Typhoon-wave-current risk assessment method based on physical field coupling numerical simulation

Abstract

This invention relates to a typhoon-wave-current hazard assessment method based on physical field coupled numerical simulation. The method includes: selecting historical or future typhoon events affecting a target area from historical typhoon trajectory data or artificially synthesized future typhoon trajectory data to generate a typhoon event set; modeling a two-dimensional typhoon mixed wind field based on the constructed typhoon event set, combined with typhoon trajectory and intensity parameters; performing a physical mechanism-based typhoon-wave-current coupled numerical simulation on the target area to generate corresponding wind, wave, and current evolution time history data; and extracting the event maximum and annual maximum values ​​from the simulated wind, wave, and current time history data, and using extreme value analysis to assess hazard risk. This invention is not limited to historical typhoon events; the entire assessment framework can be applied to the hazard assessment and analysis of future typhoon events considering the impact of climate change, improving the climate adaptability and resilience of coastal engineering infrastructure against typhoon-related complex disasters.

Description

Technical Field

[0001] This invention belongs to the field of marine engineering disaster prevention and mitigation technology, specifically relating to a typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation. Background Technology

[0002] Numerous studies and experimental data indicate that typhoons often generate giant waves and strong surface currents, creating a multi-physics coupled damage mechanism for marine engineering structures. With global warming and accelerated marine resource development, typhoon-induced wind-wave-current composite disasters have become a significant hazard to the safety of marine engineering projects. This multi-hazard coupling can lead to resonant fatigue of offshore platform jackets, coupled vibration of cross-sea bridges, failure of offshore wind power structures, and vortex-induced vibration of submarine pipelines, among other disaster effects and failure modes.

[0003] However, existing typhoon disaster assessment methods mostly target single typhoons or combined wind-wave disasters, failing to fully consider the interactions between wind, waves, and currents during typhoon evolution, as well as concurrent compound disasters. Furthermore, limitations in the quantity and quality of historical statistical data make regional multi-hazard risk assessments difficult, and the estimated return periods of extreme values ​​may have significant uncertainties, posing challenges to the reliability design of engineering structures. More importantly, existing assessment methods are largely geared towards historical typhoon events, lacking precise simulation tools for typhoon track and intensity changes under future climate scenarios. The Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC) points out that existing low-precision global climate models still have significant uncertainties in predicting the maximum intensity of typhoons at the end of the 21st century. Summary of the Invention

[0004] To address the aforementioned shortcomings, this invention proposes a high-precision, high-fidelity multi-hazard risk assessment framework based on fully coupled typhoon-wave-current numerical simulation. By establishing a typhoon meteorological field, driving coupled numerical simulation of waves and ocean currents, and combining extreme value analysis theory, the risk assessment of typhoon-related complex disasters is achieved. Furthermore, it is not limited to historical typhoon events; the entire assessment framework can also be used for disaster assessment and analysis of future typhoon events, enhancing the simulation and assessment capabilities for typhoon-related complex disasters under climate change scenarios.

[0005] The technical solution adopted in this invention is: a typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation, comprising the following steps: The first step is to select historical or future typhoon events that affect the target area from historical typhoon trajectory data or artificially synthesized future typhoon trajectory data, and generate a typhoon event set. The second step is to model a two-dimensional typhoon mixed wind field based on the constructed typhoon event set and the typhoon trajectory and intensity parameters, and generate the two-dimensional typhoon wind field corresponding to the typhoon event set. The third step is to conduct a typhoon-wave-current coupled numerical simulation based on physical mechanisms in the target area to generate wind, wave and current evolution time history data corresponding to the typhoon event set; The fourth step involves extracting the maximum event value and annual maximum value based on the simulated wind, wave, and current time history data, and then using extreme value analysis to assess the disaster risk.

[0006] Preferably, in the first step, the criteria for selecting a typhoon event are: for historical or future synthetic typhoon trajectories, the minimum distance between the selected typhoon's path and the target area is less than 250 km, and the maximum wind speed intensity throughout its lifespan is greater than 24.5 m / s.

