Gas leakage prediction method, gas leakage prediction device and electronic equipment

By establishing a gas environment model and a state transition probability matrix, the problem of insufficient accuracy and real-time performance in hydrogen leak prediction in existing technologies is solved, and simple and efficient gas leak prediction and timely prevention and control are achieved.

CN119026497APending Publication Date: 2024-11-26GUANGDONG DIANWANG GONGSI YUNFU POWER SUPPLY BUREAU
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202411015993.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Existing hydrogen leak prediction methods have poor accuracy and insufficient real-time performance, and require large data samples and complex structures, making it impossible to achieve dynamic and efficient hydrogen leak prediction.

Method used

By acquiring environmental data, a gas environment model is established, gas leakage conditions are constructed, the initial state transition probability matrix of the state space is calculated, and the risk level of future time series is predicted based on the state transition probability matrix. The risk level is then output for early warning and maintenance.

Benefits of technology

It achieves simple and efficient gas leak prediction, reduces the data volume requirements, improves prediction efficiency and real-time performance, and enables timely early warning and prevention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026497A_ABST
    Figure CN119026497A_ABST
Patent Text Reader

Abstract

The invention provides a gas leakage prediction method, a gas leakage prediction device and electronic equipment. The method comprises the following steps: acquiring environmental data, establishing a gas environment model according to the environmental data, constructing gas leakage conditions, and acquiring the current gas concentration and gas concentration change rate of the gas environment model under the gas leakage conditions, the gas leakage condition is a condition representing the strength of gas leakage; establishing a state space according to the current gas concentration and the gas concentration change rate, calculating an initial state transition probability matrix of the state space, and calculating a state transition probability matrix corresponding to a future time sequence according to the initial state transition probability matrix; and determining a risk level of gas leakage according to the state transition probability matrix, and outputting the risk level so as to carry out early warning and maintenance on the gas leakage. According to the invention, the problem that a hydrogen leakage prediction method in the prior art cannot realize efficient prediction is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of gas leakage prediction, and in particular to a gas leakage prediction method, a gas leakage prediction device, a computer-readable storage medium, and an electronic device. Background Art

[0002] With the popularization of hydrogen energy technology, hydrogen leakage accidents and the explosions, fires and other accidents caused by them are becoming more and more frequent, and the monitoring, early warning and prevention of hydrogen leakage are becoming more and more urgent. However, due to the characteristics of hydrogen with large diffusion coefficient, light molecular weight, wide explosion limit and low minimum ignition energy, general sensors can only detect the hydrogen concentration at a single point in space, and cannot intuitively display the time and space evolution characteristics of hydrogen concentration in three-dimensional space, let alone predict the subsequent development of hydrogen leakage in different scenarios, and provide limited guidance for emergency repair after hydrogen leakage. For equipment such as hydrogen fuel cell vehicles, hydrogen fuel mobile emergency power supplies, and hydrogen fuel powered ships, the many scenarios and complex boundary conditions they face during operation and hydrogenation lead to poor accuracy and lack of real-time performance in the establishment of numerical simulation models. At the same time, the prediction model often requires a large number of data samples and lacks the ability to adaptively update, which makes it impossible to achieve dynamic and efficient hydrogen leakage prediction.

[0003] Although there are some methods and early warning systems for predicting hydrogen leaks, the simulation models are often inaccurate and lack real-time performance. They also require large data samples, have complex structures, are costly, and lack the ability to adaptively update, making it impossible to achieve dynamic and efficient hydrogen leak prediction.

[0004] Therefore, an efficient hydrogen leakage prediction method is needed. Summary of the invention

[0005] The main purpose of the present application is to provide a gas leakage prediction method, a gas leakage prediction device, a computer-readable storage medium and an electronic device, so as to at least solve the problem that the hydrogen leakage prediction method in the prior art cannot make efficient predictions.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for predicting gas leakage is provided, comprising: acquiring environmental data, establishing a gas environment model according to the environmental data, constructing a gas leakage condition, and obtaining the current gas concentration and the gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing the environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage; establishing a state space according to the current gas concentration and the gas concentration change rate, and calculating an initial state transition probability matrix of the state space, and calculating a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents multiple time series after the current time series corresponding to the current gas concentration; determining the risk level of the gas leakage according to the state transition probability matrix, and outputting the risk level to warn and repair the gas leakage.

