An adaptive window fusion method for wave height capture in large wave processes

Abstract

The invention belongs to the technical field of wave observation, and proposes an adaptive window fusion method for capturing wave height in a large wave process, the method comprising: given a target area, collecting effective wave height data of multi-source satellite altimeters in the target area; determining the initial spatial resolution and initial time resolution of the wave height field in the target area; based on the time-space inverse distance weighted fusion method, generating a wave height sequence of target grid points in the target area at the initial time resolution; performing large wave process identification on the wave height sequence, capturing peak information in the large wave process; refining the fusion process of the preset time period threshold before and after the peak moment, adaptively adjusting the time resolution, and recording the dynamic change of the wave height sequence and the corresponding peak information in the large wave process at the adjusted time resolution. The method realizes the fusion of multi-source satellite data under different sea conditions, takes into account the fusion efficiency and fusion accuracy, and autonomously judges and identifies the large wave process.

Description

Technical Field

[0001] The invention belongs to the technical field of wave observation, and relates to an adaptive window fusion method and device for capturing wave height in a large wave process. Background Art

[0002] Significant wave height (SWH) is an important parameter to characterize wave conditions, and plays an important role in marine engineering planning and design, marine energy utilization, marine disaster warning and protection, etc. However, current wave observation methods have limitations in obtaining high spatial resolution wave field data. Therefore, multi-source satellite altimeter data fusion came into being.

[0003] In the study of multi-source satellite altimeter fusion wave field, the description of the daily average wave field provides a reliable data basis for the assessment of the overall sea conditions. However, the analysis of instantaneous wave fields is currently still based on fixed time points, such as data fields at one moment of the day. This fusion mode may not be able to fully capture the rapid changes in the wave field, especially the peak wave height information, when dealing with extreme weather such as typhoons. During typhoons, due to the strong and rapid changes in the wind field, the instantaneous changes in the wave field are extremely large, so the traditional instantaneous wave field fusion method may not be able to effectively capture these rapid and drastic changes. In this case, the peak information of extreme wave heights is often weakened or omitted, and it is impossible to provide sufficient accuracy support for subsequent marine monitoring and disaster prevention and mitigation. However, under normal sea conditions, maintaining a high temporal resolution fusion mode may lead to redundancy in data processing. Because the sea conditions are relatively stable and the wave field changes slowly, high-frequency fusion calculations often cannot significantly improve the accuracy of the data, but may increase the complexity of the calculation and resource consumption. Therefore, in marine engineering applications, it is necessary to reasonably select the time resolution (adaptive) of the fusion mode according to the actual sea conditions and research needs, to achieve high efficiency of data processing under different sea conditions, especially to capture the changing details of the peak wave height during large waves. This is the main technical problem solved by the present invention. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention proposes an adaptive window fusion method for capturing wave height in a large wave process. The method adopts an adaptive time resolution method to realize multi-source satellite data fusion under different sea conditions, taking into account both fusion efficiency and fusion accuracy, and independently judging and identifying large wave processes.

[0005] In order to achieve the above object, the present invention provides an adaptive window fusion method for capturing wave height in a large wave process, comprising:

[0006] Given a target area, collect multi-source satellite altimeter effective wave height data within the target area;

[0007] Determine the initial spatial resolution and initial temporal resolution of the wave height field of the target area according to the effective wave height data of the multi-source satellite altimeter;

[0008] Based on the time-space inverse distance weighted fusion method, the wave height sequence of the target grid points in the target area is generated at the initial time resolution;

[0009] Performing big wave process identification on the wave height sequence to capture peak information during the big wave process, the peak information including the wave height peak value and the peak time corresponding to the wave height peak value;

[0010] The fusion process of the preset time period thresholds before and after the peak moment is refined, and the time resolution is adaptively adjusted. At the adjusted time resolution, the dynamic changes of the wave height sequence during the big wave process and the corresponding peak information are recorded.

