CN115983475A - Sea area scale drowning personnel drift trajectory prediction method - Google Patents
Sea area scale drowning personnel drift trajectory prediction method Download PDFInfo
- Publication number
- CN115983475A CN115983475A CN202310001248.9A CN202310001248A CN115983475A CN 115983475 A CN115983475 A CN 115983475A CN 202310001248 A CN202310001248 A CN 202310001248A CN 115983475 A CN115983475 A CN 115983475A
- Authority
- CN
- China
- Prior art keywords
- drift
- wind
- trajectory
- water
- falling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 206010013647 Drowning Diseases 0.000 title claims abstract description 18
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 59
- 238000011156 evaluation Methods 0.000 claims abstract description 14
- 238000002474 experimental method Methods 0.000 claims description 50
- 239000002245 particle Substances 0.000 claims description 30
- 230000036544 posture Effects 0.000 claims description 25
- 230000008569 process Effects 0.000 claims description 19
- 238000004088 simulation Methods 0.000 claims description 19
- 238000000926 separation method Methods 0.000 claims description 14
- 238000012417 linear regression Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000009429 distress Effects 0.000 claims description 8
- 238000010586 diagram Methods 0.000 claims description 7
- 238000005259 measurement Methods 0.000 claims description 6
- 230000000750 progressive effect Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 230000009191 jumping Effects 0.000 claims description 5
- 230000001186 cumulative effect Effects 0.000 claims description 4
- 230000007613 environmental effect Effects 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 230000008878 coupling Effects 0.000 claims description 2
- 238000010168 coupling process Methods 0.000 claims description 2
- 238000005859 coupling reaction Methods 0.000 claims description 2
- 238000006073 displacement reaction Methods 0.000 claims description 2
- 230000005484 gravity Effects 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000005295 random walk Methods 0.000 claims description 2
- 239000013535 sea water Substances 0.000 claims description 2
- 238000007667 floating Methods 0.000 claims 2
- 238000004891 communication Methods 0.000 claims 1
- 230000010354 integration Effects 0.000 claims 1
- 239000010410 layer Substances 0.000 claims 1
- 230000001568 sexual effect Effects 0.000 claims 1
- 239000002344 surface layer Substances 0.000 claims 1
- 230000002349 favourable effect Effects 0.000 abstract 1
- 238000012795 verification Methods 0.000 description 9
- 238000011160 research Methods 0.000 description 8
- 238000012360 testing method Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000002950 deficient Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 208000003443 Unconsciousness Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000007175 bidirectional communication Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000005240 physical vapour deposition Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 238000013076 uncertainty analysis Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention relates to a method for predicting the drifting trajectory of a person falling into water in a sea area scale, which comprises the following steps: sea area division, drift trajectory prediction model establishment, drowning person drift trajectory prediction, drift trajectory prediction model precision and applicability evaluation. The method and the device improve the accuracy of predicting the drift trajectory of the drowning person, are favorable for improving the accurate prediction capability of the drift trajectory of the drowning person in maritime search and rescue work, and powerfully support maritime search and rescue aid decisions.
Description
Technical Field
The invention relates to a method for predicting a drifting track of a person falling into water in a sea area scale.
Background
With the continuous development of maritime transportation trade, the risk of accidents of shipwrecks such as collision and grounding is increasing (Zhang et al, 2017), which is very easy to cause a water-falling event. Maritime search and rescue comprises two processes of search and rescue, wherein a timely and scientific search plan has important significance for improving the survival probability of people falling into water. The search of people falling into water mainly comprises two key problems, namely drift track prediction and optimal search area determination, and search and rescue resource allocation and search path planning to maximize search and rescue success rate. The drift theory is the basis of a mathematical method for maritime search and rescue resource allocation and search path planning. In the maritime search and rescue process, the position of the maritime distress target is not fixed, and in the process that the search and rescue party reaches the initial position, the distress target in the maritime environment continuously drifts due to the combined action of ocean current, wind, sea waves and the like. Over time, the complexity of the marine environment and the cumulative effect of multiple uncertainties in the target drift process increase the difficulty of locating search objects, making the search work more complex (Brushett et al, 2014 2017. The gold rescue time on the sea is about 12 hours, and if the drifting track of the distress target can be predicted dynamically in time and directly arrive at the optimal search and rescue area, the search and rescue time can be greatly shortened, and the maritime search and rescue aid decision can be supported powerfully.
The sea drift trajectory prediction means that the drift trajectory of a sea distress target is rapidly, accurately and dynamically predicted, and the possible distribution probability of the distress target is given. The offshore drift trajectory prediction needs to comprehensively consider and quantify influence factors in the drift process of the target in danger, including marine environment characteristics, drift object attributes and a prediction model. The marine environment characteristics and the attributes of the drifting objects comprehensively consider two research subjects, namely a natural environment and an objective object. For people falling into water, different drift characteristics can be shown under different postures (standing upright, sitting, lying prone) or load conditions. In addition, the marine environment is complex and variable, and has uncertainty. The predictive model portion includes a correlation physics model and a probabilistic statistics based uncertainty analysis.
