POT-based damaged ship overturning probability calculation method
Technical Field
The invention relates to a POT-based damaged ship overturning probability calculation method, and belongs to the technical field of damaged ship overturning probability calculation methods.
Background
At present, a Monte Carlo method is adopted for solving the overturning probability of the damaged ship to numerically simulate a single-degree-of-freedom rolling motion equation, a four-order Runge Kutta method is adopted for solving, a short-term probability distribution table of each wave in North Atlantic is directly calculated, and then a Monte Carlo probability method and the weight of the long-term distribution of the wave are used for directly evaluating the probability. The calculation time per test is set to 1 hour (3600s), and the probability of each overturn of the ship is set to p (0)<p < 1), repeating the calculation for n times according to Bernoulli distribution (1, p), wherein the overturn time is ncThe probability of the occurrence of overturning in n times is pc=ncAnd/n, then:
when n is large enough, it is considered to obey a normal distribution Np (p, p (1-p)) in order to guarantee the probability pcThe deviation total probability p is not more than 5% of the total standard deviation p (1-p), the interval credibility reaches 95%, and each group of storm condition simulation calculation is carried out at least 1600 times. The long term overturning probability under one sea condition is calculated as follows: the short-term overturning probability is calculated for about 10s each time, each group of storm conditions is calculated 1600 times, 272 groups of storm conditions are required to be calculated in total, and the time is consumed10 x 1600 x 272/3600 x 1208h, about 50 days in time. Although the existing damaged ship overturning probability calculation method is accurate in calculation result, the convergence speed is low, the consumed time is long, a large amount of calculation amount and solution time are required to be increased for realizing the high-precision calculation result, and the method is not very suitable for quickly evaluating the safety of the ship in an emergency.
Disclosure of Invention
The POT method adopted by the invention analyzes the relation between the probability and the time by adopting Poisson distribution according to the time domain rolling calculation result of the single-degree-of-freedom rolling motion equation and the mutual independent assumption among large-amplitude rolling events. A small amount of data can be screened from a large amount of data for analysis and processing, and the purpose of rapidly solving the overturning probability is achieved.
The POT method essentially belongs to the category of extreme value theory, and is a short term of the Peak Over Threshold method, and performs data analysis processing based on a selected Threshold value. The extreme value theory overcomes the limitation of the traditional statistical method, and is firstly used for analyzing and predicting extreme events such as ocean disasters in ocean engineering in the year 1971 by Feller, so that a good result is obtained. And is therefore well suited to address problems associated with extreme motion in marine waves.
The invention discloses a POT-based damaged ship overturning probability calculation method, which has the general idea as shown in figure 1, wherein a group of sample data is obtained through model test or data simulation, all the sample data exceeding the threshold value form new sample data according to a determined threshold value, the threshold value is selected to distinguish a demarcation point (figure 1) of linear solution and nonlinear solution, ship overturning belongs to a rare problem, the sample data exceeding the threshold value is used for solving, the POT model has fewer requirements on data, valuable information existing in the original sample data cannot be filtered, the core idea of the POT model is that a small amount of data is screened from a large amount of data according to the threshold value for analysis and processing, and therefore, the calculation time can be effectively reduced.
Meanwhile, the IMO is expected to correct the current stability balance of the cabin breaking on the basis of research on the stability of the cabin breaking in the waves, so that a designer considers the influence of the stability of the cabin breaking in a design stage. With the maturity of research on stability of breaking in waves of ships, IMO is imperative to develop stability balance in waves in the future.
