An underwater vehicle ice-breaking surfacing method and system based on digital twinning
By constructing a digital twin model and combining the discrete element method and the random forest algorithm, the problems of long computation time and lack of cumulative damage assessment during the icebreaking and surfacing process of underwater vehicles were solved. Real-time risk avoidance decision-making and dynamic safety envelope generation were realized, improving the safety and adaptability of underwater vehicles.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA STATE SHIPBUILDING CORP LTD RESEARCH INSTITUTE 719
- Filing Date
- 2026-02-11
- Publication Date
- 2026-05-29
Smart Images
Figure CN122113733A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of polar marine engineering and intelligent operation and maintenance technology, and in particular to a method and system for icebreaking, surfacing and hazard avoidance of underwater vehicles based on digital twins. Specifically, it relates to a digital twin method and system for icebreaking, surfacing and hazard avoidance of underwater vehicles that integrates multiphysics simulation and data-driven models. Background Technology
[0002] As polar exploration deepens, the ability of underwater vehicles to operate under ice sheets and break through ice to surface has become a key tactical requirement. During the impact with sea ice, which causes the vehicle to bend, crush, and break apart, its hull structure is subjected to severe nonlinear impact loads.
[0003] Currently, safety assessments for icebreaking and surfacing mainly rely on the following methods: 1. Physical model testing: These are costly and difficult to reproduce extremely complex working conditions; 2. Numerical simulation methods (such as DEM or FEM): Although the calculation accuracy is high, due to the large amount of discrete fragmentation of sea ice and the nonlinear response of the structure involved in the icebreaking process, a single calculation often takes several hours or even days, which cannot meet the real-time risk avoidance decision-making needs of the vehicle during the surfacing process; 3. Conventional structural monitoring: This focuses more on transient stress and often ignores the cumulative damage and fatigue degradation caused to the hull structure by multiple surfacing operations.
[0004] In addition, existing digital twin solutions in the polar region focus more on data visualization and lack a high-fidelity mapping mechanism from microscopic ice loads to macroscopic structural failure risks, and it is difficult to achieve dynamic risk avoidance boundary delineation under multi-parameter conditions. Summary of the Invention
[0005] The purpose of this invention is to solve the problems in the prior art and to propose a method and system for underwater vehicles to break ice, surface and avoid danger based on digital twins.
[0006] This invention is achieved through the following technical solution: This invention proposes a method for underwater vehicles based on digital twins to break ice, surface, and avoid danger, the method comprising: Step S1: Construct a physical twin model of the icebreaking process: Based on the discrete element method (DEM), establish an interaction model between sea ice and the vehicle, simulate the sea ice breaking and accumulation behavior under different working conditions, and obtain time history data of icebreaking loads that change over time. Step S2: Construct a structural response twin model: Apply the icebreaking load to the finite element model of the vehicle using a spatial topology mapping mechanism, and establish a structural failure risk probability index RI. The index comprehensively considers the transient structural response at the current moment and the cumulative damage effect of historical operations. Step S3: Construct a data-driven twin model: Based on multiple sets of physical simulation samples, a random forest algorithm is used to train a surrogate prediction model, establish a nonlinear mapping relationship between working condition parameters and structural response indicators, and use cross-validation indicators to evaluate the model's credibility. Step S4: Generate safety envelope and risk avoidance strategy: Use the partial dependency analysis (PDP) algorithm to decouple the influence weight of operating condition parameters on structural risk, construct a dynamic safety envelope based on RI threshold in the multidimensional parameter space, and generate risk avoidance control commands according to the position of the real-time state point of the aircraft relative to the safety envelope. Step S5: Digital Twin Update: Based on the newly acquired simulation data or measured data, update the agent prediction model and security envelope boundary online.
[0007] Furthermore, in step S1, under a simulation environment, the motion trajectory of sea ice particles is calculated using Newton's laws of motion, and the mechanical transmission and fracture behavior within the sea ice are simulated using a parallel bonding model; by traversing all sea ice particles in contact with the aircraft, the formula is used... Obtain time-history data of icebreaking loads that vary over time. The number of particles in contact with the structure. Let be the component of the normal contact force of the j-th contacting particle in the vertical direction.
