CN118567004B

CN118567004B - A rapid forecasting method for storm surge water increase

Info

Publication number: CN118567004B
Application number: CN202411039544.9A
Authority: CN
Inventors: 陈鹏; 王永勇; 郑庆涛; 黑亮; 马筠; 周舜轩; 黄依之; 林芯伊; 刘珊霞
Original assignee: Zhujiang Hydrology Water Resource Exploration Center
Current assignee: Zhujiang Hydrology Water Resource Exploration Center
Priority date: 2024-07-31
Filing date: 2024-07-31
Publication date: 2024-11-12
Anticipated expiration: 2044-07-31
Also published as: CN118567004A

Abstract

The invention relates to the technical field of meteorology, in particular to a storm surge water increasing rapid forecasting method. The method comprises the following steps: collecting historical observation storm surge data and hydrologic data; carrying out data cleaning on the hydrologic data to generate storm weather basic data; performing difference correction processing on storm weather basic data to generate storm tide correction data; trend analysis is carried out on storm surge correction data, and first class of processing data is generated; establishing a direct empirical relation of a specific place for the first type of processing data by using a fitting curve empirical statistical method to obtain a statistical empirical formula; and establishing a simple prediction model according to the statistical empirical formula to generate the simple prediction model. The method improves storm tide prediction and analysis more comprehensively and accurately by integrating numerical simulation, data fusion, model optimization, data science and statistics, and is beneficial to improving the efficiency and level of disaster prevention and reduction work.

Description

Storm surge water increasing quick forecasting method

Technical Field

The invention relates to the technical field of meteorology, in particular to a storm surge water increasing rapid forecasting method.

Background

Storm surge model design is a comprehensive technical task and relates to a plurality of disciplines and technical means. In the field of meteorology, it is necessary to observe and analyze storm systems by using meteorological data including parameters such as wind speed, wind direction, air pressure, etc. so as to understand the formation and evolution rules thereof. In the aspect of oceanography, ocean dynamics and tide theory need to be researched, the influence of ocean environment on storm tide is understood, and a foundation is provided for model design. Numerical simulation is an important means for designing storm surge model, and simulation and prediction are carried out on the storm surge generation and propagation process by establishing a mathematical model and applying a numerical calculation method. In the aspect of data processing and analysis, measured data needs to be processed and analyzed, including data cleaning, interpolation processing and the like, and the data is analyzed and modeled by using a statistical method and a mathematical model. In addition, the technology of computer science and software engineering plays a key role in the design of storm surge models, including developing and realizing the computer software of storm surge models, writing programs, designing algorithms, constructing user interfaces and the like. In summary, storm surge model design is a complex and diversified technical task, and needs to span multiple discipline fields and integrate various technical means to realize accurate prediction and effective management of storm surge. However, the conventional storm surge prediction model technology still has the defects of incomplete data, insufficient model, insufficient uncertainty estimation, insufficient emergency processing, lack of comprehensive consideration and the like, so that the prediction accuracy and reliability are lower.

Disclosure of Invention

Based on this, it is necessary to provide a rapid forecasting method for storm surge and water increase, so as to solve at least one of the above technical problems.

In order to achieve the above purpose, a storm surge water increasing rapid forecasting method comprises the following steps:

Step S1: collecting historical observation storm surge data and hydrologic data; carrying out data preprocessing on the hydrologic data to generate storm weather basic data;

Step S2: performing difference correction processing on the historical observed storm surge data and storm weather base data to generate storm surge correction data; trend analysis is carried out on storm surge correction data, and first class of processing data is generated; establishing a direct experience relation of a specific place for the first type of processing data by a fitting curve experience statistical method to obtain a statistical experience formula; establishing a simple prediction model according to a statistical empirical formula to generate the simple prediction model; carrying out storm surge weather prediction on storm surge based on a simple prediction model to obtain a direct experience relation prediction result;

Step S3: acquiring position information data of a specific place; discretizing a specific place by utilizing historical observation storm surge data and storm weather basic data to generate a storm surge grid structure diagram; carrying out Cartesian coordinate system boundary processing on the storm tide grid structure diagram to judge storm intensity, and generating a storm tide judgment boundary condition result; carrying out numerical solution on storm surge grid structure drawing and storm surge judgment boundary condition results by utilizing FVCOM ocean modes to obtain second-class processing data; carrying out storm tide model construction processing based on the second class of processing data to generate a numerical simulation model; carrying out storm tide prediction on the numerical simulation model to obtain a numerical simulation prediction result;

Step S4: carrying out weight fusion on the direct experience relation prediction result and the numerical simulation prediction result to obtain a storm surge fusion model; carrying out harmonic analysis according to the measured tide level data to obtain a harmonic constant; and carrying out feedback optimization on the storm surge fusion model based on the harmonic constant to generate a storm surge water increasing forecast model.

According to the invention, firstly, reliable basic data is obtained through collection and cleaning treatment of historical observation data and hydrological data. And then, correction data are generated by using technical means such as difference correction, trend analysis and the like, and a direct experience relation of a specific place is established by using an experience statistical method of a fitting curve, so that a simple prediction model is established for predicting storm surge condition. On this basis, discretization processing is performed on a specific place, a grid structure diagram is generated, and boundary conditions are set so as to perform numerical simulation. And predicting storm surge by using a numerical simulation model to obtain a numerical simulation result. And then, carrying out weight fusion on the direct experience relation prediction result and the numerical simulation prediction result to generate a fusion model so as to improve the accuracy and reliability of prediction. And carrying out harmonic analysis according to the measured tide level data to obtain a harmonic constant, and further optimizing the model. Finally, a storm surge prediction model is generated through model optimization, and a more accurate and reliable technical means is provided for storm surge prediction. The method improves the accuracy and reliability of storm surge prediction by utilizing a plurality of technical means such as data cleaning, trend analysis, numerical simulation, model fusion and the like.

