Deformity wave generation method based on probability prediction model

Abstract

The invention relates to a freak wave generation method based on a probability prediction model, and belongs to the technical field of marine environment simulation, and the method comprises the steps: 1, determining an input parameter according to a target demand; step 2, obtaining multiple groups of parameters according to a prediction result of the freak wave generation probability prediction model; step 3, crushing wave prediction; step 4, determining input parameters, and screening out the input parameters from the multiple groups of parameters obtained in the step 2 according to a broken wave prediction result; and 5, realizing numerical or experimental wave generation of the freak waves based on the input parameters. According to the method, the generation probability of the freak waves can be accurately predicted before the waves are generated according to the wave generation input parameters; the freak wave generation process based on the probability prediction model does not need repeated iteration, and the freak wave generation efficiency is remarkably improved. And on the basis of the freak wave generation probability prediction model, accurate and efficient simulation of the two-dimensional freak wave is realized.

Description

Technical Field

[0001] The invention belongs to the technical field of marine environment simulation, and in particular relates to a method for simulating and generating freak waves based on a probability prediction model. Background Art

[0002] Freak waves, sometimes also called vicious waves, are extreme waves that have characteristics not possessed by regular waves, such as extremely large wave crests and wave heights, concentrated energy, and accidental occurrences. They are highly destructive and often lead to catastrophic consequences. Usually, a freak wave is defined as a wave with a maximum wave height greater than twice the significant wave height.

[0003] The wave train superposition method is a common method for generating abnormal waves. The general idea is to distribute the wave energy proportionally to a random wave train and a convergent wave train. This type of wave train superposition method is called double wave train superposition.

[0004] In a two-dimensional wave field, the wave height generated by the superposition of two wave trains is

[0005]

[0006] Where η, η1 and η2 represent the wavefront heights of the total wave train, random wave train and convergent wave train, respectively. x is the longitudinal coordinate of the wave propagation direction, t is the time, and N represents the number of component waves. i , k i ,ω i and φ i They represent the amplitude, wave number, circular frequency and phase of component wave i respectively. k and ω satisfy the wave dispersion relation ω 2 =gk tanh(kh), where h is the water depth. In a random wave train, the phase of the component wave φ i Uniformly distributed in the range of 0 to 2π (specific values ​​are generally randomly generated by computers). c and t c They respectively represent the convergence position (the position where the maximum peak height appears) and convergence time (the time when the maximum peak height appears) of the convergent wave train.

[0007] Amplitude 1i and a 2i Calculated from wave energy proportion and wave spectrum:

[0008]

[0009]

[0010] Among them, p1 and p2 are the wave energy proportions of random wave train and convergent wave train respectively, and here p1+p2=1. i ) is the circular frequency ω iThe corresponding wave spectrum energy nearby has a frequency width of △ω.

[0011] Here, the improved PM spectrum (the improved Pierson-Moskowitz Spectrum) proposed at the 15th ITTC conference is used to describe the wave field. The expression of the wave spectrum is shown in formula (3).

[0012]

[0013] Among them, the input parameter H s is the significant wave height, T 0.1 is the average period calculated from the spectral moment. 0.1 =2πm0 / m1, where m0 and m1 are the zero-order moment and first-order moment of the wave spectrum respectively. ω m and T 0.1 The relationship is: m =4.85 / T 0.1 In general, it can be considered that the component wave circular frequency ω in equations (2) and (3) i Uniformly distributed in (0,4ω m ) range. In this patent, ω i ∈(0,4ω m ).

[0014] Based on equations (1) to (3), by inputting appropriate wave parameters N, wave energy proportions p1 and p2 of random wave trains and convergent wave trains, and significant wave height H at the numerical or experimental wave generation boundary (x = 0), s , spectrum peak frequency ω m And other parameters can realize the abnormal wave generation.