Allen and Plourde first proposed a leeway model built on a statistical basis for quantifying the drift characteristics of objects in 1999. Wind induced drift refers to the movement of an object relative to the surrounding sea current due to wind acting on the aquatic portion of the object, and is generally considered to be the sum of the drift caused by wind and wave forces acting on the object (Anderson et al, 1998, brushett et al, 2014. In 2005, allen decomposed leeway into a stronger Downwind (DWL), a positive crosswind (+ CWL), and a negative crosswind component (-CWL). The marine drift experiment is the most common and reliable method for determining the wind-induced drift coefficient at present, and wind-induced drift parameters are calculated from information such as positions, atmospheric wind, sea surface currents and the like acquired by various sensors mounted on drifters or measurement carriers (Breivik and Allen,2008, allen et al, 2010. Related studies have been used as a basic guideline for national search planning tools to simulate the drifting process of floats in water, such as the french MOTHY system (Daniel et al, 2003), canadian canarp (Canadian Coast Guard, 2009), american Coast Guard SARMAP (2013), SAROPS (Kratzke et al, 2010).
In the current research, universal wind-induced drift parameters are determined mainly through experiments in a certain sea area, and the influence of different sea areas is not considered. In fact, different sea area ocean current systems and hydrological characteristics are characterized, drifting objects with the same characteristics in different sea areas have different drifting characteristics, and the uniform wind-induced drifting parameters are adopted, so that the local applicability is realized. In addition, due to the fact that the experiment process of the drowning person is limited by the difficulty of marine environment data acquisition and time cost, research on the drifting characteristics of the drowning person is deficient.
The invention provides a method for determining a wind-induced drift parameter of a person falling into water based on sea area division. The method divides the Chinese sea area into different sea areas, and the different sea areas adopt different wind-induced drift parameters, thereby solving the drift characteristic difference caused by the difference of the marine environments of the different sea areas. At present, a perfect offshore drift case database does not exist, massive information of open source channels such as the internet is utilized in the research to obtain offshore drift history cases (drift accidents and offshore drift experiments), the offshore drift case database is constructed, and the capability of performing model verification based on historical real drift events is realized. In addition, in the database expansion process, the method can be used for wind-induced drift model calibration for detailed cases of track information and marine environment data information. The time and space of the offshore drift test and the prediction verification are greatly expanded, and the problems that the offshore drift test is time-consuming and labor-consuming and is easily restricted by weather and sea conditions are solved to a certain extent.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the problems that the sea drift trajectory prediction of the drowning person is deficient and the wind-induced drift parameters do not consider the influence of different sea areas, the method for determining the wind-induced drift parameters of the drowning person based on the sea area partitions is provided, the drift trajectory prediction model is established, the trajectory prediction precision is improved, a new technical path of 'field experiment and open source information combined drive sea drift prediction verification' is provided, a sea drift case database is constructed by utilizing the mass information of open source channels such as the Internet, and the drift trajectory prediction model is verified based on historical cases, so that the time and the space of sea drift test and prediction verification are greatly expanded.
In order to solve the technical problem, the invention provides a method for predicting the drifting trajectory of the staff falling into water in the sea area, which comprises the following steps:
2.1, performing a marine drifting experiment on a person falling into water, namely performing the marine drifting experiment on simulators with different sea conditions and different postures, wherein each experiment comprises the release of the simulators integrating GPS equipment, and tracking by a mother ship to observe marine environment elements;
2.2, data preprocessing, namely obtaining a marine environment data sample and a simulated human drift track data sample;
2.3, calculating the wind-induced drift component coefficient of the person falling into the water, namely calculating six coefficients of a linear regression equation and calculating a standard deviation according to the linear regression relation between the wind-induced drift speed and the wind speed so as to determine nine coefficients of the drift trajectory prediction model;
2.4, calculating the drift velocity, namely calculating the drift velocity of the person falling into the water based on the marine environment data;
3.1, analyzing the drift wind pressure angle and the jumping characteristic of the person falling into water, namely calculating the drift wind pressure angle of the person falling into water, drawing a progressive vector diagram PVD and determining the jumping frequency;
3.2, modeling drift trajectory prediction uncertainty, namely modeling the drift trajectory prediction model and the marine environment data uncertainty;
3.3, predicting a drift trajectory, namely predicting the drift trajectory by adopting Monte Carlo particle simulation, setting the drift direction of the particles at the beginning of prediction, adjusting the drift direction of the particles according to the coupling frequency in the trajectory prediction process, and calculating the probability distribution of the simulated particles by adopting a nuclear density estimation method according to the position distribution of the simulated particles at different times;
5.1, sea area scale drift track prediction precision evaluation, namely, driving a drift track prediction model by using actually measured marine environment data to realize track prediction of people falling into water, and respectively using the accumulated separation distance and the normalized accumulated separation distance to perform precision evaluation on the result of the drift case track prediction obtained in the step 3;
5.2, evaluating the applicability of the drift trajectory prediction model, namely comparing the drift trajectory prediction model with existing models of other geographic areas in order to verify the accuracy of the model and the applicability to other areas, driving different trajectory prediction models by adopting a marine environment data set, and respectively evaluating the accuracy by using the accumulated separation distance and the distance normalization error.
The effective benefits of the invention are as follows:
(1) The invention is based on a sea area scale drowning person drift track prediction method, comprehensively considers different marine environment factors and drowning persons with different postures, and improves the drowning person drift track prediction precision.
(2) The invention provides a new technical path for 'field experiment and open source information combined drive marine drift prediction verification', and a drift track prediction model of the research is verified based on historical cases.
(3) The method and the device are beneficial to improving the accurate prediction capability of the drift track of the workers falling into water in search and rescue. Through relevant verification experiments, the method has a good application effect and is beneficial to improving the overall reliability of the track prediction.
Drawings
The invention will be further described with reference to the accompanying drawings.