The method has the beneficial effects that the development and evolution process of storm surge can be more comprehensively understood and predicted by collecting, cleaning and analyzing historical observation data and hydrologic data and utilizing a numerical simulation model and a model fusion technology. Firstly, the comprehensive analysis of the historical observation data and the hydrological data provides a reliable data basis for establishing a storm surge prediction model, so that a model prediction result is more accurate and reliable. And secondly, predicting storm tide by using a numerical simulation model, and combining a model fusion technology to fuse prediction results of different sources, so that the influence of various factors on the storm tide is considered, and the comprehensiveness and accuracy of the prediction are improved. And particularly, a direct empirical relation and a fusion model are established, so that the prediction result is more reliable, and the actual situation can be better reflected. In addition, the precision and stability of the prediction model are further improved through determination of the harmonic constant and model optimization, so that the prediction result is more reliable. Therefore, the method for comprehensively utilizing various technical means can effectively improve the accuracy of storm surge prediction, and provide more reliable early warning and forecast information for the risk brought by storm surge, thereby guaranteeing the safety and benefits of the public and related departments.

Drawings

FIG. 1 is a flow chart of the steps of a method for rapidly forecasting the water increase of storm surge;

FIG. 2 is a flowchart illustrating the detailed implementation of step S4 in FIG. 1;

FIG. 3 is a detailed flowchart illustrating the implementation of step S41 in FIG. 2;

the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

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.

To achieve the above objective, please refer to fig. 1 to 3, a method for rapidly forecasting storm surge, the method comprising the following steps:

Step S2: performing difference correction processing on storm weather basic data to generate storm tide correction data; trend analysis is carried out on storm surge correction data, and first class of processing data is generated; establishing a direct empirical relation of a specific place for the first type of processing data by using a fitting curve empirical statistical method to obtain a statistical empirical formula; establishing a simple prediction model according to a statistical empirical formula to generate the simple prediction model; carrying out storm surge meteorological prediction on storm surge meteorological according to a simple prediction model to obtain a direct experience relation prediction result;

According to the method, storm weather basic data are generated by collecting and preprocessing historical observation storm tide water increasing data and hydrological data, and difference value correction processing is carried out on the storm weather basic data, so that the integrity and consistency of the data are ensured. And generating first class processing data through trend analysis, and establishing a direct experience relation of a specific place by using a fitting curve experience statistical method, so as to construct a simple prediction model, and carrying out preliminary storm surge weather prediction to obtain a direct experience relation prediction result. And (3) performing discretization processing on the specific places by acquiring position information data of the specific places and combining historical observation data and storm weather basic data to generate a storm tide grid structure diagram. The grid structure diagram is subjected to Cartesian coordinate system boundary processing to judge storm intensity and generate storm surge judgment boundary condition results. And carrying out numerical solution on a storm surge grid structure diagram and a storm surge judgment boundary condition result by utilizing FVCOM ocean modes to obtain second-class processing data, and carrying out storm surge model construction and prediction based on the second-class processing data to generate a numerical simulation prediction result. On the basis, the direct experience relation prediction result and the numerical simulation prediction result are subjected to weight fusion to form a storm surge fusion model, so that the advantages of the two prediction methods are fully utilized, and the accuracy and the robustness of the prediction are improved. And finally, carrying out reconciliation analysis according to the actually measured tide level data to obtain a reconciliation constant, and carrying out feedback optimization on the storm tide fusion model based on the reconciliation constant to generate a storm tide water increasing forecast model. The flow not only fully utilizes the historical data and the existing meteorological and hydrological models, but also remarkably improves the accuracy and reliability of storm surge prediction through multi-level and multi-step numerical processing and model optimization, and provides powerful technical support for disaster early warning and disaster prevention and reduction.

In the embodiment of the present invention, as described with reference to fig. 1, the step flow diagram of a storm surge rapid forecasting method of the present invention is provided, and in this example, the storm surge rapid forecasting method includes the following steps:

In the embodiment of the invention, the historical observation storm surge data and the hydrologic data are collected by collecting the marine observation station, the hydrologic observation station and the meteorological observation station facilities. Data cleansing is then required, and includes removing errors, deletions or outliers, correcting the data format, and processing repeated data operations to ensure accuracy and integrity of the data. The cleaned data is storm weather basic data, and can be used for subsequent analysis and processing.

Step S2: performing difference correction processing on the historical observed storm surge data and storm weather base data to generate storm surge correction data; trend analysis is carried out on storm surge correction data, and first class of processing data is generated; establishing a direct experience relation of a specific place for the first type of processing data by a fitting curve experience statistical method to obtain a statistical experience formula; establishing a prediction model according to a statistical empirical formula to generate a direct empirical relation prediction model; carrying out storm surge weather prediction on storm surge based on the direct experience relation prediction model to obtain a direct experience relation prediction result;

In the embodiment of the invention, the storm weather basic data is processed by a difference correction method to generate storm tide correction data, and trend analysis is performed on the correction data to generate first-class processing data. And then, analyzing the first type of processing data by using a fitting curve empirical statistical method, and establishing a direct empirical relation of a specific place to obtain a statistical empirical formula. And then, a simple prediction model is established according to a statistical empirical formula, and the storm surge weather is predicted by using the prediction model to obtain a direct empirical relation prediction result.

in the embodiment of the invention, the position information data of the specific place is firstly acquired, and the accuracy and the integrity of the data are ensured. And discretizing the specific place by utilizing the historical observation storm surge data and storm weather basic data. The discretization process adopts an Arakawa B horizontal process and a vertical coordinate process method, and a storm tide grid structure diagram is generated by discretizing continuous data into grid points. These grid structure diagrams can reflect the spatial distribution and dynamic changes of storm surge in detail. And carrying out Cartesian coordinate system boundary processing on the storm surge grid structure diagram, and generating a storm surge judgment boundary condition result by evaluating the storm surge intensity of each grid point. Boundary condition results are divided into high risk and low risk regions, and decisions are made based on comprehensive assessment of historical data and weather conditions. And then carrying out numerical solution on storm surge grid structure diagram and storm surge judgment boundary condition results by utilizing FVCOM (Finite Volume Coastal Ocean Model) ocean modes. FVCOM is a high resolution ocean model that accurately simulates tidal and storm surge phenomena in offshore and coastal areas. In the numerical solution process, a momentum equation, a continuous equation and a state equation are applied, and the fluid dynamics process is simulated through a non-static pressure approximation method and a high-order differential format to obtain second-class processing data. Based on the second class of processing data, storm tide model construction processing is carried out, and a numerical simulation model capable of accurately reflecting the dynamic change of storm tide is constructed. And finally, predicting future storm surge based on the constructed numerical simulation model to obtain a numerical simulation prediction result.

