Offshore distress target drift prediction method, medium, equipment and program product
By combining marine dynamic environment field data and artificial intelligence models, the drift velocities of distressed targets at sea caused by current, wind, and waves are calculated, and a compensating drift velocity is generated. This solves the problem of insufficient dynamic update capability of drift prediction in existing technologies and achieves more accurate and timely prediction of distressed target locations.
Patent Information
- Application Number
- CN202510627739.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-15
- Publication Date
- 2025-10-24
AI Technical Summary
Existing methods for predicting the drift of distressed targets at sea are insufficient in terms of dynamic update capability and accuracy. In particular, errors accumulate significantly in long-term predictions, and they cannot effectively integrate the spatiotemporal variation characteristics of multiple factors such as real-time wind field, ocean current, and waves.
A method based on marine dynamic environmental field data is used to calculate the drift velocities caused by current, wind, and waves. Compensating drift velocities are generated through an artificial intelligence model, and a drift trajectory prediction model is established by combining the real-time target location to update the location of distressed targets at sea in real time.
It improves the prediction accuracy and timeliness of the drift trajectory of distressed targets at sea, providing more reliable search and rescue decision support.
Smart Images
Figure CN120832484A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of marine search and rescue, and in particular to a method for predicting the drift of a marine distress target, a medium, equipment and program product. BACKGROUND
[0002] With the rapid development of global marine economic activities, marine distress accidents occur frequently. Rapid and accurate positioning of distress targets (such as fallen personnel, wrecked ships or floating objects) is crucial to the success of search and rescue operations. Traditional drift prediction methods are mainly based on empirical formulas or simplified fluid mechanics models, and their accuracy is limited by the insufficient description of dynamic changes in the marine environment, such as the lack of integration of real-time wind field, ocean current, wave and other multi-factor spatio-temporal variation characteristics, especially in long-term prediction, where error accumulation is significant. In recent years, satellite remote sensing, ocean buoys and other technologies have improved data acquisition capabilities, but real-time fusion and efficient modeling of multi-source heterogeneous data still face technical bottlenecks. Therefore, there is an urgent need for an intelligent prediction method that integrates multi-dimensional environmental data and has dynamic updating capabilities to improve the prediction accuracy and timeliness of the drift trajectory of a marine distress target and provide reliable support for search and rescue decision-making. SUMMARY
[0003] The purpose of the present application is to solve the problem of poor dynamic updating capability of existing methods for predicting the drift trajectory of a marine distress target. A method for predicting the drift of a marine distress target is proposed, comprising the following steps: S1, obtaining the initial position of the marine distress target and real-time marine dynamic environmental field data; S2, calculating the drift motion velocity component caused by the marine dynamic environmental field data, and vector synthesizing the velocity component to obtain the initial drift velocity of the marine distress target; S3, using an artificial intelligence model to analyze historical data to generate a compensation drift velocity, and synthesizing the initial drift velocity of the marine distress target and the compensation drift velocity to obtain the drift velocity of the marine distress target; S4, combining the initial position of the marine distress target and the drift velocity of the marine distress target, establishing a marine distress target drift trajectory prediction model, and predicting the position of the marine distress target in real time.
[0004] Further, the initial position of the marine distress target is obtained in real time by using an emergency radio beacon.
[0005] Further, the marine dynamic environmental field data includes sea surface 10-meter height wind speed, significant wave height and ocean current speed.
[0006] Further, the drift motion velocity component includes flow-induced drift velocity, wind-induced drift velocity and wave-induced drift velocity. The flow-induced drift velocity is constructed based on the saturation rate constant, and the calculation formula of the flow-induced drift velocity is: , wherein, represents a vector representation of the flow-induced drift velocity, represents a maximum drift coefficient, represents a constant reflecting a saturation rate, represents a vector representation of the surface current velocity, represents an error of the flow-induced drift velocity; The wind-induced drift velocity is divided into three parts of the wind-induced drift velocity in the downwind direction, the right-biased wind direction, and the left-biased wind direction, and the calculation formula is: ) + + , ) + + , ) + + , wherein, represents the wind-induced drift velocity in the downwind direction, represents the wind-induced drift velocity in the right-biased wind direction, represents the wind-induced drift velocity in the left-biased wind direction, represents the wind speed at the height of 10 meters above the sea surface, , , , represents a linear term coefficient related to the wind-induced drift velocity in the downwind direction, , , , represents a linear term coefficient related to the wind-induced drift velocity in the right-biased wind direction, , , , represents a linear term coefficient related to the wind-induced drift velocity in the left-biased wind direction, wherein all the linear term coefficients related to the wind-induced drift velocity in the downwind direction, the right-biased wind direction, and the left-biased wind direction are variables, and the sizes are related to the wind speed at the height of 10 meters above the sea surface, , , respectively represent the errors of the wind-induced drift velocity in the downwind direction, the right-biased wind direction, and the left-biased wind direction; The wind-induced drift velocities in the downwind direction, the right-biased wind direction, and the left-biased wind direction are vector-synthesized to obtain the wind-induced drift velocity; The wave-induced drift velocity includes a periodic term that reflects the combined effect ratio of wave height, period, and wavelength, a directional term that adjusts the consistency of the drift direction with the wave propagation direction, and an independent periodic term that adjusts the drift velocity variation within different period ranges. The calculation formula for the wave-induced drift velocity is: ( )+ + + , in, ( ) is a periodic term, is the direction term, is an independent periodic term, The vector representation of the wave-induced drift velocity, 、 、 represents the fitting coefficient, represents the wave height of the significant wave, represents the wavelength, represents the wave direction, T represents the wave period, Indicates the error in wave-induced drift velocity.