Fig. 1 is a general flowchart of the method for predicting the drift trajectory of the person falling into water in the sea area scale according to the present invention.
FIG. 2 is a marine experiment marine environment element acquisition operation scene in the embodiment of the invention.
Fig. 3 is a working scene of the simulated human drifting experiment in the marine experiment in the embodiment of the invention.
FIG. 4 is a simulated human drift scene of different postures of a marine experiment in an embodiment of the invention.
FIG. 5 shows simulated human drift trajectories for different postures of a marine experiment in an embodiment of the invention.
FIG. 6 is an exemplary PVD graph and a jibing event annotation of a human simulator in an embodiment of the invention.
FIG. 7 is a result diagram of a verification case of the present invention in a Tan sea area.
FIG. 8 is a result chart of the verification case of the present invention in Guangdong Yuexi.
Detailed Description
The technical route and the operation steps of the invention are clearer in the following detailed description of the invention based on the attached drawings.
The embodiment is selected in the Zhoushan sea area and the Tantan sea area. Severe weather is frequent in four seasons due to the mid-latitude at the Zhoushan. Meanwhile, the terrain of the navian is complex, and each sea area along the navian is shielded and influenced by islands and mountain systems, so that the wind power of each sea area has obvious characteristics, the wind power distribution is extremely uneven, and the wind-induced drift experiment of the navian sea area has guiding significance for maritime search and rescue. In addition, maritime search and rescue in the Tantang sea area has important significance.
In this embodiment, the method for predicting the drifting trajectory of the person falling into water in the sea area scale is described by taking the above experimental area as an example, and as shown in the flowchart in fig. 1, the method specifically includes the following steps:
the method comprises the steps of firstly, dividing the offshore region of China into 18 sea areas according to the meteorological product geographical subareas of the offshore region and the peripheral sea area in the national standard of China 'Meteorological product geographical subarea', considering the geographical division, the structure and the topographic and geomorphic control factors, and combining the natural characteristics of the ocean current system area and the like.
And secondly, establishing a drift trajectory prediction model, namely researching drift motion rules of drifters in different sea areas aiming at the sea area divisions in China, and carrying out a sea drift experiment in the Zhoushan sea area and the Tantan sea area. Establishing a boat mountain sea area vertical posture falling person drift track prediction model (N _ ES _ I), a boat mountain sea area lying posture falling person drift track prediction model (N _ ES _ II), a quaistant sea area vertical posture falling person drift track prediction model (TS _ I), a quaistant sea area lying posture falling person drift track prediction model (TS _ II), a vertical posture falling person drift prediction comprehensive model (C _ I), a lying posture falling person drift prediction comprehensive model (C _ II), and concretely comprises the following steps:
2.1, carrying out 10 experiments in the drift experiment of the person falling into the water on the sea-2020-2021, wherein the accumulated experiment time is 53 hours. In 6 months and 15-18 days in 2020, field investigation and comprehensive experiments were conducted on Zhoushan Putuo and Shenghai regions. In 2021, 16-17 months 4, field investigation and comprehensive experiments were conducted in the quan sea area.
Each experiment comprises the release of a simulated person integrating GPS equipment, and the marine environmental element observation is carried out by the tracking of a mother ship. Relates to a marine drift experiment of simulators (standing upright and lying prone) with different sea conditions (large and small storms) and different postures. The mother ship carries equipment such as a GPS receiver, an ADCP (Acoustic Doppler current profiler), a meteorological station, a Beidou receiver and the like, and an experimental scene is shown in figure 2. The GPS receiver is used for recording and transmitting the position of the mother ship in real time and is arranged at the top of the mother ship; the ADCP flow meter is used for measuring flow velocity and flow direction; the meteorological station is used for measuring wind speed and wind direction, is arranged at the top of the mother ship, and has the installation height of approximately 10m; the Beidou receiver is used for bidirectional communication with the drifting buoy for maritime search and rescue. And carrying out continuous profile observation on the ocean current passing through the sea area in the process of sailing of the mother ship. In the invention, an ADCP external GNSS compass forms an ADCP flow measurement system, an absolute reference provided by the GNSS compass is adopted, the problems of measurement errors and the like caused by the influence of interference magnetic fields of a ship body and the like on an internal magnetic compass of the ADCP are solved, a sea test is carried out before a drift test, and coordinate systems of the ADCP, the GNSS compass and the ship body which are installed on the ship body are unified. Marine environmental element collection equipment and associated settings are shown in table 1.
TABLE 1 marine environmental element collection facility and related setup
The human simulation system adopts a human-shaped drifting buoy for maritime search and rescue of Qingdao sea company, the vertical posture of the human-shaped buoy body in water can be realized by adjusting the foot balancing weight, and the human simulation assembly and experimental scene is shown in figure 3. Marine experiment adopts two marine search and rescue to use shape drift buoy system, and one is for lying prone gesture (facial downwards) for the personnel's drift motion of falling into water under the simulation unconscious state, and one is upright gesture for the personnel's drift motion of falling into water under the simulation conscious state. In each experiment, after the simulated people reach a designated sea area, the simulated people with two different postures are released at the same time, the scene of the simulated people drifting on the sea is shown in fig. 4, and the drifting track is shown in fig. 5.