[0007] Preferably, in the second step, the construction of the two-dimensional typhoon mixed wind field model includes the following steps: Holland pressure field model, Georgiou wind speed field model, and typhoon inflow angle model were constructed respectively to calculate the parametric wind field wind speed and direction; the formula for calculating the parametric wind field wind speed is as follows: ; In the formula, It is the radial distance from the center of the typhoon. It's a pressure difference. It is the atmospheric pressure of the external environment. It is the central pressure of the typhoon. It is the air density constant. Latitude Coriolis parameters at the location, It is the angular velocity of the Earth's rotation; It is the speed at which the typhoon moves horizontally; This represents the clockwise angle from the direction of the typhoon's translation to the vector line connecting the typhoon's center and the calculation point; It is the radius of maximum wind speed. These are the typhoon shape parameters; In addition, the direction of wind speed in the wind field is determined using the following inflow angle model: ; ; ; In the formula, It is the angle of the vector line between the typhoon center and the calculation point, defined counterclockwise from due east. The typhoon's direction of movement is defined counterclockwise from due east. It is the relative angular momentum at the radius of maximum wind speed; Select the typhoon wind speed reanalysis dataset to obtain the wind speed vector of the reanalysis wind field; A two-dimensional mixed wind speed field of the typhoon is established based on the parametric wind field wind speed vector and the reanalysis wind field wind speed vector: ; ; ; In the formula, This is the final wind speed vector for the mixed wind field. and These represent the parametric wind field wind speed vector and the reanalysis wind field wind speed vector, respectively. These are the weighting coefficients; The wind speed in the middle is equal to 0.85. .

[0008] Preferably, in the third step, the typhoon-wave-current coupled numerical simulation adopts the SWAN+ADCIRC coupling mode, which specifically includes the following steps: (1) Extract water depth information from the GEBCO global water depth database based on the latitude and longitude range of the target area; (2) Calculate the tidal driving force at the ocean boundary based on the TPXO9 global tidal model; (3) The computational mesh of the target region was generated using the OceanMesh2D toolkit, and all meshes were unstructured triangular meshes; (4) For each typhoon to be simulated, the simulation time period is determined based on the condition that the distance between the typhoon center and the calculation area is less than 500km. (5) Take the prepared data and the typhoon two-dimensional wind field data, generate the input file corresponding to the selected typhoon event set according to the input file format of SWAN+ADCIRC mode, and perform batch simulation to obtain the corresponding wind, wave and current time history simulation results.

[0009] Preferably, in the fourth step, the maximum values ​​of the wind, wave, and current time histories obtained from the simulation of each typhoon are first extracted to form the event maximum value. V E,max , H E,max , C E,max Then, the maximum value for each year is further extracted from the event maximum value data to form the annual maximum value. V A,max , H A,max , C A,max The event maximum sample is used to fit the generalized Pareto distribution, and the threshold size is determined using the minimum RMSE method; the annual maximum sample is used to fit the Günbel distribution and the generalized extreme value distribution.

[0010] Preferably, in the fourth step, the goodness of fit of the three distributions used for fitting is tested using the KS method based on sample size correction; according to the principle of minimizing the KS statistic, the optimal distribution model is further selected from the three distribution models.

[0011] Preferably, for each grid node in the target area, the return period value corresponding to a certain return period of the three disaster-causing factors (wind, wave, and current) is calculated based on the distribution function obtained by fitting the three distribution models, and a hazard zoning map of the area for a certain return period is drawn; if the estimation result of the optimal distribution model is used for each grid node, the best estimated zoning map is obtained.

