Hydrogen concentration sensor site selection optimization method and equipment for fuel cell vehicle

Abstract

The invention discloses a hydrogen concentration sensor site selection optimization method and equipment for a fuel cell vehicle. The vehicle simulation model is established according to the structure of the target fuel cell vehicle, and the plurality of hydrogen concentration monitoring points are set to construct the database, so that the distribution condition of the hydrogen concentration in the vehicle can be comprehensively and accurately known, and the accuracy of fuel cell hydrogen leakage detection is improved. The particle swarm optimization algorithm is adopted to optimize initial arrangement, and the optimal scheme of sensor arrangement can be realized, sensor redundancy can be reduced, arrangement cost can be reduced, hydrogen leakage detection time can be shortened, and detection efficiency can be improved in combination with an objective function and constraint conditions on the premise of meeting detection requirements. According to the method, the dependence on traditional experience site selection is reduced, the problems of inaccurate detection, cost waste and the like caused by insufficient experience or misjudgment are avoided, the stability and the reliability of the hydrogen leakage detection system of the fuel cell vehicle are enhanced, and the safety of the fuel cell in the vehicle is effectively guaranteed.

Description

Technical Field

[0001] This invention relates to the field of fuel cell sensor site selection, and in particular to a method and apparatus for optimizing the site selection of hydrogen concentration sensors for fuel cell vehicles. Background Technology

[0002] Hydrogen, as a clean energy source, has been widely used in fuel cell vehicles. Fuel cell vehicles use hydrogen as fuel and generate electricity through an electrochemical reaction to power the vehicle. They offer advantages such as zero emissions and high efficiency, and are considered one of the important directions for future automotive development. However, the flammable and explosive nature of hydrogen makes hydrogen leakage a potentially significant safety hazard during the operation of fuel cell vehicles. If a hydrogen leak is not detected promptly and accurately, it can easily lead to serious accidents such as fires and explosions, threatening the lives of passengers and the normal operation of the vehicle. Therefore, accurately detecting hydrogen leaks and determining their location is crucial to ensuring the safety of fuel cell vehicles.

[0003] Traditional hydrogen concentration sensor placement in fuel cell vehicles typically relies on experience. Engineers, based on past engineering experience and a general understanding of the vehicle's structure, select specific locations where hydrogen leaks are suspected. For example, sensors are often placed near the hydrogen storage tank or at the connection points of hydrogen supply pipelines, in the hope of detecting leaks.

[0004] Experience-based sensor placement has several drawbacks. First, it easily leads to blind spots. Since hydrogen diffusion within a vehicle is influenced by various factors, such as the vehicle's internal structure and ventilation conditions, relying solely on experience-based site selection cannot comprehensively cover all possible leak areas, resulting in some hydrogen leaks going undetected. Second, it can lead to sensor redundancy. In areas that are not considered high-risk, too many sensors may be installed due to over-reliance on experience, increasing both manufacturing and maintenance costs. Furthermore, experience-based site selection does not fully consider the actual conditions of hydrogen diffusion and various complex operating scenarios, making it difficult to guarantee rapid and accurate detection of hydrogen leaks under all circumstances. This fails to effectively meet the accuracy and timeliness requirements of fuel cell vehicles for hydrogen leak detection, reducing vehicle safety and reliability. Summary of the Invention

[0005] To address the aforementioned problems, this invention proposes an optimized method for the location of hydrogen concentration sensors in fuel cell vehicles. This method effectively avoids detection blind spots, reduces sensor redundancy, lowers deployment costs, improves the accuracy and timeliness of hydrogen leak detection, and enhances the safety and reliability of fuel cell vehicles.

[0006] In a first aspect, embodiments of this disclosure provide a method for optimizing the location of a hydrogen concentration sensor in a fuel cell vehicle, the method comprising:

[0007] A vehicle simulation model is established based on the structure of the target fuel cell vehicle, and multiple hydrogen concentration monitoring points are set in the vehicle simulation model. The hydrogen concentration values ​​are detected based on each hydrogen concentration monitoring point to build a database.

[0008] The database was analyzed to determine the initial placement of hydrogen concentration sensors in the target fuel cell vehicle.

[0009] An objective function and constraints are defined, and a particle swarm optimization algorithm is used to optimize the initial layout based on the objective function and constraints to generate an optimized layout. The objective function includes maximizing scene coverage, minimizing detection time, and minimizing sensor deployment cost. The constraints include scene detection constraints, budget constraints, and detection time constraints.

[0010] Optionally, a vehicle simulation model is established based on the structure of the target fuel cell vehicle, including: analyzing key components based on the structure of the target fuel cell vehicle to identify key components, including the fuel cell stack, hydrogen storage device, and transmission device; establishing a simplified geometric model of the vehicle based on the key components using the Fluent platform, wherein the simplified geometric model is used for hydrogen leakage diffusion simulation; determining the mass, momentum, and energy conservation equations and component transport equations as the basic governing equations of the simplified geometric model for the hydrogen leakage diffusion process, and assigning values ​​to the parameters in the basic governing equations; defining key regions and secondary regions in the simplified geometric model of the vehicle, using a first specified density for meshing in the key regions and a second specified density for meshing in the secondary regions, wherein the key regions include hydrogen leak vents and vents, and the first specified density is greater than the second specified density; defining boundary conditions, including leak vents, wall boundaries, and environmental conditions; setting scene parameters, including ambient temperature, exhaust intensity, and external flow field; and iteratively calculating the simplified geometric model of the vehicle based on the boundary conditions and scene parameters to generate a vehicle simulation model.

[0011] Optionally, a database can be constructed based on hydrogen concentration values ​​obtained from each hydrogen concentration monitoring point. This includes: determining the detection time and monitoring point location corresponding to each hydrogen concentration value to establish data records for each monitoring point location; creating a blockchain data structure, and sequentially encrypting each data record with a public key, packaging it into blocks, and adding it to the data structure to form a database in the form of a blockchain.

[0012] Optionally, the database is analyzed to determine the initial arrangement of hydrogen concentration sensors in the target fuel cell vehicle, including: decrypting the database using a private key to obtain the correspondence between monitoring point locations, detection times, and hydrogen concentration values; treating each monitoring point location as a network node and defining the edges between network nodes; determining the weights of the edges between network nodes based on the hydrogen concentration values ​​and detection times to generate a network topology graph; calculating the degree centrality and betweenness centrality of each network node in the network topology graph, and determining key nodes based on the degree centrality and betweenness centrality; and using the monitoring point locations corresponding to each key node as the initial arrangement of hydrogen concentration sensors in the target fuel cell vehicle.

[0013] Optionally, based on the objective function and constraints, a particle swarm optimization algorithm is used to optimize the initial layout to generate an optimized layout. This includes: encoding the layout positions in the initial layout as particle position vectors and initializing the population based on each position vector; generating a penalty function based on the constraints and the objective function, and combining the penalty function and the objective function to generate a fitness function; and iteratively calculating each particle in the population based on the fitness function. In each iteration, individual optimal updates and global optimal updates are performed on each particle in the population. When the maximum number of iterations or the convergence condition is reached, the global optimal position is output as the optimized layout.