[0007] Optionally, calculating the initial state transition probability matrix of the state space includes: calculating the transition probability from any state in the state space to all remaining states to obtain multiple state transition probabilities; and arranging the multiple state transition probabilities in a predetermined order to form the initial state transition probability matrix.

[0008] Optionally, the state transition probability matrix corresponding to the future time series is calculated based on the initial state transition probability matrix, including: calculating the intermediate state transition probability matrix from the current time series to a time series before the future time series to obtain multiple intermediate state transition probability matrices; calculating the product of multiple intermediate state transition probability matrices to obtain the state transition probability matrix corresponding to the future time series.

[0009] Optionally, environmental data is obtained, and a gas environment model is established based on the environmental data, including: obtaining the length, width, and height dimensions of the environmental space where the gas is located, the leakage hole size, and the ventilation hole size, and establishing a grid model of the environmental space based on the length, width, and height dimensions, the leakage hole size, and the ventilation hole size; encrypting each grid in the grid model, and determining the mass flow calculation formula and component transport equation of the gas to obtain the gas environment model.

[0010] Optionally, constructing gas leakage conditions includes: obtaining the opening and closing state of the environmental space where the gas is located, the gas leakage duration and the ventilation intensity of the ventilation holes; and combining different values ​​of the opening and closing state, different values ​​of the gas leakage duration and different values ​​of the ventilation intensity to obtain a plurality of the gas leakage conditions.

[0011] Optionally, after determining the risk level of gas leakage according to the state transition probability matrix, the method further includes: obtaining actual state data corresponding to each of the future time series; verifying the risk level according to the actual state data to obtain a verification error, and determining that the risk level is accurate when the verification error is within an allowable error range.

[0012] Optionally, a state space is established according to the current gas concentration and the gas concentration change rate, including: dividing the risk levels according to the current gas concentration and the gas concentration change rate, obtaining the state corresponding to each risk level, and all the states constitute the state space.

[0013] According to another aspect of the present application, a gas leakage prediction device is provided, comprising: a first acquisition unit, used to acquire environmental data, establish a gas environment model according to the environmental data, construct a gas leakage condition, and acquire the current gas concentration and the gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing the environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage; a calculation unit, used to establish a state space according to the current gas concentration and the gas concentration change rate, and calculate an initial state transition probability matrix of the state space, and calculate a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents multiple time series after the current time series corresponding to the current gas concentration; an output unit, used to determine the risk level of the gas leakage according to the state transition probability matrix, and output the risk level to warn and repair the gas leakage.

[0014] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the gas leakage prediction methods.

[0015] According to another aspect of the present application, an electronic device is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing any one of the gas leakage prediction methods.

[0016] Apply the technical solution of the present application, obtain environmental data, establish a gas environment model based on the environmental data, construct gas leakage conditions, and obtain the current gas concentration and gas concentration change rate of the gas environment model under the gas leakage conditions; establish a state space based on the current gas concentration and gas concentration change rate, and calculate the initial state transition probability matrix of the state space, and calculate the state transition probability matrix corresponding to the future time series based on the initial state transfer matrix; determine the risk level of gas leakage based on the state transfer probability matrix, and output the risk level to warn and repair the gas leakage. Compared with the complex prediction method in the prior art, the model structure is simpler, easier to implement, and more efficient. At the same time, the requirements for the amount of data are lower, and only a small amount of feature data is required. Through subsequent continuous iteration and supplementation of measured data, it can solve the problems of few historical data, insufficient sequence integrity and low reliability, and it is also conducive to reducing the time cost of system prediction and improving prediction efficiency. Therefore, the prediction efficiency of gas leakage can be improved, and the effect of timely prediction and prevention and control can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 A hardware structure block diagram of a mobile terminal for executing a gas leakage prediction method provided in an embodiment of the present application is shown;

[0019] Figure 2 A schematic flow chart of a method for predicting gas leakage provided in an embodiment of the present application is shown;

[0020] Figure 3 A schematic diagram of the structure of a gas leakage model provided in an embodiment of the present application is shown;

