A hydrogen fuel cell ship safety protection system and method

Abstract

The present invention discloses a hydrogen fuel cell ship safety protection system and method, which relates to the field of ship safety technology, including starting multimodal sensors in various areas and embedding them into edge processing units to perform self-checking and calibration; the edge processing unit uses adaptive filtering to pre-process sensor data to obtain raw data; based on the raw data, a three-dimensional virtual model of the ship and various areas is constructed, and real-time CFD simulation is combined to simulate hydrogen diffusion and fire evolution processes to obtain risk prediction parameters and preset safety thresholds; based on active scheduling control commands, an emergency response plan is triggered according to risk level classification; the emergency response plan is verified by comparing on-site feedback information with the execution status, and a safety protection report is formed; the present invention dynamically adjusts the emergency response plan through the risk index, effectively improving the safety, response speed and protection capability of hydrogen fuel cell ships in complex environments, and reducing potential safety risks.

Description

Technical Field

[0001] The present invention relates to the field of ship safety technology, and in particular to a hydrogen fuel cell ship safety protection system and method. Background Art

[0002] As a clean and efficient energy conversion device, hydrogen fuel cells have been rapidly developed in the application of ship power systems in recent years. Hydrogen fuel cell ships have received extensive attention in ship power systems in recent years due to their zero emissions, high energy density and sustainable development potential. With the development of hydrogen energy technology, related storage, transportation and combustion control have become increasingly mature, which has improved the commercial feasibility of fuel cell ships. However, the high diffusivity and flammable and explosive properties of hydrogen make hydrogen fuel cell ships face great safety hazards during operation. Therefore, studying efficient and intelligent safety protection methods to ensure the stability and safe operation of hydrogen fuel has become one of the key issues that need to be urgently solved in this field.

[0003] The existing technical solutions still have obvious deficiencies in safety monitoring and risk prediction. In the process of data collection, preprocessing and real-time analysis, the existing multi-sensor system often cannot fully realize the high-precision fusion and calibration of various sensor information, resulting in the calculation of risk index and early warning response. There is a risk of lag and misjudgment. In addition, the existing emergency response measures generally rely on preset plans, lack of dynamic analysis and closed-loop verification of real-time feedback information on the scene, and it is difficult to achieve accurate hierarchical control and active scheduling of sudden accidents. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a hydrogen fuel cell ship safety protection method to solve the problems of inaccurate safety risk assessment, delayed emergency response and insufficient safety protection.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a hydrogen fuel cell ship safety protection method, which includes starting a multimodal sensor in each area and embedding an edge processing unit to perform self-checking and calibration;

[0008] The edge processing unit uses adaptive filtering to pre-process the sensor data to obtain raw data;

[0009] Based on the original data, a three-dimensional virtual model of the ship and each area is constructed, and combined with real-time CFD simulation to simulate the hydrogen diffusion and fire evolution process, the risk prediction parameters and preset safety thresholds are obtained;

[0010] Calculate the real-time risk index through a multi-sensor data fusion algorithm, compare it with a preset safety threshold, and output an active scheduling control command;

[0011] According to the active scheduling control command, trigger the emergency response plan according to the risk level classification;

[0012] Execute the emergency response plan, automatically control the associated hardware to perform safety operations, and collect on-site feedback information and execution status in real time;

[0013] Compare the on-site feedback information with the execution status to verify the emergency response plan and form a safety protection report.

[0014] As a preferred solution of the hydrogen fuel cell ship safety protection method described in the present invention, wherein: the risk prediction parameters and the preset safety threshold are as follows in specific steps,

[0015] Receive the original data in real time through the edge processing unit, and use the particle swarm optimization algorithm to construct and adjust the three-dimensional virtual model parameters;

[0016] Based on the multi-physics field coupling simulation framework, map the CFD control volume discretization to the qubit state, and use the variational quantum eigen coupling simulation to output the risk prediction parameters;

[0017] Use the generative adversarial network to adaptively preset the safety threshold.

[0018] As a preferred solution of the hydrogen fuel cell ship safety protection method described in the present invention, wherein: the calculation of the real-time risk index through the multi-sensor data fusion algorithm is as follows in specific steps,

[0019] Based on the multi-sensor data fusion algorithm driven by chaos theory, integrate the multi-sensor data weights;

[0020] Construct a comprehensive risk index for multi-dimensional risks across leakage, combustion, and structural failure;

[0021] Compare the comprehensive risk index with the preset safety threshold and output an active scheduling control command

[0022] As a preferred solution of the hydrogen fuel cell ship safety protection method described in the present invention, wherein: the preprocessing refers to that after each edge processing unit receives the sensor data, it is input into the adaptive Kalman filter, and the filtering parameters are dynamically adjusted based on the mean and standard deviation of the past data for high-precision denoising and standardization processing.

[0023] As a preferred solution of the hydrogen fuel cell ship safety protection method described in the present invention, wherein: the triggering of the emergency response plan according to the risk level classification is as follows in specific steps,

[0024] Dynamically divide the risk level into low, medium, high, and extreme through quantitative assessment;

[0025] The classification triggers the emergency response plan corresponding to the risk level.

[0026] As a preferred solution of the hydrogen fuel cell ship safety protection method described in the present invention, wherein: the emergency response plan refers to local isolation, automatic hydrogen supply cut-off, and fire extinguishing; the associated hardware includes solenoid valves, ventilation, fire extinguishing devices, and audible and visual alarms.

