A method for estimating extreme wind speeds with different return periods in the Northwest Pacific

Abstract

The invention discloses a method for estimating extreme wind speeds of different return periods in the northwest Pacific, including preprocessing of historical typhoon data, reconstruction of the historical typhoon wind field of the northwest Pacific by using a typhoon wind field model and extraction of the annual maximum wind speed sequence, calculation of the annual maximum wind speed sequence of each grid in the northwest Pacific by using ship sea surface wind data, calculation of the annual maximum wind speed sequence of each grid in the northwest Pacific by using satellite sea surface wind data, determination of the final annual maximum wind speed sequence samples of each grid in the northwest Pacific, estimation of extreme wind speeds of different return periods in the northwest Pacific by using different extreme value distribution models and parameter estimation methods, etc., which solves the problem of insufficient samples of the annual maximum wind speed sequence in the northwest Pacific, can simply and quickly provide the spatial distribution of extreme wind speeds of different return periods in the northwest Pacific sea area, and the method has strong universality. The method can provide a basic basis for the design of wind-resistant parameters of marine engineering and improve the reliability and safety of marine engineering operation.

Description

Technical Field

[0001] The present invention belongs to the technical field of meteorological data analysis and processing, and relates to a method for estimating extreme wind speeds in different return periods in the northwest Pacific Ocean, and in particular to a method for creating a sample sequence of the annual maximum wind speed in the northwest Pacific Ocean based on typhoon wind field reconstruction, ship sea surface wind, and satellite sea surface wind data. Background Art

[0002] In the field of marine engineering design, extreme wind speeds with different return periods are one of the key extreme environmental parameters. Due to the unique geographical location of the Northwest Pacific, the region has the largest number of typhoons and the strongest intensity in the world each year. Extreme disasters such as wind, waves, and storm surges caused by typhoons occur frequently, bringing huge risks and challenges to the construction of marine engineering and the conduct of marine activities. Therefore, it is very important to accurately estimate the extreme wind speeds with different return periods in the Northwest Pacific, which can ensure the safety and effectiveness of marine engineering structure design and save investment estimation costs.

[0003] The key links in estimating extreme wind speeds in different return periods in the northwest Pacific are sample selection, model establishment, and parameter estimation. Wind speed samples are a prerequisite. Limited by the extremely scarce sea surface wind observation data, existing studies have selected the maximum wind speed of the typhoon center in a certain area over the years as wind speed samples, but wind speed data can only be obtained when the typhoon center passes through a certain grid. Therefore, the samples obtained by this method cannot reflect the spatial differences of the entire region. Although typhoon numerical simulation technology can be used to supplement wind speed samples, typhoon wind field simulation is currently limited to individual cases or small areas, and there are relatively few typhoon wind field simulations for the entire northwest Pacific Ocean. At the same time, when some grids do not pass through typhoons, the typhoon wind field model cannot obtain the simulated wind speed of these grids, and thus the annual maximum wind speed series of these grids cannot be obtained. Existing studies mostly use empirical values ​​to supplement, but it is difficult to reflect the annual maximum wind speed characteristics of these grids. In addition, the Monte-Carlo method is used to sample virtual typhoons and bring them into the typhoon wind field model to supplement wind speed samples, which can enhance the spatial continuity of the data to a certain extent. However, this method requires high computational skills and is complex, which limits its widespread application in practical engineering. In addition, different extreme value distribution models and different parameter estimation methods estimate different extreme wind speeds during the return period, and these models and methods perform differently in different regions. Their applicability in the northwest Pacific Ocean needs to be further explored in combination with methods such as goodness of fit tests. Methods such as goodness of fit tests can be used to evaluate the applicability of models and methods, but this process requires a large amount of data support and complex statistical analysis.