[0012] This invention, building upon existing typhoon and wind / wave hazard assessments, utilizes the state-of-the-art SWAN+ADCIRC wind-wave-current coupled numerical simulation technology. It uniquely considers the assessment of ocean current-induced hazard factors, and the coupled numerical model takes into account the complex interactions between typhoon-related hazards. This overcomes the limitations of the quantity and quality of historical typhoon data, improving the accuracy and robustness of typhoon multi-hazard risk assessment. Furthermore, this invention is not limited to historical typhoon events; the entire assessment framework can also be applied to the hazard assessment and analysis of future typhoon events considering the impact of climate change, enhancing the climate adaptability and resilience of coastal engineering infrastructure against typhoon-related hazards. Attached Figure Description

[0013] Figure 1 A flowchart of the typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation provided by the present invention; Figure 2 This refers to the computational grid division and spatial distribution of water depth for the target area selected in this embodiment of the invention. Figure 3 These are 254 typhoon tracks that affected the areas of concern between 1988 and 2023; Figure 4 This is the pressure field of Typhoon Hato in 2017 at a certain moment; Figure 5 This is the wind speed field at a certain moment during Typhoon Hato in 2017; Figure 6 Empirical and theoretical distribution curves of wind, wave, and current hazards at sites P1 and P2; Figure 7 Zoning maps of the 50-year recurrence level obtained from three distribution estimates and the best estimated zoning map. Detailed Implementation

[0014] To facilitate understanding of the present invention, a more detailed description is provided below with reference to the accompanying drawings and specific embodiments. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the present invention.

[0015] This invention provides a typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation. The process is shown in Figure 1 and specifically includes the following steps: First, select the target area for disaster risk assessment. In this example, the selected area is located in the Pearl River Delta and its adjacent sea area, with a longitude range of 110.7° to 119.6° east and a latitude range of 18.0° to 25.0° north, as shown in Figure 2.

[0016] This study uses historical typhoon events for hazard risk analysis, selecting typhoon events from the CMA (China Meteorological Administration) typhoon best-path dataset from 1988 to 2023. Two criteria were used for selection: first, the minimum distance between the typhoon path and the target area was less than 250 km; second, the maximum wind speed intensity during the entire typhoon event was greater than 24.5 m / s. The resulting typhoon event set contains 254 typhoon records from these 36 years, as shown in Figure 3.

[0017] CMA's optimal path data includes the time, center latitude and longitude, and central pressure of each typhoon path. Based on this information, the Holland pressure field and Georgiou wind speed field at each path point of each typhoon in the typhoon event set can be generated, and only the data located within the selected area needs to be extracted.

[0018] The Holland pressure field is constructed using the following expression: ; In the formula, It is the radial distance from the center of the typhoon. It is the atmospheric pressure of the external environment. It is the central pressure of the typhoon. It's the air pressure difference; It is the radius of maximum wind speed. These are the typhoon shape parameters; and The regression formula is as follows: ; .

[0019] The Georgiou wind speed field is constructed using the following expression: ; In the formula, It is the air density constant. Latitude Coriolis parameters at the location, It is the angular velocity of the Earth's rotation; It is the speed at which the typhoon moves horizontally; This represents the clockwise angle from the typhoon's direction of movement to the vector line connecting the typhoon's center and the calculation point; the wind speed at a height of 10 m above sea level is represented by 0.85. Make an estimate.

[0020] This only indicates the magnitude of the wind speed; the direction of the wind speed still needs to be determined by the angle of inflow. ; ; ; In the formula, It is the angle of the vector line between the typhoon center and the calculation point, defined counterclockwise from due east. The typhoon's direction of movement is defined counterclockwise from due east. It is the relative angular momentum at the radius of maximum wind speed.

[0021] The above steps establish an accurate typhoon wind field. Additionally, reanalysis wind speed field data for the selected target area needs to be extracted from the ERA5 reanalysis dataset and then mixed with the parametric wind field to obtain the typhoon's two-dimensional mixed wind speed field. The two-dimensional mixed wind speed field is established using the following formula: ; ; ; In the formula, This is the final wind speed vector for the mixed wind field. and These represent the parametric wind field wind speed vector and the reanalysis wind field wind speed vector, respectively. These are the weighting coefficients; The wind speed in the middle is equal to 0.85. .