[0008] Furthermore, the specific method for loading the ice-breaking payload using the spatial topology mapping mechanism in step S2 is as follows: setting a spatial mapping radius threshold d. thr Calculate the Euclidean distance d between the discrete load point and the finite element node. When the distance satisfies d 2 ≤d thr 2 At that time, a topological mapping relationship is established to transform discrete loads into a distributed pressure field acting on the surface of the vehicle hull.
[0009] Furthermore, in step S2, the calculation model for the structural failure risk probability index RI is as follows:
[0010] In the formula, The von Mises equivalent stress predicted by the surrogate model. The yield strength of the material; For the predicted equivalent plastic strain, The fracture strain threshold for material failure; It is the structural cumulative damage factor; This is the cumulative damage weighting coefficient, used to adjust the weight of historical damage on the current risk assessment.
[0011] Furthermore, in step S3, a physical simulation sample set is generated based on Latin hypercube sampling, and a random forest regression model is trained as a lightweight proxy for the physical twin. This model establishes a nonlinear mapping relationship between operating parameters and structural response indicators, reducing the computation time of a single safety assessment from hours to milliseconds, thus meeting the time constraints of real-time risk avoidance.
[0012] Furthermore, the operating parameters include sea ice thickness, vehicle ascent speed, and ascent angle.
[0013] Furthermore, in step S4, the partial dependency analysis (PDP) method is used to decouple the influence weights of each operating condition parameter on structural risk, and a dynamic safety envelope with RI=1.0 as the characteristic boundary is constructed in the multidimensional parameter space. The system monitors the vehicle status in real time. When it is predicted that the status deviates from the safety envelope and enters the danger zone, the search algorithm is used to find the shortest path back to the safety zone and output the optimal risk avoidance strategy command.
[0014] The present invention also proposes an icebreaking, surfacing, and hazard-avoidance system for an underwater vehicle based on digital twins to implement the aforementioned method, the system comprising: The icebreaking process simulation module is used to construct a discrete element model of sea ice and output icebreaking load data; The structural response analysis module is used to assess the probability index of structural failure risk through a spatial topology mapping mechanism. The security status prediction module is used to build a proxy prediction model based on the random forest algorithm. The secure envelope construction module is used to generate dynamic secure envelope boundaries in a multidimensional parameter space; The risk avoidance decision output module is used to output risk avoidance strategies such as attitude adjustment or speed control based on the current risk status.
[0015] The present invention also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the aforementioned method for icebreaking, surfacing and hazard avoidance of an underwater vehicle based on digital twins.
[0016] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of the described method for icebreaking, surfacing, and hazard avoidance of an underwater vehicle based on digital twins.
[0017] The beneficial effects of this invention are: This invention proposes a digital twin-based method and system for icebreaking and surfacing avoidance of underwater vehicles, which has significant technical advantages and application value compared with existing technologies. First, by constructing a data-driven twin model based on the random forest algorithm, this invention successfully achieves millisecond-level proxy substitution for complex nonlinear finite element calculations, greatly improving the system's response speed and fundamentally solving the problem that traditional numerical methods, due to their long computation time, cannot meet the real-time avoidance decision-making requirements during icebreaking and surfacing of underwater vehicles. Second, this invention innovatively proposes a structural failure risk probability index that comprehensively considers transient stress and cumulative damage. This index not only reflects the current mechanical response but also quantitatively assesses the fatigue degradation state of the vehicle during long-term service, thus providing a more comprehensive and scientific structural safety evaluation than a single strength criterion. Furthermore, this invention utilizes partial dependency analysis (PDP) to deeply decouple the complex nonlinear coupling effects between operating parameters such as ice thickness, surfacing angle, and speed. The constructed multidimensional dynamic safety envelope boundary is intuitive and clear, providing a solid theoretical basis and auxiliary decision support for planning a feasible avoidance adjustment strategy for the vehicle in emergency situations. Finally, the present invention has a complete online update mechanism for digital twins, which can continuously evolve and correct model parameters based on measured sensor data, ensuring a high degree of consistency between the digital space and the physical environment, and significantly enhancing the system's adaptability and risk avoidance success rate in the complex and variable polar ice environment. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0019] Figure 1 This is a schematic diagram of the overall process of an icebreaking, surfacing and hazard avoidance method for an underwater vehicle based on digital twin proposed in this invention; Figure 2 This is a schematic diagram of the discrete element physical twin model constructed in this embodiment and the scene of the interaction between the vehicle and sea ice; Figure 3 This is a schematic diagram of the spatial topology mapping logic from discrete element load points to finite element nodes as described in step S2 of this embodiment; Figure 4 This is a scatter plot comparing the predicted and actual values of the random forest proxy model constructed in step S3 of this embodiment on the test set. Figure 5This is a schematic diagram of the ice thickness-dip angle multidimensional safety envelope and risk avoidance path planning constructed based on two-dimensional partial dependency analysis (PDP) in step S4 of this embodiment; Figure 6 This is a block diagram illustrating the structural composition of an icebreaking, surfacing, and hazard-avoiding underwater vehicle system based on digital twins, as proposed in this invention. Detailed Implementation
[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0021] The purpose of this invention is to overcome the shortcomings of existing technologies, such as large computational load, poor real-time performance, and lack of cumulative damage assessment mechanisms in numerical simulation, and to provide a method and system for icebreaking and surfacing avoidance of underwater vehicles based on digital twins. This invention enables rapid spatiotemporal mapping of icebreaking loads and significantly improves computational efficiency, meeting real-time requirements for structural risk prediction and safety envelope construction, thus providing a feasible risk avoidance adjustment strategy for the vehicle.
[0022] Specifically, in combination Figures 1-6 This invention proposes a method for underwater vehicles to break ice, surface, and avoid danger based on digital twins, the method comprising: Step S1: Construct a physical twin model of the icebreaking process: Based on the discrete element method (DEM), establish an interaction model between sea ice and the vehicle, simulate the sea ice breaking and accumulation behavior under different working conditions, and obtain time history data of icebreaking loads that change over time. Step S2: Construct a structural response twin model: Apply the icebreaking load to the finite element model of the vehicle using a spatial topology mapping mechanism, and establish a structural failure risk probability index RI. The index comprehensively considers the transient structural response at the current moment and the cumulative damage effect of historical operations. Step S3: Construct a data-driven twin model: Based on multiple sets of physical simulation samples, a random forest algorithm is used to train a surrogate prediction model, establish a nonlinear mapping relationship between working condition parameters and structural response indicators, and use cross-validation indicators to evaluate the model's credibility. Step S4: Generate safety envelope and risk avoidance strategy: Use the partial dependency analysis (PDP) algorithm to decouple the influence weight of operating condition parameters on structural risk, construct a dynamic safety envelope based on RI threshold in the multidimensional parameter space, and generate risk avoidance control commands according to the position of the real-time state point of the aircraft relative to the safety envelope. Step S5: Digital Twin Update: Based on the newly acquired simulation data or measured data, update the agent prediction model and security envelope boundary online.
[0023] Furthermore, in step S1, under a simulation environment, the motion trajectory of sea ice particles is calculated using Newton's laws of motion, and the mechanical transmission and fracture behavior within the sea ice are simulated using a parallel bond model; by traversing all sea ice particles in contact with the aircraft, the formula is used... Obtain time-history data of icebreaking loads that vary over time. The number of particles in contact with the structure. Let be the component of the normal contact force of the j-th contacting particle in the vertical direction.
[0024] Furthermore, the specific method for loading the ice-breaking payload using the spatial topology mapping mechanism in step S2 is as follows: setting a spatial mapping radius threshold d. thr Calculate the Euclidean distance d between the discrete load point and the finite element node. When the distance satisfies d 2 ≤d thr 2 At that time, a topological mapping relationship is established to transform discrete loads into a distributed pressure field acting on the surface of the vehicle hull.