[0004] In summary, accurate estimation of extreme wind speeds in the northwest Pacific region faces many technical challenges, including but not limited to the difficulty of data collection, the diversity of model selection, and the uncertainty of parameter estimation. How to accurately estimate extreme wind speeds with different return periods in the northwest Pacific, especially when data samples are insufficient and the uncertainty of typhoon paths is large, how to improve the accuracy and reliability of the estimation method, so as to provide a basic basis for the design of wind-resistant parameters of marine engineering and improve the reliability and safety of marine engineering operations, is still a technical problem that needs to be solved urgently. Summary of the invention

[0005] (I) Purpose of the invention

[0006] In view of the technical problems such as insufficient data samples, model limitations and parameter estimation uncertainty in the existing return period wind speed estimation methods, in order to solve at least one of the above and other technical problems in the prior art, the present invention proposes a method for estimating extreme wind speeds with different return periods in the Northwest Pacific, including preprocessing of historical typhoon data; reconstructing the historical typhoon wind field in the Northwest Pacific using a typhoon wind field model and extracting the annual maximum wind speed sequence; calculating the annual maximum wind speed sequence of each grid in the Northwest Pacific using ship surface wind data; calculating the annual maximum wind speed sequence of each grid in the Northwest Pacific using satellite surface wind data; determining the annual maximum wind speed sequence samples of each grid in the Northwest Pacific using the annual maximum wind speed sequence based on typhoon wind field reconstruction, ship surface wind and satellite surface wind; and estimating the extreme wind speeds with different return periods in the Northwest Pacific using different extreme value distribution models and parameter estimation methods. The method for estimating extreme wind speeds with different return periods in the Northwest Pacific of the present invention solves the problem of insufficient samples of the annual maximum wind speed sequence in the Northwest Pacific, can simply and quickly provide the spatial distribution of the return period wind speed in the Northwest Pacific waters, and the method has strong universality. This method can provide a basic basis for the design of wind-resistant parameters of marine engineering and improve the reliability and safety of marine engineering operation.

[0007] (II) Technical solution

[0008] In order to achieve the purpose of the invention and solve the technical problem, the technical solution adopted by the present invention to solve the problem of the return period wind speed estimation method is:

[0009] A method for estimating extreme wind speeds with different return periods in the northwest Pacific Ocean is intended to improve the accuracy and reliability of extreme wind speed estimation in marine engineering design, wherein the method comprises at least the following steps when implemented:

[0010] SS1. Preprocess the historical typhoon data, including at least collecting the historical typhoon data within the preset time range in the Northwest Pacific region, screening the typhoon samples before landfall, uniformly interpolating the typhoon path data into hourly data, and uniformly converting the central maximum wind speeds of different average time intervals;

[0011] SS2. Based on the pre-processed historical typhoon data and using the typhoon wind field model, the hourly wind field of each historical typhoon event in the Northwest Pacific is reconstructed. The annual maximum wind speed value corresponding to each grid is extracted from the hourly wind field distribution of all typhoons in the same year, forming the annual maximum wind speed sequence of each grid in the Northwest Pacific based on the reconstruction of the typhoon wind field;

[0012] SS3. Based on the ship surface wind observation data, the annual maximum wind speed of each grid in the northwest Pacific is calculated to form the annual maximum wind speed sequence of each grid in the northwest Pacific based on the ship surface wind;

[0013] SS4. Based on the satellite sea surface wind data, the maximum wind speed of each grid in the northwest Pacific Ocean is calculated to form the annual maximum wind speed sequence of each grid in the northwest Pacific Ocean based on the satellite sea surface wind.

[0014] SS5. Based on the annual maximum wind speed sequences of each grid in the northwest Pacific formed in steps SS2 to SS4, the maximum wind speed of each grid in a certain year based on typhoon wind field reconstruction, ship sea surface wind, and satellite sea surface wind is taken as the maximum wind speed of the grid in that year, and finally a comprehensive annual maximum wind speed sequence sample of each grid in the northwest Pacific is formed;

[0015] SS6. Based on the annual maximum wind speed sequence samples of each grid in the northwest Pacific obtained in step SS5, the annual maximum wind speed sequence of each grid is fitted based on different extreme value distribution models and combined with different parameter estimation methods. By comparing the goodness of fit between each model and parameter estimation method and the samples, the extreme value distribution model and parameter estimation method with the best goodness of fit are selected as the standard model of the grid to estimate the extreme wind speed of the grid with different return periods, and the spatial distribution of extreme wind speed with different return periods in the northwest Pacific is obtained accordingly.