[0027] As a preferred solution of the hydrogen fuel cell ship safety protection method described in the present invention, wherein: the safety protection report is used for risk closed-loop management and optimization iteration in the hydrogen fuel cell ship safety protection method, and the content includes the comparison between simulation and measured data, the execution record of the emergency response, the safety performance evaluation, the improvement suggestions and optimization solutions, and the digital twin verification results.

[0028] In a second aspect, the present invention provides a hydrogen fuel cell ship safety protection system, including a start-up calibration module, a data processing module, a risk prediction module, a fusion scheduling module, an emergency response module, an execution feedback module, and a verification report module;

[0029] The start-up calibration module is used to start the multi-modal sensors in each area and embed the edge processing unit, and perform self-check and calibration;

[0030] The data processing module is used for the edge processing unit to preprocess the sensor data by using adaptive filtering to obtain the original data;

[0031] The risk prediction module is used to construct three-dimensional virtual models of the ship and each area based on the original data, and combine real-time CFD simulation to simulate the hydrogen diffusion and fire evolution process to obtain risk prediction parameters and preset safety thresholds;

[0032] The fusion scheduling module is used to calculate the real-time risk index through a multi-sensor data fusion algorithm, compare it with the preset safety threshold, and output an active scheduling control command;

[0033] The emergency response module is used to trigger the emergency response plan according to the risk level based on the active scheduling control command;

[0034] The execution feedback module is used to execute the emergency response plan, automatically control the associated hardware to perform safety operations, and collect on-site feedback information and execution status in real time;

[0035] The verification report module is used to verify the emergency response plan by comparing the on-site feedback information with the execution status and form a safety protection report.

[0036] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the hydrogen fuel cell ship safety protection method described in the first aspect of the present invention is implemented.

[0037] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the hydrogen fuel cell ship safety protection method described in the first aspect of the present invention is implemented.

[0038] The beneficial effects of the present invention are as follows: Through multi-modal sensors, edge processing units, and intelligent simulation technologies, the present invention realizes the optimization of the safety protection of hydrogen fuel cell ships. First, adaptive filtering is used to perform high-precision preprocessing on sensor data, improve data quality, and reduce false alarms and missed alarms. Secondly, a real-time CFD simulation environment based on a three-dimensional virtual model is constructed, and variational quantum eigen-coupling simulation and generative adversarial networks are combined to optimize safety thresholds, making the prediction of hydrogen diffusion and fire evolution more accurate. A data fusion algorithm driven by chaos theory is used to realize the comprehensive assessment of multi-dimensional risks such as cross-leakage, combustion, and structural failure, improving the accuracy and real-time performance of risk prediction. By dynamically adjusting the emergency response plan according to the risk index, the safety, response speed, and protection ability of hydrogen fuel cell ships in complex environments are effectively improved, and potential safety risks are reduced. Finally, an intelligent scheduling mechanism is adopted to dynamically adjust the emergency response plan according to the risk index, realize the active control of hydrogen supply, ventilation, and fire extinguishing systems, and optimize safety strategies in combination with on-site feedback information. Description of the Drawings

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0040] Figure 1 It is a flowchart of the hydrogen fuel cell ship safety protection method in the present invention.

[0041] Figure 2 It is a schematic diagram of the emergency response state in the present invention.

[0042] Figure 3 It is a schematic diagram of loop evolution optimization in the present invention.

[0043] Figure 4 It is a schematic diagram of the edge processing unit calculating risk prediction in the present invention. Detailed Embodiments

[0044] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification.

[0045] In the following description, many specific details are set forth to facilitate a thorough understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0046] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that are mutually exclusive with other embodiments.

[0047] Referring to Figures 1 to 4 , this embodiment provides a safety protection method for a hydrogen fuel cell ship, including the following steps:

[0048] S1. Start the multi-modal sensors in each area and embed the edge processing unit to perform self-check and calibration.

[0049] Furthermore, deploy temperature sensors, pressure sensors, hydrogen concentration sensors, smoke sensors, and flame sensors in each area (such as fuel storage tanks, power cabins, ventilation ducts, etc.) of the hydrogen fuel cell ship to achieve real-time perception of various safety parameters such as temperature, pressure, hydrogen leakage, smoke, and flame. Achieve comprehensive monitoring, ensure the comprehensive perception ability of different environmental parameters, improve the coverage and redundancy of monitoring. Provide multi-source data input for risk assessment and safety control, and avoid potential safety hazards caused by single sensor failures and false alarms.

[0050] Among them, embed the edge processing unit at the sensor end to achieve local preprocessing of data, including filtering, feature extraction, and compressed transmission. Reduce data transmission latency, reduce network bandwidth occupancy, improve real-time performance and response speed, and be able to make preliminary decisions on-site without relying entirely on cloud computing, enhancing the autonomy and reliability of data processing.

[0051] After the sensor is started, a self-check program is executed to check the connection status of each sensor, the stability of data acquisition, and the power consumption, ensuring that all devices are working properly. Benchmark signal comparison and calibration are adopted, and historical data or built-in standard sources are used to correct the drift of the sensor, reducing the influence of environmental factors (such as temperature, humidity, and electromagnetic interference) on the measurement accuracy, ensuring the long-term stability and consistency of sensor data, and avoiding the accumulation of measurement errors caused by aging or environmental impact. The credibility of the data is improved, providing high-precision data support for subsequent risk assessment and active safety control.