[0014] Optionally, maximize scene coverage as follows:

[0015]

[0016] Where f1(X) is the scene coverage objective function, A is the set of all leaked scenes, and λ a It represents the probability of a leak occurring, x. a This means that if scene a is detected, x a =1, otherwise x a =0;

[0017] Minimize the detection time as follows:

[0018]

[0019] Where f2(X) is the detection time objective function, d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. a λ refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios. a It represents the probability of a leakage scenario occurring;

[0020] Minimize sensor deployment cost as follows:

[0021]

[0022] Where f3(X) is the objective function for total sensor cost, L is the set of all candidate sensor placement locations, and c i It is the cost of sensor i, s i The value is 1 if a sensor is installed at position i, and 0 otherwise.

[0023] Optionally, the penalty function includes penalties for scene not being covered, cost exceeding limits, and detection time exceeding limits;

[0024] The penalty for scenes not being covered is:

[0025]

[0026] Among them, P coverage For scene coverage penalty terms, α is the weighted penalty coefficient corresponding to the scene coverage penalty term, and s i The value is 1 if a sensor is installed at position i, and 0 otherwise. a This refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios;

[0027] The penalty for exceeding cost limits is as follows:

[0028]

[0029] Among them, P cost Here, β is the penalty term for exceeding cost limits, and s is the weighted penalty coefficient corresponding to the penalty term for exceeding cost limits. i The value is 1 if a sensor is installed at position i, and 0 otherwise. L is the set of all candidate sensor placement positions. i Where i is the cost of sensor i, and p is the budget;

[0030] Penalties for exceeding the detection time limit:

[0031]

[0032] Among them, P time For the penalty term for exceeding the detection time limit, γ is the weighted penalty coefficient corresponding to the penalty term for exceeding the detection time limit, and d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. α T refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios, and T max This refers to the maximum detection time allowed by the system;

[0033] The fitness function is:

[0034] F(X) = f1(X) - f2(X) - f3(X) + P coverage +P cost +P time ;

[0035] Where F(X) is the fitness function, f1(X) is the scene coverage objective function, f2(X) is the detection time objective function, f3(X) is the total sensor cost objective function, and P... coverage For scenarios not covered by penalties, P cost As a penalty for exceeding cost limits, P time This is a penalty for exceeding the detection time limit.

[0036] Optionally, the method further includes: simulating the vehicle simulation model based on the optimized layout, and determining the degree of influence of each layout position in the optimized layout during the simulation process, wherein the degree of influence includes high, medium and low; designating the layout positions with high degree of influence as key positions, and arranging redundant sensors in a parallel redundancy manner for the key positions.

[0037] Secondly, embodiments of this disclosure also provide an electronic device, the electronic device comprising:

[0038] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a hydrogen concentration sensor location optimization method for a fuel cell vehicle as described in any embodiment of this disclosure.

[0039] Thirdly, embodiments of this disclosure provide a computer storage medium storing a computer program that, when executed by a processor, implements a hydrogen concentration sensor location optimization method for a fuel cell vehicle as described in any embodiment of this disclosure.

[0040] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description.

[0041] Therefore, the present invention has the following beneficial effects:

[0042] 1. By establishing a vehicle simulation model based on the structure of the target fuel cell vehicle and setting up a database with multiple hydrogen concentration monitoring points, the distribution of hydrogen concentration inside the vehicle can be understood comprehensively and accurately, thereby improving the accuracy of hydrogen leakage detection in fuel cells.

[0043] 2. The initial layout is optimized using a particle swarm optimization algorithm. By combining objective functions such as maximizing scene coverage, minimizing detection time, and minimizing sensor deployment cost, along with corresponding constraints, the optimal sensor deployment scheme can be achieved while meeting detection requirements. This reduces sensor redundancy, lowers deployment costs, and shortens hydrogen leak detection time, thereby improving detection efficiency.

[0044] 3. It reduces reliance on traditional experience in site selection, avoids problems such as inaccurate detection and cost waste caused by insufficient experience or misjudgment, enhances the stability and reliability of the hydrogen leak detection system for fuel cell vehicles, and effectively ensures the safety of fuel cells in vehicles. Attached Figure Description

[0045] Figure 1 This is a flowchart of a hydrogen concentration sensor location optimization method for a fuel cell vehicle according to Embodiment 1 of the present invention;

[0046] Figure 2 This is a flowchart of another method for optimizing the location of hydrogen concentration sensors in fuel cell vehicles according to Embodiment 2 of the present invention;

[0047] Figure 3 This is a schematic diagram of a simplified geometric model of a vehicle according to Embodiment 2 of the present invention;

[0048] Figure 4 This is a flowchart of another method for optimizing the location of hydrogen concentration sensors in fuel cell vehicles according to Embodiment 3 of the present invention;

[0049] Figure 5 This is a flowchart of another method for optimizing the location of hydrogen concentration sensors in fuel cell vehicles according to Embodiment 4 of the present invention;

[0050] Figure 6 This is a schematic diagram of a hydrogen concentration sensor location optimization device for a fuel cell vehicle according to Embodiment 5 of the present invention.

[0051] Figure 7 This is a schematic diagram of the structure of an electronic device provided according to Embodiment Six of the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0053] In the description of this application, it should be noted that the terms "inner" and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product is in use. They are used only for the convenience of describing this application and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0054] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "setup" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0055] The present invention will now be described in detail with reference to the accompanying drawings.

[0056] Example 1

[0057] Figure 1 This document provides a flowchart of a hydrogen concentration sensor location optimization method for fuel cell vehicles according to Embodiment 1 of the present invention. This embodiment is applicable to hydrogen leakage detection scenarios around fuel cell stacks. The method can be executed by the hydrogen concentration sensor location optimization device for fuel cell vehicles provided in this disclosure. This device can be implemented in software and / or hardware and is generally integrated into a computer device. The method of this disclosure specifically includes:

[0058] S110: Establish a vehicle simulation model based on the structure of the target fuel cell vehicle, and set up multiple hydrogen concentration monitoring points in the vehicle simulation model. Detect hydrogen concentration values ​​based on each hydrogen concentration monitoring point to build a database.

[0059] Fuel cell vehicles are vehicles powered by fuel cells. Fuel cells directly convert the chemical energy generated by the chemical reaction between hydrogen and an oxidant into electrical energy, thus driving the vehicle. In fuel cell vehicles, hydrogen is a flammable and explosive gas, making accurate monitoring of its concentration crucial for safe operation. When the hydrogen concentration exceeds a safety threshold, sensors can issue an alarm signal to trigger appropriate safety measures. A vehicle simulation model is a virtual model built using specialized modeling software, such as Spaceclaim, based on the actual structure of the target fuel cell vehicle. The simulation model simplifies the main structure and components to reduce geometric complexity while retaining key features. In subsequent simulation analysis, various operating conditions and environmental conditions can be simulated within the vehicle simulation model to obtain relevant data. Hydrogen concentration monitoring points are locations within the vehicle simulation model where virtual hydrogen concentration detection devices are installed to obtain real-time hydrogen concentration values ​​at these locations. The placement of monitoring points is typically planned according to the vehicle's structural characteristics and possible hydrogen leakage paths to ensure comprehensive and accurate monitoring of hydrogen concentration changes. A database refers to a dataset formed by collecting, organizing, and storing relevant data such as hydrogen concentration values ​​detected by various hydrogen concentration monitoring points under different operating conditions.