[0016] Preferably, in the above step SS1, when performing the preprocessing operation of historical typhoon data, at least the following sub-steps should be included:

[0017] SS11. Collect and organize the historical typhoon data of the Northwest Pacific region within the preset time range. Each historical typhoon data should at least include the typhoon occurrence time, typhoon number, center longitude and latitude, center maximum wind speed and center pressure data;

[0018] SS12. Screen the pre-landing typhoon samples and determine whether each historical typhoon is a landfalling typhoon. A landfalling typhoon is defined as the state when the typhoon center passes over land. If it is a landfalling typhoon, only its pre-landfall data is retained and its post-landfall data is discarded. If it is a non-landfalling typhoon, its data is retained completely.

[0019] SS13. Interpolate the typhoon path data and adjust the typhoon path data with a recording interval greater than 1 hour to a data sequence updated once every hour through interpolation processing;

[0020] SS14. Perform uniform conversion of average wind speeds at different time intervals, and convert the central maximum wind speed data obtained at different average time intervals into the central maximum wind speed at an average time interval of 10 minutes.

[0021] Furthermore, in the above sub-step SS13, the interpolation processing adopts linear interpolation method or spline interpolation method, and adjusts the typhoon path data with a recording interval greater than 1 hour into a data sequence updated once every hour through interpolation processing to increase the probability of the typhoon wind circle passing through the grid.

[0022] Preferably, in the above step SS2, the typhoon wind field model adopts the Yan Meng typhoon wind field model, and the pre-processed hourly path data of historical typhoons in the northwest Pacific Ocean year by year are converted into hourly wind field data. For any grid g in the northwest Pacific Ocean, the hourly wind field distribution h of all typhoons in any year i in the preset time range is obtained. 1 ,h 2 ,h 3 …h m Extract the maximum wind speed value T of the i-th year corresponding to the grid g ig =max(h 1 ,h 2 ,h 3 …h m ), m is the total number of hourly wind field samples of the typhoon in the i-th year, and then the annual maximum wind speed sequence T of the grid g within the preset time range is obtained g ={T 1g ,T 2g ,T 3g ,T ig …T ng}, n is the total number of years in the preset time range, and then the annual maximum wind speed sequence of each grid in the northwest Pacific reconstructed based on the typhoon wind field is obtained in the same way.

[0023] Preferably, in the above step SS3, the ship sea surface wind observation data is ICOADS, and based on the ship sea surface wind observation data, the maximum wind speed P in any year i within a preset time range of any grid g in the northwest Pacific is counted. ig , and then obtain the annual maximum wind speed sequence P of the grid g within the preset time range g = {P 1g ,P 2g ,P 3g ,P ig …P ng}, n is the total year of the preset time range, and then the annual maximum wind speed sequence of each grid in the northwest Pacific based on the ship surface wind is formed in the same way.

[0024] Preferably, in the above step SS4, based on the satellite sea surface wind data, the maximum wind speed S in any year i in any grid g within a preset time range in the northwest Pacific Ocean is counted. ig , and then obtain the annual maximum wind speed sequence S of the grid g within the preset time range g ={S 1g ,S 2g ,S 3g ,S ig …S ng}, n is the total number of years in the preset time range, and then the annual maximum wind speed sequence of each grid in the northwest Pacific based on the satellite sea surface wind is formed in the same way.

[0025] Preferably, in the above step SS5, based on the annual maximum wind speed sequences of each grid formed in steps SS2 to SS4, the maximum value Y is taken from the maximum wind speed of any grid g in the i-th year based on typhoon wind field reconstruction, ship sea surface wind, and satellite sea surface wind. ig =max(T ig ,P ig ,S ig ) is taken as the maximum wind speed of grid g in the i-th year, and then the annual maximum wind speed sequence sample Y of grid g is formed. g = {Y 1g ,Y 2g ,Y 3g ,Y ig …Y ng}, and finally create a sample of the annual maximum wind speed series for each grid in the northwest Pacific.

[0026] Furthermore, in the above step SS5, when determining the maximum wind speed Y of any grid g in the i-th year, ig When T ig ,P ig ,S ig When any of the three is empty, the maximum value among the remaining values ​​is taken as the maximum wind speed of the grid in that year.