And 2.2, preprocessing data, namely measuring results of a meteorological station comprise wind field acquisition time, wind speed and wind direction, measuring results of a current meter comprise flow field acquisition time, and the east-west flow velocity and the south-north flow velocity of each acquisition section adopt a first layer measuring result as the surface flow velocity (0.9 m). Due to the acquisition frequency difference, wind field data and flow field data are matched in real time based on the data acquisition time of the acquisition frequency difference and the position data returned by the GPS receiver of the mother ship, and marine environment data samples at intervals of 10min are obtained; the simulated human drift trajectory data received by the Beidou receiver comprises acquisition time and WGS84 longitude and latitude. And (3) calculating the drifting speed based on the drifting position at each moment, wherein the longitude and latitude of the WGS84 must be converted into a plane coordinate, and acquiring simulated human drifting track data samples at intervals of 10 min.
And 2.3, calculating a wind-induced drift component coefficient of the person falling into the water, namely obtaining the wind-induced drift velocity of the person falling into the water by reducing the sea flow velocity from the real drift velocity, and decomposing the wind-induced drift velocity into two components of downwind velocity (DWL) and transverse wind velocity (CWL), wherein the transverse wind velocity is positive when the transverse wind velocity is positioned on the right side of the downwind direction. The formula is as follows:
L d ＝k d V wind +b d +ε d
L c+ ＝k c+ V wind +b c+ +ε c+ (1)
L c- ＝k c- V wind +b c- +ε c-
wherein L is d Is the downwind component, L c+ Is a positive cross-wind component, L c- Is a negative transverse wind direction component and has a high wind speed V equal to 10m above sea level wind The linear correlation is formed; k is a radical of formula d 、k c+ 、k c- Is a linear regression slope; b d 、b c+ 、b c- Is the intercept; epsilon d 、ε c+ 、ε c- For the error term, this regression model is an unconstrained model (y-regression). Considering the case where the sea wind speed is 0, that is, the wind-induced drift velocity is 0, the above formula becomes as follows, and is called a constraint model (0-regression).
L d ＝k d V wind +ε d
L c+ ＝k c+ V wind +ε c+ (2)
L c- ＝k c- V wind +ε c-
Further decomposed into + CWL and-CWL to the right depending on whether CWL is located downwind. In this process, DWL, + CWL, -CWL are characterized as a function of 10m wind speed using constrained and unconstrained linear regression. Nine coefficients of the drift trajectory prediction model are calculated, including the slope, intercept, and standard deviation of DWL, + CWL, -CWL, as shown in tables 2-3. The invention adopts an unconstrained model to predict the drift trajectory.
TABLE 2 unconstrained regression coefficients of drift trajectory prediction model
TABLE 3 Drift trajectory prediction model with constrained regression coefficients
2.4, calculation of the drift speed of the drowning person, namely calculating the drift speed of the drowning person based on the marine environment data, can be represented by the following formula:
V drift ＝V F-current +V leeway +V F-wave (3)
wherein, V drift As drift velocity, V F-current For the drift velocity induced by the sea surface currents, the flow induced drift velocity is approximately equal to the surface flow velocity of the sea water. V leeway Is the wind-induced drift velocity. V F-wave The drift velocity is due to waves. The action mechanism of the wave on the influence of the motion of the object in danger is complex, and the influence on the drifting object is mainly realized through the direct acting force of the wave and the Stokes drifting caused by wind-driven waves. The effect of the direct wave forces on the motion of the object is then related to the size of the object, and the forces from the waves become significant only if the length scale of the drifting object is larger than the wavelength. People falling into the water belong to small targets (less than 30m in length) and the direct force of the waves is negligible. Since the Stokes shift effect caused by wind-borne waves is mainly concentrated at the surface of the water body and is not easily separated from the wind drift velocity, its influence is usually assumed to exist in the empirical wind-induced drift coefficient and is not considered anymore.
Thirdly, predicting a drift track, namely calculating a drift wind pressure angle and a jibing frequency of the person falling into the water, and predicting the drift track by adopting Monte Carlo particle simulation, wherein the method specifically comprises the following steps:
3.1, analyzing drifting wind pressure angle and tying characteristics of the person falling into the water, namely counting the wind pressure angle according to experimental data, and calculating the + CWL probability of the person falling into the water drifting under different sea conditions and different postures, as shown in a table 4; when the wind speed is in a certain range, the direction of the wind-induced drift will shift from downwind right to left, a phenomenon known as the jumbing frequency. Calculating the downwind direction and the crosswind direction displacement of the drift of each simulated person, and drawing a Progressive Vector Diagram (PVD); and counting the times of change of the transverse wind direction per hour, namely the jumping frequency by utilizing the PVD (physical vapor deposition) progressive vector diagram.
Since the frequency of offshore marine environment changes is high, the wind pressure deflection angle is unstable, and sometimes the wind pressure angle is low (close to zero) and the direction is unstable, the sign of the CWL may change frequently. Therefore, a jibing event needs to be more strictly defined, and cannot be considered as a jibing event if the wind-induced drift floats at a small angle to the left and right of the downwind, except that the crosswind component changes from one side of the downwind to the other, as can be seen from the slope of the PVD. The method performs related parameter experiments, can effectively remove the condition that the CWL symbol changes frequently due to small and unstable wind pressure deflection angle within 10min of stabilization time, and has better experiment effect. Therefore, the following condition is added to identify a jibing event: the cross-wind component remains stable in the same direction for at least 10 minutes before and after the change of direction.