[0021] Figure 4 A schematic diagram of a hydrogen leakage monitoring, early warning and prevention system provided in an embodiment of the present application is shown;

[0022] Figure 5 A risk level schematic diagram provided by an embodiment of the present application is shown;

[0023] Figure 6 A schematic flow chart of a specific gas leakage prediction method provided in an embodiment of the present application is shown;

[0024] Figure 7 A schematic flow chart of a Markov chain prediction model in a specific gas leakage prediction method provided in an embodiment of the present application is shown;

[0025] Figure 8 A structural block diagram of a gas leakage prediction device provided in an embodiment of the present application is shown.

[0026] The above drawings include the following reference numerals:

[0027] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0028] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0029] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] As introduced in the background technology, the hydrogen leakage prediction method in the prior art requires many data samples, has a complex structure, and cannot be predicted efficiently. The prior art discloses a hydrogen leakage monitoring and early warning system and a risk control method for a hydrogen refueling station, which includes: a display unit provides a real-time three-dimensional map and a pipeline instrument flow chart; a monitoring unit is used to collect image data and environmental data; an alarm unit is used to receive image data and environmental data and perform a leakage alarm; a leakage early warning judgment unit is used to determine the leakage location and the change trend of the hydrogen concentration diffusion after the leakage; the display unit is also used to update and adjust the real-time three-dimensional map and the pipeline instrument flow chart based on the operating data of the monitoring unit, the alarm unit and the leakage early warning judgment unit. The hydrogen leakage monitoring and early warning system of the hydrogen refueling station disclosed in the present invention monitors and predicts hydrogen leakage at the hydrogen refueling station, reduces the response time when an accident occurs, gains valuable time for the handling personnel, and improves the safety factor of the hydrogen refueling station. At the same time, the leakage location and the change trend of the hydrogen concentration diffusion after the leakage can also be determined, providing a data basis for formulating corresponding response measures. This patent predicts the diffusion range and diffusion rate of hydrogen in a future set time period based on the computational fluid dynamics CFD grid algorithm. However, the hydrogen diffusion process is a transient process. The CFD calculation of the transient process is heavily dependent on transient models, computing resources and other conditions, and the accuracy of the calculation results is often poor, and the calculation takes a long time, which cannot meet the needs of real-time decision-making on hydrogen leakage on site. A method and device for detecting hydrogen leakage in a fuel cell system are also disclosed. By obtaining the decreasing rate of the hydrogen path pressure of the fuel cell system during the shutdown and hydrogen replenishment stage, it is determined that the hydrogen path is leaking when the decreasing rate exceeds the preset threshold, and then it is determined whether the hydrogen concentration sensor reading is abnormal. When the reading is abnormal, the external leakage position of the fuel cell system is determined according to the hydrogen concentration sensor reading, and when it is normal, the cross leakage position of the fuel cell system is determined by the open circuit voltage, thereby realizing real-time and accurate detection of hydrogen leakage in the fuel cell system. The device includes a hydrogen pressure acquisition module, an abnormality determination module, a hydrogen concentration judgment module, an external leakage judgment module, and a cross leakage judgment module. And a method and device for detecting hydrogen leakage in a fuel cell system. By obtaining the decreasing rate of the hydrogen path pressure of the fuel cell system during the shutdown and hydrogen replenishment stage, it is determined that the hydrogen path is leaking when the decreasing rate exceeds the preset threshold, and then it is determined whether the hydrogen concentration sensor reading is abnormal. When the reading is abnormal, the external leakage position of the fuel cell system is determined according to the reading of the hydrogen concentration sensor. When it is normal, the cross-leakage position of the fuel cell system is determined by the open circuit voltage, thereby achieving real-time and accurate detection of hydrogen leakage in the fuel cell system. The device includes a hydrogen pressure acquisition module, an abnormality determination module, a hydrogen concentration judgment module, an external leakage judgment module, and a cross-leakage judgment module. A pipeline hydrogen leakage identification method, device, computer equipment, storage medium and computer program product.The above method includes: obtaining pipeline related parameters based on preset time information and calculating the actual hydrogen change based on the pipeline related parameters; calculating the theoretical hydrogen change according to the pipeline related parameters and the preset pressure time comparison table; if the absolute value of the difference between the actual hydrogen change and the theoretical hydrogen change is outside the preset threshold range, the hydrogen change is judged to be abnormal and the relevant alarm information is sent. The method can prevent the misjudgment under normal circumstances from affecting normal operation, and timely identify before serious hydrogen leakage to ensure safety. A visual intelligent monitoring and early warning system for hydrogen leakage of hydrogen energy vehicles includes a hydrogen absorption device, an infrared thermal imaging detection module, a signal processing and display module, an alarm module and a wireless data transmission and storage module. The hydrogen absorption area of ​​the hydrogen absorption device is coated with tungsten trioxide nanoparticles, which absorb hydrogen to produce a chemical reaction and cause temperature changes. There are no tungsten trioxide nanoparticles in the reference area. The real-time temperature difference between the two can eliminate the influence of ambient temperature changes on hydrogen leakage measurement results. The surface temperature of the hydrogen absorption device is visually monitored by thermal imaging, and combined with a machine learning algorithm, the hydrogen leakage process and the evolution of the hydrogen leakage concentration are predicted to achieve advance warning of hydrogen leakage. The monitoring results are used for system intelligent upgrades to improve the prediction accuracy of the hydrogen leakage process and the advance warning capability. The invention can provide a visual, stable and intelligent hydrogen leakage monitoring and early warning solution, but the materials are complicated, the structure is complex, and the cost is high. It can be seen that although there are some methods and early warning systems for predicting hydrogen leakage in the prior art, the accuracy is often poor and the real-time performance is insufficient when establishing a simulation model. At the same time, a large data sample is often required, the structure is complex, the cost is high, and the ability of adaptive updating is lacking, and dynamic and efficient hydrogen leakage prediction cannot be achieved. In order to solve the problem of not being able to efficiently predict hydrogen leakage, the embodiments of the present application provide a method for predicting gas leakage, a device for predicting gas leakage, a computer-readable storage medium, and an electronic device.