[0027] Preferably, in the above step SS6, the annual maximum wind speed sequence of each grid is fitted using four different extreme value distributions, including Gumbel distribution, three-parameter Weibull distribution, two-parameter Weibull distribution and generalized extreme value (GEV) distribution. Each extreme value distribution uses two parameter estimation methods, moment method and probability weighted moment method, to estimate parameters, and the Kolmogorov-Smirnov (KS) test is used to evaluate the goodness of fit of each combination, and the D in the KS test result is selected. nThe combination corresponding to the smallest value is used as the optimal extreme value distribution and parameter estimation method for the grid, and the extreme wind speeds with different return periods of the grid are estimated using the optimal extreme value distribution and parameter estimation method, and then the spatial distribution of extreme wind speeds with different return periods in the northwest Pacific is obtained.

[0028] Furthermore, in the above step SS6, D n The value is the statistical result of the KS test, and its mathematical expression is D n =max{d i}, d i is the absolute value of the difference between the theoretical distribution and the empirical distribution. n The value is less than the critical value D of the KS test α When , it is considered that the theoretical distribution function of the original hypothesis fits the empirical distribution of the sample sequence well. If there are multiple combination methods that meet the significance test results, then according to the minimum D n Value selection of optimal extreme value distribution and its parameter estimation method.

[0029] (III) Technical Effect

[0030] Compared with the prior art, the method for estimating extreme wind speeds with different return periods in the northwest Pacific Ocean proposed in the present invention has the following beneficial effects:

[0031] (1) The method for estimating extreme wind speeds of different return periods in the northwest Pacific Ocean of the present invention solves the problem of extremely scarce annual maximum wind speed sequence samples in the northwest Pacific Ocean, and the annual maximum wind speed sequence is composed of annual maximum wind speed sequences based on typhoon wind field reconstruction, ship sea surface wind, and satellite sea surface wind. It avoids the problem that wind speed samples cannot be obtained by typhoon wind field simulation when no typhoon passes through certain areas. The method of the present invention significantly increases the number and coverage of samples through the comprehensive utilization of multi-source data, and improves the representativeness of the data and the estimation accuracy.

[0032] (2) The present invention selects the optimal extreme value distribution and parameter estimation method for each grid through a goodness of fit test, and then estimates the extreme wind speeds of different return periods in each grid, ensuring the applicability and accuracy of extreme wind speed estimation in different regions. This method can effectively integrate and analyze various types of data, and can accurately estimate extreme wind speeds even in areas with sparse data, and has strong universality.

[0033] (3) The extreme wind speed estimation method with different return periods of the present invention can be used to estimate the extreme wind speed with different return periods in the entire northwest Pacific Ocean, which is helpful to design more stable and safe marine engineering structures. It is of great significance to improve the reliability of marine engineering operations, strengthen marine disaster prevention and mitigation, and reduce the risks and losses caused by natural disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 It is a schematic diagram of the method for estimating extreme wind speeds with different return periods in the northwest Pacific Ocean according to the present invention.

[0036] Figure 2 This is the distribution map of the annual maximum wind speed in the northwest Pacific in 2015, where (a)-(c) are calculated from typhoon wind field reconstruction, ship sea surface wind data, and satellite sea surface wind data, respectively, and (d) is obtained by taking the maximum value of the above three types of data in the corresponding grid. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. The described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present invention, and should not be understood as limiting the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0038] In order to solve the problem of insufficient sample data of the annual maximum wind speed series in the northwest Pacific, the present invention proposes a method for estimating extreme wind speeds with different return periods in the northwest Pacific. As a specific example, Figure 1 As shown, the present invention establishes the annual maximum wind speed sequence sample of the northwest Pacific Ocean based on typhoon wind field reconstruction, ship sea surface wind, and satellite sea surface wind, and estimates the extreme wind speeds in different return periods. The method includes at least the following steps when implemented:

[0039] SS1. Preprocessing of historical typhoon data

[0040] The preprocessing of historical typhoon data at least includes collecting historical typhoon data within a preset time range in the northwest Pacific region, screening typhoon samples before landfall, uniformly interpolating typhoon path data into hourly data, and uniformly converting the central maximum wind speeds at different average time intervals into 10-minute average wind speeds.

[0041] In some preferred examples, this step, when preprocessing the historical typhoon data, is based on the best path data set of tropical cyclones in the northwest Pacific compiled by the Shanghai Typhoon Research Institute of the China Meteorological Administration, such as typhoon occurrence time, typhoon number, typhoon name, typhoon center longitude and latitude, center maximum wind speed, center pressure, etc. First, select the historical typhoons within the northwest Pacific research area; second, screen the typhoon samples, and for the landing typhoons, only take the samples before landing; third, interpolate the typhoon path data every 6 hours or every 3 hours into hourly path data according to the linear interpolation method or the spline interpolation method; finally, convert the 2-minute average center maximum wind speed in the tropical cyclone best path data set into the 10-minute average center maximum wind speed. Finally, through preprocessing, select the valid historical typhoon data.

[0042] SS2. Reconstruct the historical typhoon wind field in the northwest Pacific using the typhoon wind field model and extract the annual maximum wind speed series

[0043] Based on the preprocessed historical typhoon data and using the typhoon wind field model to simulate and reconstruct the hourly wind field of each historical typhoon event in the northwest Pacific every year, the annual maximum wind speed value corresponding to each grid is extracted from the hourly wind field distribution of all typhoons in the same year, forming the annual maximum wind speed sequence of each grid in the northwest Pacific based on the reconstruction of the typhoon wind field.

[0044] In some preferred embodiments, when implementing this step, firstly, the pre-processed historical typhoon data is brought into the Yan Meng typhoon wind field model to reconstruct the hourly wind field of each typhoon in each year. For any grid g in the northwest Pacific, the hourly wind field distribution h of all typhoons in any year i in the preset time range is obtained. 1 ,h 2 ,h 3 …h m Extract the maximum wind speed value T of the i-th year corresponding to the grid g ig =max(h 1 ,h 2 ,h 3 …h m ), m is the total number of hourly wind field samples of the typhoon in the i-th year, and then the annual maximum wind speed sequence T of the grid g within the preset time range is obtained g ={T 1g ,T 2g ,T 3g ,T ig …T ng}, n is the total number of years in the preset time range, and then the annual maximum wind speed sequence of each grid in the northwest Pacific reconstructed based on the typhoon wind field is obtained in the same way.

[0045] SS3. Calculation of the annual maximum wind speed sequence for each grid in the Northwest Pacific using ship surface wind data

[0046] Based on the ship surface wind observation data, the annual maximum wind speed of each grid in the northwest Pacific is calculated to form the annual maximum wind speed sequence of each grid in the northwest Pacific based on ship surface wind.

[0047] In some preferred embodiments, in this step, when calculating the annual maximum wind speed sequence of each grid in the northwest Pacific using the ship surface wind data, firstly, based on the ICOADS (The International Comprehensive Ocean-Atmosphere Data Set) ship surface wind data, the maximum wind speed P of any grid g in the northwest Pacific in any year i within a preset time range is counted. ig , and then obtain the annual maximum wind speed sequence P of the grid g within the preset time range g = {P 1g ,P 2g ,P 3g ,P ig …P ng}, n is the total year of the preset time range, and then the annual maximum wind speed sequence of each grid in the northwest Pacific based on the ship surface wind is formed in the same way.

[0048] SS4. Calculation of the annual maximum wind speed series for each grid in the Northwest Pacific using satellite sea surface wind data

[0049] Based on satellite sea surface wind data, the annual maximum wind speed of each grid in the northwest Pacific is calculated to form the annual maximum wind speed sequence of each grid in the northwest Pacific based on satellite sea surface wind.

[0050] In some preferred embodiments, when implementing this step, firstly, based on the satellite sea surface wind data, the maximum wind speed S of any year i in any grid g within a preset time range in the northwest Pacific Ocean is counted. ig , and then obtain the annual maximum wind speed sequence S of the grid g within the preset time range g ={S 1g ,S 2g ,S 3g ,S ig …S ng}, n is the total number of years in the preset time range, and then the annual maximum wind speed sequence of each grid in the northwest Pacific based on the satellite sea surface wind is formed in the same way.

[0051] It should be noted that since different satellites have different on-orbit times, in practical applications, the satellite sea surface wind data of the corresponding time period can be selected according to the time range of the sample.

[0052] SS5. Determine the annual maximum wind speed sequence samples for each grid in the northwest Pacific using the annual maximum wind speed sequence based on typhoon wind field reconstruction, ship surface wind, and satellite surface wind.

[0053] Based on the annual maximum wind speed sequences of each grid in the northwest Pacific formed in steps SS2 to SS4, the maximum value of the maximum wind speed of each grid in a certain year based on typhoon wind field reconstruction, ship sea surface wind, and satellite sea surface wind is taken as the maximum wind speed of the grid in that year, and finally a comprehensive annual maximum wind speed sequence sample of each grid in the northwest Pacific is formed.

[0054] In some preferred embodiments, when implementing this step, firstly, based on the reconstruction of the typhoon wind field, the ship surface wind, and the satellite surface wind in the northwest Pacific Ocean, any grid g takes the maximum value Y in the maximum wind speed in the i-th year. ig =max(T ig ,P ig ,S ig ) is taken as the maximum wind speed of grid g in the i-th year, and then the annual maximum wind speed sequence sample Y of grid g is formed. g = {Y 1g ,Y 2g ,Y 3g ,Y ig …Y ng}, and finally create a sample of the annual maximum wind speed series for each grid in the northwest Pacific.

[0055] In determining the maximum wind speed Y of grid g in year n ng When T ng ,P ng ,S ng If any of the three is empty, the maximum value among the remaining values ​​is taken as the maximum wind speed of the grid in that year. ng When it is empty, from T ng ,P ng The maximum value is taken as the maximum wind speed of the grid in that year.

[0056] Since the number, intensity and affected sea areas of typhoons each year are different, some sea areas do not have typhoons in some years, and the annual maximum wind speed of some grids obtained by typhoon wind field reconstruction has missing values ​​in some years. For example, Figure 2 a shows the spatial distribution of the maximum wind speed in the northwest Pacific in 2015 reconstructed based on the typhoon wind field. It can be seen that the annual maximum wind speed values ​​are missing in some sea areas in the northern part of the Sea of ​​Japan and the southwestern part of the South China Sea. Therefore, when the maximum wind speed in 2015 of a grid reconstructed based on the typhoon wind field is empty ( Figure 2 a), then the maximum wind speed in 2015 of the corresponding grid based on the ship surface wind ( Figure 2 b) Maximum wind speed in 2015 of the corresponding grid based on satellite sea surface wind ( Figure 2 c) and select the maximum value as the final maximum wind speed in 2015 for this grid ( Figure 2 d).

[0057] SS6. Estimation of extreme wind speeds with different return periods in the Northwest Pacific using different extreme value distribution models and parameter estimation methods

[0058] Based on the annual maximum wind speed sequence samples of each grid in the northwest Pacific obtained in step SS5, the annual maximum wind speed sequence of each grid is fitted based on different extreme value distribution models and combined with different parameter estimation methods. By comparing the goodness of fit between each model and parameter estimation method and the samples, the extreme value distribution model and parameter estimation method with the best goodness of fit are selected as the standard model of the grid to estimate the extreme wind speed of different return periods of the grid, and the spatial distribution of extreme wind speed of different return periods in the northwest Pacific is obtained accordingly.

[0059] In some preferred embodiments, this step is first implemented by using different extreme value distribution models and parameter estimation methods to estimate the annual maximum wind speed sequence samples of the extreme wind speeds in different return periods in the northwest Pacific for the grid g, and fitting them using Gumbel, 3-parameter Weibull, 2-parameter Weibull, and GEV distributions. Each extreme value distribution uses two parameter estimation methods, the moment method and the probability weighted moment method (a total of 8 combination methods), and the following KS test method is used to select the optimal combination from the 8 combination methods, so as to obtain the optimal extreme value distribution and parameter estimation method of the grid:

[0060] D n =max{d i}

[0061] Among them, D n is the KS statistical result, d i It is the absolute value of the difference between the cumulative frequency of the theoretical distribution and the empirical distribution. n Less than KS test critical D α When , it is considered that the theoretical distribution function of the original hypothesis fits the empirical distribution of the sample sequence well. If there are multiple combination methods that meet the significance test results, then according to the minimum D n Value selection of optimal extreme value distribution and its parameter estimation method.

[0062] Then, the extreme wind speeds with different return periods of the grid g can be obtained through the optimal extreme value distribution of the grid g and its parameter estimation method. Similarly, the distribution of extreme wind speeds with different return periods in the entire northwest Pacific can be obtained.

[0063] In summary, the present invention proposes a new method for estimating extreme wind speeds with different return periods in the northwest Pacific. This method constructs the annual maximum wind speed sequence samples of each grid in the northwest Pacific by integrating multi-source heterogeneous data such as historical typhoon wind field data, ship sea surface wind observation data and satellite sea surface wind remote sensing data; and adopts a variety of extreme value distribution models and parameter estimation methods for fitting, and selects the optimal extreme value distribution and parameter estimation method for each grid through KS test, so as to obtain the extreme wind speed estimation value under different return periods of the grid. This method not only solves the problem of lack of sample data, but also improves the estimation accuracy by selecting the optimal model. It has strong applicability and practical value, and provides important data support for marine engineering design site selection, offshore facility deployment, etc.

[0064] This embodiment is only a preferred specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for estimating extreme wind speeds with different return periods in the northwest Pacific, characterized in that: The method comprises at least the following steps when implemented: SS1. Preprocessing of historical typhoon data includes at least collecting historical typhoon data within a preset time range in the Northwest Pacific region, screening pre-landfall typhoon samples, uniformly interpolating typhoon path data into hourly data, and uniformly converting the central maximum wind speeds of different average time intervals. The preprocessing of historical typhoon data should at least include the following sub-steps: SS11. Collect and organize the historical typhoon data of the Northwest Pacific region within the preset time range. Each historical typhoon data should at least include the typhoon occurrence time, typhoon number, center longitude and latitude, center maximum wind speed and center pressure data; SS12. Screen the pre-landing typhoon samples and determine whether each historical typhoon is a landfalling typhoon. A landfalling typhoon is defined as the state of a typhoon center passing over land. If it is a landfalling typhoon, only its pre-landfall data is retained and its post-landfall data is discarded. If it is a non-landfalling typhoon, its data is retained intact. SS13. Interpolate the typhoon path data and adjust the typhoon path data with a recording interval greater than 1 hour to a data sequence updated every hour through interpolation; SS14. Perform uniform conversion of average wind speeds at different time intervals, and convert the central maximum wind speed data obtained at different average time intervals into the central maximum wind speed at 10-minute average time interval; SS2. Based on the pre-processed historical typhoon data and using the typhoon wind field model, the hourly wind field of each historical typhoon event in the Northwest Pacific is reconstructed. The annual maximum wind speed value corresponding to each grid is extracted from the hourly wind field distribution of all typhoons in the same year, forming the annual maximum wind speed sequence of each grid in the Northwest Pacific based on the reconstruction of the typhoon wind field; SS3. Based on the ship surface wind observation data, the maximum wind speed of each grid in the northwest Pacific Ocean is calculated to form the annual maximum wind speed sequence of each grid in the northwest Pacific Ocean based on the ship surface wind; SS4. Based on the satellite sea surface wind data, the maximum wind speed of each grid in the northwest Pacific Ocean is calculated to form the annual maximum wind speed sequence of each grid in the northwest Pacific Ocean based on the satellite sea surface wind. SS5. Based on the annual maximum wind speed sequences of each grid in the Northwest Pacific formed in steps SS2 to SS4, the maximum wind speed of each grid in a certain year based on typhoon wind field reconstruction, ship surface wind, and satellite surface wind is taken as the maximum wind speed of the grid in that year, and finally a comprehensive annual maximum wind speed sequence sample of each grid in the Northwest Pacific is formed; SS6. Based on the annual maximum wind speed series samples of each grid in the northwest Pacific obtained in step SS5, the annual maximum wind speed series of each grid is fitted based on different extreme value distribution models and combined with different parameter estimation methods. By comparing the goodness of fit between each model and parameter estimation method and the samples, the extreme value distribution model and parameter estimation method with the best goodness of fit are selected as the standard model of the grid to estimate the extreme wind speeds of different return periods of the grid, and the spatial distribution of extreme wind speeds of different return periods in the northwest Pacific is obtained accordingly.