Under most sea conditions, the probability of the jibing event is low, so that the jibing events of all samples are counted aiming at different posture simulators, the jibing frequency is calculated, and the significant effect of the jibing is the particle distribution in the middle of two directions to guide the determination of the optimal search and rescue region (Breivik et al, 2011). The horizontal arrows in fig. 6 represent the main signature change from negative (downwind left drift) to positive (downwind right drift) or positive to negative, depending on the decision criteria, of the CWL.
In the 15-hour process of the upright posture simulator experiment, the number of times of the jibing event of the simulator is 3, so the jibing frequency is 0.20. In the 19.5-hour process of the experiment of the simulation person in the lying posture, the number of times of the jitter events of the simulation person is 12, so the jitter frequency is 0.62.
TABLE 4 + CWL probability of water falling personnel drifting under different sea conditions and different postures
3.2, modeling drift trajectory prediction uncertainty, namely modeling uncertainty of a drift trajectory prediction model and marine environment data in order to describe the uncertainty of the marine test. The slope and intercept of the wind induced drift component linear regression are adjusted by adding a disturbance term. Taking DWL as an example:
wherein epsilon d ＝S yx Norm; norm-obeys normally distributed random numbers; s yx Is the standard deviation of the linear regression in cm/s.
Carrying out uncertainty modeling on marine environment data by adopting a random walk disturbance model, assuming that disturbance of a wind field obeys circular normal distribution, taking DWL as an example:
v′ n ≡(u′ n ,v′ n )∈N(0,σ w ). (5)
v 'of the total' n Is corrected wind speed u' n Is the horizontal wind speed, v' n For wind velocity in the vertical direction, σ w And N is the standard deviation of normal distribution, the number of particles simulated by the Monte card is N, and the total number of the particles is N.
3.3, predicting a drift trajectory, namely predicting the drift trajectory by adopting Monte Carlo particle simulation, wherein the drift trajectory of a person falling into water at any moment can be obtained by integral calculation of the drift velocity:
wherein t is any time, x (t) is the position of the person falling into water at the current time, and x 0 Is to start upThe water falling position, t' is the time step of the track integral calculation.
Setting the drift direction of the particles at the beginning of prediction, adjusting the drift direction of the particles according to the jibing frequency in the track prediction process, and calculating the probability distribution of different particles by adopting a nuclear density estimation method according to the position distribution of the simulated particles at different time.
And fourthly, establishing a marine drift history case library, namely acquiring a marine drift case (drift accident and marine drift experiment) by using the topic crawler by using mass information of open source channels such as the Internet and the like. Drift accidents originate from news websites, including government and professional unit websites reporting news related to marine accidents, and also including common internet commercial news portal websites. The information of the offshore drifting experiment is derived from domestic and foreign documents, including the hundred degree academy and the google academy. Constructing a marine drift case database, which mainly comprises the following steps: case summary, accident/experiment information, distress target information, trajectory information, case environment, etc., for validating the wind-induced drift model. The data information of the marine experiment and the historical drift trajectory is shown in tables 5-6.
TABLE 5 sea experiment and historical drift trajectory data information
Table 6 drift trajectory information for the marine experiments.
( The drifting track of the simulation person in the vertical posture is recorded as _1, and the drifting track of the simulation person in the lying posture is recorded as _2; average statistical conditions of wind speed and flow velocity of 10min )
And fifthly, evaluating the accuracy and the applicability of the drift trajectory prediction model, namely respectively adopting the actually measured marine environment data and the historical marine environment data set to drive the drift trajectory prediction model to realize the trajectory prediction of the person falling into water and evaluate the accuracy, and specifically comprising the following steps of:
5.1, sea area scale drift track prediction accuracy evaluation, namely, the drift track prediction model is driven by using actually measured marine environment data to realize track prediction of the person falling into the water, and accuracy evaluation is carried out on the result of the drift case track prediction obtained in the step 3 by respectively using the accumulated separation distance and the normalized accumulated separation distance. The calculation formula of the accumulated separation distance space-time distribution is as follows:
in the formula, T is the serial number of the corresponding time in the process of predicting the track, and T is the number of times in the time subinterval;
simulating a gravity center point space position vector of the particle cluster at the time of the t moment; />
The vector is the real space position vector at the t moment; t multiplied by 1 represents that the expression mode of the index is a matrix with T rows and 1 columns;
the normalized cumulative separation distance calculation formula is:
in the formula, L t The actual distance of drift of the actual floater in the time range from t-1 to t; d t The distance between the actual floater and the simulated particles at the t-th moment;
the total drift distance of the actual floater at the time t is better in prediction result, and the NCSD value is smaller.
5.2, drift trajectory prediction model applicability evaluation, namely comparing existing models of other small geographic areas which are not experimentally researched in the research, in order to verify the accuracy of the models and the applicability of the models to other areas, driving different trajectory prediction models by adopting an ocean environment data set, reanalyzing the wind field data by adopting ERA5, wherein the spatial resolution is 0.25 degrees, reanalyzing the flow field data by adopting CMEMS, and the spatial resolution is 0.083 degrees. The accuracy evaluation was performed using the cumulative separation distance and the distance normalization error, respectively. And (3) adopting two indexes of accumulated separation distance space-time distribution and distance normalization error (RASD) at the time T when the accuracy evaluation is carried out due to the loss of hourly position information of the historical case track data.