[0032] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0033] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a method for predicting gas leakage according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0034] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the gas leakage prediction method in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above network include but are not limited to the Internet, corporate intranets, local area networks, mobile communication networks and combinations thereof. The transmission device 106 is used to receive or send data via a network. The above network specific examples may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] In this embodiment, a method for predicting gas leakage running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0036] Figure 2 FIG. 1 is a flow chart of a method for predicting gas leakage according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0037] Step S201, obtaining environmental data, establishing a gas environment model according to the environmental data, constructing a gas leakage condition, and obtaining the current gas concentration and gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing the environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage;

[0038] Specifically, the present application uses hydrogen as gas for analysis, which is mainly divided into a laboratory end and a field end. The laboratory end consists of a reduced model test bench, CFD simulation (Computational Fluid Dynamics, CFD simulation for short), and a Markov chain prediction model training server. Its basic process is as follows: According to the characteristics of high diffusivity and large buoyancy of hydrogen in the air, the initial hydrogen leakage sensor is arranged. At the same time, in the CFD simulation, geometric modeling, meshing, calculation model setting, initial condition setting, numerical calculation, graphical display and other processes are performed. The CFD simulation results are corrected with the laboratory measured data, and the calculation model setting, meshing, initial conditions, and calculation methods are corrected, and then the CFD simulation model is used to optimize the laboratory sensor arrangement. This is repeated until the model calculation deviation is within the allowable range of 1%, and the sensor arrangement is relatively reasonable (areas with large hydrogen concentrations and areas with large changes in hydrogen concentrations). This CFD model is used to simulate hydrogen leakage events in real scenes, and the Markov chain prediction model is trained using the simulation results. Taking the hydrogen fuel cell cabin as an example, the environmental data includes the length, width, height and other data of the fuel cell cabin, as well as the length, width, height and other data of the fuel cell stack. It can also include other geographical data required to establish the gas environment model. There are multiple sets of leakage conditions, which will be introduced in detail below. Through the gas environment model to simulate different leakage conditions, the current gas concentration and gas concentration change rate data under different conditions are obtained. The gas concentration change rate data can be calculated by the ratio of the difference in gas concentration within a certain period of time to the time. In this way, a database of hydrogen concentration changes under different leakage conditions is preliminarily constructed.