[0015] In traditional methods, before the wave train superposition method generates the desired target distorted wave, it needs to go through a process of repeated attempts and adjustment of input parameters. In this process, when the number of component waves is small or the energy proportion of the convergent wave train is low, the probability of generating a distorted wave is very low; when the number of component waves is too large or the energy proportion of the convergent wave train is too high, the distorted wave is easy to break, and it is also difficult to generate the target wave. When the energy proportion of the convergent wave train is too large, it will also lead to obvious errors in the significant wave height. Therefore, how to select appropriate input parameters is crucial to the generation of distorted waves by wave train superposition. Summary of the invention

[0016] The present invention proposes a method for generating a freak wave based on a probability prediction model, which includes a freak wave generation probability prediction model, which is used to quantitatively predict the generation probability of a wave train superimposed freak wave under different wave-making input parameters, and has important guiding and reference significance for the efficient generation of freak waves. In a single wave condition, based on the wave-making input parameters (the number of component waves, the energy proportion of random wave trains and convergent wave trains, and the distribution range of the component wave circular frequencies), a definite freak wave generation probability can be obtained. In various conditions, the model prediction results are in good agreement with the statistical data.

[0017] In order to achieve the above object, the technical solution of the present invention provides a method for generating a freak wave based on a probability prediction model, which comprises the following steps:

[0018] Step 1: Determine input parameters according to target requirements;

[0019] Step 2, obtaining multiple groups of parameters according to the prediction results of the abnormal wave generation probability prediction model;

[0020] Step 3, breaking wave prediction;

[0021] Step 4, determining input parameters, and selecting input parameters from multiple groups of parameters obtained in step 2 according to the breaking wave prediction results;

[0022] Step 5: numerically or experimentally generate the distorted wave based on the input parameters.

[0023] Preferably, the generation process of the freak wave generation probability prediction model is as follows:

[0024] In a two-dimensional wave field, the wave height generated by the superposition of two wave trains is

[0025]

[0026] Among them, η, η1 and η2 represent the wavefront heights of the total wave train, random wave train and convergent wave train respectively; in the random wave train, the phase φ of the component wave i Uniformly distributed in the range of 0 to 2π; x c and t c They represent the convergence position and convergence time of the convergent wave train respectively;

[0027] At the convergence position of the convergent wave train, when the trough before the maximum peak of the convergent wave train appears, the wave surface height η of the random wave train 1t The probability density function of the Gaussian distribution N(0,m0) with the same mathematical expectation of 0 and variance m0 is

[0028]

[0029] Here we define the variable H1′=η 1c -η1t , from which we can get the variable H′1=η 1c -η 1t ~N(0,2m0), its probability density function

[0030]

[0031] Probability distribution function of variable H1′

[0032]

[0033] The maximum wave height H of the total wave train can be expressed as

[0034] H=η 1c -η 1t +H2=H1′+H2 (9)

[0035] Probability density function of the maximum wave height H of the total wave train

[0036]

[0037] According to equations (8) and (10), the probability of generating a distorted wave can be calculated:

[0038]

[0039] Among them, H 1s is the significant wave height of the random wave train η1, m0=p1A / (4B); According to the definition of significant wave height, H 1s is equal to the significant wave height of the total wave train generated;

[0040] Finally, in the double wave train superposition condition, the probability of generating a freak wave is

[0041]

[0042] Next, the wave height H2 of the convergent wave train is calculated by the discrete summation method;

[0043] At the convergence position x of the convergent wave train c and convergence time t c , the maximum wavefront height of the converging wave train

[0044]

[0045] In a converging wave train, the minimum wave height η before the maximum wave crest 2,min Appears at the focus time t c First 0.5T 0.1 Time, that is, t c –0.5T 0.1 ;

[0046] The zero-point method is used to calculate the wave period and wave height, and the minimum wave height at the trough of the convergent wave train is

[0047]

[0048] wave height of converging wave train

[0049]

[0050] According to equations (4), (12) and (15), the probability of generating a distorted wave P freak Depends on the energy proportion p2 (p2 = 1–p1) of the convergent wave train and the number of component waves N; can be selected (0,5ω m ) as the distribution range of the component wave circular frequencies, the corresponding probability of generating abnormal waves can be obtained.