[0022] This yields the two-dimensional typhoon meteorological field corresponding to the typhoon event set. Taking Typhoon Hato (1713) in 2017 as an example, the pressure field and wind speed field generated using the above method at 13:00 UTC on August 22, 2017 are as follows: Figure 4 and Figure 5 As shown.

[0023] Next, a typhoon-wave-current coupled numerical simulation based on physical mechanisms is performed on the selected target area.

[0024] Based on the latitude and longitude range of the target area, water depth data was extracted from the GEBCO global water depth database, as shown in Figure 2. Using the TPXO9 global tidal model, the tidal driving forces at the open boundary of the ocean in this region were calculated, including eight main tidal components (K1, K2, M2, N2, O1, P1, Q1, S2). The OceanMesh2D toolkit was used to generate an unstructured triangular computational mesh for the region. For a total of 254 typhoons to be simulated, the simulation time period was determined based on the condition that the typhoon center distance from the calculation area was less than 500 km.

[0025] To balance computational accuracy and efficiency, the mesh resolution was set to a maximum of 350 m near the coastline and a maximum of 20 km in the deep sea. Preliminary experiments were conducted to optimize the mesh, including removing small islands and smoothing irregular coastlines. The final computational mesh consists of 34,098 nodes and 58,897 triangular elements, as shown in Figure 2.

[0026] In the SWAN+ADCIRC mode, there are other parameter settings as follows, which are obtained based on experience or previous model tests: In the ADCIRC mode, the quadratic bottom friction law is adopted, and the minimum bottom friction coefficient is 0.0015; the horizontal eddy viscosity coefficient is set to 4 m. 2 / s; the drag coefficient scheme follows the Powell formula, with an upper limit set at 0.0035; the calculation time step is set to 5 seconds to ensure the Courant number is less than 1. In the SWAN model, the spectral frequency range is set between 0.03 Hz and 1.42 Hz, discretized into 40 intervals on a logarithmic scale, and the wave direction is discretized into 36 directions with a resolution of 10°; the model considers bottom friction, white wave dissipation, depth-induced wave breaking, and nonlinear wave-wave interactions (four-wave and three-wave interactions); bottom friction adopts the JONSWAP expression, and the friction coefficient is set to 0.019 m. 2 / s 3 White wave dissipation was calculated using the Komen model; the computation time step was set to 1200 seconds, consistent with the time step of the coupled model.

[0027] Using the prepared data and typhoon two-dimensional wind field data, the input files for the selected 254 typhoon events are generated according to the input file format of SWAN+ADCIRC mode. Batch numerical simulations are then performed to obtain the wind, wave, and current time history simulation results for each typhoon event.

[0028] When conducting disaster risk assessment, only extreme values ​​of the disaster are generally of concern. Therefore, the maximum values ​​of wind, waves, and current are first extracted from the time histories of wind, waves, and current in 254 typhoon events to generate a sample vector of the maximum values ​​of the events. V E,max , H E,max , C E,max Then, from the maximum event sample, the maximum value for each year is further extracted to generate an annual maximum value sample vector. V A,max , H A,max , C A,max Since a total of 254 typhoon events over 36 years were selected, the maximum number of samples for an event is 254, and the maximum number of samples for an annual event is 36.

[0029] The 254 maximum event samples were used for parameter fitting of the generalized Pareto distribution (GPD), and the threshold was determined using the minimum RMSE method. Specifically, the samples were arranged in ascending order, and the empirical probability distribution function of the samples was calculated. Then, a threshold was selected from the 50% to 90% quantiles, such that the RMSE between the probability distribution function obtained by fitting the GPD to the samples above this threshold and the empirical probability distribution function was minimized.

[0030] The 36 annual maximum value samples can be directly used for parameter fitting of the Gumbel distribution (GD) and the generalized extreme value distribution (GEVD).