[0039] Step S202, establishing a state space according to the current gas concentration and the gas concentration change rate, and calculating an initial state transition probability matrix of the state space, and calculating a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents a plurality of time series after the current time series corresponding to the current gas concentration;

[0040] Specifically, a Markov chain prediction model is designed, and the current hydrogen concentration (current gas concentration) and the rate of change of hydrogen concentration (gas concentration change rate) are used as characteristic data. The risk can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the rate of change of hydrogen concentration exceed the 4% limit. According to the division, the initial state transfer probability matrix is ​​calculated, and then the basic calculation process and the matrix probability update are completed based on the initial state transfer probability matrix to obtain the state transfer probability matrix corresponding to the future time series. Finally, the predicted risk level is obtained, that is, the situation and risk level at the next moment can be predicted based on the current hydrogen concentration and the rate of change of hydrogen concentration, so as to carry out early warning and prevention and control. And due to the stable characteristics of the Markov chain state transfer matrix, multi-step prediction can be achieved, that is, the risk level corresponding to the future time series can be predicted, and the predicted data can be saved and written into the database, so as to adaptively update the existing database.

[0041] Step S203, determining the risk level of gas leakage according to the state transition probability matrix, and outputting the risk level to provide early warning and maintenance for gas leakage.

[0042] Specifically, the Markov state prediction model obtained through training is finally embedded into the on-site hydrogen leakage monitoring, early warning and prevention and control system. When a hydrogen leakage occurs, the system will automatically judge sensor false alarms, check hydrogen leakage occurrence points, display hydrogen concentration in three dimensions in real time, predict hydrogen concentration in three dimensions in future event segments, provide hydrogen leakage repair suggestions, plan escape routes, and upload remote data to the laboratory end based on the training results. The laboratory end receives the remote data, and further corrects the CFD calculation results and graph neural network according to the deviation between the remote sensor measured data and the predicted results of the Markov chain prediction model, further improving the CFD calculation accuracy and model prediction accuracy.

[0043] Through this embodiment, environmental data can be obtained, a gas environment model can be established according to the environmental data, gas leakage conditions can be constructed, and under the gas leakage conditions, the current gas concentration and gas concentration change rate of the gas environment model can be obtained; a state space can be established according to the current gas concentration and gas concentration change rate, and the initial state transition probability matrix of the state space can be calculated, and the state transition probability matrix corresponding to the future time series can be calculated according to the initial state transfer matrix; the risk level of gas leakage can be determined according to the state transfer probability matrix, and the risk level can be output to warn and repair the gas leakage. Compared with the complex prediction method in the prior art, the model structure is simpler, easier to implement, and more efficient. At the same time, the requirements for the amount of data are lower, and only a small amount of feature data is required. Through subsequent continuous iteration and supplementation of measured data, the problems of less historical data, insufficient sequence integrity and low reliability can be solved, and it is also beneficial to reduce the time cost of system prediction and improve prediction efficiency. Therefore, the prediction efficiency of gas leakage can be improved, and the effect of timely prediction and prevention and control can be achieved.

[0044] In the specific implementation process, the above step S202 calculates the initial state transition probability matrix of the above state space, which can be implemented by the following steps: Step S2021: Calculate the transition probability from any state in the above state space to all remaining states to obtain multiple state transition probabilities; Step S2022: Arrange the multiple state transition probabilities in a predetermined order to form the above initial state transition probability matrix. The method calculates the initial state transition probability matrix through the above steps, which can represent the current state transition process.

[0045] Specifically, the risk can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the hydrogen concentration change rate exceed the 4% limit, forming four risk levels, and a state space containing four states is established according to the four risk levels: states =[1,2,3,4], where f states Indicates the risk state, 1, 2, 3, 4 correspond to the risk level. Then, according to the database data, the initial state transition probability matrix is ​​calculated as follows: Among them, T tm represents the initial state transition probability matrix, T ij represents the switching probability from state i to state j. States i and j correspond to the risk level states in the risk state space respectively, and satisfy the relationship

[0046] In some optional implementations, the above step S202 calculates the state transition probability matrix corresponding to the future time sequence according to the above initial state transition probability matrix, which can be implemented by the following steps: Step S2023: Calculate the intermediate state transition probability matrix from the above current time sequence to a time sequence before the above future time sequence, and obtain multiple intermediate state transition probability matrices; Step S2024: Calculate the product of multiple intermediate state transition probability matrices to obtain the state transition probability matrix corresponding to the above future time sequence. The method calculates the state transition probability matrix corresponding to the future time sequence through the above steps to obtain the state corresponding to the future time sequence.