[0007] Furthermore, the drift trajectory prediction model of the distress target at sea is expressed as follows: , in, The vector representation of the target position at time t, The vector representation of the initial position of the target, express The vector representation of the flow-induced drift velocity at time , express The vector representation of the wind-induced drift velocity at time , express The vector representation of the wave-induced drift velocity at time , express The vector representation of the moment-to-moment drift velocity compensation, A vector representation of the random drift due to turbulence.
[0008] Furthermore, the following method is used to adjust the linear term coefficient in the wind-induced drift velocity vector expression formula: Construct an overdetermined system of equations: ,in, , , , , , , wherein, A is a design matrix, p is a parameter vector to be optimized, b is an observation component of a bias vector; bias vector wherein, denotes a bias vector at the moment, denotes an actual position vector at the moment, denotes a position vector predicted by the offshore distress target drift trajectory prediction model at the moment; solved by a regularized least square method p= b, wherein denotes a regularization coefficient, which prevents overfitting, and I denotes a unit matrix.
[0009] Further, the historical data is analyzed using an artificial intelligence model to generate a compensation drift speed, specifically including: (1) Collecting wind speed, wave height, flow speed historical measured data, drift speed measured data, predicting the drift speed based on the wind speed, wave height, flow speed historical measured data, obtaining a prediction deviation of the drift speed; (2) Training a regression model through a machine learning algorithm, and establishing a mapping relationship between the wind speed, wave height, flow speed and the compensation drift speed based on the prediction deviation of the drift speed; (3) Inputting real-time wind, wave and flow data into the regression model to output the compensation drift speed .
[0010] The application further provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the offshore distress target drift prediction method.
[0011] The application further provides an electronic device, including a processor and a memory, the processor and the memory are connected with each other, wherein the memory is used for storing a computer program, the computer program includes computer readable instructions, the processor is configured to call the computer readable instructions, and the offshore distress target drift prediction method is executed.
[0012] The application further provides a computer program product, including computer programs / instructions, characterized in that the computer programs / instructions are executed by a processor to realize the steps of the offshore distress target drift prediction method.
[0013] The technical scheme provided by the application has the beneficial effects that: The present application obtains the flow-induced drift speed, the wind-induced drift speed and the wave-induced drift speed according to the real-time marine dynamic environment field data, generates the compensation drift speed in real time by using AI, establishes the drift trajectory prediction model of the marine distress target by combining the initial position of the marine distress target obtained in real time, and predicts the position of the marine distress target in real time. The present application combines the real-time position of the target and the real-time marine dynamic environment field data, and the compensation drift speed predicted in real time by using AI, improves the prediction accuracy and timeliness of the drift trajectory of the marine distress target. BRIEF DESCRIPTION OF DRAWINGS
[0014] Figure 1 is a flow chart of the drift prediction method of the marine distress target according to an embodiment of the present application; Figure 2 is a block diagram of an electronic device according to an exemplary embodiment of the present application. DETAILED DESCRIPTION
[0015] To make the object, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described below with reference to the drawings.
[0016] The flow chart of the drift prediction method of the marine distress target according to an embodiment of the present application is as shown in Figure 1 , and specifically includes the following steps: S1, obtaining the real-time updated initial position of the marine distress target and the real-time marine dynamic environment field data.
[0017] The marine distress target can send the target position information to the rescue agency through the satellite positioning system (such as the ship automatic identification system, the global positioning system (GPS)), the radio communication, the radar system, or the use of the unmanned aerial vehicle reconnaissance. The present application embodiment adopts the emergency radio position marker to obtain the initial position of the marine distress target in real time.
[0018] In order not to lose the target, the initial position of the marine distress target needs to be updated constantly, and the following two initial position trigger update mechanisms can be adopted: (1) fixed time trigger mode: periodically sending position data according to a preset time interval; (2) event-driven trigger mode: when the target acceleration mutation or the environmental parameter exceeds the safety threshold is detected, the position reporting is triggered immediately.
[0019] When a new position is received, the initial time is reset to the current time, and the target initial position is updated to the target position obtained at present.
[0020] The marine dynamic environment field data includes: sea surface 10-meter height wind speed, effective wave height, and ocean current speed. The 10m wind speed data can well represent the distribution characteristics and variation characteristics of the sea surface wind field. The wind produces friction on the sea surface object, which is one of the important forces for pushing the object to drift on the sea surface. The greater the wind speed, the greater the force of the wind on the object, and the faster the drift speed of the object. For objects without power devices or with small power, wind speed is the key factor determining the drift speed and direction of the object. The effective wave height refers to the actual wave height value counted according to certain rules. Since the sea surface waves are actually an irregular combination of various waves with different wave heights, periods, and directions, the wave height value of a wave does not have representativeness. Therefore, in any wave group composed of n waves, the wave heights in the wave train are arranged in descending order, and the first n / 3 waves are determined as effective waves. The wave height and period of the effective wave are equal to the average wave height and average period of the n / 3 waves. The wave height of the effective wave represents the average height of the sea wave, and the greater the wave height, the greater the impact force of the sea wave on the object. When the sea wave impacts the object, it will bring a forward thrust to the object, increasing the drift speed of the object. The ocean current speed is one of the basic driving forces for the object to drift in seawater. The object will move with the flow of seawater, and the size and direction of the ocean current speed directly determine the basic speed and direction of the object drift. If the object itself has no other power source, it will drift at the speed and direction of the ocean current. The sea surface 10-meter height wind speed, effective wave height, and ocean current speed can be obtained through satellite remote sensing data, buoy observation system, and field observation instruments.
[0021] S2, calculate the drift motion speed component caused by the marine dynamic environment field data, vector synthesize the speed component, and obtain the initial offshore distress target drift speed.
[0022] The drift motion speed component includes: current-induced drift speed, wind-induced drift speed, and wave-induced drift speed. The current-induced drift speed is constructed based on the saturation rate constant, and the calculation formula of the current-induced drift speed is: , wherein, represents the vector representation of the current-induced drift speed, represents the maximum drift coefficient, represents a constant reflecting the saturation rate, which is depicted by the decrease in drift acceleration due to inertia and drift effect, i.e. the momentum saturation phenomenon; represents the vector representation of the surface flow speed, represents the error of the current-induced drift speed. This method significantly improves the fitting effect in the scene where the sea wind speed is large.
[0023] The wind-induced drift speed is divided into three parts: wind-induced drift speed in the wind direction, right-biased wind-induced drift speed, and left-biased wind-induced drift speed, and the calculation formula is: ) +[ ]+ , ) +[ ]+ , ) +[ ]+ , wherein, represents the downwind wind-induced drift velocity, represents the right-biased wind-induced drift velocity, represents the left-biased wind-induced drift velocity, represents the sea surface 10-meter height wind speed, , , , represents the downwind wind-induced drift velocity related linear term coefficient, , , , represents the right-biased wind-induced drift velocity related linear term coefficient, , , , represents the left-biased wind-induced drift velocity related linear term coefficient, , , respectively represent the error of the downwind, right-biased wind-induced drift velocity, and left-biased wind-induced drift velocity.
[0024] wherein all the related linear term coefficients of the downwind, right-biased wind-induced drift velocity, and left-biased wind-induced drift velocity are variables, and the size is related to the sea surface 10-meter height wind speed, and the related linear term coefficients in the calculation formula of the wind-induced drift velocity are adjusted by the following method: The linear term coefficients in the vector expression formula of the wind-induced drift velocity are adjusted by the following method: An overdetermined equation set is constructed: wherein, , , , , , , Wherein, A is a design matrix, p is a parameter vector to be optimized, b is an observation component of a bias vector (and three components related to the wind direction, right wind direction and left wind direction); bias vector , wherein, represents the bias vector at the moment, represents the actual position vector at the moment, represents the position vector predicted by the offshore distress target drift trajectory prediction model of the application at the moment. At the initial moment, the bias vector is 0, and the linear term coefficient in the adjusted wind-induced drift speed is obtained based on the bias vector at the moment. the bias vector at the moment. The linear term coefficient in the adjusted wind-induced drift speed is obtained as the linear term coefficient for calculating the wind-induced drift speed at the next moment. Real-time feedback adjustment is performed on the wind-induced drift speed, so that the drift trajectory prediction of the offshore distress target can be more accurately obtained.