[0031] Taking two stations P1 and P2 in Figure 3 as examples, the parameters of the three distributions of wind, waves, and current are estimated using the maximum likelihood estimation method, and plotted together with the empirical probability distribution. Figure 6 As can be seen, the theoretical distribution fits the empirical distribution quite well. Figure 6 The distribution curves were plotted using Gumbel probability paper.

[0032] Because the sample sizes used for fitting the three distributions differ, the KS method based on sample size adjustment is employed to test the goodness of fit among the three distributions, and the distribution with the smallest KS statistic is selected as the optimal distribution. The formula for calculating the KS statistic is as follows: ; In the formula, It is an empirical probability distribution function. It is the probability distribution function obtained by fitting. This is the number of samples used for fitting.

[0033] Table 1 presents the KS statistics for wind, wave, and current hazards at sites P1 and P2, fitted using three different distributions. For the three hazard-causing factors, the distribution model was optimized based on the principle of minimizing KS; the numbers in parentheses in the table represent the optimized results.

[0034] Table 1. KS statistic values ​​for distribution fitting of wind, wave, and current hazards at two sites. .

[0035] By selecting a specific return period based on the distribution function obtained from fitting the three distribution models, the return period value corresponding to this return period of the three disaster-causing factors—wind, waves, and currents—can be calculated.

[0036] For the Gumbel distribution (GD) and the generalized extreme value distribution (GEVD), the return period is calculated using the following formula: ; In the formula, T For the recurrence period, , and These are the model's position parameters, scale parameters, and shape parameters, respectively. When the model follows a Günber distribution, when When the model is a generalized extreme value distribution, the model is a generalized extreme value distribution.

[0037] For the generalized Pareto distribution (GPD), the recurrence level is calculated using the following formula: ; In the formula, T For the recurrence period, , and These are the model's position parameters, scale parameters, and shape parameters, respectively. u For the threshold, λ The annual average number of typhoon events in the selected event set. It can be used N u / N To estimate, among which N u The number of samples exceeding the threshold. N This represents the total number of samples.

[0038] Hazard risk assessments were conducted for sites P1 and P2. Table 2 shows the 50-year and 500-year return periods of wind, wave, and current hazards calculated using the theoretical distribution functions of the two sites. These values ​​can be further used in the design of related engineering structures.

[0039] Table 2. Return periods of wind, wave, and current hazards calculated using the theoretical distribution functions of two stations. .

[0040] The hazard assessment process can be performed on a single grid node or on all nodes within a region, thereby obtaining a hazard zoning map for a given return period within that region. If the estimation results of the optimal model are used for each node, the best-estimated zoning map can be obtained.

[0041] Figure 7 shows the wind, wave, and current zoning maps with a 50-year return period calculated using the GD, GEVD, and GPD probability distributions, respectively, as well as the optimal estimated zoning map obtained by optimizing the distribution using the KS test method. This constitutes the hazard risk assessment for the entire target area.

Claims

1. A method for typhoon-wave-current hazard assessment based on physical field coupling numerical simulation, characterized in that, Includes the following steps: The first step is to select historical or future typhoon events that affect the target area from historical typhoon trajectory data or artificially synthesized future typhoon trajectory data, and generate a typhoon event set. The second step is to model a two-dimensional typhoon mixed wind field based on the constructed typhoon event set and the typhoon trajectory and intensity parameters, and generate the two-dimensional typhoon wind field corresponding to the typhoon event set. The third step is to conduct a typhoon-wave-current coupled numerical simulation based on physical mechanisms in the target area to generate wind, wave and current evolution time history data corresponding to the typhoon event set; The fourth step involves extracting the maximum event value and annual maximum value based on the simulated wind, wave, and current time history data, and then using extreme value analysis to assess the disaster risk.

2. The typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation according to claim 1, characterized in that, In the first step, the criteria for selecting typhoon events are: for historical or future composite typhoon trajectories, the minimum distance between the selected typhoon's path and the target area is less than 250 km, and the maximum wind speed intensity throughout its lifespan is greater than 24.5 m / s.