A POT-based damaged ship overturning probability calculation method is characterized by comprising the following steps:
1. time domain calculation of the water inlet process of the damaged ship:
determining a solving method according to different cabin breaking types of the ship, wherein the cabin inlet water and the outboard water of the first class cabin and the second class cabin do not interact with each other, and solving the restoring moment at different transverse inclination angles by adopting an increasing weight method; interaction between water entering the cabin and the ship body exists in the second class of cabin, influence of liquid sloshing is ignored, the liquid in the cabin is assumed to be quasi-static, and only influence of free liquid on restoring moment is considered; the third-class cabin inlet water and outboard water have interaction, the instantaneous restoring moment is solved by adopting a loss buoyancy method, and a proper flow coefficient is selected for correction in consideration of the inflow/outflow at a damaged port; applying Bernoulli equation to different situations of the water inlet cabin, integrating the break flow velocity in a time domain to obtain break flow, and performing numerical solution on the water inlet in the cabin of each time step to obtain the tilt moment of the water in the cabin to the ship body;
2. POT method-based assessment of overturning probability of damaged ship in crosswind and crosswave
According to a restoring moment curve of a damaged ship, combining a single-degree-of-freedom rolling motion equation when stormy waves coexist, and superposing irregular waves by sine waves with different amplitudes and phases; establishing an overturning probability model of a damaged ship in crosswind and crosswave; because the diffraction component in the wave disturbing force and the coupling action of the swaying on the rolling are mutually offset, the pressure integration is directly carried out on the instantaneous wet surface in the time domain by mainly considering the wave exciting moment of the Froude-Krylov part; solving the rolling motion in the crosswind and the rolling wave by using a four-order Runge-Kutta method in a time domain, analyzing and processing a time domain rolling calculation result of the damaged ship by using a POT (point of presence) method, quickly simulating the overturning of the damaged ship in the crosswind and the rolling wave, and further quickly evaluating the safety of the damaged ship;
the calculation formula of the overturning probability in the POT method is shown in the following (2) to (4), and according to a set of existing sample data, the set of sample data can be simulated or modeled according to a numerical valueObtaining model test, calculating probability lambda exceeding threshold value in sample data by combining selected threshold value1Analyzing the sample data exceeding the threshold value by adopting an extreme value theory to obtain the conditional probability lambda of the overturn occurrence2Finally, obtaining the total overturning probability according to the formula (2);
λ=λ1·λ2 (2)
where λ is the finally calculated probability of overturning, λ1Is the probability of exceeding a threshold in the roll data; lambda [ alpha ]2Is the conditional probability of overturning after the threshold value is exceeded; phi is am0Setting a threshold value; phi is am2Is a failure criterion; f. ofPOT(phi) is the probability distribution density obtained by analyzing sample data exceeding a threshold; fPOTAnd (phi) is the corresponding cumulative distribution function.
The core of the POT method adopted by the invention is that a small amount of data is selected according to the threshold value to be analyzed and processed, all observed values exceeding the threshold value in the sample are modeled, and the tail characteristic of the data is fitted, so that the rapid evaluation of the safety of the ship is realized. The method can assist a decision maker to make a quick decision and respond emergently at the critical moment of broken water inflow.
Drawings
FIG. 1: the invention relates to a POT method general idea diagram;
FIG. 2; a flow diagram of the present invention;
FIG. 3: the GZ curve of the ship under the loading condition during navigation;
FIG. 4: the GZ curve of the ship under the no-load condition;
FIG. 5: the GZ curve of the ship under the full load condition;
FIG. 6: wave height 5.5m, rolling cumulative distribution function at period 4.5 s;
FIG. 7: wave height 5.5m, roll cumulative distribution function at period 7.5 s;
FIG. 8: short-term overturning probabilities of different methods corresponding to different periods at the same wind speed;
FIG. 9: comparing long-term overturning probabilities of different methods under a full load condition;
FIG. 10: and comparing the long-term overturning probabilities of different methods under the condition of the load during sailing.