[0025] Solve p by regularized least squares method b, wherein represents a regularization coefficient to prevent overfitting, and I represents a unit matrix.
[0026] The wind-induced drift speeds in the wind direction, the right wind direction and the left wind direction are vector synthesized to obtain the wind-induced drift speed.
[0027] The wave-induced drift speed includes a period term reflecting the combined effect ratio of wave height, period and wavelength, a direction term adjusting the consistency of drift direction and wave propagation direction, and an independent period term adjusting the drift speed change in different period ranges. The calculation formula of the wave-induced drift speed is: ( )+ + + , wherein, represents the vector representation of the wave-induced drift speed, , , represents the fitting coefficient, represents the average wave height of the first 1 / 3 wave, represents the wavelength, represents the wave direction, and T represents the wave period, represents the error of the wave-induced drift speed. ( ) is a period term, which comprehensively reflects the combined effect ratio of wave height, period and wavelength; is a direction term, which adjusts the consistency of drift direction and wave propagation direction; For independent periodic terms, capture the independent influence of the period on the drift speed (such as the deep momentum transport of long-period waves), and adjust the drift speed variation in different period ranges.
[0028] S3, using an artificial intelligence (AI) model to analyze historical data to generate a compensation drift speed, which is used to correct the prediction error of wind, wave, and current, the initial offshore distress target drift speed, and the compensation drift speed are combined to form the offshore distress target drift speed.
[0029] Using an artificial intelligence model to analyze historical data to generate a compensation drift speed, specifically including: (1) Collecting wind speed, wave height, current speed historical measured data, drift speed measured data, using the method of the present application to predict the drift speed based on wind speed, wave height, current speed historical measured data, obtaining the prediction deviation of the drift speed; (2) Training a regression model (such as a gradient boosting tree (GBDT) algorithm) through a machine learning algorithm, and establishing a mapping relationship between wind speed, wave height, current speed and compensation drift speed based on the prediction deviation of the drift speed; (3) Inputting real-time wind, wave, and current data into the regression model to output the compensation drift speed .
[0030] S4, combining the initial position of the offshore distress target and the drift speed of the offshore distress target, establishing an offshore distress target drift trajectory prediction model, and real-time predicting the position of the offshore distress target.
[0031] The offshore distress target drift trajectory prediction model is represented by the formula: , Wherein, represents the initial time, represents the vector representation of the target position at time t, represents the vector representation of the initial position of the target, represents the vector representation of the current-induced drift speed at time t, represents the vector representation of the wind-induced drift speed at time t, represents the vector representation of the wave-induced drift speed at time t, represents the vector representation of the compensation drift speed at time t, represents the vector representation of the random drift caused by turbulence.
[0032] In an exemplary embodiment, a computer readable storage medium is provided, the computer readable storage medium storing a computer program, the computer program being executed by a processor to implement the offshore distress target drift prediction method described above.
[0033] Reference is made to Figure 2 In an exemplary embodiment, an electronic device is also included, comprising at least one processor, at least one memory, and at least one communication bus.
[0034] The memory has stored thereon a computer program, the computer program comprising computer readable instructions, and the processor invokes the computer readable instructions stored in the memory through the communication bus to execute the offshore distress target drift prediction method.
[0035] In an exemplary embodiment, a computer program product is also included, comprising computer program / instructions, characterized in that the computer program / instructions, when executed by the processor, implement the steps of the offshore distress target drift prediction method.
[0036] The above description of disclosed embodiments will enable one of ordinary skill in the art to make and use the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method of predicting the drift of a marine distress target, characterized in that, The method comprises the following steps: S1, acquiring real-time updated initial position of a marine distress target and real-time marine dynamic environment field data; S2, calculating a drift motion velocity component caused by the marine dynamic environment field data, vector synthesizing the velocity component to obtain an initial marine distress target drift velocity; S3, using an artificial intelligence model to analyze historical data to generate a compensation drift velocity, and synthesizing the initial marine distress target drift velocity and the compensation drift velocity to obtain a marine distress target drift velocity; S4, combining the initial position of the marine distress target and the marine distress target drift velocity to establish a marine distress target drift trajectory prediction model to predict the position of the marine distress target in real time.
2. A method of predicting the drift of a marine distress object according to claim 1, characterized in that, The initial position of the marine distress target is acquired in real time by using an emergency radio beacon.