[0011] In some embodiments, based on the spatiotemporal inverse distance weighted fusion method, the formula for obtaining the wave height sequence of the target grid points in the target area is:

[0012] ,

[0013] in, Z is the wave height sequence at the target grid points in the target area; is the satellite trajectory observation data; x , y , t are the latitude and longitude coordinates of the target grid point and the corresponding time value; , , is the latitude and longitude coordinates of the satellite trajectory observation point i and the corresponding time value, i∈(1,N), N and m are parameters; It is a weight factor that comprehensively weighs the distance factor and the time factor; It includes the time factor The distance between the satellite trajectory observation point and the target grid point; c is a factor that defines the relationship between spatial units and temporal units.

[0014] In some embodiments, in the process of identifying the big wave process of the wave height sequence, several local peaks are identified, and the several peaks are screened and filtered to obtain the peak information.

[0015] In some embodiments, multiple sample points of the wave height sequence are collected, each sample point is compared with other points in the neighborhood of the sample point, and a local maximum value is identified as a local peak value.

[0016] In some embodiments, a first threshold is set according to the wave characteristics of the target area, and the first threshold is used to screen and filter the plurality of peaks.

[0017] In some embodiments, the initial spatial resolution and the initial temporal resolution are determined on the condition that the wave height field of the target area meets a preset accuracy threshold and the effective wave height data of the multi-source satellite altimeter is not distorted during the fusion process.

[0018] In some embodiments, adaptively adjusting the temporal resolution includes adaptively increasing the temporal resolution.

[0019] In some embodiments, the effective wave height data of the multi-source satellite altimeter in the target area is data obtained by pre-processing operations.

[0020] It can be seen from the above scheme that the advantages of the present invention are:

[0021] The adaptive window fusion method for capturing wave height in a large wave process disclosed by the present invention collects effective wave height data of multi-source satellite altimeters in a given target area; and determines the initial spatial resolution and initial time resolution of the wave height field in the target area based on the effective wave height data of the multi-source satellite altimeters; then, based on the time-space inverse distance weighted fusion method, generates a wave height sequence of target grid points in the target area at the initial time resolution; and identifies the large wave process for the wave height sequence, captures the peak information in the large wave process, and the peak information includes the wave height peak value and the peak moment corresponding to the wave height peak value; then, refines the fusion process of the preset time period threshold before and after the peak moment, adaptively adjusts the time resolution, and records the dynamic changes of the wave height sequence in the large wave process and the corresponding peak information at the adjusted time resolution. The method adopts an adaptive time resolution method to realize the fusion of multi-source satellite data under different sea conditions, taking into account the fusion efficiency and fusion accuracy, and autonomously judges and identifies the large wave process, amplifies the dynamic changes of wave height under high sea conditions, and captures the peak wave height information, which plays an important role in the fields of extreme value analysis, marine structure protection and fortification standards in marine engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The figure shows a schematic diagram of the overall process of an adaptive window fusion method for capturing wave height in a large wave process provided by an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of the target area and the trajectory of the selected multi-source satellite altimeter is shown;

[0024] Figure 3The time series comparison diagrams of the effective wave height obtained by the present invention and the related technology and the effective wave height actually measured by the buoy (hourly and daily) are shown, wherein (a) is a time series comparison of the effective wave height obtained by the fixed time (daily) fusion method in the related technology and the effective wave height actually measured by the buoy (hourly and daily); (b) is a time series comparison of the effective wave height obtained by the method of the present invention and the effective wave height actually measured by the buoy (hourly);

[0025] Figure 4 A scatter point comparison diagram of the effective wave height obtained by the present invention and related technologies and the effective wave height actually measured by the buoy at the peak value is shown, wherein (a) is a scatter point comparison of the effective wave height obtained by the fixed time (daily) fusion method in the related technology and the effective wave height actually measured by the buoy at the peak value; (b) is a scatter point comparison of the effective wave height obtained by the method of the present invention and the effective wave height actually measured by the buoy at the peak value. DETAILED DESCRIPTION

[0026] In order to make the above features and effects of the present invention more clearly understood, embodiments are given below and described in detail with reference to the accompanying drawings.