[0025] Furthermore, in step S2, the calculation model for the structural failure risk probability index RI is as follows:
[0026] In the formula, The von Mises equivalent stress predicted by the surrogate model. The yield strength of the material; For the predicted equivalent plastic strain, The fracture strain threshold for material failure; It is the structural cumulative damage factor; This is the cumulative damage weighting coefficient, used to adjust the weight of historical damage on the current risk assessment.
[0027] Furthermore, in step S3, when training the surrogate prediction model, a random forest algorithm can be used to train the nonlinear mapping relationship between the buoyancy parameters and structural response indicators. Specifically, in step S3, a physical simulation sample set is generated based on Latin hypercube sampling, and a random forest regression model is trained as a lightweight surrogate for the physical twin. This model establishes a nonlinear mapping relationship between the buoyancy parameters (ice thickness, strength, velocity, and tilt angle) and structural response indicators, reducing the computation time of a single safety assessment from hours to milliseconds, thus meeting the time constraints of real-time hazard avoidance.
[0028] Furthermore, the operating parameters include sea ice thickness, vehicle ascent speed, and ascent angle.
[0029] Furthermore, in step S4, the partial dependency analysis (PDP) method is used to decouple the influence weights of each operating condition parameter on structural risk, and a dynamic safety envelope with RI=1.0 as the characteristic boundary is constructed in the multidimensional parameter space. The system monitors the vehicle status in real time. When it is predicted that the status deviates from the safety envelope and enters the danger zone, the search algorithm is used to find the shortest path back to the safety zone and output the optimal risk avoidance strategy command.
[0030] Furthermore, the specific method for constructing the dynamic safety envelope in step S4 is as follows: using the structural failure risk probability index RI=1.0 as the characteristic boundary, the safe zone and the danger zone are delineated in the multidimensional parameter space composed of the buoyancy angle, buoyancy speed and sea ice thickness.
[0031] Furthermore, in step S5, during the actual icebreaking and ascent of the vehicle, the measured data (such as strain, tilt angle, and velocity) obtained by the sensors are acquired in real time. The error correction algorithm is used to update and self-evolve the surrogate prediction model and safety envelope boundary online to ensure the real-time synchronization between the twin and the physical entity.
[0032] The present invention also proposes an icebreaking, surfacing, and hazard-avoidance system for an underwater vehicle based on digital twins to implement the aforementioned method, the system comprising: The icebreaking process simulation module is used to construct a discrete element model of sea ice and output icebreaking load data; The structural response analysis module is used to assess the probability index of structural failure risk through a spatial topology mapping mechanism. The security status prediction module is used to build a proxy prediction model based on the random forest algorithm. The secure envelope construction module is used to generate dynamic secure envelope boundaries in a multidimensional parameter space; The risk avoidance decision output module is used to output risk avoidance strategies such as attitude adjustment or speed control based on the current risk status.
[0033] This invention proposes a method and system for icebreaking and surfacing avoidance of underwater vehicles based on digital twins, belonging to the field of polar marine engineering technology. The method includes: constructing a physical twin model of sea ice breaking based on discrete element method (DEM) to obtain real-time loads; establishing a structural response twin model and using the RI index to comprehensively evaluate transient stress and cumulative damage risk; constructing a millisecond-level surrogate prediction model based on the random forest algorithm; constructing a dynamic safety envelope using partial dependency analysis and outputting the optimal avoidance strategy; and updating the twin in real time. This invention solves the problems of long computation time in numerical simulation, inability to achieve real-time avoidance, and lack of cumulative damage assessment.
[0034] Example This embodiment uses the SUBOFF underwater vehicle model as an example to simulate a scenario in which it performs an ascent maneuver in a polar ice region.
[0035] Step S1: Construct a physical twin model of the icebreaking process. Construct the sea ice field in discrete element method (DEM) simulation software. Based on typical polar sea conditions, the sea ice thickness h can be set to a range of 0.3m to 2.0m, and the sea ice bending strength... σ The value range is from 0.3 MPa to 1.0 MPa. The buoyancy velocity v can be set from 0.1 to 2.0 m / s, and the buoyancy angle... θ The range is 0 to 20°. The motion of sea ice particles is calculated based on Newton's second law, and the mechanical transmission and fracture behavior within the sea ice are simulated using a parallel bonded model. The simulation step size can be set to DT = 0.01 s. By traversing all contact particles, the formula is used... Export the three-dimensional ice-breaking load sequence files at each time step.