[0051] Preferably, in formula (5), in a random wave train, the wave surface height obeys a Gaussian distribution, while the wave height obeys a Rayleigh distribution.

[0052] Preferably, the wave height H in formula (9) represents the wave height of the random wave train and the convergent wave train at the convergence position of the convergent wave train, that is, the maximum wave height of the total wave train, and H does not refer to each wave height in the total wave train.

[0053] Preferably, the exponential term of the natural constant e in formula (10) is [-(H-H2) 2 / (4m0)].

[0054] Preferably, based on formula (15), the number of component waves N and the energy proportion p2 of the convergent wave train both affect the convergent wave train height H2. When N or p2 increases, H2 and H2 / H 1s All of them showed an obvious nonlinear increasing trend.

[0055] Preferably, step one includes the target generation probability of the significant wave height Hs, the spectrum peak frequency ωm and the distorted wave.

[0056] Preferably, the multiple groups of parameters in step 2 include the number of component waves N, the wave energy proportion p1 of the random wave train, and the wave energy proportion p2 of the convergent wave train.

[0057] Preferably, in step 3, if H2 / H 1s <0.532g / (2H 1s ), then this wave train is a non-breaking wave.

[0058] Preferably, step four includes the number of component waves N, the wave energy proportion p1 of the random wave train, and the wave energy proportion p2 of the convergent wave train.

[0059] The main advantages of the freak wave generation probability prediction model proposed in the present invention are as follows: (1) Based on the wave-making input parameters, the freak wave generation probability can be accurately predicted before the wave is generated; (2) The freak wave generation process based on this probability prediction model does not require repeated iterations, which significantly improves the efficiency of freak wave generation. Based on the freak wave generation probability prediction model, accurate and efficient simulation of two-dimensional freak waves is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 The schematic diagram of the probability of generating abnormal waves in the double wave train superposition condition;

[0061] Figure 2 It is a global schematic diagram of the superposition of convergent wave trains and random wave trains;

[0062] Figure 3 This is a partial enlarged view of the superposition of convergent wave trains and random wave trains;

[0063] Figure 4 The converged wave train height H2 and the generated significant wave height H under different p2 values ​​and N values 1s Schematic diagram of the ratio (ω i ∈(0,4ω m ));

[0064] Figure 5 Schematic diagram of the probability of generating freak waves in the double wave train superposition condition (ω i ∈(0,4ω m ));

[0065] Figure 6 It is a schematic diagram of the statistical characteristic distribution of wave train data;

[0066] Figure 7 Schematic diagram of statistical data verification of the prediction results of the freak wave generation probability prediction model (double wave train superposition condition);

[0067] Figure 8 Generate a global view of freak waves based on probabilistic prediction models;

[0068] Fig. 9 Generate a local magnified image of the deformed wave based on the probability prediction model. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] Based on the freak wave generation probability prediction model, the wave-making input parameters can be quantitatively selected according to the target generation probability of the freak wave, which can significantly improve the simulation generation efficiency of the freak wave.

[0071] Based on the freak wave generation probability prediction model, the specific process of simulating the generation of wave trains superimposed with freak waves is as follows:

[0072] (1) Determine the input parameters according to the target requirements: significant wave height H s , spectrum peak frequency ω m and target generation probability of freak waves (e.g., above 95%);

[0073] (2) Based on the prediction results of the abnormal wave generation probability prediction model ( Figure 1 ), and obtain multiple groups of parameters: the number of component waves N, the proportion of wave energy of random wave trains p1, and the proportion of wave energy of convergent wave trains p2;

[0074] (3) Breaking wave prediction: If H2 / H 1s <0.532g / (2H 1s ), then this group of waves is a non-breaking wave;

[0075] (4) Determine input parameters: select input parameters (number of component waves N, wave energy proportion p1 of random wave train, wave energy proportion p2 of convergent wave train) from the multiple groups of parameters obtained in step (2) according to the breaking wave prediction results;

[0076] (5) Numerical or experimental wave generation of abnormal waves based on input parameters.