[0027] See also Figure 1 As shown, Figure 1 The figure shows the overall flow chart of an adaptive window fusion method for capturing wave height in a large wave process provided by an embodiment of the present invention.

[0028] The adaptive window fusion method for capturing wave height in a large wave process specifically comprises the following steps:

[0029] Step S1: Given a target area, collect effective wave height data of multi-source satellite altimeters in the target area.

[0030] In this embodiment, the target area is the sea area to be measured set according to the needs, and the target area and the selected satellite altimeter trajectory are as follows: Figure 2 shown.

[0031] In this embodiment, the effective wave height data of the multi-source satellite altimeter collected in the target area is data obtained by a preprocessing operation in advance. The preprocessing operation is specifically as follows: In order to ensure the consistency of the wave height data of the multi-source satellite altimeter, it is necessary to pre-eliminate the data of unqualified quality, such as wave data under conditions such as land, rainfall, and sea ice. First, by setting an upper limit threshold, the outliers are preliminarily eliminated. Then, the multi-source satellite altimeter wave height data after the preliminarily exclusion of outliers is further divided into multiple blocks, which can form a sufficiently long segment to ensure that the data statistics are reliable. Then, according to the median absolute deviation, the outliers in each block are identified and eliminated. Specifically: MAD is defined as:

[0032] ,

[0033] in, , is the data of the ith block; b is the scaling factor of the Gaussian distribution, which is 1.4826. k is the threshold, usually set to 3, and all values ​​that do not meet the absolute deviation median condition are classified as outliers. The absolute deviation median condition is set as:

[0034] ,

[0035] Next, the block with outliers is further subdivided into multiple sub-blocks. Specifically, a certain area around each outlier can be divided into a sub-block. According to the above-mentioned identification standard of outliers in the block, check whether there are still outliers in each sub-block, and further remove the outliers. Finally, calculate the standard deviation and mean value of each sub-block data. If the ratio is greater than a preset threshold (for example, 0.5), delete the entire sub-block.

[0036] Through the above preprocessing process, the effective wave height data of multi-source satellite altimeters are obtained.

[0037] Step S2: determining the initial spatial resolution and initial temporal resolution of the wave height field in the target area based on the effective wave height data of the multi-source satellite altimeter.

[0038] The collected effective wave height data of multi-source satellite altimeters are affected by the number of satellites and the quality of the data itself. To select appropriate spatial resolution and temporal resolution, in some embodiments, the initial spatial resolution and initial temporal resolution are determined on the condition that the wave height field of the target area meets the preset accuracy threshold and the effective wave height data of the multi-source satellite altimeter is not distorted during the fusion process. This ensures that the generated wave height field meets certain accuracy conditions without causing data distortion due to excessive fusion. For example, the initial spatial resolution is set to 0.125° × 0.125°, and the initial temporal resolution is set to 1 day.

[0039] Step S3: Based on the spatiotemporal inverse distance weighted fusion method, a wave height sequence of target grid points in the target area is generated at the initial time resolution.

[0040] In this embodiment, a weighted fusion method based on the time-space inverse distance is adopted, which not only takes into account the distance factor, but also combines the time factor, and comprehensively considers the dynamic changes of waves in time and space. Therefore, it can not only fuse the daily average wave height, but also the wave height at any time.