[0060] Specifically, the controller can be modeled as a fuel cell vehicle, with a simplified model built in Spaceclaim. Here, the controller refers to the computer controller responsible for sensor location. During model building, key structural details are preserved while simplifying the vehicle body and components to reduce geometric complexity. In Fluent, a leakage scenario is set up, with pipe valve failure defined as a mass flow inlet. The wall is set as a Wall condition, and parameters such as environmental pressure, temperature, and gravitational acceleration are set. Appropriate gas and fluid models are selected. Simultaneously, the mesh is rationally divided, with increased density in critical areas and decreased density in secondary areas to ensure simulation accuracy and computational efficiency.

[0061] Furthermore, setting up multiple hydrogen concentration monitoring points in the vehicle simulation model involves arranging 5000 monitoring points in the fluid domain according to the height of the vehicle compartment, with sparser points at the bottom, medium density in the middle, and high density at the top, to comprehensively capture the distribution and variation patterns of hydrogen concentration. After completing multi-scenario simulations, the monitoring point data can be exported to construct a database containing information such as sensor numbers and corresponding detection times, providing basic data for subsequent optimization.

[0062] For example, the monitoring points can be arranged as follows: Divide the carriage into four layers according to its height. At the bottom layer, set up relatively sparse monitoring points, accounting for 20% of the total monitoring points, approximately 1000 points, evenly distributed to form a sparse array. In the layer above the bottom layer, gradually increase the density of monitoring points, arranging 30% of the monitoring points, approximately 1500 points, forming a medium-density array to capture the concentration distribution in the central area. Set up the densest monitoring points on the remaining two layers, approximately 2500 points, forming a high-density array to capture the concentration distribution in the upper area, for a total of 5000 points, evenly distributed throughout the carriage to record the hydrogen concentration distribution during the simulation. After the multi-scenario simulation is completed, export these 5000 points to provide timely and accurate basic data for subsequent intelligent algorithm optimization.

[0063] S120: Analyze the database to determine the initial placement of hydrogen concentration sensors in the target fuel cell vehicle.

[0064] Specifically, the controller can analyze the constructed database, extract the detection time and response of the monitoring point locations in each scenario, and identify locations with wide coverage and fast response time. This information can then be used as a guide to determine the initial layout scheme of the hydrogen concentration sensors in the target fuel cell vehicle.

[0065] S130: Set the objective function and constraints. Based on the objective function and constraints, use the particle swarm optimization algorithm to optimize the initial layout to generate an optimized layout. The objective function includes maximizing scene coverage, minimizing detection time, and minimizing sensor deployment cost. The constraints include scene detection constraints, budget constraints, and detection time constraints.

[0066] Among them, Particle Swarm Optimization (PSO) is a swarm intelligence-based optimization algorithm that finds the optimal solution by simulating the behavior of flocks of birds or schools of fish. In this method, each possible sensor placement scheme is treated as a particle, which continuously moves and updates its position in the search space. The merits of each particle are evaluated based on the objective function and constraints, ultimately finding the optimal sensor placement scheme. Maximizing scene coverage means ensuring that the proportion of hydrogen leak scenarios that the sensors can detect is as high as possible, guaranteeing timely detection of hydrogen leaks under various possible leakage conditions. Minimizing detection time means minimizing the time from the occurrence of a hydrogen leak to its detection and alarm issuance, enabling rapid response and reducing safety risks. Minimizing sensor placement cost means reducing the expenses required to install sensors, including the cost of the sensors themselves and installation costs, while meeting detection requirements.

[0067] Optionally, maximize scene coverage as follows:

[0068]

[0069] Where f1(X) is the scene coverage objective function, A is the set of all leaked scenes, and λ a It represents the probability of a leak occurring, x. α This means that if scene a is detected, x a =1, otherwise x a =0;

[0070] Minimize the detection time as follows:

[0071]

[0072] Where f2(X) is the detection time objective function, d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. α λ refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios. a It represents the probability of a leakage scenario occurring;

[0073] Minimize sensor deployment cost as follows:

[0074]

[0075] Where f3(X) is the objective function for total sensor cost, L is the set of all candidate sensor placement locations, and c i It is the cost of sensor i, s i The value is 1 if a sensor is installed at position i, and 0 otherwise.

[0076] It is important to note that in multi-objective optimization problems, setting the objective function is only the first step; setting the constraints is equally crucial. Constraints limit the range of feasible solutions, ensuring that no infeasible or unreasonable solutions are generated during the optimization process. Typically, constraints can include equality constraints, such as physical laws and mass conservation; inequality constraints, such as resource limitations and cost budgets; and boundary constraints, such as upper and lower bounds of variables. Appropriate constraint setting not only improves solution efficiency but also ensures that the optimization results meet practical application requirements. Therefore, after setting the objective function, it is essential to define constraints appropriately, considering the problem background, practical needs, and computational feasibility, to obtain the optimal solution that satisfies all requirements.

[0077] Constraints refer to the restrictions imposed on the optimization process to ensure that the optimization results meet actual needs and feasibility. Constraints include scenario detection constraints, budget constraints, and detection time constraints. Scenario detection constraints require that at least one sensor can detect each possible hydrogen leak scenario to ensure comprehensive detection. Budget constraints stipulate that the total cost of sensor deployment cannot exceed a pre-set budget to avoid excessive costs. Detection time constraints set the maximum time limit for sensors to detect hydrogen leaks, ensuring that leaks can be detected promptly within the specified time.

[0078] Specifically, the following constraints can be set:

[0079] Scene detection constraints:

[0080]

[0081] Among them, L a The set of sensor locations detected in scenario a, where A is the set of all leak scenarios, x a This means that if scene a is detected, x a =1, otherwise x a =0,s i The value is 1 if a sensor is installed at position i, and 0 otherwise. This constraint means that for each leakage scenario a, the value is 0 only if at least one sensor can detect the leak. a The talent value is 1.

[0082] Budget constraints:

[0083]

[0084] Where L is the set of all candidate sensor placement locations, c i It is the cost of sensor i, s i The value is 1 if a sensor is installed at position i, and 0 otherwise. p is the budget, which means that the total cost of the sensor cannot exceed the budget p.

[0085] Detection time upper limit constraint:

[0086]

[0087] Where, d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. a T refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios, and T max This refers to the maximum detection time allowed by the system.

[0088] Optionally, the method further includes: simulating the vehicle simulation model based on the optimized layout, and determining the degree of influence of each layout position in the optimized layout during the simulation process, wherein the degree of influence includes high, medium and low; designating the layout positions with high degree of influence as key positions, and arranging redundant sensors in a parallel redundancy manner for the key positions.

[0089] Specifically, the controller will re-simulate the vehicle simulation model based on the optimized layout scheme. The purpose of this second simulation is to deeply analyze the actual role and influence of each sensor placement location in the hydrogen leak detection process. During the simulation, the controller can determine the degree of influence of each placement location. For example, the controller can monitor the response speed of sensors at different locations to changes in hydrogen concentration under various leak scenarios. Locations with fast response speeds and timely and accurate reflection of hydrogen leak conditions are likely to have a higher degree of influence. The controller can also analyze the contribution of sensor locations to the overall scene coverage. If the absence of a sensor at a certain location significantly reduces scene coverage, then that location has a high degree of influence. Finally, the controller will classify the degree of influence of each placement location into three levels: high, medium, and low.

[0090] Furthermore, the controller identifies high-impact locations as critical locations. Sensor malfunctions or detection errors at these critical locations could lead to serious consequences, such as the inability to detect hydrogen leaks in a timely manner, potentially causing safety accidents. Therefore, for critical locations, redundant sensors can be deployed in parallel. Parallel redundancy refers to installing multiple additional sensors at the same critical location. These sensors operate independently but are all used to detect the hydrogen concentration at that location. When one sensor fails, the other parallel sensors continue to function normally, monitoring the hydrogen concentration and ensuring reliable hydrogen leak detection capabilities at critical locations. For example, at a critical location near a hydrogen storage device, where initially only one sensor might be deployed, two or more identical sensors could be installed in parallel to address its importance. This way, even if one sensor fails, the others can still detect hydrogen leaks promptly, ensuring continuous and accurate detection and improving the reliability and safety of the hydrogen leak detection system.

[0091] The technical solution of this invention establishes a vehicle simulation model based on the structure of the target fuel cell vehicle and constructs a database by setting multiple hydrogen concentration monitoring points. This allows for a comprehensive and accurate understanding of the hydrogen concentration distribution within the vehicle, improving the accuracy of fuel cell hydrogen leak detection. The initial layout is optimized using a particle swarm optimization algorithm. Combined with objective functions such as maximizing scene coverage, minimizing detection time, and minimizing sensor deployment costs, along with corresponding constraints, the optimal sensor deployment scheme can be achieved while meeting detection requirements. This reduces sensor redundancy, lowers deployment costs, and shortens hydrogen leak detection time, improving detection efficiency. It reduces reliance on traditional experience-based site selection, avoiding inaccurate detection and cost waste caused by insufficient experience or misjudgment. This enhances the stability and reliability of the fuel cell vehicle hydrogen leak detection system, effectively ensuring the safety of the fuel cell in the vehicle.

[0092] Example 2

[0093] Figure 2 This is a flowchart illustrating a hydrogen concentration sensor location optimization method for a fuel cell vehicle according to Embodiment 2 of the present invention. Based on Embodiment 1, this embodiment establishes a vehicle simulation model according to the structure of the target fuel cell vehicle and sets multiple hydrogen concentration monitoring points within the simulation model. Hydrogen concentration values ​​are detected based on each monitoring point to construct a database for further explanation. Figure 2 As shown, the method includes:

[0094] S210: Conduct a critical component analysis based on the structure of the target fuel cell vehicle to identify the critical components, including the fuel cell stack, hydrogen storage device, and transmission device.

[0095] Key components include the fuel cell stack, hydrogen storage device, and transmission device. The fuel cell stack is the core component that converts the chemical energy of hydrogen and oxygen into electrical energy. The hydrogen storage device stores high-pressure hydrogen, and the transmission device is responsible for transporting hydrogen from the storage device to the fuel cell stack. Identifying these key components allows for accurate determination of the critical factors affecting hydrogen leakage and diffusion during subsequent modeling.

[0096] S220: A simplified geometric model of the vehicle is built based on key components using the Fluent platform. This simplified geometric model is used for the diffusion simulation of hydrogen leakage.

[0097] Specifically, using the Fluent platform, a simplified geometric model of the vehicle can be constructed for simulating hydrogen leakage and diffusion based on the actual structure and dimensions of key components. During the modeling process, the vehicle's geometry is simplified, focusing on retaining key components and areas closely related to hydrogen leakage and diffusion, while ignoring details with minimal impact on the simulation results. This reduces model complexity and improves computational efficiency. For example, Figure 3 The present invention provides a structural schematic diagram of a simplified geometric model of a vehicle, wherein 1 is a sunroof, 2 is a hydrogen storage tank, 3 is a top exhaust port, 4 is a ventilation opening, 5 is a window, and 6 is a bottom exhaust port.

[0098] S230: For the hydrogen leakage and diffusion process, the mass, momentum, and energy conservation equations and the component transport equations are determined as the basic control equations for the simplified geometric model of the vehicle, and values ​​are assigned to the parameters in the basic control equations.

[0099] The mass conservation equation ensures that the mass of hydrogen does not appear or disappear during the hydrogen leakage and diffusion process. The momentum conservation equation describes the motion and forces acting on hydrogen in the flow field. The energy conservation equation considers the energy changes of hydrogen during diffusion, such as heat transfer. The component transport equation describes the concentration distribution and diffusion process of hydrogen in the air.

[0100] Specifically, the controller can assign specific values ​​to each parameter in the basic control equation based on the actual physical characteristics and operating conditions. For example, physical parameters such as the density, specific heat capacity, and viscosity of hydrogen, as well as operating parameters such as ambient temperature and pressure.

[0101] S240: Defines the critical and secondary regions in the simplified geometric model of the vehicle. The critical regions are meshed using a first specified density, and the secondary regions are meshed using a second specified density.

[0102] The critical areas include hydrogen leak points and ventilation openings, where the first designated density is greater than the second designated density. Hydrogen leak points are the source of hydrogen entering the flow field, while ventilation openings affect the diffusion and exhaust of hydrogen inside the vehicle. Secondary areas refer to areas other than the critical areas, and their impact on hydrogen leakage and diffusion is relatively small.

[0103] Specifically, using a higher mesh density in critical areas allows for more accurate capture of hydrogen flow details and concentration changes near leaks and vents, improving simulation accuracy. A lower mesh density is used in less important areas to reduce computational load and improve efficiency. Simultaneously, computational resources are used in a balanced manner while maintaining simulation accuracy.

[0104] S250: Define boundary conditions, which include leak outlets, wall boundaries, and environmental conditions.

[0105] Specifically, the controller can define the location, shape, and size of the hydrogen leak, and set the initial conditions for the leak, such as leakage rate and leakage pressure. The leak boundary conditions affect the amount of hydrogen leaked and the initial diffusion state. The controller can define the wall boundary conditions inside and outside the vehicle, such as the smoothness of the walls and the presence of adsorption or chemical reactions. Wall boundary conditions affect the flow and diffusion of hydrogen near the walls. Environmental conditions include parameters such as ambient temperature, pressure, and airflow velocity. Environmental conditions have a significant impact on the hydrogen diffusion process; for example, different ambient temperatures affect the density and diffusion coefficient of hydrogen, thus altering the diffusion rate and range.

[0106] S260: Set scene parameters, including ambient temperature, exhaust intensity, and external flow field.

[0107] Ambient temperature affects the physical properties and diffusion behavior of hydrogen. Higher temperatures intensify the molecular motion of hydrogen, accelerating diffusion, while lower temperatures slow it down. Exhaust ventilation intensity determines the airflow inside the vehicle and plays a crucial role in hydrogen diffusion and removal. Stronger ventilation accelerates hydrogen removal and reduces hydrogen concentration inside the vehicle, while weaker ventilation may lead to hydrogen accumulation, increasing safety risks. The external flow field refers to the airflow conditions outside the vehicle, such as wind speed and direction. The external flow field affects the direction and extent of hydrogen diffusion after a leak, especially during vehicle operation, where its influence is more significant.

[0108] S270: Iteratively calculate the simplified geometric model of the vehicle based on boundary conditions and scene parameters to generate a vehicle simulation model.

[0109] Specifically, in each iteration, the controller solves the basic control equations and updates the distribution of physical quantities such as velocity, pressure, and concentration in the flow field. The iteration residual value is used to measure the degree of difference between the calculation result of each iteration and the result of the previous iteration. When the iteration residual value is less than a preset residual threshold, it indicates that the calculation result has converged and reached a certain accuracy requirement.

[0110] S280: Set up multiple hydrogen concentration monitoring points in the vehicle simulation model, and build a database based on the hydrogen concentration values ​​detected at each monitoring point.

[0111] Optionally, a database can be constructed based on hydrogen concentration values ​​obtained from each hydrogen concentration monitoring point. This includes: determining the detection time and monitoring point location corresponding to each hydrogen concentration value to establish data records for each monitoring point location; creating a blockchain data structure, and sequentially encrypting each data record with a public key, packaging it into blocks, and adding it to the data structure to form a database in the form of a blockchain.

[0112] Specifically, the monitoring point location can be a sensor number or a coordinate within the vehicle simulation model. The controller integrates the determined hydrogen concentration value, the corresponding detection time, and the monitoring point location information to form a complete data record. When building a blockchain for hydrogen concentration monitoring data, this basic data structure framework must first be established. To ensure data security and privacy, each established data record can be encrypted using public-key cryptography. The controller packages the public-key encrypted data record into a block and then adds the newly generated block to the previously created blockchain data structure. During the addition process, the new block is linked to the previous block through the hash value of the previous block in its block header, forming a chain structure, ultimately gradually building a blockchain-style database. As more monitoring point data records are continuously generated and encrypted, the blockchain database will continuously grow and update, completely recording the hydrogen concentration information of the vehicle at different times and locations.

[0113] The technical solution of this invention analyzes key components of the target fuel cell vehicle structure, establishes a simplified geometric model of the vehicle using the Fluent platform, and accurately simulates the hydrogen leakage and diffusion process by combining mass, momentum, and energy conservation equations and component transport equations. By rationally dividing the mesh, defining boundary conditions, and setting scene parameters, the accuracy and reliability of the model are improved. The blockchain-based database established on this basis can securely and reliably store hydrogen concentration monitoring data. This allows for a deeper understanding of hydrogen leakage patterns and concentration distribution characteristics, providing a solid data foundation for subsequent optimization.

[0114] Example 3

[0115] Figure 4 This is a flowchart of a hydrogen concentration sensor location optimization method for a fuel cell vehicle according to Embodiment 3 of the present invention. This embodiment specifically describes the initial arrangement of hydrogen concentration sensors in the target fuel cell vehicle based on the above embodiments. Figure 4 As shown, the method includes:

[0116] S310: Decrypt the database using the private key to obtain the correspondence between monitoring point location, detection time, and hydrogen concentration value.

[0117] Specifically, since the database uses blockchain technology and public-key cryptography to store data, it first needs to be decrypted using a private key. The private key, paired with the public key used to encrypt the data, is held by an authorized user. Decryption using the private key allows the reading of the encrypted information in the database, thereby obtaining the accurate correspondence between monitoring point locations, detection times, and hydrogen concentration values.

[0118] S320: Treat the locations of each monitoring point as network nodes and define the edges between network nodes.

[0119] Specifically, the controller can treat each monitoring point location as a network node, representing different hydrogen concentration monitoring locations within the vehicle. The controller also defines edges between these network nodes; the existence of edges indicates a correlation between different monitoring points. This correlation can be determined based on factors such as hydrogen diffusion paths and airflow effects, meaning that changes in hydrogen concentration may influence each other across these related monitoring points.

[0120] S330: Determine the weights of edges between network nodes based on hydrogen concentration values ​​and detection time to generate a network topology graph.

[0121] Specifically, the controller can determine the weight of edges between network nodes based on hydrogen concentration values ​​and detection time. If the time interval between hydrogen concentration changes between two monitoring points is short, and the concentration changes follow a certain pattern, such as simultaneous increases or decreases, it indicates a close correlation between the two monitoring points, and the edge weight is high. Conversely, if the time interval between hydrogen concentration changes between two monitoring points is long, and the concentration changes show no obvious correlation, the edge weight is low. For example, if monitoring points A and B detect significant increases in hydrogen concentration successively within a short period, the edge weight between them is relatively high; if the hydrogen concentration changes between monitoring points C and D are unrelated in both time and value, the edge weight is low. Ultimately, the controller can assign a weight value to each edge that reflects the degree of correlation between monitoring points, generating a network topology graph that visually displays the relationships between the monitoring points.

[0122] S340: Calculate the degree centrality and betweenness centrality of each network node in the network topology graph, and determine the key nodes based on the degree centrality and betweenness centrality.

[0123] Degree centrality measures the number of direct connections a node has with other nodes; a higher degree centrality indicates a more extensive network of connections. Betweenness centrality reflects the frequency with which a node appears on the shortest paths between other nodes in the network. The controller calculates both degree centrality and betweenness centrality for each node and combines these two metrics to assess the node's importance. Ultimately, nodes with high combined degree centrality and betweenness centrality are identified as critical nodes.

[0124] S350: The locations of the monitoring points corresponding to each key node are used as the initial layout of the hydrogen concentration sensors in the target fuel cell vehicle.

[0125] Specifically, the monitoring points corresponding to key nodes were chosen as the initial locations for hydrogen concentration sensors in the target fuel cell vehicle because these locations are crucial in the entire hydrogen concentration monitoring network. Placing sensors in these locations makes it more likely that hydrogen leaks will be detected in a timely and comprehensive manner, providing effective data support for subsequent safety measures. This led to the initial determination of a relatively reasonable sensor placement scheme in the vehicle.

[0126] The technical solution of this invention obtains the corresponding relationship by decrypting the database, constructs the network structure and topology, calculates the node centrality to determine the key nodes and uses them as the initial arrangement of sensors, which can provide accurate data for hydrogen leak detection, avoid subjective judgment, reduce blindness and waste of resources, enhance monitoring capabilities and ensure the safety of fuel cell vehicles.

[0127] Example 4

[0128] Figure 5 This is a flowchart of a hydrogen concentration sensor location optimization method for a fuel cell vehicle according to Embodiment 4 of the present invention. This embodiment specifically describes the initial arrangement of hydrogen concentration sensors in the target fuel cell vehicle based on the above embodiments. Figure 5 As shown, the method includes:

[0129] S410: Define the objective function and constraints, where the objective function includes maximizing scene coverage, minimizing detection time, and minimizing sensor deployment cost, and the constraints include scene detection constraints, budget constraints, and detection time constraints.

[0130] S420: Encode the placement positions in the initial arrangement into particle position vectors, and initialize the population based on each position vector.

[0131] Specifically, in the particle swarm optimization algorithm, the initial sensor placement positions are first encoded and converted into particle position vectors. For example, if there are multiple preset candidate sensor placement positions in a vehicle, these positions can be numbered in a certain order. Assuming there are n candidate positions, each particle's position vector is an n-dimensional vector composed of 0s and 1s, where "1" indicates placing the sensor at the corresponding numbered position, and "0" indicates not placing it. Through encoding, the initial placement scheme can be represented by particle position vectors. Based on the position vectors, a certain number of particles can be randomly generated to initialize the population. Each particle represents a possible sensor placement scheme, and these particles constitute the initial search space of the algorithm, laying the foundation for finding better placement schemes later.

[0132] S430: Generate a penalty function based on constraints and objective function, and combine the penalty function and objective function to generate a fitness function.

[0133] Optionally, maximize scene coverage as follows:

[0134]

[0135] Where f1(X) is the scene coverage objective function, A is the set of all leaked scenes, and λ a It represents the probability of a leak occurring, x. a This means that if scene a is detected, x a =1, otherwise x a =0;

[0136] Minimize the detection time as follows:

[0137]

[0138] Where f2(X) is the detection time objective function, d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. a λ refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios. a It represents the probability of a leakage scenario occurring;

[0139] Minimize sensor deployment cost as follows:

[0140]

[0141] Where f3(X) is the objective function for total sensor cost, L is the set of all candidate sensor placement locations, and c i It is the cost of sensor i, si The value is 1 if a sensor is installed at position i, and 0 otherwise.

[0142] The constraints include scenario detection constraints, budget constraints, and detection time constraints. Scenario detection constraints require that at least one sensor can detect each leakage scenario; budget constraints stipulate that the total sensor cost cannot exceed the budget; and detection time constraints limit the maximum allowable detection time of the system. When the sensor placement scheme represented by a certain particle does not meet these constraints, a penalty function is applied.

[0143] Optionally, the penalty function includes penalties for scene not being covered, cost exceeding limits, and detection time exceeding limits;

[0144] The penalty for scenes not being covered is:

[0145]

[0146] Among them, P coverage For scene coverage penalty terms, α is the weighted penalty coefficient corresponding to the scene coverage penalty term, and s i The value is 1 if a sensor is installed at position i, and 0 otherwise. a This refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios;

[0147] The penalty for exceeding cost limits is as follows:

[0148]

[0149] Among them, P cost Here, β is the penalty term for exceeding cost limits, and s is the weighted penalty coefficient corresponding to the penalty term for exceeding cost limits. i The value is 1 if a sensor is installed at position i, and 0 otherwise. L is the set of all candidate sensor placement positions. i Where i is the cost of sensor i, and p is the budget;

[0150] Penalties for exceeding the detection time limit:

[0151]

[0152] Among them, P time For the penalty term for exceeding the detection time limit, γ is the weighted penalty coefficient corresponding to the penalty term for exceeding the detection time limit, and d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. a T refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios, and T maxThis refers to the maximum detection time allowed by the system;

[0153] The fitness function is:

[0154] F(X) = f1(X) - f2(X) - f3(X) + P coverage +P cost +P time ;

[0155] Where F(X) is the fitness function, f1(X) is the scene coverage objective function, f2(X) is the detection time objective function, f3(X) is the total sensor cost objective function, and P... coverage For scenarios not covered by penalties, P cost As a penalty for exceeding cost limits, P time This is a penalty for exceeding the detection time limit.

[0156] It should be noted that since the objective function involves scene coverage, detection time, and deployment cost, and these three factors have significantly different dimensions and numerical ranges, directly weighting them would lead to one objective dominating the optimization direction. Therefore, the objective function value needs to be normalized, uniformly mapped to the [0,1] interval, and then a fitness function is constructed by combining it with a penalty term. After normalization, the objectives are fused through weighted coefficients, and a penalty term is added. The final fitness function is:

[0157]

[0158] Where F(X) is the fitness function, The objective function for normalized scene coverage is... It is the normalized detection time objective function. This refers to the normalized objective function for the total sensor cost, P. coverage For scenarios not covered by penalties, P cost As a penalty for exceeding cost limits, P time For the penalty item for exceeding the detection time limit, w1, w2, and w3 are weighting coefficients that can be adjusted according to actual needs.

[0159] In summary, the normalization process makes coverage, time, and cost comparable on the same scale, preventing a single objective from dominating the optimization due to its large numerical range. By flexibly adjusting the priority of each objective through weighting coefficients, it can adapt to different scenario requirements. Furthermore, normalization results in smoother fluctuations in the objective function value, reducing drastic oscillations during particle velocity updates and helping the algorithm converge to the global optimum.

[0160] S440: Based on the fitness function, iterative calculations are performed on each particle in the population. In each iteration, individual optimal updates and global optimal updates are performed on each particle in the population. When the maximum number of iterations or the convergence condition is reached, the global optimal position is output as the optimized arrangement.

[0161] Specifically, in each iteration, the fitness value of each particle is first calculated, and the fitness function is used to evaluate the quality of the sensor placement scheme represented by that particle. Then, an individual optimal update is performed, comparing the current fitness value of each particle with the fitness value of its historical best position, Pbest. If the fitness value of the current position is better, the particle's Pbest is updated, recording the current position as the particle's historical best position. Next, a global optimal update is performed, comparing the fitness values ​​of all particles' Pbest to find the optimal solution, the global optimal solution, Gbest. If a particle's current optimal solution is better than the existing Gbest, the global optimal solution is updated. In multi-objective optimization, Gbest is not a single solution, but a representative scheme selected from the Pareto optimal solution set to guide all particles towards a non-dominated direction.

[0162] Specifically, the particle updates its velocity and position according to the velocity update formula and the position update formula. The velocity update formula is as follows:

[0163]

[0164] in, Let be the velocity of particle i in the t-th iteration, which is the optimization scheme in the optimization process. Let Pbest be the position of particle i in the t-th iteration, i.e., the current sensor arrangement. i Gbest refers to the optimal position found by particle i in the history of the optimization process, which can be understood as the individual optimal solution. Gbest is the current optimal solution of the entire swarm, which can be understood as the global optimal solution. w is the inertia weight, which controls the ability to maintain the particle's previous velocity. The larger the weight, the more stable the particle is; the smaller the weight, the more flexible the particle is. c1 and c2 are learning factors, which adjust the tendency of the particle to move closer to the optimal solution, that is, adjust the scheme towards the direction of optimization of the optimal solution. r1 and r2 are random numbers between [0,1], which increase the diversity of the search and avoid synchronization and premature convergence.

[0165] The position update formula is:

[0166]

[0167] in, Let be the position of particle i in the t-th iteration, i.e., the current sensor arrangement. Let i be the velocity of particle i in the (t+1)th iteration. Let be the position of particle i in the (t+1)th iteration.

[0168] In summary, when the preset maximum number of iterations is reached, or certain convergence conditions are met, such as minimal changes in particle position or fitness value after multiple iterations, the iteration stops, and the global optimal position is output. The sensor layout scheme corresponding to the global optimal position is the optimized layout scheme, which can achieve a good balance between maximizing scene coverage, minimizing detection time, and minimizing sensor layout cost, thus realizing optimized sensor layout.

[0169] The technical solution of this invention, by encoding the initial layout, generating the fitness function, and iteratively updating it using the particle swarm optimization algorithm, finally outputs the optimized layout. It can integrate multiple objectives and constraints, efficiently search for the ideal layout scheme of the sensors, reduce the blindness and cost of layout, reduce the dependence on traditional experience, enhance the stability and reliability of the hydrogen leak detection system, and ensure the safety of fuel cells.

[0170] Example 5

[0171] Figure 6 This is a schematic diagram of a hydrogen concentration sensor location optimization device for a fuel cell vehicle according to Embodiment 5 of the present invention. This device can be implemented using software and / or hardware, and is generally integrated into the electronic device performing the method. For example... Figure 6 As shown, the device includes: a simulation model establishment and database construction module 510, used to: establish a vehicle simulation model based on the structure of the target fuel cell vehicle, set multiple hydrogen concentration monitoring points in the vehicle simulation model, and detect hydrogen concentration values ​​based on each hydrogen concentration monitoring point to construct a database;

[0172] The initial layout determination module 520 is used to: analyze the database to determine the initial layout of the hydrogen concentration sensors in the target fuel cell vehicle;

[0173] The optimized layout generation module 530 is used to: set objective functions and constraints, and optimize the initial layout using particle swarm optimization algorithm based on the objective functions and constraints to generate an optimized layout. The objective functions include maximizing scene coverage, minimizing detection time, and minimizing sensor deployment cost. The constraints include scene detection constraints, budget constraints, and detection time constraints.

[0174] Optionally, the simulation model establishment and database construction module 510 specifically includes: a vehicle simulation model establishment unit, used for: analyzing key components based on the structure of the target fuel cell vehicle to identify key components, including the fuel cell stack, hydrogen storage device, and transmission device; establishing a simplified geometric model of the vehicle based on the key components using the Fluent platform, wherein the simplified geometric model is used for hydrogen leakage diffusion simulation; determining the mass, momentum, and energy conservation equations and component transport equations as the basic control equations of the simplified geometric model for the hydrogen leakage diffusion process, and assigning values ​​to the parameters in the basic control equations; defining key and secondary regions in the simplified geometric model of the vehicle, using a first specified density for meshing in the key region and a second specified density for meshing in the secondary region, wherein the key region includes the hydrogen leak and the ventilation opening, and the first specified density is greater than the second specified density; defining boundary conditions, including the leak, wall boundaries, and environmental conditions; setting scene parameters, including ambient temperature, exhaust intensity, and external flow field; and iteratively calculating the simplified geometric model of the vehicle based on the boundary conditions and scene parameters to generate a vehicle simulation model.

[0175] Optionally, the simulation model establishment and database construction module 510 specifically includes: a database construction unit, used to: determine the detection time and monitoring point location corresponding to each hydrogen concentration value, so as to establish data records for each monitoring point location; create a blockchain data structure, and sequentially package each data record into a block after being encrypted with a public key, and add it to the data structure to form a database in the form of a blockchain.

[0176] Optionally, the initial layout determination module 520 is specifically used for: decrypting the database using a private key to obtain the correspondence between monitoring point locations, detection times, and hydrogen concentration values; treating each monitoring point location as a network node and defining the edges between network nodes; determining the weights of the edges between network nodes based on the hydrogen concentration value and detection time to generate a network topology graph; calculating the degree centrality and betweenness centrality of each network node in the network topology graph, and determining key nodes based on the degree centrality and betweenness centrality; and using the monitoring point locations corresponding to each key node as the initial layout of the hydrogen concentration sensors in the target fuel cell vehicle.

[0177] Optionally, the optimized layout generation module 530 is specifically used for: encoding the layout positions in the initial layout into particle position vectors, and initializing the population based on each position vector; generating a penalty function based on constraints and an objective function, and combining the penalty function and the objective function to generate a fitness function; performing iterative calculations on each particle in the population based on the fitness function, and in each iteration, performing individual optimal updates and global optimal updates on each particle in the population, and outputting the global optimal position as the optimized layout when the maximum number of iterations or the convergence condition is reached.

[0178] Optionally, maximize scene coverage as follows:

[0179]

[0180] Where f1(X) is the scene coverage objective function, A is the set of all leaked scenes, and λ a It represents the probability of a leak occurring, x. a This means that if scene a is detected, x a =1, otherwise x a =0;

[0181] Minimize the detection time as follows:

[0182]

[0183] Where f2(X) is the detection time objective function, d a,i The detection time of sensor i in scene a, x a,i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. a λ refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios. a It represents the probability of a leakage scenario occurring;

[0184] Minimize sensor deployment cost as follows:

[0185]

[0186] Where f3(X) is the objective function for total sensor cost, L is the set of all candidate sensor placement locations, and c i It is the cost of sensor i, s i The value is 1 if a sensor is installed at position i, and 0 otherwise.

[0187] Optionally, the penalty function includes penalties for scene not being covered, cost exceeding limits, and detection time exceeding limits;

[0188] The penalty for scenes not being covered is:

[0189]

[0190] Among them, P coverage For scene coverage penalty terms, α is the weighted penalty coefficient corresponding to the scene coverage penalty term, and s i The value is 1 if a sensor is installed at position i, and 0 otherwise. a This refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios;

[0191] The penalty for exceeding cost limits is as follows:

[0192]

[0193] Among them, P cost Here, β is the penalty term for exceeding cost limits, and s is the weighted penalty coefficient corresponding to the penalty term for exceeding cost limits. i The value is 1 if a sensor is installed at position i, and 0 otherwise. L is the set of all candidate sensor placement positions. i Where i is the cost of sensor i, and p is the budget;

[0194] Penalties for exceeding the detection time limit:

[0195]

[0196] Among them, P time For the penalty term for exceeding the detection time limit, γ is the weighted penalty coefficient corresponding to the penalty term for exceeding the detection time limit, and d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. α T refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios, and T max This refers to the maximum detection time allowed by the system;

[0197] The fitness function is:

[0198] F(X) = f1(X) - f2(X) - f3(X) + P coverage +P cost +P time ;

[0199] Where F(X) is the fitness function, f1(X) is the scene coverage objective function, f2(X) is the detection time objective function, f3(X) is the total sensor cost objective function, and P... coverage For scenarios not covered by penalties, P cost As a penalty for exceeding cost limits, P time This is a penalty for exceeding the detection time limit.

[0200] Optionally, the device further includes: a redundant sensor arrangement module, used to: simulate the vehicle simulation model based on the optimized arrangement, and determine the degree of influence of each arrangement position in the optimized arrangement during the simulation process, wherein the degree of influence includes high, medium and low; designate the arrangement position with a high degree of influence as the key position, and arrange redundant sensors in parallel redundancy for the key position.

[0201] The technical solution of this invention establishes a vehicle simulation model based on the structure of the target fuel cell vehicle and constructs a database by setting multiple hydrogen concentration monitoring points. This allows for a comprehensive and accurate understanding of the hydrogen concentration distribution within the vehicle, improving the accuracy of fuel cell hydrogen leak detection. The initial layout is optimized using a particle swarm optimization algorithm. Combined with objective functions such as maximizing scene coverage, minimizing detection time, and minimizing sensor deployment costs, along with corresponding constraints, the optimal sensor deployment scheme can be achieved while meeting detection requirements. This reduces sensor redundancy, lowers deployment costs, and shortens hydrogen leak detection time, improving detection efficiency. It reduces reliance on traditional experience-based site selection, avoiding inaccurate detection and cost waste caused by insufficient experience or misjudgment. This enhances the stability and reliability of the fuel cell vehicle hydrogen leak detection system, effectively ensuring the safety of the fuel cell in the vehicle.

[0202] The hydrogen concentration sensor location optimization device for fuel cell vehicles provided in this embodiment of the invention can execute the hydrogen concentration sensor location optimization method for fuel cell vehicles provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0203] Example 6

[0204] Figure 7 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0205] like Figure 7As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0206] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0207] Processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing unit (CPU), graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 executes the various methods and processes described above, such as a method for optimizing the location of hydrogen concentration sensors in a fuel cell vehicle. That is: a vehicle simulation model is established based on the structure of the target fuel cell vehicle, and multiple hydrogen concentration monitoring points are set in the vehicle simulation model. Based on the hydrogen concentration values ​​detected by each hydrogen concentration monitoring point, a database is constructed; the database is analyzed to determine the initial arrangement of hydrogen concentration sensors in the target fuel cell vehicle; an objective function and constraints are set, and based on the objective function and constraints, a particle swarm optimization algorithm is used to optimize the initial arrangement to generate an optimized arrangement. The objective function includes maximizing scene coverage, minimizing detection time, and minimizing sensor arrangement cost. The constraints include scene detection constraints, budget constraints, and detection time constraints.