[0047] In the specific implementation process, after the preliminary state transition probability matrix is ​​established, the risk state of 2 to n time series can be predicted based on the state transition probability matrix. tm,n =T tm ·T tm,2 ·...·T tm,n-1 , where T tm,n is the state transition probability matrix of the nth time series. The risk level of the current time series is judged by the transition probability relationship of this probability matrix, so as to take corresponding early warning and prevention measures. At the same time, the actual state data after each step of prediction will be written into the database, and the state transition probability matrix T will continue to be dynamically updated. tm,n The Markov chain prediction model is trained repeatedly using the dynamic database of simulation results until the training is completed after reaching the allowable error range.

[0048] In some optional embodiments, the above step S201 obtains environmental data, and establishes a gas environment model based on the above environmental data, which can be achieved by the following steps: Step S2011: Obtain the length, width and height dimensions of the environmental space where the above gas is located, the leakage hole dimensions and the ventilation hole dimensions, and establish a grid model of the above environmental space based on the length, width and height dimensions, the leakage hole dimensions and the ventilation hole dimensions; Step S2012: Encrypt each grid in the above grid model, and determine the mass flow calculation formula and component transport equation of the above gas to obtain the above gas environment model. The method establishes a simulation model through the above steps, so that gas leakage, especially hydrogen leakage, in real situations can be simulated to predict future leakage.

[0049] Specifically, the environment space can be a fuel cell cabin, and a simulation model of the fuel cell cabin is established, such as Figure 3As shown in the figure, the fuel cell passenger ship is 30 meters long, 10 meters wide, and has 150 passenger seats. Its power source is mainly 40 120kW proton exchange membrane fuel cell groups, with a total installed power of 4.92MW (including 120kW reserve power). The length, width and height of a single stack are 1.07 meters, 0.762 meters and 1.99 meters respectively. It is placed in the fuel cell room on the deck. The fuel cell room, control cabin and cabin are all adjacent to each other on the second floor of the ship. A grid model of the fuel cell cabin is established according to the size of the fuel cell cabin, the exhaust port and the hatch. The entire calculation process is carried out using transient numerical calculation. In order to reduce the calculation time, save computing resources, and make the simulation easy to converge, the Icem tool (a professional CAE pre-processing software with powerful CAD model repair capabilities) is used to divide the structured grid, and the grids of the horizontal and longitudinal sections where the leakage holes and ventilation holes are located are encrypted, with a total of 358,949 grids. The leakage mode is continuous leakage caused by the rupture of the hydrogen supply pipeline in the fuel cell compartment. The hydrogen uses the ideal gas model, that is, PV = nRT. The outlet flow state is supersonic flow, so the hydrogen mass flow calculation formula is: Where Q is the mass flow rate of hydrogen leakage, P0 is the ambient pressure, P is the initial pressure, k is the gas adiabatic index, C d is the gas leakage coefficient, A is the area of ​​the leak hole, M is the molar mass of hydrogen, R is the gas constant, and T is the temperature. During the leakage and diffusion process of hydrogen, the gas flows continuously, satisfies the conservation of mass and energy, and each component satisfies the component transport equation. The Realize k-ε turbulence model is selected based on the characteristics of the leak hole. The Realize k-ε turbulence model is described by the following equation: Where k is the turbulent kinetic energy, unit J, ε is the turbulent dissipation rate, ρ is the density, unit kg / m3, u j is the component of the velocity vector in the j direction, in m / s; x j is the transmission distance in the j direction, in m, μ, μ t are the dynamic viscosity coefficient and turbulent viscosity coefficient of the fluid, in kg / (m·s); C1, C2, C 1ε ,C 3ε is the transmission dissipation constant; σ k ,σ ε are the turbulent Prandtl numbers of turbulent kinetic energy and its dissipation rate, respectively; v is the correction term for turbulent dissipation rate; G k is the term for turbulent kinetic energy generated by the mean velocity gradient, G b is the term for turbulent kinetic energy generated by buoyancy, Y M =2ρεMa 2 The dissipation term caused by wave expansion in compressible turbulence, Ma is the turbulent Mach number, S k , Sε It is a user-defined heat source item. The boundary conditions are set as the ambient pressure and temperature are set to standard atmospheric pressure (101.325 kPa) and 300 K. The influence of gravity is considered and the speed is set to 9.8 m / s 2 , vertically downward. The pressure inside the fractured tube is 1MPa, and the fracture surface is a circular hole with a diameter of 20mm.