[0041] Based on the time-space inverse distance weighted fusion method, the formula for generating the wave height sequence of the target grid points in the target area at the initial time resolution is:

[0042] ,

[0043] in, Z is the wave height sequence at the target grid points in the target area; is the satellite trajectory observation data; x , y , t are the latitude and longitude coordinates of the target grid point and the corresponding time value; , , is the latitude and longitude coordinates of the satellite trajectory observation point i and the corresponding time value, i∈(1,N), N and m are parameters; It is a weight factor that comprehensively weighs the distance factor and the time factor; It includes the time factor The distance between the satellite trajectory observation point and the target grid point; c It is a factor used to define the relationship between spatial units and time units, which is the ratio of spatial units to time units.

[0044] Step S4, performing a big wave process identification on the wave height sequence to capture peak information during the big wave process, wherein the peak information includes a wave height peak value and a peak time corresponding to the wave height peak value.

[0045] In some embodiments, in step S4, a plurality of sample points of the wave height sequence are first collected, and each sample point is compared with other points in the neighborhood of the sample point to identify the local maximum value as a local peak value. For example, for a certain sample point x(i), if its value is greater than the values ​​of other points x(i-1), x(i-2), x(i-3)... and x(i+1), x(i-2), x(i-3)... in its neighborhood, then x(i) is regarded as a local peak value. In this way, a plurality of local peak values ​​are obtained, and the plurality of peak values ​​are further screened and filtered to obtain the peak value information. In some embodiments, a first threshold value can be set according to the wave characteristics of the target area, and the plurality of peak values ​​can be screened and filtered using the first threshold value. In one embodiment, the first threshold value is set to 2.5 m.

[0046] Step S5, refine the fusion process of the preset time period thresholds before and after the peak moment, adaptively adjust the time resolution, and record the dynamic changes of the wave height sequence and the corresponding peak information during the big wave process at the adjusted time resolution.

[0047] In this embodiment, after the peak information is identified, the peak value and peak time of the wave height recorded by the target grid point are further located, and the fusion process of the preset time period threshold before and after the peak moment is refined, and the time resolution is adaptively adjusted. In one embodiment, the time resolution can be gradually improved, but the present invention is not limited to this. Then, at the adjusted time resolution, the dynamic changes of the wave height sequence during the big wave process and the corresponding peak information are re-recorded to form a complete system with both low sea state data at low time resolution and high sea state data at high time resolution.

[0048] To verify the effect of the adaptive window fusion method for wave height capture in large wave processes of the present invention, especially under the influence of extreme weather such as typhoons. Figure 3 The fused wave time series during the 2018 typhoon Maria are shown, where (a) is a time series comparison between the effective wave height obtained by the fixed-time (daily) fusion method in the relevant technology and the effective wave height measured by the buoy (hourly and daily); (b) is a time series comparison between the effective wave height obtained by the method of the present invention and the effective wave height measured by the buoy (hourly). The results show that although the fused wave height at a fixed time (daily) in the relevant technology is consistent with the daily measured wave height, the peak information displayed in the hourly measured wave height sequence is missed by the fixed-time fusion method. The method of the present invention accurately captures this peak wave height and refines this big wave process around July 11, 2018.

[0049] At the same time, in order to verify the reliability of the peak information capture of the present invention during the big wave process, Figure 4 The scatter point comparison of the peak wave height in the measured wave height sequence of the buoy and the fusion result is presented, where (a) is a scatter point comparison of the effective wave height obtained by the fixed-time (daily) fusion method of the related technology and the measured effective wave height of the buoy at the peak; (b) is a scatter point comparison of the effective wave height obtained by the method of the present invention and the measured effective wave height of the buoy at the peak. The results show that the former significantly underestimates the measured data (the deviation is -0.624 m), especially for wave heights above 4 m, which indicates that the fixed-time fusion method used in the related technology will weaken or omit the peak information. Compared with the former, the results of the present invention are more consistent with the measured data, with a deviation of only 0.168m, which is suitable for capturing the peak wave height in large wave processes.