The verification example:
the following description is continued with this example in order to verify the reliability of the method of the invention. In order to verify the accuracy of the method for predicting the drifting trajectory of the staff falling into the water at the sea area scale, the accuracy of the trajectory prediction is verified in the quan sea area. In 2021, 17 days in 4 months, in a quan sea area, aiming at a simulated person in a lying state, drift time is 6 hours, and five groups of experiments are set. The first group adopts a model of Allen et al (1999), sets the probability of + CWL to be 50%, and the jibing frequency to be 0.04, which is used for comparing the accuracy of the model of the invention and the existing research model; the groups 2 to 5 adopt the model of the invention, set different + CWL probabilities and jibing frequencies, and discuss the influence of the model coefficients on the track prediction. Group 2 sets the + CWL probability to 50%, ignoring the tying frequency; group 3 sets the + CWL probability to 50% and the tying frequency to 0.62; group 4 sets the + CWL probability to 80% and the tying frequency to 0.62; group 5 sets the + CWL probability to 80%, ignoring the jibing frequency.
In order to analyze the track prediction precision, the central tracks of all the particle tracks are drawn for the track prediction result, and the final particle probability heat map is compared with the actual drift track for analysis. The experimental results are shown in fig. 7, and the error calculation results are shown in table 7. The result shows that the particle track simulated by the drift track prediction model of the research is more consistent with the actual observation track, and the model simulation error can be reduced to a certain extent by properly setting the + CWL probability.
TABLE 7 comparison of prediction accuracy of Tan sea area drifting trajectory
To verify the applicability of the model to other regions, comparisons were made between literature models of other small geographic regions not experimentally studied in the present invention. Two groups of experiments are set according to the experimental case of the simulation human drifting in the upright state in Guangdong Yuexi sea area in 2 and 17 days in 2011. The probability of + CWL in the first group is set to be 50%, the jibing frequency is a default value of 0.04, and the wind-induced drift coefficient adopts an empirical value of Allen and the like; the second group adopts the vertical comprehensive model of the invention, the probability of + CWL is set to be 58%, and the tying frequency is set to be 0.20.
The experimental results are shown in fig. 8, and the error calculation results are shown in table 8. The result shows that the wind-induced drift coefficient can effectively improve the track prediction precision; meanwhile, the model provided by the invention is also applicable to marine environment data with lower space-time resolution.
TABLE 8 Guangdong Yuexi sea area Drift trajectory prediction accuracy comparison
In addition to the above embodiments, the present invention may have other embodiments. All technical solutions formed by adopting equivalent substitutions or equivalent transformations fall within the protection scope of the present invention.
Claims (10)
1. A method for predicting drifting tracks of people falling into water in sea area scale comprises the following steps:
step 1, sea area division, namely dividing a target sea area into a plurality of sea area divisions by considering geographic division, construction, topographic and geomorphic control factors and combining natural characteristics of sea current system areas and the like;
step 2, establishing a drift trajectory prediction model, namely researching drift motion rules of drowning people in different sea areas according to the divided sea areas, carrying out a sea drift experiment in different sea areas, and establishing a drift trajectory prediction model of drowning people in different sea areas and different postures, wherein the drift trajectory prediction model specifically comprises the following steps:
2.1, performing a marine drifting experiment on a person falling into water, namely performing the marine drifting experiment on simulators with different sea conditions and different postures, wherein each experiment comprises the release of the simulators integrating GPS equipment, and tracking by a mother ship to observe marine environment elements;
2.2, preprocessing data, namely acquiring a marine environment data sample and a simulated human drift track data sample;
2.3, calculating the wind-induced drift component coefficient of the person falling into the water, namely calculating six coefficients of a linear regression equation and calculating a standard deviation according to the linear regression relation between the wind-induced drift speed and the wind speed so as to determine nine coefficients of the drift trajectory prediction model;
2.4, calculating drift speed, namely calculating the drift speed of the person falling into the water based on the marine environment data;
step 3, drift trajectory prediction, namely calculating a drift wind pressure angle and a jibing frequency of a person falling into water, and predicting the drift trajectory by adopting Monte Carlo particle simulation, wherein the method specifically comprises the following steps:
3.1, analyzing the drift wind pressure angle and the jumping characteristic of the person falling into water, namely calculating the drift wind pressure angle of the person falling into water, drawing a progressive vector diagram PVD and determining the jumping frequency;
3.2, modeling drift trajectory prediction uncertainty, namely modeling the drift trajectory prediction model and the marine environment data uncertainty;
3.3, predicting a drift trajectory, namely predicting the drift trajectory by adopting Monte Carlo particle simulation, setting the drift direction of the particles at the beginning of prediction, adjusting the drift direction of the particles according to the coupling frequency in the trajectory prediction process, and calculating the probability distribution of the simulated particles by adopting a nuclear density estimation method according to the position distribution of the simulated particles at different times;
step 4, establishing a marine drift history case database, namely acquiring marine drift case information by utilizing an open source channel, wherein the marine drift case comprises drift accidents and a marine drift experiment and is used for verifying an established drift trajectory prediction model, and the marine drift case database mainly comprises: case summary, accident/experiment information, distress target information, track information and case environment;
step 5, drift trajectory prediction model precision and applicability evaluation, namely, respectively adopting the actually measured marine environment data and the historical marine environment data set to drive the drift trajectory prediction model, so as to realize trajectory prediction of the person falling into water and perform precision evaluation, and specifically comprises the following steps:
5.1, sea area scale drift track prediction precision evaluation, namely, driving a drift track prediction model by using actually measured marine environment data to realize track prediction of people falling into water, and respectively using the accumulated separation distance and the normalized accumulated separation distance to perform precision evaluation on the result of the drift case track prediction obtained in the step 3;
5.2, evaluating the applicability of the drift trajectory prediction model, namely comparing the drift trajectory prediction model with existing models of other geographic areas in order to verify the accuracy of the model and the applicability to other areas, driving different trajectory prediction models by adopting a marine environment data set, and respectively evaluating the accuracy by using the accumulated separation distance and the distance normalization error.