[0077] 1. Deformation wave generation probability prediction model:

[0078] In a two-dimensional wave field, the wave height generated by the superposition of two wave trains is

[0079]

[0080] Among them, η, η1 and η2 represent the wavefront heights of the total wave train, random wave train and convergent wave train respectively. In the random wave train, the phase φ of the component wave i Uniformly distributed in the range of 0 to 2π (specific values ​​are generally randomly generated by computers). c and t c They respectively represent the convergence position (the position where the maximum peak height appears) and convergence time (the time when the maximum peak height appears) of the convergent wave train.

[0081] First, we study the superposition problem of the wavefront height of the random wave train and the convergent wave train at the convergence position and convergence time of the convergent wave train (such as Figure 2 and Figure 3 shown).

[0082] It is generally believed that the change of the wave height of a random wave train (especially a random wave train composed of a series of cosine regular waves) over time and space is a stationary random process with ergodicity. Figure 3 In the example, at the convergence position and convergence time of the convergent wave train, the wavefront height η of the random wave train is 1c It obeys the Gaussian distribution (also known as normal distribution) N(0,m0) with mathematical expectation 0 and variance m0, denoted by η 1c ~N(0,m0). η 1c Probability density function

[0083]

[0084] Wherein, m0 is the zero-order moment of the variance spectrum (here the wave spectrum), and for the improved PM spectrum, m0=p1A / (4B).

[0085] It is important to distinguish here that in a random wave train, the wave surface height follows a Gaussian distribution, while the wave height follows a Rayleigh distribution.

[0086] Similarly, at the convergence position of the convergent wave train, when the trough before the maximum peak of the convergent wave train appears, the wave surface height η of the random wave train is 1t The probability density function of the Gaussian distribution N(0,m0) with the same mathematical expectation of 0 and variance m0 is

[0087]

[0088] Here we define the variable H1′=η 1c -η 1t Note that the variable H1′ is independent of a specific wave height. Thus, the variable H′1=η 1c -η 1t ~N(0,2m0), its probability density function

[0089]

[0090] Cumulative probability function of variable H1′

[0091]

[0092] The maximum wave height H of the total wave train can be expressed as

[0093] H=η 1c -η 1t +H2=H1′+H2 (9)

[0094] It should be noted that the wave height H here refers to the wave height of the random wave train and the convergent wave train at the convergence position of the convergent wave train, that is, the maximum wave height of the total wave train. H does not refer to the individual wave heights in the total wave train.

[0095] Probability density function of the maximum wave height H of the total wave train

[0096]

[0097] It should be noted that the exponential term of the natural constant e in formula (10) is [-(H-H2) 2 / (4m0)] instead of [-H 2 / (4m0)]. Therefore, equation (10) does not indicate that the total wave height H obeys Gaussian distribution.

[0098] According to equations (8) and (10), the probability of generating a distorted wave can be calculated:

[0099]

[0100] Among them, H 1s is the significant wave height of the random wave train η1, m0=p1A / (4B). According to the definition of significant wave height, H 1s Equal to the significant wave height of the total wave train generated.

[0101] Finally, in the double wave train superposition condition, the probability of generating a freak wave is

[0102]

[0103] Next, the wave height H2 of the convergent wave train is calculated by the discrete summation method.

[0104] At the convergence position x of the convergent wave train c and convergence time t c , the maximum wavefront height of the converging wave train

[0105]

[0106] At the same time, this section analyzes the wave surface height η at the deepest point of the converging wave train trough based on a large number of numerical simulation results. 2,min The moment of occurrence. According to the average period T 0.1 The definition and related analysis results of the convergent wave train, the minimum wave surface height η before the maximum wave crest 2,min Generally appears at the focusing moment t c First 0.5T 0.1 Time, that is, t c –0.5T 0.1 (This section will continue Figure 4Provides verification.)

[0107] According to the IAHR recommendation, the zero-crossing method is used to calculate the wave period and wave height.

[0108]

[0109] wave height of converging wave train

[0110]

[0111] Based on formula (15), the wave height H2 of the converged wave train and the generated significant wave height H under different p2 values ​​and N values ​​are 1s The ratio H2 / H 1s like Figure 2 When N = 100, 250, 500 and 1000, p2 = 0.05, 0.10, 0.20 and 0.30, this section uses the simulation results of 16 sets of wave data (recorded as Figure 4 The 16 dots in the figure are compared with the result of formula (15), and the two are in good agreement, which verifies the accuracy of formula (15) in calculating the wave height of the convergent wave train.

[0112] At the same time, it can be found that the number of component waves N and the energy proportion p2 of the convergent wave train have a significant effect on the convergent wave train height H2. When N or p2 increases, H2 and H2 / H 1s All of them showed an obvious nonlinear increasing trend.

[0113] However, as mentioned above, when the input N value or p2 value is too large, the generated convergent wave train will inevitably break. It is very important to choose the appropriate input parameters. In deep water conditions, the critical breaking wave steepness H is w / λ(where H w For non-breaking convergent wave trains, the wave steepness must satisfy H2 / λ<0.142, and the wavelength calculation formula λ=gT 0.1 2 / (2π), we can get H2 / H 1s <0.532g / (ω m 2 ·H 1s ). Then the necessary and sufficient condition for non-breaking freak waves is 2 <H / H 1s <0.532g / (ω m 2 ·H 1s ).

[0114] About H s and H 1s The relationship between H sH is the input target significant wave height; 1s is the significant wave height of the random wave train, which is also equal to the significant wave height of the total wave train generated. s and H 1s Satisfy the relationship The larger the p1 value, the greater the significant wave height H of the total wave train generated. 1s The closer to the target value H s When p1 = 0.95, the error of the significant wave height is -2.5%; when p1 = 0.90, the error of the significant wave height is -5.1%; when p1 = 0.80, the error of the significant wave height is -10.6%; and when p1 = 0.70, the error of the significant wave height can reach -16.3%. This is why it is recommended to select a larger p1 value (to reduce the proportion of the converged wave train energy).

[0115] According to equations (4), (12) and (15), the probability of generating a distorted wave P freak It depends on the energy proportion p2 (p2 = 1 – p1) of the convergent wave train and the number of component waves N (related to △ω). i The values ​​are evenly distributed in (0,4ω m ) interval, the probability of generating abnormal waves corresponding to different p2 values ​​and N values ​​is as follows Figure 3 As shown. Some scholars also choose (0,5ω m ) is used as the distribution range of the component wave circular frequencies. Based on this method, the corresponding abnormal wave generation probability can also be obtained.

[0116] 2. Statistical verification of the probability prediction model:

[0117] This section verifies the accuracy of the prediction results of the freak wave generation probability prediction model based on the statistical results of a large amount of wave data. For each working condition, 10,000 groups of wave trains are generated with random values ​​according to the proportion of the converged wave train energy and the number of component waves N, and the number of freak waves in them, nf, is counted. According to the law of large numbers, the statistical occurrence probability of freak wave (hereinafter referred to as statistical probability) is nf / 10000.

[0118] Taking the double wave train superposition condition as an example, the specific method and steps for obtaining the statistical probability of freak wave generation are as follows.

[0119] (1) According to the verification conditions, determine the input parameters N, p1, p2, H s ,ω m And the distribution range of the component wave circular frequency. Among them, the spectrum peak circular frequency ω m The effect on the probability of freak wave generation is negligible. For non-breaking waves, H sThe value of H / H also has no effect on the probability of generating abnormal waves (because the ratio H / H 1s With H s not relevant).