In the embodiment of the invention, the direct experience relation prediction result and the numerical simulation prediction result are subjected to weight fusion, a weighted average and regression analysis method is used, and the weight value is determined according to the historical observation data and the model performance evaluation result. And then, carrying out harmonic analysis on the hydrologic data according to the fusion model, processing the hydrologic data by using a statistical analysis method, and carrying out analysis by combining the result of the fusion model. The purpose of the harmonic analysis is to determine the harmonic constants that are used to adjust the fusion model to make it more realistic. Finally, carrying out feedback optimization on the storm surge fusion model based on the harmonic constant, and optimizing the model by using a genetic algorithm and a simulated annealing algorithm technology so as to improve the prediction precision and applicability of the model. The objective of the optimization is to generate a storm surge prediction model that can be used to predict future storm surge changes.

Preferably, step S1 comprises the steps of:

step S11: acquiring historical observation storm surge data;

step S12: obtaining measured tide level data through a tide station and a hydrological observation station;

step S13: acquiring meteorological data through a meteorological observation station and satellite remote sensing, wherein the meteorological data comprises water depth, air pressure, wind speed and wind direction and typhoon moving paths;

Step S14: acquiring astronomical tide meter data through a marine meteorological department;

Step S15: data integration is carried out on the actually measured tide level data, the meteorological data and the astronomical tide meter data to obtain hydrological data;

Step S16: and (5) carrying out data cleaning on the hydrologic data to generate storm weather basic data.

According to the invention, through obtaining various data dimensions, such as the real tide level, weather, typhoon moving path and astronomical tide table data, not only are data sources enriched, but also a more comprehensive and finer prediction model is facilitated to be established. By comprehensively utilizing various data sources, the accuracy and reliability of storm surge prediction can be improved, so that complex marine meteorological environments can be better dealt with. In addition, by establishing a basic model, a reliable basis is provided for subsequent numerical simulation and prediction, and important support is provided for more accurate storm surge water increase prediction.

In the embodiment of the invention, the acquisition of the historical observed storm surge moisturizing data mainly involves the utilization of various meteorological and hydrological observation equipment, such as tidal stations, hydrological observation stations, meteorological observation stations, satellite remote sensing and the like, so as to acquire related data. The tide station and the hydrological observation station continuously record ocean water level change data through devices such as a tide level meter, a water level meter and the like; the meteorological observation station records meteorological element data in real time through barometer, anemometer, anemoscope and other devices; and satellite remote sensing obtains large-scale meteorological data through a satellite sensor. The acquired historical observation data needs to be subjected to quality control and correction, and the accuracy and reliability of the data are ensured. When processing the measured tide level data, quality control is firstly needed to be carried out on the data, including checking whether abnormal values or missing values exist in the data or not, and corresponding data correction is carried out. This involves the technical means of data interpolation, outlier processing, etc. And then, analyzing the data to extract periodic variation characteristics of the tide so as to facilitate subsequent data processing and model establishment. The processing of the meteorological data comprises the steps of data cleaning, preprocessing, quality control and the like. The cleaning mainly removes noise and abnormal values in the data, and the preprocessing includes data interpolation, smoothing processing and the like so that the data can be effectively utilized by the model. The quality control ensures the accuracy and the integrity of the data, and can be detected and corrected by adopting a statistical method or a professional algorithm. Acquisition and processing of astronomical tidal table data involves acquiring data files from the marine meteorological department or related institution and parsing and sorting. For data in the astronomical tidal table, interpolation, fitting or extrapolation is required to accommodate the needs of the predictive model. And finally, integrating and cleaning the processed actually measured tide level data, meteorological data and astronomical tide table data to generate storm meteorological basic data. The process of integration involves techniques of data format conversion, data alignment, data matching, etc. to ensure consistency and availability of data. Meanwhile, quality inspection is required to be carried out on the data, so that the generated basic data can meet the requirements of a follow-up storm surge prediction model.

Preferably, step S2 comprises the steps of:

Step S21: performing difference correction processing on storm weather basic data to generate storm tide correction data;

Trend analysis is carried out on storm surge correction data to generate first-class processing data, wherein the trend analysis comprises denoising and smoothing processing, application time sequence analysis, regression analysis and spectrum analysis; denoising and smoothing the storm surge correction data by using an exponential smoothing method to generate storm surge smoothing data; the storm surge correction data is used as time series data to be input into an MA moving average model, so as to generate storm surge MA data; carrying out meteorological variable analysis on storm surge correction data by utilizing a multivariate analysis method to generate storm surge multivariate data; carrying out frequency domain analysis on storm surge correction data by utilizing Fourier transformation to generate storm surge frequency domain data; carrying out data merging processing on storm tide smoothing data, storm tide MA data, storm tide multivariate data and storm tide frequency domain data to generate a first class processing model;

Step S23: the method comprises the steps of establishing a direct empirical relation in a specific place on first class of processing data through a fitting curve empirical statistical method to obtain a statistical empirical formula;

step S24: establishing a prediction model according to a statistical empirical formula to generate a direct empirical relation prediction model;

Step S25: and carrying out storm tide weather prediction on the storm tide based on the direct experience relation prediction model to obtain a direct experience relation prediction result.