2. The method for predicting the drifting trajectory of the staff falling into the water in the sea area according to claim 1, wherein the method comprises the following steps: in step 2.1, the mother ship carries a GPS receiver, an ADCP current meter, a weather station and a Beidou receiver; the system comprises a GPS receiver, an ADCP (advanced digital control panel) flow meter, a weather station, a Beidou receiver and a simulated man buoy, wherein the GPS receiver is used for recording and transmitting the position of a mother ship in real time, the ADCP flow meter is used for measuring the flow speed and the flow direction, the weather station is used for measuring the wind speed and the wind direction, and the Beidou receiver is used for carrying out two-way communication with the simulated man buoy; in the process of an offshore experiment, a mother ship tracks the drift trajectory of a simulation person and is used for measuring marine environmental elements of a sea area near the simulation person; in addition, the mother ship carries a data acquisition computer which is used as a receiver for GPS information, marine environment data and simulated people track data and stores the data in a database; after the experiment reaches the designated sea area, the simulative persons in different postures are released at the same time.
3. The method for predicting the drifting trajectory of the staff falling into the water in the sea area according to claim 1, wherein the method comprises the following steps: in step 2.2, the measurement results of the meteorological station comprise wind field acquisition time, wind speed and wind direction, the measurement results of the current meter comprise flow field acquisition time, east-west flow velocity and south-north flow velocity of each acquisition section, and the first layer of measurement results are used as surface layer flow velocity; based on data acquisition time of a weather station and a current meter and position data returned by a mother ship GPS receiver, real-time matching of wind field data and flow field data is carried out; the simulated person drift trajectory data received by the Beidou receiver comprises acquisition time and WGS84 longitude and latitude, the WGS84 longitude and latitude are converted into plane coordinates, and the real drift speed of the simulated person is calculated based on drift positions at all times.
4. The method for predicting the drifting trajectory of people falling into water in the sea area scale according to claim 1, wherein the method comprises the following steps: in step 2.3, the wind-induced drift velocity of the person falling into the water is obtained by reducing the sea flow velocity from the true drift velocity, and the wind-induced drift velocity is decomposed into two components, namely a downwind velocity DWL and a crosswind velocity CWL, wherein the crosswind velocity is positive when the crosswind velocity is located on the right side of the downwind direction, and the formula is as follows:
L d ＝k d V wind +b d +ε d
L c+ ＝k c+ V wind +b c+ +ε c+
L c- ＝k c- V wind +b c- +ε c-
wherein L is d Is the downwind component, L c+ Is a positive cross-wind component, L c- Is a negative transverse wind direction component and has a high wind speed V equal to 10m above sea level wind The linear correlation is formed; k is a radical of d 、k c+ 、k c- Is a linear regression slope; b d 、b c+ 、b c- Is an intercept; epsilon d 、ε c+ 、ε c- For the error term, the above formula is called unconstrained model linear regression;
further decomposing into positive crosswind speed + CWL and negative crosswind speed-CWL according to whether CWL is located at the leeward and to the right side. In the process, DWL, + CWL, -CWL are represented as a function of 10m wind speed by adopting constrained and unconstrained linear regression, and six coefficients of a drift trajectory prediction model are calculated, wherein the six coefficients comprise: and calculating the slope and intercept of the DWL, + CWL and-CWL, calculating the standard deviation, and determining nine coefficients of the drift trajectory prediction model.
5. The method for predicting the drifting trajectory of the staff falling into the water in the sea area according to claim 4, wherein the method comprises the following steps: in step 2.4, the drift speed of the person falling into the water is expressed by the following formula:
V drift ＝V F-current +V leeway +V F-wave
wherein, V drift As drift velocity, V F-current The flow-induced drift velocity is approximately equal to the surface flow velocity of the seawater; v leeway Is the wind-induced drift velocity; v F-wave The drift velocity is due to waves.
6. The method for predicting the drifting trajectory of the staff falling into the water in the sea area according to claim 5, wherein the method comprises the following steps: in the step 3.1, a wind pressure angle refers to an included angle between a wind-induced drifting direction and a wind direction, the downwind direction and the transverse wind direction displacement of the drifting of each simulation person are calculated, and a progressive vector diagram PVD is drawn; and counting the change times of the transverse wind direction per hour by using a progressive vector diagram PVD, namely the jibing frequency.
7. The method for predicting the drifting trajectory of the staff falling into the water in the sea area according to claim 6, wherein the method comprises the following steps: in the step 3.2, the slope and intercept of the wind-induced drift component linear regression are adjusted by adding a disturbance term; taking DWL as an example:
wherein epsilon d ＝S yx Norm; norm is a random number which follows normal distribution; s yx Is a wireStandard deviation of sexual regression in cm/s;
carrying out uncertainty modeling on marine environment data by adopting a random walk disturbance model, assuming that disturbance of a wind field obeys circular normal distribution, taking DWL as an example:
v′ n ≡(u′ n ,v′ n )∈N(0,σ w ).
wherein, v' n Is corrected wind speed u' n Is the horizontal wind speed, v' n Is the wind speed in the vertical direction, sigma w Is the standard deviation of normal distribution, N is the number of particles simulated by Monte Carr, and N is the total number of particles.