[0120] (2) Set the initial value of the number of freak wave occurrences nf to 0 and set the parameter v = 0. Calculate ω i 、S(ω i ) and H 1s .

[0121] (3) Generate N random numbers in the range of 0 to 2π by computer as the phases of each component wave in the random wave train.

[0122] (4) Calculate the total wave train at the convergence position x c The wavefront time history η(x c ,t).

[0123] (5) Calculate the total wave train at the convergence position x c The maximum wave height H i .

[0124] (6) Set the parameter v = v + 1. If H i / H 1s >2.0, let nf=nf+1.

[0125] (7) Repeat steps (3) to (6) until v = 10000. The statistical probability of generating a freak wave is calculated to be nf / 10000.

[0126] In the double wave train superposition condition, when N = 250, p1 = 0.85, p2 = 0.15, H s =0.12m,ω m =2.0rad / s,ω i ∈(0,4ω m ), the statistical characteristics of the generated wave train data are as follows Figure 4 shown. Figure 4 There are 10,000 black dots in the figure, corresponding to 10,000 sets of simulated wave train data. 1s The ratio H / H 1s , most of the wave trains (7556 groups) are distributed in the range of (2.0,3.0), and the statistical probability P(2.0 <H / H 1s <3.0) is 0.7556. In addition, the statistical probability P(1.0 <H / H 1s <2.0) is 0.1560, P(3.0 <H / H 1s <4.0) is 0.0881, P(H / H 1s>4.0) is 0.0003. In this case, the statistical probability of abnormal wave generation nf / 10000 is 84.40% (=P(2.0 <H / H 1s <3.0)+P(3.0 <H / H 1s <4.0)+P(H / H 1s >4.0)). According to the prediction results of the abnormal wave generation probability prediction model, the prediction probability corresponding to this working condition is 82.88%, which is only 1.83% error compared with the statistical probability (84.40%).

[0127] The accuracy of the freak wave generation probability prediction model in the double wave train superposition condition is verified as follows: Figure 7 In the figure, the statistical probability value is represented by a square, and the color of the square represents the energy proportion of the convergent wave train, which is consistent with the energy proportion represented by the line color.

[0128] In general, the prediction results obtained by the abnormal wave generation probability prediction model proposed in this patent are in good agreement with the statistical probability, and the input parameters can be accurately determined within the range of 0.6 to 1.0 for the abnormal wave generation probability, thus achieving the established goal of efficient and accurate wave generation by superimposing abnormal waves by wave trains.

[0129] 3. Generate abnormal waves based on probability prediction model simulation:

[0130] Considering the case of double wave train superposition, in order to make the probability of freak wave generation large enough (here set to be above 95.0%) and avoid wave breaking as much as possible, the input parameters are set as follows: s =0.12m,ω m =3.62rad / s(T 0.1 =1.34s), p1=0.80, p2=0.20, N=250, ω i ∈(0,4ω m ). According to the probability prediction model prediction results, P freak The converged wave train height H2 is 98.6%, which means that the distorted wave can be generated efficiently. 1s The ratio is H2 / H 1s =2.79, which is significantly smaller than the predicted value of 0.532g / (ω m 2 ·H 1s )=3.71. The wave train generated based on the probability prediction model converges at the position x c The wave height at Figure 8 and Fig. 9 shown.

[0131] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating abnormal waves based on a probability prediction model, characterized in that: The following steps are involved: Step 1: Determine input parameters according to target requirements; Step 2, obtaining multiple groups of parameters according to the prediction results of the abnormal wave generation probability prediction model; Step 3, breaking wave prediction; Step 4, determining input parameters, and selecting input parameters from multiple groups of parameters obtained in step 2 according to the breaking wave prediction results; Step 5: numerical or experimental wave generation of the distorted wave is realized based on the input parameters.