3. The typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation according to claim 1, characterized in that, In the second step, the construction of the two-dimensional typhoon mixed wind field model includes the following steps: Holland pressure field model, Georgiou wind speed field model, and typhoon inflow angle model were constructed respectively to calculate the parametric wind field wind speed and direction; the formula for calculating the parametric wind field wind speed is as follows: ； In the formula, It is the radial distance from the center of the typhoon. It's a pressure difference. It is the atmospheric pressure of the external environment. It is the central pressure of the typhoon. It is the air density constant. Latitude Coriolis parameters at the location, It is the angular velocity of the Earth's rotation; It is the speed at which the typhoon moves horizontally; This represents the clockwise angle from the direction of the typhoon's translation to the vector line connecting the typhoon's center and the calculation point; It is the radius of maximum wind speed. These are the typhoon shape parameters; In addition, the direction of wind speed in the wind field is determined using the following inflow angle model: ； ； ； In the formula, It is the angle of the vector line between the typhoon center and the calculation point, defined counterclockwise from due east. The typhoon's direction of movement is defined counterclockwise from due east. It is the relative angular momentum at the radius of maximum wind speed; Select the typhoon wind speed reanalysis dataset to obtain the wind speed vector of the reanalysis wind field; A two-dimensional mixed wind speed field of the typhoon is established based on the parametric wind field wind speed vector and the reanalysis wind field wind speed vector: ； ； ； In the formula, This is the final wind speed vector for the mixed wind field. and These represent the wind speed vectors of the parametric wind field and the reanalysis wind field, respectively. These are the weighting coefficients; The wind speed in the middle is equal to 0.

85. .

4. The method for assessing the multi-hazard risk of typhoons based on multiphysics coupled numerical simulation according to claim 1, characterized in that, In the third step, the typhoon-wave-current coupled numerical simulation adopts the SWAN+ADCIRC coupling mode, which specifically includes the following steps: (1) Extract water depth information from the GEBCO global water depth database based on the latitude and longitude range of the area of ​​interest; (2) Calculate the tidal driving force at the ocean boundary based on the TPXO9 global tidal model; (3) The computational mesh of the target region was generated using the OceanMesh2D toolkit, and all meshes were unstructured triangular meshes; (4) For each typhoon to be simulated, the simulation time period is determined based on the condition that the distance between the typhoon center and the calculation area is less than 500 km. (5) Take the prepared data and the typhoon two-dimensional wind field data, generate the input file corresponding to the selected typhoon event set according to the input file format of SWAN+ADCIRC mode, and perform batch simulation to obtain the corresponding wind, wave and current time history simulation results.

5. The typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation according to claim 1, characterized in that, In the fourth step, the maximum values ​​of the wind, wave, and current time histories obtained from the simulation of each typhoon are first extracted to form the event maximum value. V E,max , H E,max , C E,max Then, the maximum value for each year is further extracted from the event maximum value data to form the annual maximum value. V A,max , H A,max , C A,max The event maximum sample is used to fit the generalized Pareto distribution, and the threshold size is determined using the minimum RMSE method; the annual maximum sample is used to fit the Günbel distribution and the generalized extreme value distribution.

6. The typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation according to claim 5, characterized in that, In the fourth step, the goodness of fit of the three distributions used for fitting is tested using the KS method based on sample size correction; according to the principle of minimizing the KS statistic, the optimal distribution model is further selected from the three distribution models.

7. The typhoon-wave-current hazard assessment method based on physical field coupling numerical simulation according to claim 6, characterized in that, For each grid node in the computational region, the return period value corresponding to a certain return period of the three disaster-causing factors (wind, wave, and current) is calculated based on the distribution function obtained by fitting the three distribution models, and a hazard zoning map of the region for a certain return period is drawn. If the estimation result of the optimal distribution model is used for each grid node, the best estimated zoning map is obtained.

Images