[0052] Preferably, by arranging multi-modal sensors in each area of the hydrogen fuel cell ship and combining with an edge processing unit for local data preprocessing, while performing self-check and calibration, the environmental state can be sensed in real time and accurately, and the accuracy, stability, and reliability of the data can be ensured. This not only improves the risk identification ability of the hydrogen fuel cell ship but also greatly improves the response speed and intelligent level of the entire safety protection, laying a solid foundation for efficient safety decision-making and emergency response.

[0053] S2. The edge processing unit preprocesses the sensor data using adaptive filtering to obtain the original data.

[0054] Furthermore, the original data collected by the sensor usually contains problems such as environmental noise, measurement errors, and signal drift. Directly using the unprocessed data may lead to misjudgment or decision-making delay. Localized data processing is performed in the edge processing unit close to the sensor, reducing the cloud computing burden and improving the response speed. Ensuring the integrity and accuracy of the data before transmission provides input for subsequent simulation modeling and risk analysis.

[0055] Among them, preprocessing means that after each edge processing unit receives the sensor data, it is input into an adaptive Kalman filter, and the filtering parameters are dynamically adjusted based on the mean and standard deviation of past data for high-precision denoising and standardization processing.

[0056] Specifically, through adaptive Kalman filtering, the filtering parameters are dynamically adjusted according to historical data, enabling the filtering process to adapt to different environmental conditions (such as sea vibrations, temperature changes, etc.). Variational fuzzy mean filtering is used to optimize the non-linear noise, further reducing the measurement error, improving the extraction accuracy of weak signals, eliminating burst interference, random noise, and drift phenomena in the data, improving the temporal and spatial consistency of the data, enhancing the stability and robustness of the sensor network, and avoiding the influence of error accumulation on the risk calculation results.

[0057] The sensor data after filtering optimization is the original data, and its error has been controlled within a reasonable range, which can be directly used to construct the three-dimensional virtual model of the ship and CFD simulation calculation. Ensure that the calculation results of subsequent hydrogen diffusion simulation, fire evolution analysis and risk assessment are more accurate, reduce misjudgment or false triggering of emergency response caused by abnormal data, provide reliable data support for intelligent safety decision-making, and improve the safety monitoring accuracy of hydrogen fuel cell ships.

[0058] By performing adaptive filtering preprocessing on the sensor data through the edge processing unit, it is possible to clean noise, compensate for errors, and improve data accuracy in real time, ensuring the stability and reliability of the sensor data. This process provides data support for subsequent risk analysis, simulation calculation, and active safety control, thereby enhancing the intelligence level, real-time response ability, and risk prediction accuracy of the safety protection of hydrogen fuel cell ships, effectively reducing potential safety hazards, and achieving more efficient safety protection.

[0059] S3. Construct three-dimensional virtual models of the ship and each area based on the original data, and combine real-time CFD simulation to simulate the hydrogen diffusion and fire evolution processes to obtain risk prediction parameters and preset safety thresholds.

[0060] The risk prediction parameters and preset safety thresholds are as follows:

[0061] Furthermore, the edge processing unit receives the original data in real time, and uses the particle swarm optimization algorithm to construct and adjust the parameters of the three-dimensional virtual model;

[0062] Map the original data into the three-dimensional virtual model in real time, accurately reproduce the physical states of each area of the ship (such as hydrogen concentration distribution, structural stress, temperature field, etc.), and provide a high-fidelity data basis for subsequent simulation and risk prediction. By integrating the three-dimensional virtual models of key areas such as fuel storage tanks and pipelines, it supports cross-module coupling analysis, such as the collaborative simulation of hydrogen leakage paths and ventilation responses, and avoids the limitations of single-area analysis. Through global data fusion, the three-dimensional virtual model can detect hidden coupling risks (such as the superposition effect of pipeline leakage and structural stress) at an early stage, provide a high-precision simulation environment for subsequent accident scenario simulations (such as hydrogen leakage and diffusion, fire spread), so as to formulate emergency strategies in advance and reduce the uncertainty in actual accidents.

[0063] Specifically, construct three-dimensional virtual models of the ship and its functional areas through the original data (temperature, pressure, hydrogen concentration, smoke, flame, etc.), including fuel storage tanks, pipelines, ventilation and power cabins, etc. Through geometric modeling, use lidar (LiDAR) to scan the ship structure to generate a resolution ≤ 1 cm 3For the point cloud model, key components such as the fuel tank and ventilation ducts are extracted through U-Net semantic segmentation. The hydrogen concentration gradient and structural stress distribution in the original data are mapped to the nodes of the three-dimensional virtual model. Through interpolation algorithms (cubic spline interpolation), high-fidelity reproduction of physical fields, temperature fields, and pressure fields is achieved. An NVIDIA Jetson AGX Orin edge processing unit is deployed, and a lightweight CFD solver (customized based on OpenFOAM) is run, with a single simulation taking ≤5 ms.