2. According to the method for generating abnormal waves based on a probability prediction model according to claim 1, it is characterized in that: The generation process of the freak wave generation probability prediction model is as follows: In a two-dimensional wave field, the wave height generated by the superposition of two wave trains is Among them, η, η1 and η2 represent the wavefront heights of the total wave train, random wave train and convergent wave train respectively; in the random wave train, the phase φ of the component wave i Uniformly distributed in the range of 0 to 2π; x c and t c represent the convergence position and convergence time of the convergent wave train respectively; At the convergence position of the convergent wave train, when the trough before the maximum peak of the convergent wave train appears, the wave surface height η of the random wave train 1t The probability density function of the Gaussian distribution N(0,m0) with the same mathematical expectation of 0 and variance m0 is Here we define the variable H1′=η 1c -η 1t , from which we can get the variable H′1=η 1c -η 1t ~N(0,m0), its probability density function Probability distribution function of variable H1′ The maximum wave height H of the total wave train can be expressed as H=h 1c -or 1t +H2=H1′+H2 (9) Probability density function of the maximum wave height H of the total wave train According to equations (8) and (10), the probability of generating a distorted wave can be calculated: Among them, H 1s is the significant wave height of the random wave train η1, m0=p1A / (4B); According to the definition of significant wave height, H 1s is equal to the significant wave height of the total wave train generated; Finally, in the double wave train superposition condition, the probability of generating a freak wave is Next, the wave height H2 of the convergent wave train is calculated by the discrete summation method; At the convergence position x of the convergent wave train c and convergence time t c , the maximum wavefront height of the converging wave train In a converging wave train, the minimum wave height η before the maximum wave crest 2,min Appears at the focus time t c First 0.5T 0.1 Time, i.e. t c –0.5T 0.1 ; The zero-point method is used to calculate the wave period and wave height, and the minimum wave height at the trough of the convergent wave train is wave height of converging wave train According to equations (4), (12) and (15), the probability of generating a distorted wave P freak Depends on the energy proportion p2 (p2 = 1–p1) of the convergent wave train and the number of component waves N; can be selected (0,5ω m ) as the distribution range of the component wave circular frequencies, the corresponding probability of generating abnormal waves can be obtained.

3. The method for generating a freak wave based on a probability prediction model according to claim 2, characterized in that: In formula (5), in a random wave train, the wave surface height obeys the Gaussian distribution, while the wave height obeys the Rayleigh distribution.

4. According to claim 3, a method for generating a freak wave based on a probability prediction model is characterized in that: The wave height H in formula (9) represents the wave height of the random wave train and the convergent wave train at the convergence position of the convergent wave train, that is, the maximum wave height of the total wave train. H does not refer to the individual wave heights in the total wave train.

5. The method for generating a freak wave based on a probability prediction model according to claim 4, characterized in that: The exponential term of the natural constant e in formula (10) is [-(H-H2) 2 / (4m0)].

6. The method for generating a freak wave based on a probability prediction model according to claim 5, characterized in that: Based on formula (15), the number of component waves N and the energy proportion p2 of the convergent wave train both affect the convergent wave train height H2. When N or p2 increases, H2 and H2 / H 1s All of them showed an obvious nonlinear increasing trend.

7. The method for generating a freak wave based on a probability prediction model according to claim 1, characterized in that: Step 1 includes the target generation probability of the sense wave height Hs, the spectrum peak frequency ωm and the distorted wave.

8. The method for generating a freak wave based on a probability prediction model according to claim 7, characterized in that: The multiple groups of parameters in step 2 include the number of component waves N, the wave energy proportion of random wave trains p1, and the wave energy proportion of convergent wave trains p2.

9. The method for generating a freak wave based on a probability prediction model according to claim 8, characterized in that: In step 3, if H2 / H 1s <0.532g / (2H 1s ), then this wave train is a non-breaking wave.

10. The method for generating a freak wave based on a probability prediction model according to claim 9, characterized in that: Step 4 includes the number of component waves N, the proportion of wave energy of random wave trains p1, and the proportion of wave energy of convergent wave trains p2.

Images