According to the method, through detailed description of trend analysis processes of storm surge correction data, including denoising and smoothing, time sequence analysis, regression analysis, spectrum analysis and other methods, accuracy and comprehensiveness of data processing are improved, and further, a direct experience relation of a specific place is established through fitting curve experience statistical method to form a prediction model, so that accuracy and reliability of storm surge weather prediction are improved effectively.

In the embodiment of the invention, the accuracy and reliability of storm surge weather prediction are improved through detailed processing and analysis steps. Performing difference correction processing on storm weather basic data to generate storm tide correction data; denoising and smoothing the storm surge correction data by an exponential smoothing method to generate storm surge smoothing data; the storm surge correction data is used as time series data to be input into an MA moving average model, so as to generate storm surge MA data; carrying out meteorological variable analysis on storm surge correction data by utilizing a multiple regression analysis method to generate storm surge multiple data; and carrying out frequency domain analysis on the storm surge correction data through Fourier transformation to generate storm surge frequency domain data. And carrying out data merging processing on the storm surge smoothing data, the storm surge MA data, the storm surge multivariate data and the storm surge frequency domain data to generate a first type of processing model. And then, establishing a direct empirical relation of the first type of processing data in a specific place by using a fitting curve empirical statistical method to obtain a statistical empirical formula. And finally, establishing a prediction model according to a statistical empirical formula, generating a direct empirical relation prediction model, and performing weather prediction on storm surge based on the model to obtain a direct empirical relation prediction result. Through the steps, advanced data processing technology and analysis methods are utilized, including difference correction, exponential smoothing, time sequence analysis, regression analysis and spectrum analysis, so that the accuracy and the comprehensiveness of data processing are improved, and the performance and the reliability of a storm surge weather prediction model are also remarkably improved.

Preferably, step S3 comprises the steps of:

Step S31: acquiring position information data of a specific place;

Step S32: performing discretization processing on the position information data of the specific place by utilizing the historical observation storm surge data and storm weather basic data to generate a storm surge grid structure diagram, wherein the discretization processing comprises Arakawa B horizontal processing and vertical coordinate processing;

Step S33: carrying out Cartesian coordinate system boundary judgment on storm intensity on a storm surge grid structure chart based on a preset storm surge boundary judgment condition to generate a storm surge judgment condition result, wherein the storm surge judgment condition result is divided into a high risk area and a low risk area; when the judging result is a high-risk area, performing smaller time step numerical simulation storm precision resolution on the high-risk area to generate high-risk area simulation data; when the judging result is a low-risk area, carrying out storm surge limited volume average characteristic analysis through storm surge water increasing data to generate low-risk area simulation data;

Step S34: performing numerical solution on the high-risk area simulation data and the low-risk area simulation data by utilizing FVCOM ocean modes to obtain second-class processing data;

step S35: performing model construction processing based on the second class of processing data to generate a numerical simulation model;

step S36: and importing the second type of processing data into a numerical simulation model to conduct storm surge prediction, and obtaining a numerical simulation prediction result.

The method and the device ensure the accuracy and the integrity of the geographic information by acquiring the position information data of the specific place. And discretizing the position information data of the specific place by utilizing the historical observation storm surge data and storm weather basic data. The discretization process adopts the Arakawa B horizontal processing and vertical coordinate processing method, and generates a detailed storm surge grid structure diagram by discretizing continuous data into grid points. The gridding data representation method can effectively capture the spatial distribution and dynamic change characteristics of storm surge. Based on the preset storm tide boundary judgment conditions, boundary judgment is carried out on a storm tide grid structure diagram by using a Cartesian coordinate system, storm intensity of each grid point is estimated, and therefore storm tide judgment condition results are generated. Based on these results, the regions are divided into high risk regions and low risk regions. Performing numerical simulation of smaller time steps for the high-risk area to improve storm surge accuracy resolution and generate high-risk area simulation data; and for the low-risk area, performing storm surge finite volume average characteristic analysis through storm surge moisturizing data to generate low-risk area simulation data. And carrying out numerical solution on the high-risk area simulation data and the low-risk area simulation data by utilizing FVCOM (Finite Volume Coastal Ocean Model) marine modes. FVCOM is a high resolution ocean model that accurately simulates tidal and storm surge phenomena in offshore and coastal areas. In the numerical solution process, a momentum equation, a continuous equation and a state equation are applied, and the fluid dynamics process is simulated through a non-static approximation method and a high-order differential format to generate second-class processing data. Based on the data, construction processing of a storm surge model is carried out, and a numerical simulation model capable of accurately reflecting the dynamic change of the storm surge is generated. And finally, importing the second type of processing data into a numerical simulation model to conduct storm surge prediction, and obtaining a numerical simulation prediction result. The complete technical flow is from data acquisition, discretization processing, boundary judgment and numerical solution to model construction and prediction through a systematic and scientific processing method, not only makes full use of historical data and the existing meteorological and hydrological models, but also combines an advanced numerical simulation technology, remarkably improves the accuracy and reliability of storm surge prediction, and provides scientific basis and technical support for disaster early warning and disaster prevention and reduction. The method has the effectiveness of comprehensively processing the data and accurately simulating the complex physical process, and finally realizes the high-precision prediction and evaluation of the dynamic change of storm surge.

In the embodiment of the invention, firstly, the position information data of a specific place is acquired, and the data geographic positioning accuracy is ensured. And discretizing the position information data of the specific place by utilizing the historical observation storm surge data and storm weather basic data. The discretization processing adopts an Arakawa B horizontal processing method and a vertical coordinate processing method, and the continuous data are discretized into grid points to generate a storm surge grid structure diagram, so that the spatial distribution and dynamic change of storm surge can be reflected in detail. And carrying out boundary judgment on a storm surge grid structure diagram by using a Cartesian coordinate system based on a preset storm surge boundary judgment condition, and evaluating storm intensity of each grid point so as to generate a storm surge judgment condition result. According to the determination result, the region is divided into a high risk region and a low risk region. Performing numerical simulation of smaller time steps for the high-risk area to improve storm surge accuracy resolution and generate high-risk area simulation data; and for the low-risk area, performing storm surge finite volume average characteristic analysis through storm surge moisturizing data to generate low-risk area simulation data. And carrying out numerical solution on the high-risk area simulation data and the low-risk area simulation data by utilizing FVCOM (Finite Volume Coastal Ocean Model) marine modes. FVCOM is a high resolution ocean model that accurately simulates tidal and storm surge phenomena in offshore and coastal areas. In the numerical solution process, a momentum equation, a continuous equation and a state equation are applied, and the fluid dynamics process is simulated through a non-static approximation method and a high-order differential format to generate second-class processing data. Based on the data, construction processing of a storm surge model is carried out, and a numerical simulation model capable of accurately reflecting the dynamic change of the storm surge is generated. And finally, importing the second type of processing data into a numerical simulation model to conduct storm surge prediction, and obtaining a numerical simulation prediction result. The series of steps form a complete technical flow from data acquisition, discretization processing, boundary judgment, numerical solution to model construction and prediction through a systematic and scientific processing method.

Preferably, step S32 includes the steps of:

Step S321: carrying out direction analysis on the position information data of the specific place to obtain horizontal direction data of the specific place and vertical direction data of the specific place;

step S322: carrying out Arakawa B level processing on the horizontal direction data of the specific place by utilizing storm weather basic data to generate grid level data;

Step S323: performing vertical coordinate discretization processing on the vertical direction data of the specific place by utilizing storm weather basic data to generate grid vertical data;

step S324: performing data integration on the grid horizontal data and the grid vertical data to obtain grid integrated data;

step S325: and carrying out interpolation processing on the grid integrated data to generate a storm tide grid structure diagram.

The invention fully utilizes various data sources by comprehensively utilizing the technologies of direction analysis, difference value processing, data integration, interpolation and the like, and improves the accuracy and the reliability of the data. Meanwhile, the generated storm surge grid structure diagram is used as the input of a numerical simulation model, so that the accuracy and the reliability of the model are improved, and the storm surge condition can be predicted more accurately. The application of the comprehensive technology also saves the time and cost for processing the data, and accelerates the model establishment and the generation speed of the prediction result.

In an embodiment of the invention, the data in the horizontal and vertical directions are determined by performing a direction analysis on the position information of a specific place by using a Geographic Information System (GIS) technology. Then, the Arakawa B horizontal processing is adopted, the earth surface is divided into grids by the processing method, and the horizontal direction data of a specific place is processed by adopting a specific C-shaped grid layout, so that the stability and the accuracy of numerical calculation are ensured. And meanwhile, vertical coordinate discretization processing, such as isobaric plane coordinates and depth coordinates, is adopted to process vertical direction data of a specific place so as to adapt to the requirements of a numerical simulation model. And integrating the grid data processed in the horizontal and vertical directions, and integrating and combining the data by utilizing data processing software so as to facilitate subsequent processing and analysis. And finally, processing the integrated grid data by adopting an interpolation technology, and using linear interpolation, bilinear interpolation and cubic spline interpolation methods to fill the blank area among the data and generate a storm surge grid structure diagram.

Preferably, step S34 includes the steps of:

step S341: importing FVCOM the high-risk region simulation data and the low-risk region simulation data into a marine mode, and carrying out numerical solution through a momentum equation, a continuous equation and a state equation to generate high-risk region fine data and low-risk region fine data;

step S342: performing storm tide high-level difference format optimization based on the high-risk area fine data and the low-risk area fine data to generate storm tide high-level difference data; carrying out high-grid resolution fluid dynamics setting on storm tide high-level difference data to generate storm tide grid resolution data;

step S343: carrying out FVCOM ocean mode resolution analysis on storm tide grid resolution data to generate numerical model data;

Step S344: and performing time integration processing of the numerical model data by using a time integration method by adopting the combination of implicit expression and explicit expression to obtain second-class processing data.

According to the invention, simulation data of a high risk area and a low risk area are imported FVCOM (Finite Volume Coastal Ocean Model) into a marine mode, and numerical solution is carried out through a momentum equation, a continuous equation and a state equation. These equations are the core equations of fluid dynamics describing the motion and interaction of the fluid, and numerically solving these equations can generate high risk region fine data and low risk region fine data. This step ensures the accuracy and reliability of the data simulation by applying high-precision numerical methods and high-performance computing techniques. Based on the fine data, storm tide high level difference format optimization is performed, and storm tide high level difference data is generated. The high-order differential format is a numerical analysis method, and can improve the accuracy and stability of numerical solution. And further performing high-grid resolution fluid dynamics setting on the high-level difference data to generate storm tide grid resolution data. This process ensures a fine depiction of different areas of storm surge, especially in high risk areas, by means of an adaptive grid technique. And carrying out FVCOM ocean mode resolution analysis on the storm tide grid resolution data to generate numerical model data. Resolution analysis enables a numerical model to more accurately simulate the spatial and temporal variation characteristics of storm surge by refining the grid and optimizing the computational parameters. And finally, performing implicit and explicit combined time integration processing on the numerical model data by using a time integration method. The combination of the implicit method and the explicit method can improve the calculation efficiency and the accuracy while guaranteeing the calculation stability, and finally the second-class processing data is obtained. Through the series of steps, a complete technical flow is formed from the preliminary numerical solution of the high-risk and low-risk area data, the high-order differential optimization and the grid resolution setting to the time integration processing.

In an embodiment of the invention, simulation data of the high risk area and the low risk area are imported FVCOM (Finite Volume Coastal Ocean Model) into a marine mode, and numerical solution is performed through a momentum equation, a continuous equation and a state equation. These equations are the core equations of fluid dynamics describing the motion, mass conservation and state change of the fluid, by solving these equations, high risk region fine data and low risk region fine data are generated. This step utilizes high precision numerical methods, such as finite volume methods, to ensure accuracy and reliability of the data simulation. And carrying out storm tide high-level difference format optimization based on the generated high-risk area fine data and low-risk area fine data, and generating storm tide high-level difference data. The high-order differential format is a high-level numerical analysis method, and can improve the accuracy and stability of numerical solutions. And further performing high-grid resolution fluid dynamics setting on the high-level difference data to generate storm tide grid resolution data. The high grid resolution setting ensures fine characterization of storm surge by an adaptive grid technique, and particularly in high risk areas, higher spatial resolution is achieved by refining the grid. And carrying out FVCOM ocean mode resolution analysis on the storm tide grid resolution data to generate numerical model data. The resolution process involves optimizing the computational parameters and refining the grid to improve the accuracy of the numerical model to simulate storm surge spatial and temporal variations. FVCOM's high resolution capability enables it to accurately simulate complex hydrodynamic processes in coastal and offshore areas.

Finally, in step S344, the numerical model data is subjected to a time integration process that combines implicitly and explicitly using a time integration method. The time integration method is a key step in numerical simulation, the implicit method has good stability, the explicit method has higher calculation efficiency, and the combination of the two methods can improve the calculation efficiency while ensuring the stability. By this hybrid time integration method, the second type of processed data is finally obtained.

Preferably, step S4 comprises the steps of:

Step S41: carrying out weight fusion on the direct experience relation prediction result and the numerical simulation prediction result to obtain a storm surge fusion model;

step S42: performing model performance evaluation on the storm surge fusion model to obtain a model performance evaluation result;

step S43: carrying out harmonic analysis according to the measured tide level data to obtain harmonic constants, wherein the harmonic constants comprise two harmonic constants of amplitude and delay angle; converting the measured tide level data into measured tide level frequency domain data by utilizing Fourier transformation to generate storm tide frequency; generating storm surge harmonic analysis results by using a least square fitting method for storm surge tidal frequency; carrying out harmonic constant analysis based on storm surge harmonic analysis results to generate two harmonic constants of amplitude and delay angle;

step S44: and carrying out feedback optimization on the storm surge fusion model based on the harmonic constant to generate a storm surge water increasing forecast model.

According to the method, the storm surge fusion model is obtained by carrying out weight fusion on the direct experience relation prediction result and the numerical simulation prediction result, so that the prediction capability of two different methods is fully utilized, and the comprehensive accuracy of prediction is improved. And then, carrying out model performance evaluation on the fusion model, and objectively evaluating the prediction capability and reliability of the model so as to provide guidance for further optimization. The model can be effectively corrected by carrying out harmonic analysis on hydrologic data to obtain a harmonic constant, so that the adaptability and stability of the model can be improved. Finally, the fusion model is subjected to feedback optimization based on the harmonic constants, so that the prediction capability and accuracy of the model are further improved, and a solid foundation is laid for generating a storm surge water increasing prediction model.

As an example of the present invention, referring to fig. 2, step S4 described in this example includes.

In the embodiment of the invention, the weights of the direct experience relation predicted result and the numerical simulation predicted result are determined by using a weight distribution algorithm, and the direct experience relation predicted result and the numerical simulation predicted result are fused according to the weights by a weighted average fusion algorithm. And then, constructing a model function of the fused result to generate a storm surge fusion model.

In the embodiment of the invention, the statistical index and Root Mean Square Error (RMSE) technical means are adopted for carrying out model performance evaluation on the storm surge fusion model. And carrying out generalization capability and robustness verification by adopting a verification method, carrying out error analysis to explore error sources and distribution rules, carrying out contrast display on model prediction results and observation data by utilizing a visualization tool, and checking significance and statistical effects of the model by using a statistical hypothesis checking method. The comprehensive technical means can comprehensively evaluate the performance of the storm surge fusion model, and provide reference basis for optimization and improvement of the model.

In an embodiment of the present invention, the measured tidal level data is converted into frequency domain data by fourier transformation, and the tidal signal of the time domain is converted into a representation of the frequency domain, thereby clearly observing the distribution characteristics of tidal frequencies. And then fitting by using a least square method to fit the tidal frequency components so as to obtain an optimal fitting curve, thereby being beneficial to accurately identifying and quantifying the amplitude and phase information of different frequency components. After the fitting result is obtained, the amplitude and the delay angle of the tidal component are extracted from the fitting curve by a harmonic analysis method, and the data form a harmonic constant, wherein the amplitude represents the amplitude of the tide and the delay angle represents the delay condition of the tide phase. Through the process, the characteristics of the tidal signal can be comprehensively understood, and important references are provided for subsequent prediction model establishment and data analysis.

In the embodiment of the invention, the relation between the harmonic constant and the measured tide level data is determined by regression analysis, and the model parameters are adjusted by an optimization algorithm so as to improve the fitting precision of the model. In addition, the model correction method can be used for correcting the output result of the model, so that the output result is more in line with actual observation data. The application of the technical means can effectively optimize the storm surge water increasing forecast model, and improves the forecast precision and reliability.

Preferably, step S41 includes the steps of:

Step S411: generating a two-dimensional graph curve of the direct experience relation prediction result and the numerical simulation prediction result, and generating a direct experience relation prediction two-dimensional graph and a numerical simulation prediction two-dimensional graph; linear regression fitting degree analysis is carried out on the two-dimensional graph based on the direct experience relation prediction and the numerical simulation prediction to obtain a weight value of the direct experience relation prediction result and a weight value of the numerical simulation prediction result;

step S412: carrying out confidence interval weighted summation on the weight value of the direct experience relation prediction result and the weight value of the numerical simulation prediction result by using a weighted average method to obtain a fusion prediction result;

Step S413: and constructing a storm surge fusion model according to a storm surge construction function, so as to obtain a storm surge fusion model, wherein the storm surge construction function is as follows:

In the formula, In order to increase the water content of the water-absorbing agent,As an increment of the storm flow,In order for the coriolis force to be a coriolis force,The acceleration of the gravity is that,For the purpose of observing the air pressure at the point,For the air pressure at the observed point,For the variation of air pressure in the direction perpendicular to the shoreline,In order to be a vertical wind stress,In order to accumulate the effect on the coastal wind,For the overall height of the observation point,For the initial height of the observation point,In order to observe the initial water depth of the point,Is the density of the seawater, and the seawater is the density of the seawater,Is the length of the connecting line from the center of the tropical cyclone to the center of the length of the forecast station.

According to the invention, the simulation data of the high risk area and the low risk area are imported FVCOM (Finite Volume Coastal Ocean Model) into a marine mode, and the numerical solution is carried out through a momentum equation, a continuous equation and a state equation. These equations are the core equations of fluid dynamics describing the motion, mass conservation and state change of the fluid, by solving these equations, high risk region fine data and low risk region fine data are generated. This step utilizes high precision numerical methods, such as finite volume methods, to ensure accuracy and reliability of the data simulation. And carrying out storm tide high-level difference format optimization based on the generated high-risk area fine data and low-risk area fine data, and generating storm tide high-level difference data. The high-order differential format is a high-level numerical analysis method, and can improve the accuracy and stability of numerical solutions. And further performing high-grid resolution fluid dynamics setting on the high-level difference data to generate storm tide grid resolution data. The high grid resolution setting ensures fine characterization of storm surge by an adaptive grid technique, and particularly in high risk areas, higher spatial resolution is achieved by refining the grid. The resolution of the fine grid can capture the small change of storm surge in space, and the accuracy of the whole simulation is improved.

And carrying out FVCOM ocean mode resolution analysis on the storm tide grid resolution data to generate numerical model data. The resolution process involves optimizing the computational parameters and refining the grid to improve the accuracy of the numerical model to simulate storm surge spatial and temporal variations. FVCOM's high resolution capability enables it to accurately simulate complex hydrodynamic processes in coastal and offshore areas, thereby providing a more detailed and reliable storm surge dynamics prediction. Such high resolution modeling is particularly important for identifying and assessing the potential impact areas of storm surge, helping to take effective precautions and countermeasures prior to the occurrence of a disaster. And finally, performing implicit and explicit combined time integration processing on the numerical model data by using a time integration method. The time integration method is a key step in numerical simulation, the implicit method has good stability, the explicit method has higher calculation efficiency, and the combination of the two methods can improve the calculation efficiency while ensuring the stability. By this hybrid time integration method, the second type of processed data is finally obtained. The combination of the implicit time integration method and the explicit time integration method not only improves the efficiency of numerical simulation, but also ensures the stability and accuracy of results, and particularly in long-time scale simulation, the evolution process of storm surge can be better captured.

As an example of the present invention, referring to fig. 3, step S41 described in this example includes.

In the embodiment of the invention, firstly, data visualization processing is required to be carried out on the direct experience relation prediction result and the numerical simulation prediction result, and the change trend of the two types of prediction results is shown by generating a two-dimensional graph curve. This step utilizes a two-dimensional mapping technique to visualize the various types of data points in a coordinate system to visually observe and compare the differences and consistency of the two types of predictions. After the two-dimensional graph curve is generated, linear regression fitting degree analysis is performed based on the graph data. Linear regression is a statistical analysis method used to evaluate the linear relationship between two variables. In the process, regression analysis is carried out on the data of the direct empirical relationship prediction two-dimensional graph and the numerical simulation prediction two-dimensional graph, and regression coefficients and fitting goodness are calculated to evaluate the linear correlation and fitting degree of the two types of prediction results. This step involves not only the basic linear regression algorithm, but also the least squares method may be used to optimize the regression model, improving the fitting accuracy. And further calculating to obtain the weight value of the direct experience relation predicted result and the weight value of the numerical simulation predicted result through linear regression fitting degree analysis. The determination of the weight value is based on the analysis results of the goodness of fit and the regression coefficient, and the contribution degree of the two types of prediction results is comprehensively considered by adopting a weighted average method.

In an embodiment of the present invention, the direct empirical relationship prediction results and the numerical simulation prediction results are weighted and summed using a weighted average method based on the previously calculated weight values. The weighted average method is a common statistical technique, and can reflect the importance and contribution of each prediction result in the overall prediction by giving different weights to different data. The concept of confidence interval is further introduced to carry out weighted summation processing. Confidence intervals are a statistical method for describing the reliable range of estimates. By calculating confidence intervals for the direct empirical relationship prediction results and the numerical simulation prediction results, respectively, the stability and reliability of the prediction results can be evaluated. In the weighted summation process, the confidence interval is considered to improve the reliability of the fusion prediction result. Specifically, when the prediction results are weighted and summed, the weight value is adjusted according to the width degree of the confidence interval, namely, the narrower the confidence interval is, the more reliable the prediction results are, and the higher the weight value is relatively; conversely, the wider the confidence interval, the lower the reliability of the predicted outcome and the lower the weight value.

In the formula, In order to increase the water content of the water-absorbing agent,As an increment of the storm flow,In order for the coriolis force to be a coriolis force,The acceleration of the gravity is that,For the purpose of observing the air pressure at the point,For the air pressure at the observed point,For the variation of air pressure in the direction perpendicular to the shoreline,In order to be a vertical wind stress,In order to accumulate the effect on the coastal wind,For the total water depth of the observation point,In order to observe the static water depth of the point,To observe the tidal level value of the point relative to the mean sea level,Is the density of the seawater, and the seawater is the density of the seawater,Is the length of the connecting line from the center of the tropical cyclone to the center of the length of the forecast station.