[0064] The particle swarm optimization (PSO) algorithm is used to iteratively optimize the parameters of the three-dimensional virtual model to improve the modeling accuracy, enabling it to adapt to different operating environments and load changes, achieving digital reproduction of the ship's structure, equipment layout, and safety areas, and providing accurate environmental information. The expression is as follows:

[0065]

[0066] Among them, ρ t is the material density distribution at time t (kg / m 3 ), ρ t+1 is the material density distribution at time t + 1 (kg / m 3 ), η represents the learning rate, represents the gradient operator, J is the structural stress-strain energy function, σ is the noise coefficient (set based on the standard deviation of sensor errors), is the standard normal distribution, E is the elastic modulus (GPa), v is the Poisson's ratio (dimensionless), ∈ n is the equivalent plastic strain at the nth integration node, ε is the mean principal strain, V n represents the volume element at the nth integration node, n is the integration node index variable, and N is the total number of integration nodes.

[0067] It should be noted that by minimizing the J structural stress-strain energy function through the gradient term , the material density distribution can better fit the actual structural mechanics characteristics. The noise term introduces random perturbations to prevent the three-dimensional virtual model from falling into local optima, and at the same time simulates sensor data noise to enhance the adaptability to environmental interference. Through material parameters such as the elastic modulus E and Poisson's ratio v, the stress-strain relationship of the structure under different loads is quantified to provide a mechanical behavior benchmark. Combining the equivalent plastic strain ∈ n and the mean principal strain ε, the strain distributions at the local (integration node) and global (overall structure) levels are comprehensively evaluated to support refined risk assessment. Through parameter optimization, the three-dimensional virtual model can adapt to different operating states, ensure long-term monitoring accuracy, provide a high-precision simulation basis for real-time hydrogen leakage detection, gas diffusion prediction, and heat transfer analysis, and be able to simulate different accident scenarios in advance to improve the foresight of safety management.

[0068] Preferably, the particle swarm optimization algorithm quickly locates the optimal solutions of model parameters (such as material density distribution, structural stiffness coefficient, etc.) through the swarm intelligence search strategy, avoiding model mismatch caused by manual adjustment or fixed parameters. During the operation of the ship, the particle swarm optimization algorithm dynamically adjusts the parameters according to real-time sensor data (such as load changes, temperature fluctuations) to ensure that the model is always consistent with the actual working conditions. Through online optimization, the particle swarm optimization algorithm enables the three-dimensional virtual model to maintain high accuracy during long-term monitoring, reducing the risks of false alarms and missed alarms. Automated parameter optimization reduces the dependence on expert experience, and can still maintain the effectiveness of the three-dimensional virtual model especially under complex working conditions (such as severe sea conditions, high-load navigation).

[0069] Based on the multi-physics field coupling simulation framework, the CFD control volume discretization is mapped to the quantum bit state, and variational quantum eigen coupling simulation is used to output risk prediction parameters;

[0070] Furthermore, by encoding the control volumes of the CFD grid (such as parameters like fluid velocity, pressure, temperature, etc.) into the superposition state of quantum bits, and utilizing the parallelism and entanglement characteristics of quantum states, the physical state combinations of multiple control volumes can be represented simultaneously, significantly reducing the computational complexity. CFD simulation requires a large amount of computing resources to process high-resolution grids, while the superposition state of quantum bits can exponentially compress data representation to achieve higher-precision simulation with limited quantum resources.

[0071] Specifically, when CFD deals with multi-physics field coupling (such as fluid-thermal-structure coupling), the computational amount grows exponentially with the dimension. Quantum mapping can transform complex coupling problems into the joint evolution of quantum states, reducing the computational complexity and making real-time simulation possible. With the same hardware resources, quantum bit mapping can achieve a finer resolution than classical grids. For example, it can model the tiny turbulent effects of the hydrogen leakage path, improving the spatial resolution of risk prediction (such as from meter level to centimeter level).

[0072] The variational quantum eigen (VQE) algorithm is used to transform the multi-physics field coupling simulation framework. The multi-physics field coupling equations (such as Navier-Stokes equation, heat conduction equation, structural mechanics equation) are transformed into quantum Hamiltonians, and the quantum states are iteratively optimized through VQE to solve their eigenstates (i.e., steady-state or transient solutions), thereby predicting key parameters such as hydrogen diffusion, temperature distribution, and structural stress.

[0073] VQE approximates the eigen - solution through a parameterized quantum circuit (such as Ansatz), which can run on quantum computers available in the near future (NISQ era), avoiding dependence on idealized quantum error - correction devices. Through the joint evolution of quantum superposition states, VQE can explore multiple solution spaces in parallel and achieve exponential acceleration for specific problems (such as solving high - dimensional partial differential equations). For non - linear coupling processes such as combustion and structural deformation caused by hydrogen leakage, the quantum eigen - solution can more accurately capture non - linear interactions, such as the coupling effect between the flame front and structural stress.

[0074] Preferably, a multi - physical - field coupling simulation framework is used to integrate the fluid field, thermal field, and structural field. In quantum simulation, the quantum states of the fluid field, thermal field, and structural field are real - time correlated through quantum entanglement, avoiding the timing error caused by decoupled calculations of each field. The local high temperature caused by hydrogen leakage will affect the structural stress through the coupling of quantum states, and then feedback to the change of flow resistance in the fluid field, forming a closed - loop simulation. Step - by - step simulation requires artificial assumptions of the interaction conditions between fields, while quantum coupling simulation can natively support multi - field dynamic coupling through the joint evolution of quantum states, improving the physical consistency of risk prediction. The sudden increase in cabin pressure caused by hydrogen leakage may trigger structural deformation, which in turn changes the leakage path. Quantum simulation can synchronously model and predict a more realistic accident evolution path.

