-12-29 | 2022-04-12 | 湖南省柘溪电力集团有限公司 | Cross-platform data processing method |
CN114529815A (en) * | 2022-02-10 | 2022-05-24 | 中山大学 | Deep learning-based traffic detection method, device, medium and terminal |
CN116819655A (en) * | 2023-08-29 | 2023-09-29 | 深圳市银河系科技有限公司 | Data management method and system based on natural language processing technology |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0972764A (en) * | 1995-09-04 | 1997-03-18 | Hitachi Ltd | System for predicting cross-sectional profile of river bed |
CN108647807A (en) * | 2018-04-10 | 2018-10-12 | 武汉理工大学 | The prediction technique of river discharge |
CN111159149A (en) * | 2019-12-13 | 2020-05-15 | 国网浙江省电力有限公司紧水滩水力发电厂 | River flow prediction method based on three-dimensional convolutional neural network |
-
2020
- 2020-06-29 CN CN202010605978.6A patent/CN111860970B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0972764A (en) * | 1995-09-04 | 1997-03-18 | Hitachi Ltd | System for predicting cross-sectional profile of river bed |
CN108647807A (en) * | 2018-04-10 | 2018-10-12 | 武汉理工大学 | The prediction technique of river discharge |
CN111159149A (en) * | 2019-12-13 | 2020-05-15 | 国网浙江省电力有限公司紧水滩水力发电厂 | River flow prediction method based on three-dimensional convolutional neural network |
Non-Patent Citations (3)
Title |
---|
NOORI R: "Assessment of input variables determination on the SVM model performance using PCA, Gamma test, and forward selection techniques for monthly stream flow prediction", 《JOURNAL OF HYDROLOGY》 * |
王建中等: "基于遗传算法与支持向量机的日流量预测", 《水电能源科学》 * |
蔡超等: "大规模数据的分块SCAD惩罚回归分析", 《数理统计与管理》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114330123A (en) * | 2021-12-29 | 2022-04-12 | 湖南省柘溪电力集团有限公司 | Cross-platform data processing method |
CN114529815A (en) * | 2022-02-10 | 2022-05-24 | 中山大学 | Deep learning-based traffic detection method, device, medium and terminal |
CN116819655A (en) * | 2023-08-29 | 2023-09-29 | 深圳市银河系科技有限公司 | Data management method and system based on natural language processing technology |
CN116819655B (en) * | 2023-08-29 | 2023-11-03 | 深圳市银河系科技有限公司 | Data management method and system based on natural language processing technology |
Also Published As
Publication number | Publication date |
---|---|
CN111860970B (en) | 2022-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111860970B (en) | River flow prediction method and device, electronic equipment and storage medium | |
CN110210513B (en) | Data classification method and device and terminal equipment | |
KR101004495B1 (en) | Method of noise estimation using incremental bayes learning | |
US11355097B2 (en) | Sample-efficient adaptive text-to-speech | |
US20230267730A1 (en) | Image abnormality detection model training | |
US20240046955A1 (en) | Voice extraction method and apparatus, and electronic device | |
US11501107B2 (en) | Key-value memory network for predicting time-series metrics of target entities | |
CN111737640A (en) | Water level prediction method, device and computer readable storage medium | |
CN112767959A (en) | Voice enhancement method, device, equipment and medium | |
WO2019138897A1 (en) | Learning device and method, and program | |
CN114974292A (en) | Audio enhancement method and device, electronic equipment and readable storage medium | |
Zhou et al. | A new online Bayesian NMF based quasi-clean speech reconstruction for non-intrusive voice quality evaluation | |
CN114818864A (en) | Gesture recognition method based on small samples | |
CN114118570A (en) | Service data prediction method and device, electronic equipment and storage medium | |
CN111797220B (en) | Dialog generation method, apparatus, computer device and storage medium | |
CN113590941A (en) | Target information recommendation method and device, electronic equipment and storage medium | |
CN111680218B (en) | User interest identification method and device, electronic equipment and storage medium | |
Urteaga et al. | Sequential estimation of hidden ARMA processes by particle filtering—part I | |
Andrieu et al. | A particle filter for model based audio source separation | |
CN116882536A (en) | Rainfall data prediction method and device, electronic equipment and storage medium | |
CN112885367B (en) | Fundamental frequency acquisition method, fundamental frequency acquisition device, computer equipment and storage medium | |
CN107766944B (en) | System and method for optimizing system function flow by utilizing API analysis | |
Elvira et al. | Gauss-Hermite quadrature for non-Gaussian inference via an importance sampling interpretation | |
US20240170003A1 (en) | Audio Signal Enhancement with Recursive Restoration Employing Deterministic Degradation | |
Holubnychyi et al. | Presenting AR Model Error in Terms of Geman-McClure Function for Prediction of Processes in Telecommunications |
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 |