3. The method of claim 1, wherein, The marine dynamic environment field data comprises a 10-meter-height wind speed on the sea surface, an effective wave height, and a marine flow velocity.
4. A method of predicting the drift of a marine distress beacon according to claim 3, wherein, The drift motion velocity component comprises a flow-induced drift velocity, a wind-induced drift velocity, and a wave-induced drift velocity. The flow-induced drift velocity is constructed based on a saturation rate constant, and a calculation formula of the flow-induced drift velocity is as follows: ， wherein, denotes a vector representation of the flow-induced drift velocity, denotes a maximum drift coefficient, denotes a constant reflecting a saturation rate, denotes a vector representation of the surface layer flow velocity, denotes an error of the flow-induced drift velocity; The wind-induced drift velocity is divided into three parts, i.e., a downwind drift velocity, a right-biased wind drift velocity, and a left-biased wind drift velocity, and a calculation formula of the wind-induced drift velocity is as follows: ) +[ ]+ ， ) +[ ]+ ， ) +[ ]+ ， wherein, denotes the downwind wind-induced drift speed, denotes the right-bias wind-induced drift speed, denotes the left-bias wind-induced drift speed, denotes the sea surface 10-meter height wind speed, , , , denotes the downwind wind-induced drift speed related linear term coefficient, , , , denotes the right-bias wind-induced drift speed related linear term coefficient, , , , denotes the left-bias wind-induced drift speed related linear term coefficient, wherein all the related linear term coefficients of the downwind, right-bias and left-bias wind-induced drift speeds are variables, the sizes of which are related to the sea surface 10-meter height wind speed, , , denote the errors of the downwind, right-bias and left-bias wind-induced drift speeds, respectively; The downwind drift velocity, the right-biased wind drift velocity, and the left-biased wind drift velocity are vector synthesized to obtain the wind-induced drift velocity. The wave-induced drift velocity comprises a period term reflecting a combined effect ratio of a wave height, a period, and a wavelength, a direction term adjusting consistency of a drift direction and a wave propagation direction, and an independent period term adjusting drift velocity variation in different period ranges, and a calculation formula of the wave-induced drift velocity is as follows: ( )+ + + ， wherein is a periodic term, is a directional term, is an independent periodic term, denotes a vector representation of the wave-induced drift velocity, , , denotes a fitting coefficient, denotes the wave height of the effective wave, denotes the wavelength, denotes the wave direction, T denotes the wave period, denotes the error of the wave-induced drift velocity. 5. A method of predicting the drift of a marine distress beacon according to claim 4, wherein, The marine distress target drift trajectory prediction model is expressed by a formula as follows: ， wherein, a vector representation of the position of the target at time t, a vector representation of the initial position of the target, a vector representation of a vector representation of the flow-induced drift velocity at time t, a vector representation of a vector representation of the wind-induced drift velocity at time t, a vector representation of a vector representation of the wave-induced drift velocity at time t, a vector representation of a vector representation of the compensation drift velocity at time t, a vector representation of the random drift occurring due to turbulence effects.
6. A method of predicting the drift of a marine distress beacon according to claim 5, wherein, The linear term coefficient in the vector expression formula of the wind-induced drift velocity is adjusted by using the following method: Construct overdetermined equation system: wherein, ， ， ， ， ， ， Wherein, A is a design matrix, p is a parameter vector to be optimized, and b is an observation value component of a bias vector. bias vector wherein denotes a bias vector at a time instant, denotes an actual position vector at a time instant, denotes a position vector predicted by the drift trajectory prediction model for the marine distress target at a time instant; Solve p= by regularized least squares method b, where represents the regularization coefficient to prevent overfitting, and I represents the identity matrix.
7. The method of claim 1, wherein, The artificial intelligence model is used to analyze the historical data to generate the compensation drift velocity, and the method specifically comprises the following steps: (1) collecting historical measured data of wind speed, wave height, flow velocity, and drift velocity, predicting the drift velocity based on the historical measured data of wind speed, wave height, and flow velocity to obtain a prediction deviation of the drift velocity; (2) training a regression model by using a machine learning algorithm, and establishing a mapping relationship between the wind speed, the wave height, the flow velocity, and the compensation drift velocity based on the prediction deviation of the drift velocity; (3) input real-time wind, wave, and current data into the regression model to output a compensated drift speed .
8. A computer-readable storage medium, the computer-readable storage medium storing a computer program, characterized in that: The computer program is executed by the processor to realize the method of any one of claims 1-7.
9. An electronic device, comprising: The computer program is executed by the processor to realize the method of any one of claims 1-7.
10. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to realize the steps of the method of any one of claims 1-7.