[0075] Among them, the output risk prediction parameters include hydrogen concentration distribution, fire spread rate, and structural failure probability. After quantum simulation is accelerated, it can complete the classical simulation that originally took several hours within seconds, providing real - time risk parameters for the edge processing unit to support rapid decision - making. The output risk parameters can be directly input into the digital twin models in the invention (such as fuel tank and pipeline models) to dynamically adjust their safety thresholds or emergency response strategies. Improve the initiative of emergency response: For example, if it is predicted in advance that the hydrogen concentration in a certain area will exceed the explosion limit after 10 seconds, local isolation can be immediately triggered, avoiding the lag of relying on sensor threshold alarms.

[0076] Use a generative adversarial network to adaptively preset the safety threshold.

[0077] Furthermore, the generative adversarial network GAN consists of a generator and a discriminator, and optimizes the output data distribution of the generator through adversarial training. The generator takes historical risk data (such as hydrogen concentration, temperature, etc.) as input to generate safety thresholds; the discriminator guides the optimization of the generator by comparing the generated thresholds with actual accident data. Through adversarial training, the generative adversarial network makes the safety threshold dynamically adapt to the real - time data distribution. For example, during ship navigation, when the environmental temperature rises and the hydrogen diffusion rate increases, the generative adversarial network can adaptively lower the safety threshold (such as from 4% to 3.5%) to trigger an emergency response in advance. In addition, the unsupervised learning feature of GAN can directly utilize historical accident data (without manual annotation), significantly reducing the cost of manual intervention.

[0078] S4. Calculate the real-time risk index through a multi-sensor data fusion algorithm, compare it with the preset safety threshold, and output an active scheduling control command.

[0079] Furthermore, based on the multi-sensor data fusion algorithm driven by chaos theory, integrate the weights of multi-sensor data;

[0080] Calculate the weights in real time through the parameterized chaos theory-driven multi-sensor data fusion algorithm, and dynamically adjust them with the mutation of sensor data (such as noise and outliers). For example, when the data of a certain sensor fluctuates violently, the weight of the corresponding sensor data is reduced to suppress noise interference. The chaotic weight avoids periodic oscillation of the weight through sensitive dependence. Through the sensitive dependence of the Logistic map, the weight can adapt to mutant data. For example, in a fire scenario, the weights of temperature and pressure synchronously rise to 0.6 to accurately identify the combustion risk. The divergence of the chaotic weight is suppressed through a feedback mechanism to ensure stability, and the weight update and data fusion are completed within the period. At the same time, the anti-noise ability is significantly improved, which is suitable for real-time safety monitoring in complex working conditions such as ships. It improves the accuracy of risk prediction and more realizes the intelligent priority sorting of multi-sensor data through dynamic weight allocation, providing a highly reliable and adaptive solution for ship safety.

[0081] Construct a comprehensive risk index covering multi-dimensional risks of leakage, combustion, and structural failure;

[0082] Furthermore, construct a comprehensive risk index covering multi-dimensional risks of leakage, combustion, and structural failure, and the expression is as follows:

[0083]

[0084] Among them, R is the comprehensive risk index, k represents different risk dimensions, k = 1 represents the leakage risk dimension, k = 2 represents the combustion risk dimension, and k = 3 represents the structural failure risk dimension. Represents the cumulative effect from time 0 to t, λ k (t) is the risk attenuation coefficient, w k Is the dynamic weight of different dimension intensities k, Represents the risk index, dt represents the infinitesimal increment of the time variable, f k Represents the risk intensity of different dimension intensities k.

[0085] Specifically, the risk contributions of the three dimensions of leakage, combustion, and structural failure are unified and quantified through integral operations, avoiding information islands caused by independent evaluation of each dimension. For example, when a hydrogen leak causes a fire, the synergistic effect of the leakage term and the combustion term can be comprehensively reflected. When the leakage concentration (leakage term) and the structural stress (structural failure term) increase simultaneously, the weight contributions of the two can be integrated to trigger an emergency response in advance.

[0086] When k = 1, the integral reflects the cumulative exposure of hydrogen concentration from the leakage starting point to the current time t, avoiding misjudgment of instantaneous concentration fluctuations.

[0087] When k = 2, the integral Through the exponential amplifies the mutation effect of the flame intensity (such as the sudden increase in risk at the moment of spark ignition).

[0088] When k = 3, the integral accumulates the structural stress-strain energy, predicting the long-term deterioration trend (such as the corrosion of the storage tank wall thickness).

[0089] Through integral operation, the risk index is more sensitive to early minor perturbations (such as initial leakage). Compared with the threshold trigger mechanism, it can give an early warning of potential accidents 30 seconds in advance.

[0090] Compare the comprehensive risk index with the preset safety threshold and output an active scheduling control command.

[0091] Furthermore, the real-time calculated comprehensive risk index is compared with the preset safety threshold of the generative adversarial network to judge the risk level (such as normal, warning, dangerous, emergency). Automatically adjust the safety threshold under different environmental conditions to avoid false alarms or missed alarms caused by improper setting of fixed thresholds. Through the comparison of the comprehensive risk index and the preset safety threshold, a quantitative assessment of the safety status is realized, improving the intelligent level of safety protection. By setting a highly adaptable safety threshold, risks can be accurately identified in a complex and changeable environment, reducing false triggering situations, improving safety and operating efficiency, providing a scientific basis for active scheduling control, and making the triggering of safety measures more reasonable and accurate.