[0208] In some embodiments, a hydrogen concentration sensor location optimization method for a fuel cell vehicle can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the hydrogen concentration sensor location optimization method for a fuel cell vehicle described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform a hydrogen concentration sensor location optimization method for a fuel cell vehicle by any other suitable means (e.g., by means of firmware).

[0209] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0210] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0211] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0212] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0213] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0214] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and VPS servers, such as high management difficulty and weak business scalability.

[0215] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0216] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for optimizing the location of a hydrogen concentration sensor in a fuel cell vehicle, characterized in that, include: A vehicle simulation model is established based on the structure of the target fuel cell vehicle, and multiple hydrogen concentration monitoring points are set in the vehicle simulation model. The hydrogen concentration value is detected based on each hydrogen concentration monitoring point to build a database. The database is analyzed to determine the initial arrangement of hydrogen concentration sensors in the target fuel cell vehicle; An objective function and constraints are defined, and the initial layout is optimized using a particle swarm optimization algorithm based on the objective function and constraints to generate an optimized layout. The objective function includes maximizing scene coverage, minimizing detection time, and minimizing sensor deployment cost. The constraints include scene detection constraints, budget constraints, and detection time constraints.

2. The method according to claim 1, characterized in that, The process of establishing a vehicle simulation model based on the structure of the target fuel cell vehicle includes: The key components are analyzed based on the structure of the target fuel cell vehicle to identify the key components, which include the fuel cell stack, hydrogen storage device, and transmission device. A simplified vehicle geometry model is built using the Fluent platform based on the key components, wherein the simplified vehicle geometry model is used for hydrogen leakage diffusion simulation. For the hydrogen leakage and diffusion process, the mass, momentum, energy conservation equations and component transport equations are determined as the basic control equations for the simplified geometric model of the vehicle, and values ​​are assigned to the parameters in the basic control equations. Define key regions and secondary regions in the simplified geometric model of the vehicle. Mesh the key regions using a first specified density and the secondary regions using a second specified density. The key regions include hydrogen leak vents and ventilation vents. The first specified density is greater than the second specified density. Define boundary conditions, which include the leak point, wall boundaries, and environmental conditions; Set scene parameters, including ambient temperature, exhaust intensity, and external flow field; The simplified geometric model of the vehicle is iteratively calculated based on boundary conditions and scene parameters to generate a vehicle simulation model.

3. The method according to claim 1, characterized in that, The process of acquiring hydrogen concentration values ​​based on each of the hydrogen concentration monitoring points to construct a database includes: Determine the detection time and monitoring point location corresponding to each hydrogen concentration value in order to establish data records for each monitoring point location; Create a blockchain data structure, and then package each data record into a block after encrypting it with a public key and add it to the data structure to form a database in the form of a blockchain.

4. The method according to claim 3, characterized in that, The analysis of the database to determine the initial arrangement of hydrogen concentration sensors in the target fuel cell vehicle includes: The database is decrypted using a private key to obtain the correspondence between monitoring point locations, detection times, and hydrogen concentration values. Each monitoring point location is treated as a network node, and the edges between the network nodes are defined. The weights of the edges between network nodes are determined based on the hydrogen concentration value and the detection time to generate a network topology graph; Calculate the degree centrality and betweenness centrality of each network node in the network topology graph, and determine the key nodes based on the degree centrality and betweenness centrality; The locations of the monitoring points corresponding to each of the key nodes are used as the initial arrangement of hydrogen concentration sensors in the target fuel cell vehicle.

5. The method according to claim 1, characterized in that, The step of optimizing the initial layout using a particle swarm optimization algorithm based on the objective function and constraints to generate an optimized layout includes: The placement positions in the initial arrangement are encoded as particle position vectors, and the population is initialized based on each of the aforementioned position vectors; A penalty function is generated based on the constraints and the objective function, and the penalty function and the objective function are combined to generate a fitness function; Based on the fitness function, iterative calculations are performed on each particle in the population. In each iteration, individual optimal updates and global optimal updates are performed on each particle in the population. When the maximum number of iterations or the convergence condition is reached, the global optimal position is output as the optimized arrangement.

6. The method according to claim 5, characterized in that, The maximum scene coverage rate is: Where f1(X) is the scene coverage objective function, A is the set of all leaked scenes, and λ a It represents the probability of a leak occurring, x. a This means that if scene a is detected, x a =1, otherwise x a =0; The minimum detection time is: Where f2(X) is the detection time objective function, d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. a λ refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios. a It represents the probability of a leakage scenario occurring; The minimum sensor deployment cost is: Where f3(X) is the objective function for total sensor cost, L is the set of all candidate sensor placement locations, and c i It is the cost of sensor i, s i The value is 1 if a sensor is installed at position i, and 0 otherwise.

7. The method according to claim 6, characterized in that, The penalty function includes penalties for scene coverage failure, cost exceeding limits, and detection time exceeding limits; The following penalty items are not covered in the scenario: Among them, P coverage For scene coverage penalty terms, α is the weighted penalty coefficient corresponding to the scene coverage penalty term, and s i The value is 1 if a sensor is installed at position i, and 0 otherwise. a This refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios; The penalty for exceeding the cost limit is as follows: Among them, P cost Here, β is the penalty term for exceeding cost limits, and s is the weighted penalty coefficient corresponding to the penalty term for exceeding cost limits. i The value is 1 if a sensor is installed at position i, and 0 otherwise. L is the set of all candidate sensor placement positions. i Where i is the cost of sensor i, and p is the budget; The penalty for exceeding the detection time limit: Among them, P time For the penalty term for exceeding the detection time limit, γ is the weighted penalty coefficient corresponding to the penalty term for exceeding the detection time limit, and d a，i The detection time of sensor i in scene a, x a，i This means that if sensor i is the first sensor to detect scene a, then the value is 1; otherwise, it is 0. a T refers to the set of sensor locations detected in scenario a, where A is the set of all leakage scenarios, and T max This refers to the maximum detection time allowed by the system; The fitness function is: F(X)=f1(X)-f2(X)-f3(X)+P coverage +P cost +P time ; Where F(X) is the fitness function, f1(X) is the scene coverage objective function, f2(X) is the detection time objective function, f3(X) is the total sensor cost objective function, and P... coverage For scenarios not covered by penalties, P cost As a penalty for exceeding cost limits, P time This is a penalty for exceeding the detection time limit.

8. The method according to claim 1, characterized in that, The method further includes: The vehicle simulation model is simulated based on the optimized layout, and the degree of influence of each layout position in the optimized layout is determined during the simulation process. The degree of influence includes high, medium and low. The locations with the highest impact are designated as critical locations, and redundant sensors are deployed in parallel for these critical locations.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

10. A computer storage medium, characterized in that, The computer storage medium stores computer instructions that are used to cause a processor to execute the method of any one of claims 1-8.

Images