8. The method for predicting the drifting trajectory of people falling into water in the sea area scale according to claim 7, wherein the method comprises the following steps: in step 3.3, the drift trajectory of the person falling into the water at any time is calculated by the following formula:
wherein t is any time, x (t) is the position of the person falling into water at the current time, and x 0 For the initial water-falling position, t' is the time step of the trajectory integration calculation.
9. The method for predicting the drifting trajectory of the staff falling into the water in the sea area according to claim 1, wherein the method comprises the following steps: step 4, obtaining a marine drift history case by adopting a theme crawler; the drift accidents originate from news websites, including government and professional unit websites reporting news related to marine accidents and internet commercial news portal websites; the information of the offshore drifting experiment is derived from domestic and foreign documents, including the hundred degree academy and the google academy.
10. The method for predicting the drifting trajectory of the staff falling into the water in the sea area according to claim 1, wherein the method comprises the following steps: in step 5.1, the calculation formula of the accumulated separation distance space-time distribution is as follows:
in the formula, T is the serial number of the corresponding time in the process of predicting the track, and T is the number of times in the time subinterval;
simulating a gravity center point space position vector of the particle cluster at the time of the t moment; />
The vector is the real space position vector at the t moment; t multiplied by 1 represents that the expression mode of the index is a matrix with T rows and 1 columns;
the normalized cumulative separation distance calculation formula is:
in the formula, L t The actual distance of drift of the actual floater in the time range from t-1 to t; d t The distance between the actual floating object and the simulated particles at the t-th moment;
the predicted result is better and the NCSD value is smaller for the total drift distance of the actual floating object at the t moment. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310001248.9A CN115983475A (en) | 2023-01-03 | 2023-01-03 | Sea area scale drowning personnel drift trajectory prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310001248.9A CN115983475A (en) | 2023-01-03 | 2023-01-03 | Sea area scale drowning personnel drift trajectory prediction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115983475A true CN115983475A (en) | 2023-04-18 |
Family
ID=85959397
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310001248.9A Pending CN115983475A (en) | 2023-01-03 | 2023-01-03 | Sea area scale drowning personnel drift trajectory prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115983475A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117114207A (en) * | 2023-10-23 | 2023-11-24 | 山东科技大学 | Marine personnel drift track prediction method |
-
2023
- 2023-01-03 CN CN202310001248.9A patent/CN115983475A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117114207A (en) * | 2023-10-23 | 2023-11-24 | 山东科技大学 | Marine personnel drift track prediction method |
| CN117114207B (en) * | 2023-10-23 | 2024-02-06 | 山东科技大学 | Marine personnel drift track prediction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11488069B2 (en) | Method for predicting air quality with aid of machine learning models | |
| US9262723B2 (en) | Predicting climate data using climate attractors derived from a global climate model | |
| CN103197299B (en) | Extraction and quantitative analysis system of weather radar radial wind information | |
| EP1971883A2 (en) | A statistical-deterministic approach to natural disaster prediction | |
| Bi et al. | Arctic Sea ice area export through the Fram Strait estimated from satellite-based data: 1988–2012 | |
| Ashkboos et al. | Ens-10: A dataset for post-processing ensemble weather forecasts | |
| CN115983475A (en) | Sea area scale drowning personnel drift trajectory prediction method | |
| CN103914737B (en) | A kind of existing the weather information computational methods of power transmission and transformation line full line | |
| Falocchi et al. | A method to determine the characteristic time‐scales of quasi‐isotropic surface‐layer turbulence over complex terrain: A case‐study in the Adige Valley (Italian Alps) | |
| CN106097389A (en) | A kind of river flow measuring method based on image steganalysis | |
| Zhu et al. | An ensemble trajectory prediction model for maritime search and rescue and oil spill based on sub-grid velocity model | |
| CN115115138A (en) | Offshore storm surge water increase prediction method | |
| Abeysirigunawardena et al. | Extreme wind regime responses to climate variability and change in the inner south coast of British Columbia, Canada | |
| CN116908114B (en) | Remote sensing monitoring method for river basin granule organic carbon flux | |
| Wu et al. | Modeling the leeway drift characteristics of persons-in-water at a sea-area scale in the seas of China | |
| Figurski et al. | The impact of initial and boundary conditions on severe weather event simulations using a high-resolution WRF model. Case study of the derecho event in Poland on 11 August 2017 | |
| KR102411074B1 (en) | Devices and Methods for Detection and Prediction of Cold Water Mass Using Artificial Intelligence and Satellite Imaging | |
| Zhu et al. | Probabilistic modelling of surface drift prediction in marine disasters based on the NN–GA and ARMA model | |
| Spaulding et al. | A hierarchy of stochastic particle models for search and rescue (SAR): Application to predict surface drifter trajectories using HF radar current forcing | |
| Farfán et al. | Observations and forecasts from the landfall of tropical cyclones John, Lane, and Paul (2006) over northwestern Mexico | |
| Dong et al. | Quality control algorithm for marine meteorological data based on interest degree association rules | |
| Lu et al. | Verification of tropical cyclones (TC) wind structure forecasts from global NWP models and ensemble prediction systems (EPSs) | |
| Zhou et al. | A station-data-based model residual machine learning method for fine-grained meteorological grid prediction | |
| Fu et al. | Committee No I. 1: Environment | |
| CN106779206A (en) | A kind of short-term trend Forecasting Methodology of the wave wave direction based on analyze data again |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |