CN117113854B - Salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation - Google Patents
Salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation Download PDFInfo
- Publication number
- CN117113854B CN117113854B CN202311345170.9A CN202311345170A CN117113854B CN 117113854 B CN117113854 B CN 117113854B CN 202311345170 A CN202311345170 A CN 202311345170A CN 117113854 B CN117113854 B CN 117113854B
- Authority
- CN
- China
- Prior art keywords
- tide
- data
- salty
- salt
- wind
- 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
- 150000003839 salts Chemical class 0.000 title claims abstract description 221
- 238000004088 simulation Methods 0.000 title claims abstract description 60
- 238000013277 forecasting method Methods 0.000 title claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims abstract description 79
- 238000012545 processing Methods 0.000 claims abstract description 69
- 238000000034 method Methods 0.000 claims abstract description 68
- 238000004364 calculation method Methods 0.000 claims abstract description 52
- 230000008521 reorganization Effects 0.000 claims abstract description 52
- 238000005457 optimization Methods 0.000 claims abstract description 33
- 238000012795 verification Methods 0.000 claims abstract description 8
- 238000013135 deep learning Methods 0.000 claims abstract description 6
- 230000008859 change Effects 0.000 claims description 144
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 70
- 230000002159 abnormal effect Effects 0.000 claims description 54
- 238000012876 topography Methods 0.000 claims description 35
- 230000008569 process Effects 0.000 claims description 30
- 238000012937 correction Methods 0.000 claims description 27
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 10
- 238000013075 data extraction Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 9
- 230000008439 repair process Effects 0.000 claims description 7
- 238000010168 coupling process Methods 0.000 claims description 6
- 230000010354 integration Effects 0.000 claims description 6
- 230000006978 adaptation Effects 0.000 claims description 5
- 230000008878 coupling Effects 0.000 claims description 5
- 238000005859 coupling reaction Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 230000005856 abnormality Effects 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 claims description 4
- 238000003909 pattern recognition Methods 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 2
- 230000001932 seasonal effect Effects 0.000 description 17
- 230000000737 periodic effect Effects 0.000 description 12
- 230000007774 longterm Effects 0.000 description 11
- 238000007619 statistical method Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 239000012530 fluid Substances 0.000 description 4
- 238000012300 Sequence Analysis Methods 0.000 description 3
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000000704 physical effect Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 239000011780 sodium chloride Substances 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 230000003746 surface roughness Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 239000013049 sediment Substances 0.000 description 1
- 238000012732 spatial analysis Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000013526 transfer learning Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of salt tide forecasting, in particular to a salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation. The method comprises the following steps: acquiring data of a salty tide wind field, and performing history reorganization analysis to obtain a salty tide reorganization history wind field set; constructing a three-dimensional salty tide numerical model and performing model adjustment verification processing to obtain a three-dimensional salty tide numerical optimization model; acquiring parameter data of a three-dimensional salt tide numerical optimization model, and carrying out river bed roughness extraction calculation and time-by-time interpolation processing to obtain a salt tide river bed roughness field data set; deep learning prediction analysis is carried out through ConvLSTM, and three-dimensional salty tide numerical simulation is carried out by utilizing a three-dimensional salty tide numerical optimization model according to a prediction result and tide forecast outside sea tide level result data of a target forecast period, so that a three-dimensional salty tide prediction field data set is obtained; and performing corresponding salt tide forecasting work according to the three-dimensional salt tide forecasting field data set. The method can realize high-precision and real-time salt tide forecasting.
Description
Technical Field
The invention relates to the technical field of salt tide forecasting, in particular to a salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation.
Background
Salty tide is a tidal phenomenon occurring at the junction of sea and river, and has an important influence on the ecological environment and economic activity in coastal areas. The traditional salty tide forecasting method is mainly based on an empirical formula and a statistical model, has limited accuracy and timeliness and poor adaptability to complex environmental changes, and often cannot capture complex nonlinear relations and space-time dependencies, so that a forecasting result is inaccurate.
Disclosure of Invention
Based on this, the present invention needs to provide a salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation to solve at least one of the above technical problems.
In order to achieve the above purpose, a salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation comprises the following steps:
step S1: acquiring data of a salty tide wind field, and performing historical reorganization analysis on the data of the salty tide wind field to obtain a salty tide reorganization historical wind field set;
step S2: acquiring the water content data of the salty tide and the topography data of the salty tide, and carrying out salty tide simulation by utilizing the water content data of the salty tide and the topography data of the salty tide to construct a three-dimensional salty tide numerical model; performing adjustment verification on the three-dimensional salty tide numerical model to obtain a three-dimensional salty tide numerical optimization model;
Step S3: acquiring the salt tide hydrologic combination parameter data, and extracting and calculating the river bed roughness of the salt tide hydrologic combination parameter data to obtain a river bed roughness field of each salt tide hydrologic combination; performing time-by-time face interpolation processing on each salt tide hydrologic combined river bed roughness field to obtain a salt tide river bed roughness field data set;
step S4: deep learning prediction analysis is carried out on the salt tide reorganization historical wind field set and the salt tide riverbed roughness field data set through ConvLSTM, so that a salt tide target period wind field data set and a salt tide target period roughness field data set are obtained;
step S5: obtaining tide forecast outside sea tide level result data of a target forecast period, and performing three-dimensional salty tide numerical simulation on a salty tide target period wind field data set, a salty tide target period roughness field data set and tide forecast outside sea tide level result data of the target forecast period by using a three-dimensional salty tide numerical optimization model to obtain a three-dimensional salty tide forecast field data set; and performing corresponding salt tide forecasting work according to the three-dimensional salt tide forecasting field data set.
According to the invention, firstly, the data of the salty tide wind field is obtained through a reliable data source, the data comprise the wind direction and wind speed information of salty tide areas, and a data base can be provided for subsequent analysis and prediction by obtaining accurate data of the salty tide wind field. Meanwhile, through historical backtracking and reorganization analysis on the data of the salty tide wind field, the change condition of the wind direction and the wind speed of the salty tide region can be researched and known, the change condition comprises seasonal change, annual change, long-term trend, abnormal fluctuation and the like of the wind direction and the wind speed, the typical wind field range, wind field distribution characteristics, influence of climate change on the wind field and the like of the salty tide region can be known, and the obtained reorganization historical wind field set can provide wind field data of different time periods in the salty tide generation process and is the basis for carrying out salty tide prediction and numerical simulation. Second, by acquiring hydrological data and topographic data of a salty tide area, these include river discharge, tidal information, water level observation data, topographic data, and the like. By acquiring these data, a basic data set related to the salt tide can be established, providing accurate and complete input data for subsequent salt tide simulation and numerical modeling. And constructing a three-dimensional salty tide numerical model by carrying out salty tide simulation by using the acquired salty tide wind field data, salty tide hydrologic data and salty tide topographic data. Through simulation calculation, the change of parameters such as water level, flow speed, salinity and the like of the salt tide in time and space can be predicted, so that the deep understanding of the dynamic process of the salt tide is facilitated, and quantitative description and prediction of the salt tide phenomenon are provided. The parameters and parameterization scheme of the model are optimized by adjusting and verifying the constructed three-dimensional salty tide numerical model, so that the model can more accurately simulate salty tide process, and the prediction accuracy and reliability are improved. Then, by acquiring parameter data of the combination of the salt water and tide, the parameter data may include data related to hydrologic characteristics, tidal conditions, river channel characteristics and the like, and extracting and calculating the river bed roughness of each combination of the salt water and tide through calculation and analysis, wherein the calculation of the river bed roughness is based on factors such as the characteristics of the bottom bed, the flow velocity and the physical properties of fluid, the resistance characteristics of water flow in the river channel can be obtained through calculation, and then the simulation result of the salt water and tide hydrologic process is affected. In addition, discrete measurement point data are converted into continuous river bed roughness data through carrying out time-by-time face interpolation processing on each salt tide hydrologic combination river bed roughness field, so that a salt tide river bed roughness field data set can be obtained and used for subsequent salt tide prediction and numerical simulation. Next, predictive analysis was performed on the salt tide reorganized historical wind field set and the salt tide riverbed roughness field data set by using the ConvLSTM deep learning model. The ConvLSTM model is capable of capturing space-time dependencies and non-linearities in the time series data, so that historical wind and river bed roughness information can be used to predict the salt-tide wind and roughness for a target period. The prediction results include a wind field data set and a roughness field data set for a target period of salt tide, thereby providing inputs and references for salt tide prediction and numerical modeling. Finally, by using a three-dimensional salty tide numerical optimization model to carry out numerical simulation on tide forecast outside sea tide level result data of a target forecast period and a wind field data set and a rough field data set of a salty tide target period, the wind field and the rough field can be combined with the outside sea tide level data, and salty tide process of the target forecast period can be simulated, so that a three-dimensional salty tide forecast field data set is obtained. According to the three-dimensional salt tide prediction field data set, corresponding salt tide prediction work is carried out, including tide level prediction, salt tide level early warning and the like, the prediction results can provide spatial distribution and time sequence change of the salt tide process, salt tide prediction information and response measures are provided for related departments and decision makers, and therefore accuracy and timeliness of salt tide prediction are improved.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of a non-limiting implementation, made with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow chart of the steps of a salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation;
FIG. 2 is a detailed step flow chart of step S1 in FIG. 1;
fig. 3 is a detailed step flow chart of step S13 in fig. 2.
Detailed Description
The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.
Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.
It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
In order to achieve the above objective, referring to fig. 1 to 3, the present invention provides a salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation, the method comprises the following steps:
step S1: acquiring data of a salty tide wind field, and performing historical reorganization analysis on the data of the salty tide wind field to obtain a salty tide reorganization historical wind field set;
step S2: acquiring the water content data of the salty tide and the topography data of the salty tide, and carrying out salty tide simulation by utilizing the water content data of the salty tide and the topography data of the salty tide to construct a three-dimensional salty tide numerical model; performing adjustment verification on the three-dimensional salty tide numerical model to obtain a three-dimensional salty tide numerical optimization model;
Step S3: acquiring the salt tide hydrologic combination parameter data, and extracting and calculating the river bed roughness of the salt tide hydrologic combination parameter data to obtain a river bed roughness field of each salt tide hydrologic combination; performing time-by-time face interpolation processing on each salt tide hydrologic combined river bed roughness field to obtain a salt tide river bed roughness field data set;
step S4: deep learning prediction analysis is carried out on the salt tide reorganization historical wind field set and the salt tide riverbed roughness field data set through ConvLSTM, so that a salt tide target period wind field data set and a salt tide target period roughness field data set are obtained;
step S5: obtaining tide forecast outside sea tide level result data of a target forecast period, and performing three-dimensional salty tide numerical simulation on a salty tide target period wind field data set, a salty tide target period roughness field data set and tide forecast outside sea tide level result data of the target forecast period by using a three-dimensional salty tide numerical optimization model to obtain a three-dimensional salty tide forecast field data set; and performing corresponding salt tide forecasting work according to the three-dimensional salt tide forecasting field data set.
In the embodiment of the present invention, please refer to fig. 1, which is a schematic flow chart of steps of a salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation of the present invention, in this example, the steps of the salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation include:
Step S1: acquiring data of a salty tide wind field, and performing historical reorganization analysis on the data of the salty tide wind field to obtain a salty tide reorganization historical wind field set;
the embodiment of the invention firstly acquires the data of the salty tide wind field through a meteorological observation station, satellite remote sensing, a meteorological model and other ways. Then, wind direction and wind speed historical backtracking processing is carried out on the salty tide wind field data so as to analyze data such as historical evolution, periodic change, seasonal characteristics and the like of wind directions and wind speeds of salty tide areas, and the salty tide historical wind direction and wind speed data are subjected to reorganization processing according to time steps, so that a salty tide reorganization historical wind field set is finally obtained.
Step S2: acquiring the water content data of the salty tide and the topography data of the salty tide, and carrying out salty tide simulation by utilizing the water content data of the salty tide and the topography data of the salty tide to construct a three-dimensional salty tide numerical model; performing adjustment verification on the three-dimensional salty tide numerical model to obtain a three-dimensional salty tide numerical optimization model;
according to the embodiment of the invention, hydrologic and topographic features of a salty tide area in different time and space are acquired through a hydrologic observation station, a tide level observation station and satellite remote sensing, so that salty tide hydrologic data and salty tide topographic data are obtained. And then, carrying out salt tide simulation by comprehensively considering the conditions of wind fields, hydrology, terrains and the like by using salt tide wind field data, salt tide hydrologic data and salt tide topography data so as to simulate the changes of parameters such as the water level, the flow speed, the salinity and the like of salt tide in time and space so as to construct a three-dimensional salt tide numerical model. And finally, adjusting and verifying the constructed three-dimensional salty tide numerical model to compare the fitting degree of the simulation result and the actually measured observation data, and optimizing the model to finally obtain the three-dimensional salty tide numerical optimization model.
Step S3: acquiring the salt tide hydrologic combination parameter data, and extracting and calculating the river bed roughness of the salt tide hydrologic combination parameter data to obtain a river bed roughness field of each salt tide hydrologic combination; performing time-by-time face interpolation processing on each salt tide hydrologic combined river bed roughness field to obtain a salt tide river bed roughness field data set;
according to the embodiment of the invention, parameters related to hydrologic conditions, such as water level, flow rate, bottom friction force and the like, are acquired from hydrologic observation sites, remote sensing data and the like, so that the salty tide hydrologic combination parameter data is obtained. And then, analyzing the salt tide hydrologic combination parameter data, extracting data related to the river bed roughness, performing association matching processing on the data and the space coordinates corresponding to the data in the geographic space, and representing the distribution condition of the river bed roughness in the form of the geographic space to obtain the salt tide hydrologic combination river bed roughness field. And finally, carrying out time-by-time smooth interpolation on each salt tide hydrologic combination river bed roughness field at different time steps and different spatial positions to obtain a continuous salt tide river bed roughness field, and finally obtaining a salt tide river bed roughness field data set.
Step S4: deep learning prediction analysis is carried out on the salt tide reorganization historical wind field set and the salt tide riverbed roughness field data set through ConvLSTM, so that a salt tide target period wind field data set and a salt tide target period roughness field data set are obtained;
According to the embodiment of the invention, a model is firstly constructed through ConvLSTM, a salty tide reorganization historical wind field set and a salty tide river bed roughness field data set are input into the constructed ConvLSTM model as input data for deep learning prediction analysis, and the salty tide wind field and roughness field data of a target period are predicted through training and optimizing network weight parameters of the model, so that a salty tide target period wind field data set and a salty tide target period roughness field data set are finally obtained.
Step S5: obtaining tide forecast outside sea tide level result data of a target forecast period, and performing three-dimensional salty tide numerical simulation on a salty tide target period wind field data set, a salty tide target period roughness field data set and tide forecast outside sea tide level result data of the target forecast period by using a three-dimensional salty tide numerical optimization model to obtain a three-dimensional salty tide forecast field data set; and performing corresponding salt tide forecasting work according to the three-dimensional salt tide forecasting field data set.
According to the embodiment of the invention, the tidal forecast outside sea level result data of the target forecast period is obtained to serve as the boundary condition of the model, the three-dimensional salty tide numerical optimization model is used for carrying out three-dimensional salty tide numerical simulation on the salty tide target period wind field data set, the salty tide target period rough rate field data set and the tidal forecast outside sea level result data of the target forecast period, and the wind field, the rough rate field and the outside sea level data are combined to simulate the salty tide process of the target forecast period, so that the three-dimensional salty tide prediction field data set is obtained. And then, according to the three-dimensional salty tide prediction field data set obtained by simulation, carrying out salty tide prediction work such as corresponding tide level prediction, salty tide level early warning and the like.
According to the invention, firstly, the data of the salty tide wind field is obtained through a reliable data source, the data comprise the wind direction and wind speed information of salty tide areas, and a data base can be provided for subsequent analysis and prediction by obtaining accurate data of the salty tide wind field. Meanwhile, through historical backtracking and reorganization analysis on the data of the salty tide wind field, the change condition of the wind direction and the wind speed of the salty tide region can be researched and known, the change condition comprises seasonal change, annual change, long-term trend, abnormal fluctuation and the like of the wind direction and the wind speed, the typical wind field range, wind field distribution characteristics, influence of climate change on the wind field and the like of the salty tide region can be known, and the obtained reorganization historical wind field set can provide wind field data of different time periods in the salty tide generation process and is the basis for carrying out salty tide prediction and numerical simulation. Second, by acquiring hydrological data and topographic data of a salty tide area, these include river discharge, tidal information, water level observation data, topographic data, and the like. By acquiring these data, a basic data set related to the salt tide can be established, providing accurate and complete input data for subsequent salt tide simulation and numerical modeling. And constructing a three-dimensional salty tide numerical model by carrying out salty tide simulation by using the acquired salty tide wind field data, salty tide hydrologic data and salty tide topographic data. Through simulation calculation, the change of parameters such as water level, flow speed, salinity and the like of the salt tide in time and space can be predicted, so that the deep understanding of the dynamic process of the salt tide is facilitated, and quantitative description and prediction of the salt tide phenomenon are provided. The parameters and parameterization scheme of the model are optimized by adjusting and verifying the constructed three-dimensional salty tide numerical model, so that the model can more accurately simulate salty tide process, and the prediction accuracy and reliability are improved. Then, by acquiring parameter data of the combination of the salt water and tide, the parameter data may include data related to hydrologic characteristics, tidal conditions, river channel characteristics and the like, and extracting and calculating the river bed roughness of each combination of the salt water and tide through calculation and analysis, wherein the calculation of the river bed roughness is based on factors such as the characteristics of the bottom bed, the flow velocity and the physical properties of fluid, the resistance characteristics of water flow in the river channel can be obtained through calculation, and then the simulation result of the salt water and tide hydrologic process is affected. In addition, discrete measurement point data are converted into continuous river bed roughness data through carrying out time-by-time face interpolation processing on each salt tide hydrologic combination river bed roughness field, so that a salt tide river bed roughness field data set can be obtained and used for subsequent salt tide prediction and numerical simulation. Next, predictive analysis was performed on the salt tide reorganized historical wind field set and the salt tide riverbed roughness field data set by using the ConvLSTM deep learning model. The ConvLSTM model is capable of capturing space-time dependencies and non-linearities in the time series data, so that historical wind and river bed roughness information can be used to predict the salt-tide wind and roughness for a target period. The prediction results include a wind field data set and a roughness field data set for a target period of salt tide, thereby providing inputs and references for salt tide prediction and numerical modeling. Finally, by using a three-dimensional salty tide numerical optimization model to carry out numerical simulation on tide forecast outside sea tide level result data of a target forecast period and a wind field data set and a rough field data set of a salty tide target period, the wind field and the rough field can be combined with the outside sea tide level data, and salty tide process of the target forecast period can be simulated, so that a three-dimensional salty tide forecast field data set is obtained. According to the three-dimensional salt tide prediction field data set, corresponding salt tide prediction work is carried out, including tide level prediction, salt tide level early warning and the like, the prediction results can provide spatial distribution and time sequence change of the salt tide process, salt tide prediction information and response measures are provided for related departments and decision makers, and therefore accuracy and timeliness of salt tide prediction are improved.
Preferably, step S1 comprises the steps of:
step S11: acquiring data of a salty tide wind field;
step S12: carrying out wind direction data extraction and wind speed data extraction on the salty tide wind field data to obtain salty tide wind direction data and salty tide wind speed data;
step S13: carrying out wind direction history backtracking analysis on the salty tide wind direction data to obtain salty tide historical wind direction data;
step S14: performing wind speed history backtracking analysis on the salty tide wind speed data to obtain salty tide history wind speed data;
step S15: and performing time alignment and interpolation reorganization processing on the salty tide historical wind direction data and the salty tide historical wind speed data to obtain a salty tide reorganization historical wind field set.
As an embodiment of the present invention, referring to fig. 2, a detailed step flow chart of step S1 in fig. 1 is shown, in which step S1 includes the following steps:
step S11: acquiring data of a salty tide wind field;
according to the embodiment of the invention, the salty tide wind field data is obtained through a meteorological observation station, satellite remote sensing, meteorological models and other approaches.
Step S12: carrying out wind direction data extraction and wind speed data extraction on the salty tide wind field data to obtain salty tide wind direction data and salty tide wind speed data;
according to the embodiment of the invention, the information data related to the wind direction of the salty tide region is extracted by extracting the wind direction data of the salty tide wind field data, so as to obtain the salty tide wind direction data, and meanwhile, the information data related to the wind speed of the salty tide region is extracted by extracting the wind speed data, so as to obtain the salty tide wind speed data.
Step S13: carrying out wind direction history backtracking analysis on the salty tide wind direction data to obtain salty tide historical wind direction data;
according to the embodiment of the invention, wind direction history backtracking processing is carried out on the salty tide wind direction data by using a corresponding statistical technology, so that factors such as wind direction distribution, trend, abnormal fluctuation correction and the like are analyzed, accurate historical wind direction data are obtained, and salty tide historical wind direction data are finally obtained.
Step S14: performing wind speed history backtracking analysis on the salty tide wind speed data to obtain salty tide history wind speed data;
according to the embodiment of the invention, firstly, the weather change influence analysis and the restoration processing are carried out on the salty tide wind speed data so as to eliminate abnormal influence caused by possible instability, fluctuation or seasonal change in the salty tide wind speed data, then, the wind speed change pattern recognition analysis and the space radial analysis are carried out on the restored salty tide wind speed data, the change patterns, the periodic characteristics and the possible trend changes of the wind speed on different time scales and space scales can be recognized, and finally, the historical trend retrospective analysis is carried out on the salty tide wind speed data so as to deeply understand the data of the historical evolution, the periodic changes, the seasonal characteristics and the like of the wind speed of salty tide areas, and finally, the salty tide historical wind speed data is obtained.
Step S15: and performing time alignment and interpolation reorganization processing on the salty tide historical wind direction data and the salty tide historical wind speed data to obtain a salty tide reorganization historical wind field set.
According to the embodiment of the invention, the salty tide historical wind direction data and salty tide historical wind speed data are subjected to time alignment processing to ensure that the salty tide historical wind direction data and the salty tide historical wind speed data have the same time step or time interval, the missing data or incomplete data are subjected to interpolation to fill up the missing value, so that continuous salty tide historical wind direction and wind speed data are obtained, then continuous salty tide historical wind direction and wind speed data are subjected to reorganization processing according to the time step, and finally the salty tide reorganization historical wind field set is obtained.
According to the invention, the data of the salty tide wind field is obtained through reliable data sources such as a meteorological observation station, satellite remote sensing, a meteorological model and the like, the data comprise wind direction and wind speed information of salty tide areas, and a data base can be provided for subsequent analysis and prediction by obtaining accurate data of the salty tide wind field. Meanwhile, wind direction data extraction and wind speed data extraction are carried out on the salty tide wind field data, wind direction and wind speed information can be extracted from the salty tide wind field data, so that the direction and the intensity of wind in salty tide areas can be accurately judged, and the salty tide wind field data can be used for subsequent historical backtracking analysis, salty tide simulation construction and other aspects, and the wind direction and wind speed data extraction is an important basis for understanding wind field characteristics and meteorological changes in salty tide areas. Secondly, through carrying out historical backtracking analysis on the salty tide wind direction data, the wind direction change condition of salty tide areas can be researched and known, the method comprises the steps of analyzing seasonal change, annual change, long-term trend, abnormal fluctuation and the like of wind directions, and the acquisition of salty tide historical wind direction data is beneficial to researching the characteristics of wind fields and influence factors of wind directions, so that basis is provided for related application of salty tide areas. Then, by performing historical backtracking analysis on the salty tide wind speed data, the wind speed change situation of salty tide areas can be explored and known, including analysis of seasonal change, annual change, long-term trend, climate change influence and the like of wind speeds, and by acquiring salty tide historical wind speed data, the typical wind speed range, wind speed distribution characteristics, influence of climate change on wind speeds and the like of salty tide areas can be known, so that basic data is provided for related applications and risk assessment. Finally, the salty tide historical wind direction data and salty tide historical wind speed data are aligned in time, namely, the consistency of the salty tide historical wind direction data and the salty tide historical wind speed data in time is ensured, interpolation and reorganization are carried out on the salty tide historical wind direction data and the salty tide historical wind speed data to fill possibly missing data or incomplete data, so that a processed data set can provide salty tide wind field data with continuous time and good consistency, and a reliable data base is provided for aspects of weather research, wind energy evaluation, salty tide forecasting and the like in salty tide areas.
Preferably, step S13 comprises the steps of:
step S131: frequency calculation is carried out on the salty tide wind direction data by utilizing a wind direction frequency calculation formula, so as to obtain a salty tide wind direction frequency value;
step S132: according to the salt tide wind direction frequency value, carrying out time change trend exploration on the salt tide wind direction data so as to generate a salt tide wind direction change mode graph;
step S133: detecting abnormal fluctuation of the graph of the wind direction change pattern of the salty tide to generate abnormal fluctuation points of the wind direction change of the salty tide;
step S134: performing abnormal correction processing on the abnormal fluctuation points of the change of the wind direction of the salty tide to obtain a correction chart of a change mode curve of the wind direction of the salty tide;
step S135: and carrying out historical statistics retrospective analysis on the correction chart of the wind direction change pattern curve of the salty tide to obtain the historical wind direction data of the salty tide.
As an embodiment of the present invention, referring to fig. 3, a detailed step flow chart of step S13 in fig. 2 is shown, in which step S13 includes the following steps:
step S131: frequency calculation is carried out on the salty tide wind direction data by utilizing a wind direction frequency calculation formula, so as to obtain a salty tide wind direction frequency value;
according to the embodiment of the invention, a proper wind direction frequency calculation formula is constructed by combining the angle of the salty tide wind direction, the transverse and longitudinal space coordinates of salty tide wind direction data in a frequency space, the transverse wave vector and longitudinal wave vector, the frequency space phase imaginary number unit, the fluctuation influence amplitude and fluctuation influence angular frequency of wind direction frequency components and related parameters, so that the frequency calculation is carried out on the salty tide wind direction data, the occurrence condition of the relative distribution of different wind directions in salty tide areas is counted, and finally the salty tide wind direction frequency value is obtained.
Step S132: according to the salt tide wind direction frequency value, carrying out time change trend exploration on the salt tide wind direction data so as to generate a salt tide wind direction change mode graph;
according to the embodiment of the invention, the data of the salty tide wind direction is identified and analyzed according to the calculated frequency value of the salty tide wind direction, so that the main mode or trend of the salty tide wind direction change is identified, the frequency change trend conditions of different wind directions in different time periods are explored and revealed, and finally, a graph of the salty tide wind direction change mode is drawn and generated according to a time sequence.
Step S133: detecting abnormal fluctuation of the graph of the wind direction change pattern of the salty tide to generate abnormal fluctuation points of the wind direction change of the salty tide;
according to the embodiment of the invention, the abnormal fluctuation detection is carried out on the graph of the change mode of the salty tide wind direction by using a statistical method or a time sequence analysis method so as to detect the data points which are obviously deviated from the normal wind direction change mode, and finally the abnormal fluctuation points of the salty tide wind direction change are generated.
Step S134: performing abnormal correction processing on the abnormal fluctuation points of the change of the wind direction of the salty tide to obtain a correction chart of a change mode curve of the wind direction of the salty tide;
according to the embodiment of the invention, the influence of the abnormal fluctuation point of the change of the salty wind direction on the overall trend of the salty wind direction change mode graph is eliminated or corrected by using a smoothing algorithm or a filtering algorithm to correct the abnormal fluctuation point of the change of the salty wind direction in the salty wind direction change mode graph, and finally the salty wind direction change mode graph is obtained.
Step S135: and carrying out historical statistics retrospective analysis on the correction chart of the wind direction change pattern curve of the salty tide to obtain the historical wind direction data of the salty tide.
According to the embodiment of the invention, the historical trend statistical analysis is carried out on the salty tide wind direction change mode curve correction chart by using a statistical method and a historical trend analysis method so as to analyze factors such as distribution, change trend, periodic trend and the like of salty tide historical wind directions and finally obtain salty tide historical wind direction data.
According to the method, firstly, frequency calculation is carried out on the salty tide wind direction data by using a proper wind direction frequency calculation formula, so that the occurrence frequency of different wind directions in salty tide areas can be determined, the occurrence frequency of each wind direction in observation data is counted, the relative distribution situation of different wind directions can be known through wind direction frequency calculation, and therefore the main wind direction mode of salty tide areas is identified, and basic data are provided for subsequent wind direction change analysis. Secondly, by using the calculated frequency value of the wind direction of the salty tide, the time change trend of wind direction data of the salty tide area can be explored, and seasonal change, annual change and long-term trend of the wind direction of the salty tide area can be revealed by observing the frequency change conditions of different wind directions in different time periods. The generated graph of the change pattern of the wind direction of the salty tide can intuitively show the change rule of the wind direction along with time, and a visual tool and a visual reference are provided for further analysis. Then, by performing abnormal fluctuation detection on the graph of the salty-tide wind direction change pattern, the abnormal fluctuation detection is to identify an abnormal situation in the salty-tide wind direction change pattern. By analyzing the graph of the change pattern of the salty tide wind direction, abnormal fluctuation points which possibly exist, namely data points which are obviously deviated from the normal wind direction change pattern, can be detected, and the detection of the abnormal fluctuation points is helpful for finding out emergencies, abnormal weather conditions and the like and provides clues for subsequent abnormal correction. Then, by performing abnormality correction processing on abnormal fluctuation points of the detected changes in the direction of the salty wind, the influence of these abnormal points on the overall trend can be eliminated or corrected. The corrected salty tide wind direction change pattern curve correction chart reflects the change trend of wind direction more accurately, and can provide more reliable historical wind direction data. By means of anomaly correction, data deviation caused by observation errors, equipment faults or other anomaly factors can be eliminated, and accordingly reliability and quality of data are improved. Finally, historical statistics retrospective analysis is carried out on the curve correction chart of the change pattern of the salty tide wind direction, so that historical wind direction data of salty tide areas can be obtained. Analysis includes studying the overall distribution of the corrected graph, frequency statistics, wind direction preferences, and the like. Through historical statistics backtracking analysis, the historical evolution, periodic change, seasonal characteristics and the like of the wind direction of the salty tide area can be deeply known, and basic data is provided for subsequent interpolation and reorganization processing.
Preferably, the wind direction frequency calculation formula in step S131 is specifically:
；
in the method, in the process of the invention,for the value of the wind direction frequency of salty tide, +.>The angle of the wind direction of salty tide>For the transverse spatial coordinates of the wind direction data of the salt tide in the frequency space, < >>Longitudinal spatial coordinates in frequency space for the data of the wind direction of the salt tide, < >>For the transverse wave vector in the frequency space of the salty-tide wind direction data, < >>Longitudinal wave vector in frequency space for the data of the wind direction of the salty tide, < >>Is the imaginary unit of frequency space phase +.>For the number of wind-direction frequency components in the salty-tide wind-direction data, +.>Is the +.f. in the wind direction data of salty tide>The fluctuation of the individual wind direction frequency components affects the amplitude, +.>Is the +.f. in the wind direction data of salty tide>The fluctuation of the individual wind direction frequency components affects the angular frequency, < >>Is the correction value of the wind direction frequency value of the salty tide.
According to the invention, a wind direction frequency calculation formula is constructed for calculating the frequency of the salty tide wind direction data, and the wind direction frequency calculation formula can reflect the distribution condition of the salty tide wind direction data in different angles through the integral operation of the salty tide wind direction data in a frequency space, so that the understanding of the trend and the characteristics of the wind direction is facilitated. Spatial coordinates in the formulaAnd->Representing the position of the salty-tide wind direction data in the frequency space, whereas the transverse wave vector +. >And longitudinal wave vector->The corresponding frequency component is represented. By considering the space coordinates and wave vectors, the formula can analyze and weigh wind direction components of different frequencies. At the same time +.>And->The fluctuation of each frequency component in the salty tide wind direction data is respectively expressed to influence the amplitude and the angular frequency, and the parameters reflect the fluctuation degree and the periodicity of different frequency components on the wind direction data, so that the characteristic and the rule of wind direction change are helpful to be understood. In addition, the correction value of the salty-tide wind direction frequency value is also adjusted by introducing the correction value. By correcting the frequency value of the wind direction of the salty tide, the accuracy of frequency calculation can be further improved, and the result is closer to the actual situation. The formula fully considers the salt tide wind direction frequency value +.>Angle of salty tide wind direction +.>Lateral spatial coordinates of the salty-tide wind direction data in the frequency space +.>Longitudinal spatial coordinates of the salt tide wind direction data in the frequency space +.>Transverse wave vector of the salty-tide wind direction data in the frequency space +.>Longitudinal wave vector of the salty-tide wind direction data in the frequency space +.>Frequency space phase imaginary unit>Number of wind-direction frequency components in the salty-tide wind-direction data +.>The first >Fluctuation of the frequency component of the individual wind direction affects the amplitude +.>The first>Fluctuation of the frequency component of the individual wind direction affects the angular frequency +.>Correction value of the wind direction frequency value of salty tide +.>Wherein by frequency space phase imaginary unit +.>Number of wind-direction frequency components in the salty-tide wind-direction data +.>The first>Fluctuation of the frequency component of the individual wind direction affects the amplitude +.>The first>Fluctuation of the frequency component of the individual wind direction affects the angular frequency +.>Angle of salty tide wind direction +.>Constitutes a wind direction frequency component fluctuation influence term function relation +.>According to the value of the wind direction frequency of salty tide +.>The interrelationship between the parameters constitutes a functional relationship:
；
the formula can realize the frequency calculation process of the salty-tide wind direction data, and simultaneously, the correction value of the salty-tide wind direction frequency value is usedThe introduction of the wind direction frequency calculation formula can be adjusted according to actual conditions, so that the accuracy and the applicability of the wind direction frequency calculation formula are improved.
Preferably, step S14 comprises the steps of:
step S141: performing climate change influence analysis on the salty tide wind speed data to obtain salty tide wind speed climate change influence factor data;
according to the embodiment of the invention, the weather change index is used for carrying out statistical analysis on the salt tide wind speed data, and the influence correlation between the weather change and the salt tide wind speed data is explored so as to identify and quantify the influence degree of different weather change index factors on the salt tide wind speed data, and finally the salt tide wind speed and weather change influence factor data is obtained.
Step S142: performing influence fluctuation restoration processing on the salty-tide wind speed data according to the salty-tide wind speed and climate change influence factor data to obtain salty-tide wind speed abnormality influence restoration data;
according to the embodiment of the invention, firstly, the abnormal fluctuation data in the salty-tide wind speed data is analyzed according to the salty-tide wind speed and climate change influence factor data, then the abnormal fluctuation data in the salty-tide wind speed data is repaired by using a smoothing algorithm or a filtering algorithm, so that abnormal influence fluctuation caused by instability, fluctuation or seasonal change in the salty-tide wind speed data is eliminated, and finally the salty-tide wind speed abnormal influence repair data is obtained.
Step S143: performing wind speed change pattern recognition analysis on the repair data of the abnormal influence of the wind speed of the salty tide to obtain data of a wind speed change pattern of the salty tide;
according to the embodiment of the invention, the wind speed change mode identification is carried out on the repair data of the abnormal influence of the wind speed of the salty tide by using a time sequence analysis technology so as to identify the main change mode of the wind speed of the salty tide, including the identification and analysis of long-term trend, seasonal period, periodic oscillation and the like, and finally the salty tide wind speed change mode data is obtained.
Step S144: carrying out spatial radial analysis on the repair data of the abnormal influence of the wind speed of the salty tide to obtain spatial scale change mode data of the wind speed of the salty tide;
According to the embodiment of the invention, the spatial analysis method is used for carrying out radial distribution analysis and variation function analysis on the repair data of the abnormal influence of the wind speed of the salty tide, the spatial scale change mode of the wind speed of the salty tide is explored, the change rules of the wind speed of the salty tide at different spatial positions are analyzed, and finally the spatial scale change mode data of the wind speed of the salty tide is obtained.
Step S145: and carrying out historical trend backtracking analysis on the data of the air speed change mode of the salty tide and the data of the air speed space scale change mode of the salty tide to obtain the historical air speed data of the salty tide.
According to the embodiment of the invention, the historical trend backtracking analysis is carried out on the salty tide wind speed change mode data and the salty tide wind speed space scale change mode data so as to analyze the historical evolution trend of long-term trend, seasonal change, periodic change and the like of salty tide wind speed, and finally salty tide historical wind speed data is obtained.
According to the invention, by analyzing the influence of climate change on the salty tide wind speed data, the climate change factors influencing the salty tide wind speed can be determined, and the long-term wind speed observation data are counted and analyzed to determine the climate change factors which possibly influence the wind speed, such as seasonal change, annual change and the like. By analyzing the climate change influence factor data of the salty tide wind speed, the long-term trend and change mode of the climate change on the wind speed of salty tide areas can be revealed. Secondly, according to the climatic change influence factor data of the salty tide wind speed, influence fluctuation restoration processing is carried out on the salty tide wind speed data, and the aim of eliminating abnormal influence caused by instability, fluctuation or seasonal change possibly existing in the salty tide wind speed data is fulfilled, so that the quality and accuracy of the data are improved, and the restored salty tide wind speed abnormal influence restoration data can reflect the real change condition of the wind speed. Then, by performing wind speed change pattern recognition analysis on the salty-tide wind speed abnormality influence restoration data, a change pattern of wind speed can be recognized, which includes recognition and analysis of long-term trends, seasonal periods, periodic oscillations, and the like, and the generated salty-tide wind speed change pattern data can provide detailed information of wind speed change, including change patterns, periodic characteristics, and possibly trend changes on different time scales. Then, by carrying out spatial radial analysis on the repaired data of the abnormal influence of the wind speed of the salt tide, a spatial scale change mode of the wind speed can be revealed, and the spatial scale change mode comprises analysis on the distribution, the difference and the change trend of the wind speed in different spatial positions of the salt tide region. The generated data of the spatial scale change mode of the wind speed of the salty tide can provide information of the change of the wind speed along with the spatial position, and the distribution characteristics and the spatial correlation of the wind speed at different geographic positions are researched. Finally, by performing historical trend backtracking analysis on the salty tide wind speed change mode data and the salty tide wind speed space scale change mode data, including research on long-term trend, periodic change, seasonal characteristics and the like of wind speed, more accurate salty tide historical wind speed data can be obtained. Through historical trend backtracking analysis, the historical evolution, periodic change, seasonal characteristics and the like of the wind speed in the salty tide area can be deeply known, and important references and basic data are provided for subsequent interpolation and reorganization processing.
Preferably, step S2 comprises the steps of:
step S21: acquiring water data of a salty tide and topography data of the salty tide;
according to the embodiment of the invention, the data comprising tide observation data, water level data, flow speed data and the like are obtained from a hydrological observation station, a tide level observation station or a tide research institution to describe hydrological characteristics of a tide area in different time and space so as to obtain the tide hydrological data, and meanwhile, the satellite remote sensing data, measurement data or digital sea chart and other sources are used for obtaining the topographic characteristic data comprising the submarine topography, the shoreline morphology, the tide river channel and the like so as to obtain the tide topography data.
Step S22: carrying out salty tide simulation by using salty tide wind field data, salty tide hydrologic data and salty tide topography data to construct a three-dimensional salty tide numerical model;
according to the embodiment of the invention, the conditions of wind fields, hydrology, terrains and the like are comprehensively considered by using the data of the wind fields of the salty tides, the hydrologic data of the salty tides and the data of the terrains, so that the salty tides are simulated, and the changes of parameters such as the water level, the flow velocity and the salinity of the salty tides in time and space are simulated, so that a three-dimensional salty tides numerical model is constructed.
Step S23: carrying out typical time sequence integration analysis on the data of the salty tide wind field, the salty tide hydrologic data and the salty tide topography data to obtain typical three-dimensional salty tide combination parameter data;
According to the embodiment of the invention, the data of the salty tide wind field, the salty tide hydrologic data and the salty tide topography data are subjected to time sequence analysis, a representative typical time period is selected for data integration processing, typical characteristic parameters of salty tide are extracted, and parameter data such as water level, flow rate and salinity are included, so that typical three-dimensional salty tide combination parameter data are finally obtained.
Step S24: performing characteristic pattern analysis on the typical three-dimensional salty tide combination parameter data to obtain typical three-dimensional salty tide combination data characteristics;
according to the embodiment of the invention, the characteristic pattern analysis is carried out on the typical three-dimensional salty tide combination parameter data by using a statistical analysis division method and a clustering analysis division method so as to identify and describe the data characteristics and rules of the typical salty tide combination, and finally the typical three-dimensional salty tide combination data characteristics are obtained.
Step S25: and carrying out adjustment verification on the three-dimensional salty tide numerical model according to the characteristic of the typical three-dimensional salty tide combination data so as to obtain a three-dimensional salty tide numerical optimization model.
According to the embodiment of the invention, the constructed three-dimensional salty tide numerical model is adjusted and verified according to the characteristic of the typical three-dimensional salty tide combination data obtained through analysis, and firstly, parameters of the three-dimensional salty tide numerical model, including salty tide wind field, hydrology, topography parameters and the like, are adjusted so that the simulation result of the model more accords with actual observation and typical characteristics. And then, carrying out numerical simulation verification on the three-dimensional salty tide numerical model, comparing the fitting degree of the simulation result and the actually measured observation data or the typical characteristic data, and carrying out optimization treatment on the model to finally obtain the three-dimensional salty tide numerical optimization model.
The method comprises the steps of firstly, acquiring hydrological data and topographic data of a salty tide area, wherein the hydrological data comprise tide observation data, water level data, flow velocity data and the like, and are used for describing hydrological characteristics of the salty tide area in different time and space. The topographic data mainly comprises data of topographic features such as submarine topography, shoreline morphology, tidal river channel and the like. By acquiring these data, a basic data set related to the salt tide can be established, providing accurate and complete input data for subsequent salt tide simulation and numerical modeling. Secondly, a salt tide simulation is performed by using the acquired salt tide wind field data, salt tide hydrologic data and salt tide topography data to construct a three-dimensional salt tide numerical model. The three-dimensional salty tide numerical model is a simulation model established based on a physical principle and a mathematical equation, and the comprehensive influence of a plurality of factors such as wind field, hydrology and topography is considered. Through simulation calculation, the change of parameters such as water level, flow speed, salinity and the like of the salt tide in time and space can be predicted, so that the deep understanding of the dynamic process of the salt tide is facilitated, and quantitative description and prediction of the salt tide phenomenon are provided. Then, by carrying out typical time sequence integration analysis on the data of the salty tide wind field, the data of the salty tide hydrologic system and the data of the salty tide topography, representative typical salty tide combination parameter data comprising water level, flow rate, salinity and the like can be obtained by integrating the data of different periods, and the data integration is helpful for grasping the time change rule and the spatial distribution characteristic of salty tide, so as to provide a foundation for subsequent model analysis and adjustment. The characteristic pattern analysis is carried out on the characteristic three-dimensional salt tide combination parameter data, and the characteristic and rule of the characteristic salt tide combination can be identified and described through the statistical analysis and pattern extraction methods, wherein the characteristic pattern analysis comprises the steps of analyzing parameters such as amplitude, phase and period of salt tide, finding out a representative characteristic pattern, helping to understand the complexity and the variability of the salt tide deeply through the analysis, and providing a reference basis for further model adjustment and optimization. And finally, adjusting and verifying the constructed three-dimensional salt tide numerical model according to the characteristic of the typical three-dimensional salt tide combination data so as to obtain an optimized model. By comparing the model simulation results with the characteristic salt tide combination data characteristics, the accuracy and applicability of the model can be evaluated. If the model results are in good agreement with the actual data, the model can be considered to have high reliability and predictive power. If there is a difference, parameters, boundary conditions, etc. of the model can be adjusted and corrected to improve the accuracy and reliability of the model. The finally obtained optimization model can more accurately simulate and predict the dynamic characteristics of the salt tide, and provides more reliable tools and bases for salt tide research and application.
Preferably, step S22 comprises the steps of:
step S221: performing anomaly filtering processing on the salty tide wind field data, the salty tide hydrologic data and the salty tide topographic data to obtain salty tide wind field anomaly processing data, salty tide hydrologic anomaly processing data and salty tide topographic anomaly processing data;
according to the method, abnormal filtering processing is carried out on the salty tide wind field data, the salty tide hydrologic data and the salty tide topographic data by using methods such as moving average and high-pass filtering, so that short-term fluctuation and noise of the salty tide wind field data, the salty tide hydrologic data and the salty tide topographic data are removed, and relatively stable salty tide wind field abnormal processing data, salty tide hydrologic abnormal processing data and salty tide topographic abnormal processing data are finally obtained.
Step S222: performing space-time interpolation on the salt-tide wind field anomaly handling data, the salt-tide hydrologic anomaly handling data and the salt-tide topography anomaly handling data to obtain salt-tide wind field interpolation data, salt-tide hydrologic interpolation data and salt-tide topography interpolation data;
according to the embodiment of the invention, smooth interpolation processing is carried out on the abnormal processing data of the salty tide wind field, the abnormal processing data of the salty tide hydrologic system and the abnormal processing data of the salty tide topography in time and space by using methods such as kriging interpolation, inverse distance weighted interpolation and the like so as to fill the gaps among the data and the space distribution of the smooth data, and finally the interpolation data of the salty tide wind field, the interpolation data of the salty tide hydrologic system and the interpolation data of the salty tide topography are obtained.
Step S223: coupling the salty tide wind field interpolation data, the salty tide hydrologic interpolation data and the salty tide topography interpolation data to construct initial conditions and boundary conditions, and obtaining the initial conditions and boundary conditions of the numerical model;
according to the embodiment of the invention, the data are ensured to be consistent in space and time by coupling the interpolation data of the salty tide wind field, the interpolation data of the salty tide hydrologic system and the interpolation data of the salty tide topography, and the initial conditions and the boundary conditions of the model required by construction are analyzed and constructed, so that the initial conditions and the boundary conditions of the numerical model are finally obtained.
Step S224: and carrying out salt tide simulation on the initial conditions and the boundary conditions of the numerical model based on a finite difference method to construct a three-dimensional salt tide numerical model.
According to the embodiment of the invention, the initial condition and the boundary condition of the numerical model are subjected to differential calculation by using a finite difference method, and numerical calculation is performed according to a discrete differential equation and simulation parameters of the numerical model, so that the dynamic evolution processes of the spread, change and the like of the salt tide are simulated, and the three-dimensional salt tide numerical model is finally constructed.
According to the invention, the data of the salty tide wind field, the salty tide hydrologic data and the salty tide topography data are subjected to abnormal filtering treatment, and the purpose of abnormal filtering is to remove noise and abnormal values in the original data, so that the data are smoother and more reliable. Through filtering processing, main change trend and periodic characteristics in salty tide data can be extracted, random fluctuation and abnormal deviation of the data are reduced, and a more reliable data base is provided for subsequent data processing and analysis. Secondly, by performing space-time interpolation on the abnormally processed data of the salty tide wind field, the salty tide hydrologic data and the salty tide topographic data, the space-time interpolation is a method for calculating the value of the missing position or time through an interpolation algorithm based on limited data points or grids. The interpolation processing can fill in the gaps among the data and smooth the spatial distribution of the data, and obtain continuous salty tide wind field, hydrology and topography data, so that the interpolation data can better reflect the time-space change rule of salty tide phenomenon, and provide a more complete and accurate data base for subsequent model construction and analysis. Then, the data of the salty tide wind field, the salty tide hydrologic data and the salty tide topography data obtained through interpolation processing are coupled, and the coupling process refers to the integration of different types of data together so as to construct the initial conditions and the boundary conditions required by the numerical model. Through coupling processing, the data of each point in space can be related to each other, the consistency of the data and the requirement of a model are ensured, and the initial condition and the boundary condition provide accurate and consistent input for subsequent numerical simulation, so that the model can more truly reflect the characteristics and dynamic changes of the salt tide process. And finally, carrying out salt tide simulation on the initial condition and the boundary condition of the numerical model obtained through coupling based on a finite difference method, wherein the finite difference method can be used for discretizing a calculation region into grids, converting a partial differential equation into a differential equation based on Taylor series expansion and differential approximation, and finally obtaining a numerical solution by solving the differential equation through iteration. By carrying out differential calculation on the initial conditions and the boundary conditions, the dynamic evolution process of the salt tide can be simulated, and the three-dimensional spatial distribution and time-varying characteristics of the salt tide are obtained, so that a numerical model constructed by simulation can provide quantitative description and prediction on the salt tide phenomenon, and therefore important tools and references are provided for the research and application of the salt tide.
Preferably, step S3 comprises the steps of:
step S31: acquiring the salt tide hydrologic combination parameter data, and extracting the river bed roughness parameters of the salt tide hydrologic combination parameter data to obtain the relevant parameter data of the river bed roughness of each salt tide hydrologic combination;
according to the embodiment of the invention, parameters related to hydrologic conditions, such as water level, flow rate, bottom friction and the like, are acquired from hydrologic observation stations, remote sensing data and the like to obtain the salt tide hydrologic combination parameter data, and then the salt tide hydrologic combination parameter data is analyzed to extract parameter data related to the river bed roughness, so that the parameter data related to the river bed roughness of each salt tide hydrologic combination is finally obtained.
Step S32: calculating the river bed roughness value of each salty water text combination river bed by using a river bed roughness calculation formula;
according to the embodiment of the invention, a proper river bed roughness calculation formula is constructed by combining the total depth of a river bed, the river bed roughness coefficient, the water flow resistance influence coefficient, the index influence adjustment parameter, the river bed water flow speed, the gravity acceleration, the river bed river water depth, the Euler normal height, the time variation parameter and the related parameters to calculate the river bed roughness, and finally the river bed roughness value of each salty water text combination is obtained.
Step S33: performing field source association processing on the river bed roughness values of the saline-water culture combinations and the corresponding space coordinates through a geographic information system to obtain river bed roughness fields of the saline-water culture combinations;
according to the embodiment of the invention, the correlation matching processing is carried out on the combined river bed roughness values of the salty water and the space coordinates corresponding to the combined river bed roughness values in the geographic space by using the geographic information system, the distribution condition of the river bed roughness is represented in the form of the geographic space by the combined river bed roughness values of the salty water according to the data processing function of the geographic information system, and finally, the combined river bed roughness field of the salty water is obtained.
Step S34: performing time-by-time face interpolation processing on each salt tide hydrologic combination river bed roughness field by using a time-by-time interpolation algorithm to obtain salt tide river bed roughness field interpolation result data;
according to the embodiment of the invention, time-by-time interpolation algorithms such as kriging interpolation, inverse distance weighted interpolation and the like are used for carrying out time-by-time face smooth interpolation on each salt tide hydrologic combination river bed roughness field at different time steps and different space positions so as to obtain continuous salt tide river bed roughness fields, and finally, interpolation result data of the salt tide river bed roughness fields is obtained according to required time intervals and space resolution.
Step S35: and carrying out block sampling on the interpolation result data of the salty tide river bed roughness field to obtain a salty tide river bed roughness field data set.
According to the embodiment of the invention, firstly, the salty tide river bed roughness field interpolation result data is subjected to block processing according to actual requirements, the salty tide river bed roughness field interpolation result data is divided according to a certain space range, each block area is randomly sampled to obtain a discrete salty tide river bed roughness field data point set, and then the salty tide river bed roughness field data point set obtained by sampling is arranged to finally obtain the salty tide river bed roughness field data set.
The invention first obtains parameter data of a salt tide hydrologic combination, wherein the parameters can comprise data related to hydrologic characteristics, tidal conditions, river channel characteristics and the like. By collecting and collating these parameter data, a parameter library of salt tide hydrologic combinations can be established. Meanwhile, parameter information related to the river bed roughness is extracted by analyzing and processing the parameter data of the combination of the salty water and the water, wherein the parameters related to the river bed roughness comprise parameters such as the total depth of a river channel of the river bed, the rough coefficient of the river bed, the influence coefficient of water flow resistance and the like, and the parameters can be used for describing the characteristics and influence factors of the river bed in the combination of the salty water and the water, so that a foundation is provided for the subsequent calculation and interpolation of the river bed roughness. And secondly, calculating the river bed roughness value of each salty water hydrologic combination by using a proper river bed roughness calculation formula and calculating the river bed roughness related parameter data of each salty water hydrologic combination according to the adopted calculation formula and method and combining the parameter data. The calculation of the river bed roughness is based on factors such as the characteristics of the bottom bed, the flow velocity, the physical properties of the fluid and the like, and the resistance characteristics of the water flow in the river channel can be obtained through calculation, so that the simulation result of the salt tide hydrologic process is affected. And then, carrying out field source association processing on the river bed roughness values of the saline tide hydrologic combinations and corresponding space coordinates by using a Geographic Information System (GIS) to obtain a river bed roughness field of the saline tide hydrologic combinations. By correlating the river bed roughness values with the spatial coordinates, a river bed roughness profile, i.e., a river bed roughness field, at different locations in the river channel can be generated to show the spatial variation of the roughness, which can be provided to subsequent model simulations and analyses for revealing the response and impact of the salt-tide hydrology on the river bed roughness profile. Next, the time-by-time face interpolation processing is performed on each of the salt-tide hydrologic combination riverbed roughness fields by using a time-by-time interpolation algorithm, which is a time-and space-based interpolation method for generating continuous time and space lattice point data from limited observation point data. Interpolation results of the river bed roughness of the salty tide on different time steps and different space positions can be obtained through time-by-time interpolation processing, and the interpolation result data can provide space-time change information of the river bed roughness in the salty tide hydrologic process, so that the dynamic change rules of the salty tide in aspects of water level, flow velocity, flood evolution and the like can be analyzed. Finally, the continuous river bed roughness field data can be divided into a plurality of block areas by carrying out block sampling on the salty tide river bed roughness field interpolation result data, and each area contains a specific river channel section or space range. Through block sampling, a salty tide river bed roughness data set of a specific area can be extracted, and further analysis and application are facilitated.
Preferably, the river bed roughness calculation formula in step S33 is specifically:
；
in the method, in the process of the invention,combining the river bed roughness values for each salt tide hydrology,/->Is the total depth of the riverway of the riverbed>River bed roughness coefficient as the river bed roughness rate, < ->Water flow resistance influence coefficient for river bed roughness, < ->Adjusting parameters for the exponential influence of the river bed roughness, < ->For the water flow speed of the river bed, < > is->Acceleration of gravity, ++>Is the depth of river water of the river bed>Is the Euler normal height of the riverbed, < >>Time-varying parameter for Euler normal height of river bed,/->Correction values for the river bed roughness values are combined for each salt tide hydrology.
The invention constructs a river bed roughness calculation formula for calculating the river bed roughness of the relevant parameter data of the combination river bed of the hydrologic combination river bed of each salty tide, and the river bed roughness calculation formula comprehensively considers factors such as the surface roughness of the river bed, the water flow speed, the water depth, the river bed height and the like, and the roughness characteristics of the river bed can be more comprehensively described by taking the factors into the calculation formula, so that the roughness value can reflect the actual situation more accurately. Meanwhile, the river bed roughness is an important parameter for describing the resistance of the water flow passing through the river channel, and the resistance of the water flow in the river channel can be quantitatively evaluated by calculating the river bed roughness value, so that the hydrodynamic characteristics of the water flow, including speed distribution, energy loss and the like, are known, and the river bed roughness value has important significance for the application of hydraulic analysis, river channel design, hydrologic simulation and the like. The river bed roughness calculation formula is established based on a physical process and a fluid mechanics principle, and the river bed surface roughness and water flow resistance are closely related, so that the processes of water flow transmission, sediment transport, suspended matter diffusion and the like in a river channel can be more accurately simulated, and the reliability and fidelity of river bed roughness calculation are improved. In addition, by introducing the correction value, the accuracy of calculation of the river bed roughness can be further improved, so that the result is closer to the actual situation. The formula fully considers the roughness value of the river bed of each combination of the saline tide hydrology Total depth of river bed and river channel>River bed roughness coefficient of river bed roughness +.>Water flow resistance influence coefficient of river bed roughness +.>Index-influencing adjustment parameter of river bed roughness +.>River bed water flow speed->Gravitational acceleration->River depth of riverway>Riverbed Euler normal height +.>Time-varying parameter of the Euler normal height of the riverbed +.>Correction value of the combined river bed roughness value of each salt tide hydrology +.>Combining river bed roughness values according to the hydrology of each salt tide>The interrelationship between the parameters constitutes a functional relationship:
；
the river bed roughness calculation formula can realize the river bed roughness calculation process of the related parameter data of the river bed roughness of each salty water text combination, and simultaneously, the correction value of the river bed roughness value is calculated through each salty water text combinationThe introduction of the formula can be adjusted according to actual conditions, so that the accuracy and applicability of a river bed roughness calculation formula are improved.
Preferably, step S4 comprises the steps of:
step S41: constructing a ConvLSTM salty tide prediction model by ConvLSTM, and carrying out self-adaptive attention adjustment treatment on the ConvLSTM salty tide prediction model by introducing a space-time attention mechanism according to a salty tide integer historical wind field set and a salty tide river bed roughness field data set to obtain a ConvLSTM salty tide strong attention prediction model;
According to the embodiment of the invention, a ConvLSTM salty tide prediction model is constructed by combining a ConvLSTM algorithm framework with a space-time modeling method of a convolutional neural network and a long-short-term memory network, a salty tide reorganization historical wind field set and a salty tide river bed roughness field data set are used as input data of the model, and the ConvLSTM salty tide prediction model is subjected to self-adaptive attention adjustment by introducing a space-time attention mechanism, so that the capturing capacity of the ConvLSTM salty tide prediction model on important space-time characteristics is improved, and finally the ConvLSTM salty tide strong attention prediction model is obtained.
Step S42: the hidden layers of the ConvLSTM salty tide intensity attention prediction model are subjected to cross-layer feature fusion connection, and the ConvLSTM salty tide intensity attention prediction model is subjected to migration learning and domain adaptation to optimize network weight parameters of the ConvLSTM salty tide intensity attention prediction model, so that a ConvLSTM salty tide prediction optimization model is obtained;
according to the embodiment of the invention, the characteristics of different hidden layers of the ConvLSTM salty tide intensity attention prediction model are subjected to cross-layer fusion, so that the effective capturing capability of the ConvLSTM salty tide intensity attention prediction model on different time scales and spatial characteristics is enhanced, more abundant space-time characteristics are extracted, the performance of the model is further optimized, meanwhile, the ConvLSTM salty tide intensity attention prediction model parameters are migrated to different data sets or environments to learn and improve the performance, domain adaptation is performed under the different data sets or environments, and the better generalization performance on a target domain is realized by modeling and adjusting the difference between a source domain and the target domain, so that the ConvLSTM salty tide prediction optimization model is finally obtained.
Step S43: carrying out time step stacking treatment on the salty tide reorganization historical wind field set and the salty tide riverbed roughness field data set to obtain a salty tide reorganization historical wind field time course characteristic diagram and a salty tide riverbed roughness field time course characteristic diagram;
according to the embodiment of the invention, the continuous salty tide reorganization historical wind field set and salty tide river bed roughness field data set are stacked in time sequence, wherein each time step represents the salty tide reorganization historical wind field set and salty tide river bed roughness field data set at one time point, and finally the salty tide reorganization historical wind field time course characteristic map and salty tide river bed roughness field time course characteristic map are obtained.
Step S44: carrying out time step module prediction processing on the salty tide reorganization historical wind field time course feature map through a ConvLSTM salty tide prediction optimization model to obtain salty tide reorganization historical wind field prediction results and wind field image channel numbers of each layer of time steps;
according to the embodiment of the invention, the salty tide reorganization historical wind field time course characteristic diagram is used as the input of a ConvLSTM salty tide prediction optimization model, a modularized function of the ConvLSTM salty tide prediction optimization model is utilized to divide and process time steps of each hidden layer, and prediction is carried out according to time sequence information of the salty tide reorganization historical wind field time course characteristic diagram, so that a salty tide reorganization historical wind field prediction result and a wind field image channel number of each layer of time steps are finally obtained.
Step S45: carrying out time step module prediction processing on the time course characteristic map of the salt tide river bed roughness field by using a ConvLSTM salt tide prediction optimization model to obtain a salt tide river bed roughness field prediction result and a river bed roughness field image channel number of each layer of time steps;
according to the embodiment of the invention, the time step module division processing is carried out on each hidden layer by taking the salty tide river bed roughness field time course characteristic diagram as the input of a ConvLSTM salty tide prediction optimization model and utilizing the modularized function of the ConvLSTM salty tide prediction optimization model, and the prediction is carried out according to the time sequence information of the salty tide river bed roughness field time course characteristic diagram, so that the salty tide river bed roughness field prediction result and the river bed roughness field image channel number of each layer of time step are finally obtained.
Step S46: and carrying out model target period integrated prediction on the salty tide integer historical wind field prediction result and the wind field image channel number of each layer of time steps and the salty tide river bed roughness field prediction result and the river bed roughness field image channel number of each layer of time steps to obtain a salty tide target period wind field data set and a salty tide target period roughness field data set.
According to the embodiment of the invention, the salty tide integer historical wind field prediction result and the wind field image channel number of each layer of time step are used as the input of the next layer to be integrated and fused in an iteration mode, target time period prediction is carried out on the salty tide integer historical wind field prediction result and the wind field image channel number to obtain a salty tide target time period wind field data set, and then the salty tide river bed roughness field prediction result and the river bed roughness field image channel number of each layer of time step are used as the input of the next layer to be integrated and fused in an iteration mode, and target time period prediction is carried out on the salty tide target time period roughness field data set.
The ConvLSTM salty tide prediction model is built by using ConvLSTM, and the ConvLSTM is a model framework combining a convolutional neural network (Convolutional Neural Network) and a Long Short-Term Memory network (Long Short-Term Memory). By constructing a ConvLSTM salt tide prediction model, the salt tide phenomenon can be predicted. In the training process, a salty tide reorganization historical wind field set and a salty tide riverbed roughness field data set are used as input, and a ConvLSTM network is used for learning the mode and the rule of space-time information. Meanwhile, in order to pay attention to important space-time characteristics in salt tide prediction better, a space-time attention mechanism is introduced to adaptively adjust the attention degree of a ConvLSTM salt tide prediction model, so that the capturing capacity of the model to the important space-time characteristics can be improved, and the accuracy and reliability of prediction are enhanced. Second, further optimization was performed by focusing on the ConvLSTM salt tide power prediction model. Firstly, integrating the features in the hidden layer by a cross-layer feature fusion connection mode so as to enhance the effective capturing and expressing capability of the model on different time scales and spatial features. And then optimizing the ConvLSTM salty tide intensity attention prediction model by using a migration learning and domain adaptation method. The transfer learning uses the trained model parameters as initial values, accelerates the model training process and improves the generalization capability of the model. Domain adaptation is used to address the adaptability problem of the model in different data sets or environments, thereby improving the effect and stability of the model in practical application. In addition, by performing time-step stacking processing on the salt tide reorganization history wind field set and the salt tide riverbed roughness field data set, the model is enabled to learn and utilize time sequence information in the time sequence data by taking relevant data of a plurality of continuous time steps as input. Through time step stacking processing, a time course characteristic map of a salt tide integer historical wind field and a time course characteristic map of a salt tide river bed roughness field can be generated, wherein the characteristic map at each moment represents the spatial distribution condition of the wind field or the river bed roughness field at the moment, and the time course characteristic map has time continuity for the input of a model and can better capture the time-space change rule of the salt tide. And then, carrying out time step module prediction processing on the salty tide reorganization historical wind field time course feature map by utilizing an optimized ConvLSTM salty tide prediction model, and predicting the wind field condition of the next time step according to the existing time sequence feature so as to accurately obtain salty tide reorganization historical wind field prediction results of all the time steps, wherein the wind field image channel numbers of different time steps are included, and the prediction results can reflect the time-space change feature of the salty tide reorganization historical wind field in a specific period and provide important basis for salty tide prediction. And the optimized ConvLSTM salt tide prediction model is used for carrying out time step module prediction processing on the salt tide river bed roughness field time course characteristic map, so that the river bed roughness field condition of the next time step can be predicted according to the existing time sequence characteristic. And predicting by a time step module to obtain a prediction result of the salt tide river bed roughness field and the number of image channels of the river bed roughness field in each time step, wherein the prediction results can reflect the time-space change characteristics of the salt tide river bed roughness field in a specific period, thereby providing an important basis for salt tide prediction. Finally, by carrying out integrated prediction on the prediction results of each time step, the prediction results of the model on the target period can be comprehensively considered, so as to obtain more accurate and stable prediction values. The wind field data set and the rough field data set of the salty tide target period can provide detailed time-space information, comprise wind field distribution and river bed rough characteristics of salty tide in a prediction period, and have important significance for salty tide prediction, analysis and decision making.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. The salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation is characterized by comprising the following steps of:
step S1: acquiring data of a salty tide wind field, and performing historical reorganization analysis on the data of the salty tide wind field to obtain a salty tide reorganization historical wind field set;
step S2: acquiring the water content data of the salty tide and the topography data of the salty tide, and carrying out salty tide simulation by utilizing the water content data of the salty tide and the topography data of the salty tide to construct a three-dimensional salty tide numerical model; performing adjustment verification on the three-dimensional salty tide numerical model to obtain a three-dimensional salty tide numerical optimization model;
Step S3: acquiring the salt tide hydrologic combination parameter data, and extracting and calculating the river bed roughness of the salt tide hydrologic combination parameter data to obtain a river bed roughness field of each salt tide hydrologic combination; performing time-by-time face interpolation processing on each salt tide hydrologic combined river bed roughness field to obtain a salt tide river bed roughness field data set;
step S4: deep learning prediction analysis is carried out on the salt tide reorganization historical wind field set and the salt tide riverbed roughness field data set through ConvLSTM, so that a salt tide target period wind field data set and a salt tide target period roughness field data set are obtained;
step S5: obtaining tide forecast outside sea tide level result data of a target forecast period, and performing three-dimensional salty tide numerical simulation on a salty tide target period wind field data set, a salty tide target period roughness field data set and tide forecast outside sea tide level result data of the target forecast period by using a three-dimensional salty tide numerical optimization model to obtain a three-dimensional salty tide forecast field data set; and performing corresponding salt tide forecasting work according to the three-dimensional salt tide forecasting field data set.
2. The method for salt tide prediction based on ConvLSTM and three-dimensional numerical modeling according to claim 1, wherein step S1 comprises the steps of:
step S11: acquiring data of a salty tide wind field;
Step S12: carrying out wind direction data extraction and wind speed data extraction on the salty tide wind field data to obtain salty tide wind direction data and salty tide wind speed data;
step S13: carrying out wind direction history backtracking analysis on the salty tide wind direction data to obtain salty tide historical wind direction data;
step S14: performing wind speed history backtracking analysis on the salty tide wind speed data to obtain salty tide history wind speed data;
step S15: and performing time alignment and interpolation reorganization processing on the salty tide historical wind direction data and the salty tide historical wind speed data to obtain a salty tide reorganization historical wind field set.
3. The method for salt tide prediction based on ConvLSTM and three-dimensional numerical modeling according to claim 2, wherein step S13 comprises the steps of:
step S131: frequency calculation is carried out on the salty tide wind direction data by utilizing a wind direction frequency calculation formula, so as to obtain a salty tide wind direction frequency value;
step S132: according to the salt tide wind direction frequency value, carrying out time change trend exploration on the salt tide wind direction data so as to generate a salt tide wind direction change mode graph;
step S133: detecting abnormal fluctuation of the graph of the wind direction change pattern of the salty tide to generate abnormal fluctuation points of the wind direction change of the salty tide;
step S134: performing abnormal correction processing on the abnormal fluctuation points of the change of the wind direction of the salty tide to obtain a correction chart of a change mode curve of the wind direction of the salty tide;
Step S135: and carrying out historical statistics retrospective analysis on the correction chart of the wind direction change pattern curve of the salty tide to obtain the historical wind direction data of the salty tide.
4. The method for forecasting salty tide based on ConvLSTM and three-dimensional numerical simulation according to claim 3, wherein the wind direction frequency calculation formula in the step S131 is specifically:
；
in the method, in the process of the invention,for the value of the wind direction frequency of salty tide, +.>The angle of the wind direction of salty tide>For the transverse spatial coordinates of the wind direction data of the salt tide in the frequency space, < >>Longitudinal spatial coordinates in frequency space for the data of the wind direction of the salt tide, < >>For the transverse wave vector in the frequency space of the salty-tide wind direction data, < >>Longitudinal wave vector in frequency space for the data of the wind direction of the salty tide, < >>Is the imaginary unit of frequency space phase +.>Wind-induced frequency components in the wind-direction data of salty tidesQuantity of->Is the +.f. in the wind direction data of salty tide>The fluctuation of the individual wind direction frequency components affects the amplitude, +.>Is the +.f. in the wind direction data of salty tide>The fluctuation of the individual wind direction frequency components affects the angular frequency, < >>Is the correction value of the wind direction frequency value of the salty tide.
5. The method for salt tide prediction based on ConvLSTM and three-dimensional numerical modeling according to claim 2, wherein step S14 comprises the steps of:
Step S141: performing climate change influence analysis on the salty tide wind speed data to obtain salty tide wind speed climate change influence factor data;
step S142: performing influence fluctuation restoration processing on the salty-tide wind speed data according to the salty-tide wind speed and climate change influence factor data to obtain salty-tide wind speed abnormality influence restoration data;
step S143: performing wind speed change pattern recognition analysis on the repair data of the abnormal influence of the wind speed of the salty tide to obtain data of a wind speed change pattern of the salty tide;
step S144: carrying out spatial radial analysis on the repair data of the abnormal influence of the wind speed of the salty tide to obtain spatial scale change mode data of the wind speed of the salty tide;
step S145: and carrying out historical trend backtracking analysis on the data of the air speed change mode of the salty tide and the data of the air speed space scale change mode of the salty tide to obtain the historical air speed data of the salty tide.
6. The method for salt tide prediction based on ConvLSTM and three-dimensional numerical modeling according to claim 1, wherein step S2 comprises the steps of:
step S21: acquiring water data of a salty tide and topography data of the salty tide;
step S22: carrying out salty tide simulation by using salty tide wind field data, salty tide hydrologic data and salty tide topography data to construct a three-dimensional salty tide numerical model;
Step S23: carrying out typical time sequence integration analysis on the data of the salty tide wind field, the salty tide hydrologic data and the salty tide topography data to obtain typical three-dimensional salty tide combination parameter data;
step S24: performing characteristic pattern analysis on the typical three-dimensional salty tide combination parameter data to obtain typical three-dimensional salty tide combination data characteristics;
step S25: and carrying out adjustment verification on the three-dimensional salty tide numerical model according to the characteristic of the typical three-dimensional salty tide combination data so as to obtain a three-dimensional salty tide numerical optimization model.
7. The method for salt tide prediction based on ConvLSTM and three-dimensional numerical modeling according to claim 6, wherein step S22 comprises the steps of:
step S221: performing anomaly filtering processing on the salty tide wind field data, the salty tide hydrologic data and the salty tide topographic data to obtain salty tide wind field anomaly processing data, salty tide hydrologic anomaly processing data and salty tide topographic anomaly processing data;
step S222: performing space-time interpolation on the salt-tide wind field anomaly handling data, the salt-tide hydrologic anomaly handling data and the salt-tide topography anomaly handling data to obtain salt-tide wind field interpolation data, salt-tide hydrologic interpolation data and salt-tide topography interpolation data;
step S223: coupling the salty tide wind field interpolation data, the salty tide hydrologic interpolation data and the salty tide topography interpolation data to construct initial conditions and boundary conditions, and obtaining the initial conditions and boundary conditions of the numerical model;
Step S224: and carrying out salt tide simulation on the initial conditions and the boundary conditions of the numerical model based on a finite difference method to construct a three-dimensional salt tide numerical model.
8. The method for salt tide prediction based on ConvLSTM and three-dimensional numerical modeling according to claim 1, wherein step S3 comprises the steps of:
step S31: acquiring the salt tide hydrologic combination parameter data, and extracting the river bed roughness parameters of the salt tide hydrologic combination parameter data to obtain the relevant parameter data of the river bed roughness of each salt tide hydrologic combination;
step S32: calculating the river bed roughness value of each salty water text combination river bed by using a river bed roughness calculation formula;
step S33: performing field source association processing on the river bed roughness values of the saline-water culture combinations and the corresponding space coordinates through a geographic information system to obtain river bed roughness fields of the saline-water culture combinations;
step S34: performing time-by-time face interpolation processing on each salt tide hydrologic combination river bed roughness field by using a time-by-time interpolation algorithm to obtain salt tide river bed roughness field interpolation result data;
step S35: and carrying out block sampling on the interpolation result data of the salty tide river bed roughness field to obtain a salty tide river bed roughness field data set.
9. The salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation according to claim 8, wherein the river bed roughness calculation formula in step S32 is specifically:
；
in the method, in the process of the invention,combining the river bed roughness values for each salt tide hydrology,/->Is the total depth of the riverway of the riverbed>River bed roughness coefficient as the river bed roughness rate, < ->Water flow resistance influence coefficient for river bed roughness, < ->Adjusting parameters for the exponential influence of the river bed roughness, < ->For the water flow speed of the river bed, < > is->Acceleration of gravity, ++>Is the depth of river water of the river bed>Is the Euler normal height of the riverbed, < >>Time-varying parameter for Euler normal height of river bed,/->Correction values for the river bed roughness values are combined for each salt tide hydrology.
10. The method for salt tide prediction based on ConvLSTM and three-dimensional numerical modeling according to claim 1, wherein step S4 comprises the steps of:
step S41: constructing a ConvLSTM salty tide prediction model by ConvLSTM, and carrying out self-adaptive attention adjustment treatment on the ConvLSTM salty tide prediction model by introducing a space-time attention mechanism according to a salty tide integer historical wind field set and a salty tide river bed roughness field data set to obtain a ConvLSTM salty tide strong attention prediction model;
step S42: the hidden layers of the ConvLSTM salty tide intensity attention prediction model are subjected to cross-layer feature fusion connection, and the ConvLSTM salty tide intensity attention prediction model is subjected to migration learning and domain adaptation to optimize network weight parameters of the ConvLSTM salty tide intensity attention prediction model, so that a ConvLSTM salty tide prediction optimization model is obtained;
Step S43: carrying out time step stacking treatment on the salty tide reorganization historical wind field set and the salty tide riverbed roughness field data set to obtain a salty tide reorganization historical wind field time course characteristic diagram and a salty tide riverbed roughness field time course characteristic diagram;
step S44: carrying out time step module prediction processing on the salty tide reorganization historical wind field time course feature map through a ConvLSTM salty tide prediction optimization model to obtain salty tide reorganization historical wind field prediction results and wind field image channel numbers of each layer of time steps;
step S45: carrying out time step module prediction processing on the time course characteristic map of the salt tide river bed roughness field by using a ConvLSTM salt tide prediction optimization model to obtain a salt tide river bed roughness field prediction result and a river bed roughness field image channel number of each layer of time steps;
step S46: and carrying out model target period integrated prediction on the salty tide integer historical wind field prediction result and the wind field image channel number of each layer of time steps and the salty tide river bed roughness field prediction result and the river bed roughness field image channel number of each layer of time steps to obtain a salty tide target period wind field data set and a salty tide target period roughness field data set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311345170.9A CN117113854B (en) | 2023-10-18 | 2023-10-18 | Salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311345170.9A CN117113854B (en) | 2023-10-18 | 2023-10-18 | Salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117113854A CN117113854A (en) | 2023-11-24 |
| CN117113854B true CN117113854B (en) | 2024-02-02 |
Family
ID=88809350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311345170.9A Active CN117113854B (en) | 2023-10-18 | 2023-10-18 | Salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117113854B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117709242B (en) * | 2024-02-05 | 2024-04-09 | 珠江水利委员会珠江水利科学研究院 | Estuary salty tide tracing strength evaluation method |
| CN118312775A (en) * | 2024-04-01 | 2024-07-09 | 华南农业大学 | Salt tide prediction method based on LSTM-GRU-CNN integrated model |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107992960A (en) * | 2017-11-03 | 2018-05-04 | 中山大学 | Hekou District salty tide traces back chlorosity Simulation prediction method |
| CN109086933A (en) * | 2018-08-03 | 2018-12-25 | 中山大学 | A kind of salty tide Medium-long Term Prediction method |
| WO2019200743A1 (en) * | 2018-04-17 | 2019-10-24 | 平安科技（深圳）有限公司 | Forest fire prediction method, apparatus, computer device, and storage medium |
| AU2020102630A4 (en) * | 2019-11-01 | 2021-01-21 | North China Electric Power University | Method for acquiring roughness of river channels under gradually-varied flow conditions |
| CN113887819A (en) * | 2021-10-20 | 2022-01-04 | 珠江水资源保护科学研究所 | Method and equipment for forecasting salt tide in estuary dry season |
| CN114862015A (en) * | 2022-04-29 | 2022-08-05 | 中国科学院南海海洋研究所 | Typhoon wind speed intelligent prediction method based on FEN-ConvLSTM model |
| CN116776539A (en) * | 2023-05-09 | 2023-09-19 | 珠江水利委员会珠江水利科学研究院 | Salt tide forecasting method and system based on cross wavelet analysis |
-
2023
- 2023-10-18 CN CN202311345170.9A patent/CN117113854B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107992960A (en) * | 2017-11-03 | 2018-05-04 | 中山大学 | Hekou District salty tide traces back chlorosity Simulation prediction method |
| WO2019200743A1 (en) * | 2018-04-17 | 2019-10-24 | 平安科技（深圳）有限公司 | Forest fire prediction method, apparatus, computer device, and storage medium |
| CN109086933A (en) * | 2018-08-03 | 2018-12-25 | 中山大学 | A kind of salty tide Medium-long Term Prediction method |
| AU2020102630A4 (en) * | 2019-11-01 | 2021-01-21 | North China Electric Power University | Method for acquiring roughness of river channels under gradually-varied flow conditions |
| CN113887819A (en) * | 2021-10-20 | 2022-01-04 | 珠江水资源保护科学研究所 | Method and equipment for forecasting salt tide in estuary dry season |
| CN114862015A (en) * | 2022-04-29 | 2022-08-05 | 中国科学院南海海洋研究所 | Typhoon wind speed intelligent prediction method based on FEN-ConvLSTM model |
| CN116776539A (en) * | 2023-05-09 | 2023-09-19 | 珠江水利委员会珠江水利科学研究院 | Salt tide forecasting method and system based on cross wavelet analysis |
Non-Patent Citations (1)
| Title |
|---|
| 长江口咸潮入侵预报系统的设计与应用;陶英佳等;华东师范大学学报(自然科学版)(02);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117113854A (en) | 2023-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117113854B (en) | Salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation | |
| Plant et al. | Prediction and assimilation of surf-zone processes using a Bayesian network: Part I: Forward models | |
| CN112099110B (en) | Ocean internal wave forecasting method based on machine learning and remote sensing data | |
| CN110580328B (en) | Method for repairing underground water level monitoring value loss | |
| Stojanovic et al. | A self-tuning system for dam behavior modeling based on evolving artificial neural networks | |
| Shi et al. | Probabilistic real-time deep-water natural gas hydrate dispersion modeling by using a novel hybrid deep learning approach | |
| CN109636171A (en) | A kind of comprehensive diagnos and risk evaluating method that regional vegetation restores | |
| CN105546352A (en) | Natural gas pipeline tiny leakage detection method based on sound signals | |
| CN103198215A (en) | Method for predicting subsidence of soil outside pit based on differential evolution support vector machine | |
| CN117077554B (en) | Three-dimensional salty tide forecasting method based on ConvGRU | |
| CN118152990B (en) | Online current measurement system for hydrologic tower | |
| Wang et al. | Integration of data-driven and physics-based modeling of wind waves in a shallow estuary | |
| CN115471679A (en) | Method and intelligent system for synchronously assimilating water level and flow of natural river | |
| Ehlers et al. | Using a simple post-processor to predict residual uncertainty for multiple hydrological model outputs | |
| Khairudin et al. | In-Depth review on machine learning models for long-term flood forecasting | |
| CN117520718B (en) | Tidal river hydrologic data processing method and system | |
| Rajabi-Kiasari et al. | Forecasting of absolute dynamic topography using deep learning algorithm with application to the Baltic Sea | |
| Wu | A Novel Artificial Neural Network Ensemble Model Based on K--Nearest Neighbor Nonparametric Estimation of Regression Function and Its Application for Rainfall Forecasting | |
| Al Mehedi et al. | Unraveling The Complexities of Urban Flood Hydraulics Through AI | |
| Nawab et al. | Comparison of ANN global horizontal irradiation predictions with satellite global horizontal irradiation using statistical evaluation | |
| HADJI | A coupled models Hydrodynamics-Multi headed Deep convolutional neural network for rapid forecasting large-scale flood inundation | |
| Xin et al. | Deep-learning approach based on multi-data fusion for damage recognition of marine platforms under complex loads | |
| CN116204836B (en) | Method for screening suitable forest areas of mangrove forest on estuary beach | |
| Du et al. | Neural network models for seabed stability: a deep learning approach to wave-induced pore pressure prediction | |
| CN117634325B (en) | Method and system for identifying extremum event of data-limited estuary area and analyzing composite flood disasters |
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 |