[0036] Step S2: Construct a structural response twin model. Read the rigid body node file and finite element mesh file of the aircraft. The spatial mapping radius threshold Distance0 = 0.5m can be set. Traverse the discrete load points and calculate their Euclidean distance d from the finite element nodes. When d... 2 When the value is ≤0.25, a mapping relationship is established to convert the discrete impact force into a distributed pressure field acting on the shell surface, generating an amplitude file.
[0037] For structural response calculations, this embodiment can assume that the aircraft hull material is high-strength steel, and its yield strength... =785MPa, failure strain =0.15. Assuming the aircraft is in the middle of its service life, a cumulative damage weighting coefficient can be set. =0.2. The system calculates the structural failure risk probability index in real time based on the following formula:
[0038] in, ( ) represents the predicted von Mises stress at the current moment. This is the historical fatigue factor estimated based on Miner's linear cumulative damage theory.
[0039] Step S3: Construct a data-driven twin model. A Latin hypercube sampling method can be used to design 100 sets of working condition samples for physical twin simulation. Working condition parameters (ice thickness, ice strength, ascent speed, ascent angle) are extracted as input features, and the RI index at the corresponding time point is used as the output label. A random forest algorithm can be used for model training. In this embodiment, the predicted model obtained through training has a coefficient of determination R0. 2 The accuracy reaches 0.835, meeting the engineering precision requirements.
[0040] Step S4: Generate a safety envelope and avoidance strategy. Partial dependency analysis (PDP) is used to calculate the interaction between ice thickness h and buoyancy angle θ on the structural risk index RI. In this embodiment, a safety assessment threshold of RI=1.0 can be set. When the vehicle's sensors detect that the predicted RI for the current operating parameters (e.g., ice thickness 1.8m, buoyancy angle 15°) is 1.15, the vehicle is determined to have entered a danger zone. The system automatically retrieves the safety envelope boundary and calculates that, under the current ice thickness, the buoyancy angle needs to be adjusted to within 8° to return to the safe zone (RI<1.0). This strategy is then output as a avoidance suggestion to assist the vehicle's attitude adjustment.
[0041] Step S5: Digital Twin Update. During the ascent, the vehicle receives real-time strain data from key measuring points via surface-mounted strain gauges. The system calculates the residual between the predicted strain and the measured strain, and uses a Bayesian update algorithm to correct the weight parameters of the random forest model online, thereby updating the risk avoidance boundary.
[0042] The present invention also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the aforementioned method for icebreaking, surfacing and hazard avoidance of an underwater vehicle based on digital twins.
[0043] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of the described method for icebreaking, surfacing, and hazard avoidance of an underwater vehicle based on digital twins.
[0044] The memory in this application embodiment can be volatile memory or non-volatile memory, or it can include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory used in the methods described in this invention is intended to include, but is not limited to, these and any other suitable types of memory.
[0045] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., high-density digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).
[0046] In implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software. The steps of the method disclosed in the embodiments of this application can be directly implemented by a hardware processor, or by a combination of hardware and software modules in the processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, detailed descriptions are omitted here.
[0047] It should be noted that the processor in the embodiments of this application can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiments can be completed by the integrated logic circuitry in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied as execution by a hardware decoding processor, or as a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above methods.
[0048] The above provides a detailed description of the icebreaking, surfacing, and hazard avoidance method and system for underwater vehicles based on digital twins proposed in this invention. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.