[0092] Among them, according to the comparison result of the comprehensive risk index and the safety threshold, combined with the preset emergency response strategy, an active scheduling control command is generated. Different risk levels trigger different control measures. For example: low risk, only conduct data monitoring and update the safety status; medium risk, trigger ventilation to reduce the hydrogen concentration; high risk, implement active safety measures such as local isolation, automatic cut-off of hydrogen supply, and activation of fire extinguishing devices; extreme risk, issue an emergency alarm and notify the ship control center for manual intervention.

[0093] Through the hierarchical response mechanism, the emergency measures can accurately match the risk level, avoid problems of overreaction or insufficient safety measures, make automatic safety decisions, reduce the dependence on manual intervention, and improve the real-time performance and accuracy of safety response. Ensure that hydrogen fuel cell ships can respond quickly in sudden safety incidents, reducing personal injuries and equipment damage. Make the safety control intelligent, automated and scalable, adaptable to different ship structures and operating conditions, and improve the universality.

[0094] Preferably, by comparing the comprehensive risk index with a preset safety threshold and outputting an active scheduling control command, a real-time safety protection mechanism with high intelligence and security is realized. Among them, the multi-sensor fusion algorithm driven by chaos theory ensures the accuracy of risk assessment, the setting of the preset safety threshold improves the adaptability of safety judgment, and active scheduling control enables timely triggering of appropriate safety measures. Finally, this step ensures that the hydrogen fuel cell ship has higher safety in a complex marine environment, reduces the risks of hydrogen leakage, fire and explosion accidents, and improves the intelligent level of the overall ship safety management.

[0095] S5. According to the active scheduling control command, trigger the emergency response plan according to the risk level classification.

[0096] Furthermore, the risk level is dynamically divided through quantitative assessment. By quantitatively comparing and evaluating the comprehensive risk index and the preset safety threshold, the current risk level is determined. The comprehensive risk index < 0.2 is low risk; 0.2 <= comprehensive risk index < 0.5 is medium risk; 0.5 <= comprehensive risk index < 0.8 is high risk; the comprehensive risk index >= 0.8 is extreme risk. The risk level is divided into four levels, namely low, medium, high and extreme, so as to match different emergency response strategies. Through quantitative risk assessment, the accuracy of emergency response is ensured, and misjudgment or delayed response caused by subjective judgment is avoided. Dynamic adjustment according to the risk level enables the safety protection measures to neither intervene excessively nor lag in response, ensuring the continuity and safety of ship operation. It is applicable to hydrogen fuel cell ships in complex marine environments, enabling the safety response mechanism to adapt to different environmental conditions and accident scenarios, improving the universality of safety control, facilitating integration into ship intelligent management, and realizing more intelligent risk management and accident response.

[0097] Trigger the emergency response plan corresponding to the corresponding risk level.

[0098] Specifically, for low risk (normal range), only monitoring and recording are carried out, continuously tracking data changes, and no additional physical measures are taken; for medium risk (potential threat), trigger the warning mode, start ventilation to dilute the hydrogen concentration, and notify the operator to conduct an inspection; for high risk (dangerous state), execute partial isolation, automatically cut off the hydrogen supply, and at the same time start fire prevention to prevent combustion and explosion accidents; for extreme risk (emergency accident), directly enter the full emergency mode, activate all safety protection measures, including audible and visual alarms, solenoid valve closing, fire extinguishing device activation, sending emergency signals to shore-based management, etc., and record the accident process data for subsequent analysis.

[0099] By triggering emergency responses in a hierarchical manner, it ensures that emergency measures match the severity of the accident, avoiding overreactions that may affect the operation of the ship. At the same time, it ensures rapid response in a real emergency to reduce accident losses. Combining with the automatic control of the ship, it ensures that the execution of the emergency plan does not rely on manual operations, improving real-time performance and reliability. It is applicable to different types of hydrogen fuel ships and can optimize the hierarchical strategy according to factors such as ship structure, fuel storage location, and operating environment. In the offshore remote operation mode, it can automatically trigger appropriate safety measures with limited crew intervention, enhancing the ship's autonomous safety protection ability.

[0100] Preferably, by triggering the emergency response plan in a hierarchical manner according to the active dispatching control command, an intelligent emergency management mechanism based on dynamic risk assessment is realized. It can accurately identify the risk level and automatically match the appropriate emergency response strategy to ensure that the hydrogen fuel cell ship will neither overreact and affect normal operation during a safety incident nor cause the accident to escalate due to a lag in response. Ultimately, this step significantly improves the safety of hydrogen fuel cell ships, the intelligent level of emergency handling, and enhances the overall adaptive ability and accident prevention ability.

[0101] S6. Execute the emergency response plan, automatically control the associated hardware to perform safety operations, and collect on-site feedback information and execution status in real time.

[0102] Furthermore, by sending control instructions, it automatically activates the associated hardware (such as solenoid valves, ventilation, fire extinguishing devices, audible and visual alarms, etc.) to perform corresponding safety operations. By automating the control of safety equipment, it avoids misoperations, delays, or human errors that may be caused by manual intervention, improving the efficiency and accuracy of emergency responses. It can quickly execute critical safety operations within a time range of milliseconds. When accidents such as hydrogen leakage or fire occur, it can quickly cut off the hydrogen source, suppress the fire, and reduce the accident risk.

[0103] Specifically, collect on-site feedback information and execution status to form a closed-loop control. Real-time collect on-site feedback information through multi-modal sensors (such as temperature sensors, pressure sensors, hydrogen concentration sensors, etc.), detect the environmental changes after the execution of emergency measures, and monitor the execution status of the associated hardware. For example: check whether the solenoid valve is fully closed, whether the fire extinguishing device is spraying the fire extinguishing agent normally, whether the ventilation reaches the expected air change rate, etc. Input the collected on-site feedback information and execution status into the edge processing unit for real-time analysis and dynamic adjustment. When necessary, re-execute the adjusted emergency strategy to ensure the effectiveness of safety operations.

[0104] Preferably, by implementing an emergency response plan, automatically controlling associated hardware to perform safety operations, and collecting on-site feedback information and execution status in real time, an intelligent, automated, and closed-loop feedback safety management system is achieved. This step ensures that when an emergency occurs on a hydrogen fuel cell ship, emergency measures can be accurately and efficiently implemented, and dynamic adjustment can be made through data feedback to ensure the effectiveness of safety measures. Ultimately, this method improves the intelligent level of ship safety management, reduces the need for manual intervention, and enhances the adaptive ability in complex environments.

[0105] S7. Compare the on-site feedback information with the execution status to verify the emergency response plan and form a safety protection report.

[0106] Furthermore, the safety protection report is used for risk closed-loop management and optimization iteration in the safety protection method of hydrogen fuel cell ships. The content includes the comparison of simulation and measured data, the execution record of the emergency response, the safety performance evaluation, improvement suggestions and optimization plans, and the digital twin verification results.

[0107] Specifically, real-time feedback information of various areas of the ship is obtained through multi-modal sensors (temperature, pressure, hydrogen concentration, smoke, flame, etc.), including environmental status, equipment operation status, and changes in safety parameters. The execution status is collected, including the closed state of solenoid valves, the working condition of fire extinguishing devices, the ventilation efficiency, etc., to judge whether the emergency response measures are implemented as expected. Data analysis methods, such as time series analysis, anomaly detection, and adaptive threshold evaluation, are used to cross-compare the execution status data with the on-site feedback information to verify the actual effect of the emergency response strategy. Combining the three-dimensional virtual model with the measured feedback information, analyzing the data change trends before, during, and after the accident, evaluating the execution effect of the safety strategy, recording the execution of the emergency response, including response time, execution status, fault records, and corrective measures, etc., to form a complete safety disposal log. Based on the data analysis results, safety optimization suggestions are put forward, such as adjusting the emergency response logic, optimizing control parameters, upgrading hardware equipment, or enhancing detection accuracy, etc. Through digital twin technology, the prediction of the comprehensive risk index is trained using historical data to improve the accuracy of future emergency response plans and achieve continuous optimization.

[0108] Preferably, by comparing the on-site feedback information with the execution status to verify the emergency response plan and form a safety protection report, a closed-loop safety management system based on data feedback is achieved. This step ensures that the emergency response plan of the hydrogen fuel cell ship can not only be accurately implemented but also be evaluated and optimized after each execution, thus enhancing the overall safety. Ultimately, this method enhances the scientific nature of the ship safety strategy, improves the traceability of accident handling, and promotes data-based safety optimization iteration, providing an important guarantee for the safe operation of hydrogen fuel cell ships.

[0109] This embodiment also provides a safety protection system for a hydrogen fuel cell ship, including: a startup calibration module, a data processing module, a risk prediction module, a fusion scheduling module, an emergency response module, an execution feedback module, and a verification report module; the startup calibration module is used to start multi-modal sensors in each area and embed an edge processing unit, and perform self-check and calibration; the data processing module is used for the edge processing unit to preprocess the sensor data by using adaptive filtering to obtain the original data; the risk prediction module is used to construct three-dimensional virtual models of the ship and each area based on the original data, and combine real-time CFD simulation to simulate the hydrogen diffusion and fire evolution processes to obtain risk prediction parameters and preset safety thresholds; the fusion scheduling module is used to calculate the real-time risk index through a multi-sensor data fusion algorithm, compare it with the preset safety threshold, and output an active scheduling control command; the emergency response module is used to trigger an emergency response plan according to the risk level in accordance with the active scheduling control command; the execution feedback module is used to execute the emergency response plan, automatically control the associated hardware to perform safety operations, and collect on-site feedback information and execution status in real time; the verification report module is used to compare the on-site feedback information with the execution status to verify the emergency response plan and form a safety protection report.

[0110] This embodiment also provides a computer device, which is applicable to the case of the safety protection method for a hydrogen fuel cell ship, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the safety protection method for a hydrogen fuel cell ship proposed in the above embodiment.

[0111] This computer device can be a terminal, and this computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication) or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covered on the display screen, or a button, a trackball or a touchpad set on the shell of the computer device, or an external keyboard, a touchpad or a mouse, etc.

[0112] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for realizing the safety protection of a hydrogen fuel cell ship as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM for short), electrically erasable programmable read-only memory (EEPROM for short), erasable programmable read-only memory (EPROM for short), programmable read-only memory (PROM for short), read-only memory (ROM for short), magnetic memory, flash memory, magnetic disk or optical disc.

[0113] In summary, through multi-modal sensors, an edge processing unit, and intelligent simulation technology, the present invention realizes the optimization of the safety protection of hydrogen fuel cell ships. First, adaptive filtering is used to perform high-precision preprocessing on sensor data, improve data quality, and reduce false alarms and missed alarms. Secondly, a real-time CFD simulation environment based on a three-dimensional virtual model is constructed, and variational quantum eigen-coupling simulation and generative adversarial networks are combined to optimize the safety threshold, making the prediction of hydrogen diffusion and fire evolution more accurate. A data fusion algorithm driven by chaos theory is used to realize the comprehensive assessment of multi-dimensional risks such as cross-leakage, combustion, and structural failure, improving the accuracy and real-time performance of risk prediction. By dynamically adjusting the emergency response plan according to the risk index, the safety, response speed, and protection ability of hydrogen fuel cell ships in complex environments are effectively improved, and potential safety risks are reduced. Finally, an intelligent scheduling mechanism is adopted to dynamically adjust the emergency response plan according to the risk index, realize the active control of hydrogen supply, ventilation, and fire extinguishing systems, and optimize the safety strategy in combination with on-site feedback information.

[0114] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A safety protection method for a hydrogen fuel cell ship, characterized in that: including Start the multimodal sensors in each area and embed the edge processing unit to perform self-check and calibration; The edge processing unit preprocesses the sensor data using adaptive filtering to obtain the raw data; Based on the raw data, construct 3D virtual models of the ship and each area, and combine with real-time CFD simulation to simulate the hydrogen diffusion and fire evolution processes to obtain risk prediction parameters and preset safety thresholds; Calculate the real-time risk index through a multi-sensor data fusion algorithm, compare it with the preset safety threshold, and output an active scheduling control command; According to the active scheduling control command, trigger the emergency response plan according to the risk level classification; Execute the emergency response plan, automatically control the associated hardware to perform safety operations, and collect on-site feedback information and execution status in real time; Compare the on-site feedback information with the execution status to verify the emergency response plan and generate a safety protection report; The steps for obtaining the risk prediction parameters and preset safety thresholds are as follows Receive the raw data in real time through the edge processing unit, and use the particle swarm optimization algorithm to construct and adjust the 3D virtual model parameters; Based on the multi-physics field coupling simulation framework, discretize the CFD control volume and map it to the qubit state, and use variational quantum eigen-coupling simulation to output the risk prediction parameters; Use the generative adversarial network to adaptively preset the safety threshold.

2. The safety protection method for a hydrogen fuel cell ship according to claim 1, wherein: The steps for calculating the real-time risk index through the multi-sensor data fusion algorithm are as follows Based on the multi-sensor data fusion algorithm driven by chaos theory, integrate the weights of multi-sensor data; Construct a comprehensive risk index for multi-dimensional risks across leakage, combustion, and structural failure; Compare the comprehensive risk index with the preset safety threshold and output an active scheduling control command.

3. The safety protection method for a hydrogen fuel cell ship according to claim 1, wherein: The preprocessing means that after each edge processing unit receives the sensor data, it is input into an adaptive Kalman filter, and the filtering parameters are dynamically adjusted based on the mean and standard deviation of historical data for high-precision denoising and standardization processing.

4. The hydrogen fuel cell ship safety protection method according to claim 1, wherein: The steps for triggering the emergency response plan according to the risk level classification are as follows Dynamically divide the risk level into low, medium, high, and extreme through quantitative evaluation; Trigger the emergency response plans for the corresponding risk levels in a hierarchical manner.

5. The safety protection method for a hydrogen fuel cell ship according to claim 1, characterized in that: The emergency response plan refers to local isolation, automatic hydrogen supply cut-off, and fire extinguishing; the associated hardware includes solenoid valves, ventilation, fire extinguishing devices, and audible and visual alarms.

6. The safety protection method for a hydrogen fuel cell ship according to claim 1, characterized in that: The safety protection report is used for risk closed-loop management and optimization iteration in the safety protection method for hydrogen fuel cell ships, and the content includes the comparison of simulation and measured data, the execution records of emergency responses, safety performance evaluations, improvement suggestions and optimization plans, and digital twin verification results.

7. A safety protection system for a hydrogen fuel cell ship, based on the safety protection method for a hydrogen fuel cell ship according to any one of claims 1 to 6, characterized in that: including a start calibration module, a data processing module, a risk prediction module, a fusion scheduling module, an emergency response module, an execution feedback module, and a verification report module; The start calibration module is used to start the multimodal sensors in each area and embed the edge processing unit to perform self-check and calibration; The data processing module is used for the edge processing unit to preprocess the sensor data using adaptive filtering to obtain the raw data; The risk prediction module is used to construct 3D virtual models of the ship and each area based on the original data, and combine real-time CFD simulation to simulate the hydrogen diffusion and fire evolution process, so as to obtain risk prediction parameters and preset safety thresholds; The fusion scheduling module is used to calculate the real-time risk index through a multi-sensor data fusion algorithm, compare it with the preset safety threshold, and output an active scheduling control command; The emergency response module is used to trigger the emergency response plan according to the risk level based on the active scheduling control command; The execution feedback module is used to execute the emergency response plan, automatically control the associated hardware to perform safety operations, and collect on-site feedback information and execution status in real time; The verification report module is used to verify the emergency response plan by comparing the on-site feedback information with the execution status, and form a safety protection report.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the hydrogen fuel cell ship safety protection method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the hydrogen fuel cell ship safety protection method according to any one of claims 1 to 6.

Images