The invention calculates the ratio of the integral of the vertical wind stress in the accumulated action of the coastal wind to the integral height of the observation point and the integral of the sea water density, and obtains the water increment through the difference between the ratio of the sea water with the air pressure difference and the integral of the storm flow increment. The ratio of the air pressure difference to the seawater is the ratio of the air pressure difference between the observed point and the observed point to the seawater density. The integral of storm tide increment is the product of the ratio of the Coriolis force to the gravity acceleration and the integral of storm flow in the accumulated action of the storm wind. This formula provides a framework for studying the development and evolution of tropical cyclones and for assessing the degree of risk to the observation point. The method is not only beneficial to research and design in scientific research and engineering application, but also can guide related departments to take precautions and countermeasures, and reduces the loss caused. The accuracy and timeliness of early warning and forecasting can be improved through the result obtained through formula calculation, so that the safety of the environment, life and property is effectively protected, and the stable and sustainable development of society is maintained.

Preferably, the storm surge prediction model in step S44 is as follows:

In the formula, Is the water quantity, the water is used as the water,Is the difference in air pressure, and is,Is the density of the seawater, and the seawater is the density of the seawater,Is water-increasing by air pressure,Is the undetermined coefficient of linear quay wind energy,For the coefficient of uncertainty of the shore-bound wind energy (linear quay wind energy and the coefficient of uncertainty of the shore-bound wind energy are both related to the angle of the tropical cyclone path direction to the shoreline),To forecast the wind speed of a site 10m from the sea surface,In order to observe the static water depth of the point,To observe the tidal level value of the point relative to the mean sea level,The included angle between the connecting line of the forecasting site and the tropical cyclone center and the eastern direction is formed; To forecast site location and an included angle between the shoreline and the east direction.

The invention obtains the water increasing amount by calculating the sum of the influence of the air pressure difference on the water increasing amount and the influence of different wind speeds on the water increasing amount. Wherein the influence of the air pressure difference on the water increasing amount is detailed as the ratio of the air pressure difference to the product of the sea water density and the gravity acceleration. The influence of wind speed on the water increasing amount is detailed as the wind speedThe ratio of the square term of (a) to the sum of the initial height of the observation point and the initial water quantity of the observation point, and the product of linear quay times the wind energy and the cosine value of (the angle between the central connecting line of the forecasting site and the tropical cyclone and the east minus the angle between the shoreline of the forecasting site and the east). Wind speedThe ratio of the square term of (1) to the sum of the initial height of the observation point and the initial water quantity of the observation point, the product of the sum and the wind energy in the shore direction, and the product of the sum (the angle between the connecting line of the forecasting site and the tropical cyclone center and the east direction minus the angle between the shore line of the forecasting site and the east direction) sine value.

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

1. The storm surge water increasing rapid forecasting method is characterized by comprising the following steps:

2. The storm surge rapid forecast method of claim 1, wherein step S1 includes the steps of:

step S11: acquiring historical observation storm surge data;

3. The storm surge rapid forecast method of claim 1, wherein step S2 comprises the steps of:

Step S21: performing difference correction processing on the historical observed storm surge data and storm weather base data to generate storm surge correction data;

step S22: trend analysis is carried out on storm surge correction data to generate first-class processing data, wherein the trend analysis comprises denoising and smoothing processing, application time sequence analysis, regression analysis and spectrum analysis; denoising and smoothing the storm surge correction data by using an exponential smoothing method to generate storm surge smoothing data; the storm surge correction data is used as time series data to be input into an MA moving average model, so as to generate storm surge MA data; carrying out meteorological variable analysis on storm surge correction data by utilizing a multivariate analysis method to generate storm surge multivariate data; carrying out frequency domain analysis on storm surge correction data by utilizing Fourier transformation to generate storm surge frequency domain data; carrying out data merging processing on storm tide smoothing data, storm tide MA data, storm tide multivariate data and storm tide frequency domain data to generate a first class processing model;

4. The storm surge rapid forecast method of claim 1, wherein step S3 includes the steps of:

Step S31: acquiring position information data of a specific place;

5. The storm surge rapid forecast method of claim 4, wherein step S32 includes the steps of:

Step S322: carrying out Arakawa B level processing on the horizontal direction data of the specific place by utilizing the historical observation storm surge data and storm weather basic data to generate grid level data;

Step S323: performing vertical coordinate discretization processing on the vertical direction data of the specific place by utilizing the historical observation storm surge data and storm weather basic data to generate grid vertical data;

6. The storm surge rapid forecast method of claim 4, wherein step S34 includes the steps of:

step S341: importing FVCOM the high-risk region simulation data and the low-risk region simulation data into a marine mode, and carrying out physical quantity numerical solution through a momentum equation, a continuous equation and a state equation to generate high-risk region fine data and low-risk region fine data;

7. The storm surge rapid forecast method of claim 1, wherein step S4 comprises the steps of:

Step S44: and carrying out feedback optimization on the storm surge fusion model based on the amplitude and the delay angle to generate a storm surge water increasing forecast model.

8. The storm surge rapid forecast method of claim 7, wherein step S41 includes the steps of:

9. The rapid storm surge prediction method according to claim 7, wherein the calculation formula of the storm surge prediction model in step S44 is as follows:

In the formula, In order to increase the water content of the water-absorbing agent,Is the difference in air pressure, and is,Is the density of the seawater, and the seawater is the density of the seawater,Is water-increasing by air pressure,Is the undetermined coefficient of linear quay wind energy,For the coefficient of uncertainty of the shore-wise wind energy,To forecast the wind speed of a site 10m from the sea surface,In order to observe the static water depth of the point,To observe the tidal level value of the point relative to the mean sea level,The included angle between the connecting line of the forecasting site and the tropical cyclone center and the eastern direction is formed; To forecast site location and an included angle between the shoreline and the east direction.