[0050] In some optional implementations, the above step S201 constructs the gas leakage condition, which can be achieved by the following steps: Step S2011: Obtain the open and closed state of the environment space where the above gas is located, the gas leakage duration, and the ventilation intensity of the ventilation hole; Step S2012: Combine different values ​​of the above open and closed state, different values ​​of the above gas leakage duration, and different values ​​of the above ventilation intensity to obtain multiple gas leakage conditions. The method constructs the gas leakage condition by the above steps, so that the real gas leakage scene can be simulated, thereby obtaining gas leakage data under different scenes.

[0051] Specifically, according to the opening and closing state of the hatch, the ventilation intensity of the vent and the leakage duration, a variety of leakage condition combinations are set, as shown in Table 1. In Fluent, it is set to pressurize the pressure outlet, and the gauge pressure is 0MPa. Because the front hatch leads to the cabin control room, in order to ensure that there is not too much hydrogen leakage to the control room and the cabin, the front hatch is half-open when it is opened, with a width and height of 0.3 meters and 2 meters respectively. It is set as a pressure outlet in Fluent, and the gauge pressure is 0MPa. The exhaust conditions of the vent can be set to natural ventilation and forced ventilation. The exhaust efficiency of forced ventilation is the mass flow rate of the gas, which is set as the mass flow outlet in Fluent. According to common ventilation conditions, boundary conditions are set, and gas monitoring points are inserted. In the transient numerical calculation process, the time step is fixed to 0.002s. According to the simulation results, a hydrogen leakage database is established, and the changes in hydrogen concentration over time at each hydrogen detection point under various conditions are established, and a hydrogen concentration change library under different leakage conditions is preliminarily constructed.

[0052] Table 1

[0053]

[0054] In some optional implementations, after the above step S203 determines the risk level of gas leakage according to the above state transition probability matrix, the above method further includes the following steps: step S204: obtaining the actual state data corresponding to each of the above future time series; step S205: verifying the above risk level according to the above actual state data to obtain a verification error, and determining that the above risk level is accurate when the above verification error is within the error tolerance range. The method verifies the prediction results of the Markov state prediction model through the above steps, so that the accuracy of the model can be verified and further modified according to the verification results.

[0055] Specifically, the Markov chain prediction model is repeatedly trained using the dynamic database of simulation results until the training is completed after the error tolerance is reached. The Markov state prediction model obtained through training is embedded into the on-site hydrogen leakage monitoring, early warning and prevention and control system. Figure 4 The system includes the stern, fuel cell compartment, control room, cabin and bow. The control room includes a central warning unit, and the fuel cell compartment includes a hydrogen fuel cell stack. Each of the above modules contains one or more hydrogen sensors. When a hydrogen leak occurs, the system will, based on the training results, have the functions of automatic sensor false alarm judgment, hydrogen leak point investigation, three-dimensional real-time display of hydrogen concentration, three-dimensional prediction of future hydrogen concentration events, hydrogen leak repair suggestions, escape route planning, and remote data upload to the laboratory. The laboratory receives the remote data and further corrects the CFD calculation results and graph neural network according to the deviation between the remote sensor measured data and the predicted results of the Markov chain prediction model, further improving the CFD calculation accuracy and model prediction accuracy.