Claims
1. A method for icebreaking, surfacing, and hazard avoidance of an underwater vehicle based on digital twins, characterized in that, The method includes: Step S1: Construct a physical twin model of the icebreaking process: Based on the discrete element method (DEM), establish an interaction model between sea ice and the vehicle, simulate the sea ice breaking and accumulation behavior under different working conditions, and obtain time history data of icebreaking loads that change over time. Step S2: Construct a structural response twin model: Apply the icebreaking load to the finite element model of the vehicle using a spatial topology mapping mechanism, and establish a structural failure risk probability index RI. The index comprehensively considers the transient structural response at the current moment and the cumulative damage effect of historical operations. Step S3: Construct a data-driven twin model: Based on multiple sets of physical simulation samples, a random forest algorithm is used to train a surrogate prediction model, establish a nonlinear mapping relationship between working condition parameters and structural response indicators, and use cross-validation indicators to evaluate the model's credibility. Step S4: Generate safety envelope and risk avoidance strategy: Use the partial dependency analysis (PDP) algorithm to decouple the influence weight of operating condition parameters on structural risk, construct a dynamic safety envelope based on RI threshold in the multidimensional parameter space, and generate risk avoidance control commands according to the position of the real-time state point of the aircraft relative to the safety envelope. Step S5: Digital Twin Update: Based on the newly acquired simulation data or measured data, update the agent prediction model and security envelope boundary online.
2. The method according to claim 1, characterized in that, In step S1, under simulation conditions, the trajectory of sea ice particles is calculated using Newton's laws of motion, and the mechanical transmission and fracture behavior within the sea ice are simulated using a parallel bonding model. By iterating through all sea ice particles in contact with the aircraft, the formula... Obtain time-history data of icebreaking loads that vary over time. The number of particles in contact with the structure. Let be the component of the normal contact force of the j-th contacting particle in the vertical direction.
3. The method according to claim 1, characterized in that, The specific method for loading the ice-breaking payload using the spatial topology mapping mechanism in step S2 is as follows: set a spatial mapping radius threshold d. thr Calculate the Euclidean distance d between the discrete load point and the finite element node. When the distance satisfies d 2 ≤d thr 2 At that time, a topological mapping relationship is established to transform discrete loads into a distributed pressure field acting on the surface of the vehicle hull.
4. The method according to claim 1, characterized in that, In step S2, the calculation model for the structural failure risk probability index RI is as follows: In the formula, The von Mises equivalent stress predicted by the surrogate model. The yield strength of the material; For the predicted equivalent plastic strain, The fracture strain threshold for material failure; It is the structural cumulative damage factor; This is the cumulative damage weighting coefficient, used to adjust the weight of historical damage on the current risk assessment.
5. The method according to claim 1, characterized in that, In step S3, a physical simulation sample set is generated based on Latin hypercube sampling, and a random forest regression model is trained as a lightweight proxy for the physical twin. This model establishes a nonlinear mapping relationship between operating parameters and structural response indicators, reducing the computation time of a single safety assessment from hours to milliseconds, thus meeting the time constraints of real-time risk avoidance.
6. The method according to claim 5, characterized in that, The operating parameters include sea ice thickness, vehicle ascent speed, and ascent angle.
7. The method according to claim 1, characterized in that, In step S4, the partial dependency analysis (PDP) method is used to decouple the influence weights of each working condition parameter on structural risk, and a dynamic safety envelope with RI=1.0 as the characteristic boundary is constructed in the multidimensional parameter space. The system monitors the vehicle's status in real time. When it predicts that the status has deviated from the safe envelope and entered the danger zone, it uses a search algorithm to find the shortest path back to the safe zone and outputs the optimal risk avoidance strategy command.
8. A digital twin-based underwater vehicle icebreaking, surfacing, and hazard avoidance system for implementing the method of any one of claims 1 to 7, characterized in that, The system includes: The icebreaking process simulation module is used to construct a discrete element model of sea ice and output icebreaking load data; The structural response analysis module is used to assess the probability index of structural failure risk through a spatial topology mapping mechanism. The security status prediction module is used to build a proxy prediction model based on the random forest algorithm. The secure envelope construction module is used to generate dynamic secure envelope boundaries in a multidimensional parameter space; The risk avoidance decision output module is used to output risk avoidance strategies such as attitude adjustment or speed control based on the current risk status.
9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-7.
10. A computer-readable storage medium for storing computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the method according to any one of claims 1-7.