2. The method according to claim 1, characterized in that: In the above sub-step SS13, the interpolation process uses linear interpolation or spline interpolation to adjust the typhoon path data with a recording interval greater than 1 hour into a data sequence updated once every hour through interpolation to increase the probability of the typhoon wind circle passing through the grid.

3. The method according to claim 1, characterized in that In the above step SS2, the typhoon wind field model adopts the Yan Meng typhoon wind field model, which converts the pre-processed hourly path data of historical typhoons in the northwest Pacific Ocean year by year into hourly wind field data. g , any year from the preset time range i Hourly wind distribution of all typhoons in h 1, h 2, h 3… h m Extract the corresponding grid g No. i Maximum wind speed of the year T ig =max( h 1, h 2, h 3… h m ), m For the i The total number of hourly wind field samples of the annual typhoon is obtained, and then the grid is obtained within the preset time range. g The annual maximum wind speed series T g ={ T 1g , T 2g , T 3g , T ig … T ng }, n is the total year of the preset time range, and then the annual maximum wind speed sequence of each grid in the northwest Pacific reconstructed based on the typhoon wind field is obtained in the same way.

4. The method according to claim 1, characterized in that In the above step SS3, the ship sea surface wind observation data is ICOADS. Based on the ship sea surface wind observation data, statistics are collected for any grid in the northwest Pacific. g Any year within the preset time range i Maximum wind speed P ig , and then get the grid within the preset time range g The annual maximum wind speed series P g ={ P 1g , P 2g , P 3g , P ig … P ng }, n The total years of the preset time range are then used to form the annual maximum wind speed sequence of each grid in the northwest Pacific based on the ship surface wind in the same way.

5. The method according to claim 1, characterized in that In the above step SS4, based on the satellite sea surface wind data, statistics are collected for any grid in the northwest Pacific. g Any year within the preset time range i Maximum wind speed S ig , and then get the grid within the preset time range g The annual maximum wind speed series S g ={ S 1g , S 2g , S 3g , S ig … S ng }, n The total years of the preset time range are then used to form the annual maximum wind speed series of each grid in the northwest Pacific based on the satellite sea surface wind in the same way.

6. The method according to claim 1, characterized in that In the above step SS5, based on the annual maximum wind speed sequence of each grid formed in steps SS2 to SS4, from any grid based on typhoon wind field reconstruction, ship sea surface wind, satellite sea surface wind g In the i The maximum wind speed in the year Y ig =max( T ig , P ig , S ig ) as a grid g In the i The maximum wind speed of the year, thus forming the grid g The annual maximum wind speed series sample Y g ={ Y 1g , Y 2g , Y 3g , Y ig … Y ng }, and finally create a sample of the annual maximum wind speed series for each grid in the northwest Pacific.

7. The method according to claim 6, characterized in that In the above step SS5, when determining any grid g In the i Maximum wind speed of the year Y ig When T ig , P ig , S ig When any of the three is empty, the maximum value among the remaining values ​​is taken as the maximum wind speed of the grid in that year.

8. The method according to claim 1, characterized in that In the above step SS6, the annual maximum wind speed series of each grid is fitted using four different extreme value distributions, including Gumbel distribution, three-parameter Weibull distribution, two-parameter Weibull distribution and GEV distribution. Each extreme value distribution uses two parameter estimation methods, moment method and probability weighted moment method, to estimate the parameters, and uses KS Test to evaluate the goodness of fit of each combination and select KS The smallest test result D n The combination corresponding to the values ​​is used as the optimal extreme value distribution and parameter estimation method for the grid, and the extreme wind speeds with different return periods of the grid are estimated using the optimal extreme value distribution and parameter estimation method, and then the spatial distribution of extreme wind speeds with different return periods in the northwest Pacific is obtained.

9. The method according to claim 8, characterized in that In the above step SS6, D n The value is the statistical result of the KS test, and its mathematical expression is D n =max{ d i }, d i is the absolute value of the difference between the theoretical distribution and the empirical distribution. D n The value is less than the critical value of the KS test D α When , it is considered that the theoretical distribution function of the original hypothesis fits the empirical distribution of the sample sequence well. If there are multiple combination methods that meet the significance test results, then according to the minimum D n Value selection of optimal extreme value distribution and its parameter estimation method.

Images