[0050] In view of this, the present invention discloses an adaptive window fusion method for capturing wave height in a large wave process. The method collects effective wave height data of a multi-source satellite altimeter in a given target area; and determines the initial spatial resolution and the initial time resolution of the wave height field in the target area according to the effective wave height data of the multi-source satellite altimeter; then, based on a spatiotemporal inverse distance weighted fusion method, generates a wave height sequence of target grid points in the target area at the initial time resolution; and performs large wave process identification on the wave height sequence to capture peak information in the large wave process, the peak information including the wave height peak value and the peak moment corresponding to the wave height peak value; and then refines the fusion process of a preset time period threshold before and after the peak moment, adaptively adjusts the time resolution, and records the dynamic change of the wave height sequence and the corresponding peak information in the large wave process at the adjusted time resolution. This method uses adaptive time resolution to achieve multi-source satellite data fusion under different sea conditions, taking into account both fusion efficiency and fusion accuracy, and independently judges and identifies large wave processes, amplifies the dynamic changes of wave height under high sea conditions, and captures peak wave height information. It plays an important role in extreme value analysis in marine engineering, marine structure protection and fortification standards.

[0051] In addition, it should be noted that the adaptive window fusion method for capturing wave height in a large wave process disclosed in the present invention is not limited to being executed in the order of the above steps S1-S5. The steps can be reordered, added or deleted. As long as the expected results of the technical solution of the present invention can be achieved, the present invention is not limited here.

[0052] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An adaptive window fusion method for capturing wave height in large wave processes, characterized in that: Include: Given a target area, collect multi-source satellite altimeter effective wave height data within the target area; Determine the initial spatial resolution and initial temporal resolution of the wave height field of the target area according to the effective wave height data of the multi-source satellite altimeter; Based on the time-space inverse distance weighted fusion method, the wave height sequence of the target grid points in the target area is generated at the initial time resolution; Performing big wave process identification on the wave height sequence to capture peak information during the big wave process, the peak information including the wave height peak value and the peak time corresponding to the wave height peak value; The fusion process of the preset time period thresholds before and after the peak moment is refined, and the time resolution is adaptively adjusted. At the adjusted time resolution, the dynamic changes of the wave height sequence during the big wave process and the corresponding peak information are recorded.

2. The method according to claim 1, characterized in that in, Based on the time-space inverse distance weighted fusion method, the formula for obtaining the wave height sequence of the target grid points in the target area is: ， in, Z is the wave height sequence at the target grid points in the target area; is the satellite trajectory observation data; x , y , t are the latitude and longitude coordinates of the target grid point and the corresponding time value; , , is the latitude and longitude coordinates of the satellite trajectory observation point i and the corresponding time value, i∈(1,N), N and m are parameters; It is a weight factor that comprehensively weighs the distance factor and the time factor; It includes the time factor The distance between the satellite trajectory observation point and the target grid point; c It is a factor used to define the relationship between spatial units and time units, which is the ratio of spatial units to time units.

3. The method according to claim 1, characterized in that: In the process of identifying the big wave process of the wave height sequence, several local peaks are identified, and the peak information is obtained by screening and filtering the several peaks.

4. The method according to claim 3, characterized in that A plurality of sample points of the wave height sequence are collected, each sample point is compared with other points in the neighborhood of the sample point, and a local maximum value is identified as a local peak value.

5. The method according to claim 3 or 4, characterized in that: A first threshold is set according to the wave characteristics of the target area, and the plurality of peaks are screened and filtered using the first threshold.

6. The method according to claim 1, characterized in that The initial spatial resolution and the initial temporal resolution are determined on the condition that the wave height field of the target area meets the preset accuracy threshold and the effective wave height data of the multi-source satellite altimeter is not distorted during the fusion process.

7. The method according to claim 1, characterized in that Adaptively adjusting the time resolution includes adaptively improving the time resolution.

8. The method according to claim 1, characterized in that: The effective wave height data of the multi-source satellite altimeter in the target area are data obtained through pre-processing operations.

Images