FIG. 11: and comparing the long-term overturning probabilities of different methods under the condition of no-load navigation.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 2, the method for calculating the overturn probability of the damaged ship based on the POT comprises the following steps:
1. time domain calculation of the water inlet process of the damaged ship:
determining a solving method according to different cabin breaking types of the ship, wherein the cabin inlet water and the outboard water of the first class cabin and the second class cabin do not interact with each other, and solving the restoring moment at different transverse inclination angles by adopting an increasing weight method; interaction between water entering the cabin and the ship body exists in the second class of cabin, influence of liquid sloshing is ignored, the liquid in the cabin is assumed to be quasi-static, and only influence of free liquid on restoring moment is considered; the third-class cabin inlet water and outboard water have interaction, the instantaneous restoring moment is solved by adopting a loss buoyancy method, and a proper flow coefficient is selected for correction in consideration of the inflow/outflow at a damaged port; applying Bernoulli equation to different situations of the water inlet cabin, integrating the break flow velocity in a time domain to obtain break flow, and performing numerical solution on the water inlet in the cabin of each time step to obtain the tilt moment of the water in the cabin to the ship body;
2. POT method-based assessment of overturning probability of damaged ship in crosswind and crosswave
According to a restoring moment curve of a damaged ship, combining a single-degree-of-freedom rolling motion equation when stormy waves coexist, and superposing irregular waves by sine waves with different amplitudes and phases; establishing an overturning probability model of a damaged ship in crosswind and crosswave; because the diffraction component in the wave disturbing force and the coupling action of the swaying on the rolling are mutually offset, the pressure integration is directly carried out on the instantaneous wet surface in the time domain by mainly considering the wave exciting moment of the Froude-Krylov part; solving the rolling motion in the crosswind and the rolling wave by using a four-order Runge-Kutta method in a time domain, performing probability analysis on a time domain rolling calculation result of the damaged ship by using a POT (point of arrival) method, quickly simulating the overturning of the damaged ship in the crosswind and the rolling wave, and further quickly evaluating the safety of the damaged ship;
the calculation formula of the overturning probability in the POT method is shown in the following (2) to (4), according to a group of existing sample data, the group of sample data can be obtained according to numerical simulation or model test, and the probability lambda exceeding the threshold value in the sample data is obtained by combining the selected threshold value calculation1Analyzing the sample data exceeding the threshold value by adopting an extreme value theory to obtain the conditional probability lambda of the overturn occurrence2Finally, obtaining the total overturning probability according to the formula (2);
λ=λ1·λ2 (2)
where λ is the finally calculated probability of overturning, λ1Is the probability of exceeding a threshold in the roll data; lambda [ alpha ]2Is the conditional probability of overturning after the threshold value is exceeded; phi is am0Setting a threshold value; phi is am2Is a failure criterion; f. ofPOT(phi) is the probability distribution density obtained by analyzing sample data exceeding a threshold; fPOTAnd (phi) is the corresponding cumulative distribution function.
The GZ curves for a given vessel in the final position of integrity and breakage under different loading conditions are shown in the following figures. Fig. 3 shows the underway load condition, fig. 4 shows the empty load condition, and fig. 5 shows the full load condition.
Fig. 6 is a cumulative distribution function obtained by the POT method when the wave height is 5.5m and the wave period is 4.5s, fig. 7 is a cumulative distribution function obtained by the POT method when the wave height is 5.5m and the wave period is 7.5s, and it can be seen that the type of the distribution function changes with the change of the period. The variation of the short-term overturning probability with the period for the same wind speed is as shown in fig. 8, and since the rolling period of the ship is around T-5.5, the short-term overturning probability at that position is maximized. And the overturning probability obtained by adopting the probability POT method is increased along with the decrease of the threshold value and the failure criterion, wherein the threshold value is selected to be 32 degrees of the ultimate static inclination angle of the ship, the criterion 1 is 70 degrees of the stationarity disappearance angle of the ship, and the criterion 2 criterion is 35 degrees considering that the 70 degrees of the failure criterion is biased to danger for the ships such as a fishery vessel. The result shows that the failure criterion is closely related to the magnitude of the probability of overturning, and the lower the failure criterion, the more data points that satisfy the condition, the greater the probability of overturning, and vice versa. The criterion of criterion 2 is better matched with the Monte-Carlo calculation result, and the calculation result of the POT method is more safe, so that criterion 2 is selected as the failure criterion. Compared with the traditional Monte-Carlo method, the POT method rapidly provides the short-term overturning probability under a certain load condition according to the time domain rolling result obtained by the energy method and by combining the given proper threshold value and the failure criterion, takes the calculation of the short-term overturning probability when the wave height is 5.5m and the wave period is 7.5s as an example, firstly determines the number exceeding the threshold value in 3600 data points to be 3399 according to the given threshold value, and obtains the lambda10.944, in combination with a given failure criterion 2, i.e., #m2At 35 °, λ is obtained according to the cumulative distribution function given in fig. 72The final calculation yields λ 0.927 as 0.982. And the like for other sea state calculations. And aiming at the calculated short-term overturning probability, obtaining the long-term overturning probability of the sea area according to the North Atlantic wave distribution diagram.
FIG. 9 is a comparison of the POT method under a full load condition with the overturning probability results of Monte-Carlo, FIG. 10 is a comparison of the POT method under a underway load condition with the overturning probability results of Monte-Carlo, and FIG. 11 is a comparison of the POT method under an empty load condition with the overturning probability results of Monte-Carlo. The POT method is mainly characterized by being capable of rapidly calculating the overturning probability of damaged ships when the ships are paralyzed and being suitable for rapid assessment of the safety of the ships in emergency.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.