[0056] In some other optional implementations, the above step S201 establishes a state space according to the above current gas concentration and the above gas concentration change rate, which can be achieved by the following steps: divide the risk level according to the above current gas concentration and the above gas concentration change rate, obtain the state corresponding to each risk level, and form the above state space with all the above states. The method establishes a state space through the above steps, so that the transition probability between different states can be calculated, thereby predicting the state corresponding to a certain time sequence in the future, judging whether danger occurs, and performing risk prediction.

[0057] In the specific implementation process, the risks can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the hydrogen concentration change rate exceed the 4% limit, such as Figure 5 As shown, the horizontal axis a represents the current hydrogen concentration, and the vertical axis is the hydrogen concentration change rate Δa, which is expressed as Among them, f(a, Δa) represents the risk level, the independent variable a represents the current hydrogen concentration, and Δa represents the rate of change of hydrogen concentration, forming the above four risk levels, namely, danger level 1, danger level 2, danger level 3 and danger level 4, that is, the four values ​​of f(a, Δa) above, and a state space containing four states is established according to the four risk levels: f states =[1,2,3,4], where f states Indicates risk status, 1, 2, 3, 4 correspond to risk levels.

[0058] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the gas leakage prediction method of the present application will be described in detail below in conjunction with specific embodiments.

[0059] This embodiment relates to a specific method for predicting gas leakage, such as Figure 6 As shown, the following steps are included:

[0060] Step S1: building a hydrogen leakage simulation model (gas environment model) for a hydrogen fuel cell ship and constructing a simulation model database;

[0061] Step S2: Markov chain prediction model;

[0062] Step S3: Preliminarily calculating the Markov chain state transition probability matrix (initial state transition probability matrix) through simulation database data;

[0063] Step S4: acquiring real-time data of the hydrogen sensor and calculating and updating the Markov chain state transition probability matrix, and dynamically updating the database data (state transition probability matrix);

[0064] Step S5: predicting the future risk level (risk level of gas leakage) according to the risk level model and the Markov chain state transition probability matrix;

[0065] Step S6: The early warning system assesses risks and issues early warnings.

[0066] This embodiment also relates to a flow chart of a Markov chain prediction model in a specific gas leakage prediction method, such as Figure 7 As shown, the following steps are included:

[0067] Step S7: constructing a risk state space according to the risk partition model (establishing a state space);

[0068] Step S8: Preliminarily construct a state transition probability matrix (initial state transition probability matrix) according to the simulation database;

[0069] Step S9: judging the current state through sensor data;

[0070] Step S10: Calculate and predict the future risk state through the Markov chain state transition probability matrix, dynamically update the database data and the state transition probability matrix, and update the state transition probability matrix in step S8;

[0071] Step S11: The early warning system assesses risks and issues early warnings.

[0072] The present application embodiment also provides a prediction device for gas leakage. It should be noted that the prediction device for gas leakage of the present application embodiment can be used to execute the prediction method for gas leakage provided by the present application embodiment. The device is used to implement the above-mentioned embodiments and preferred embodiments, and the descriptions have been made and will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0073] The following is an introduction to the gas leakage prediction device provided in the embodiments of the present application.

[0074] Figure 8 Schematic diagram of a gas leakage prediction device according to an embodiment of the present application. Figure 8 As shown, the device comprises:

[0075] The first acquisition unit 10 is used to acquire environmental data, establish a gas environment model according to the environmental data, construct a gas leakage condition, and acquire the current gas concentration and the gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing the environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage;

[0076] Specifically, the present application uses hydrogen as gas for analysis, which is mainly divided into a laboratory end and a field end. The laboratory end consists of a reduced model test bench, CFD simulation (Computational Fluid Dynamics, CFD simulation for short), and a Markov chain prediction model training server. Its basic process is as follows: According to the characteristics of high diffusivity and large buoyancy of hydrogen in the air, the initial hydrogen leakage sensor is arranged. At the same time, in the CFD simulation, geometric modeling, meshing, calculation model setting, initial condition setting, numerical calculation, graphical display and other processes are performed. The CFD simulation results are corrected with the laboratory measured data, and the calculation model setting, meshing, initial conditions, and calculation devices are corrected, and then the CFD simulation model is used to optimize the laboratory sensor arrangement. This is repeated until the model calculation deviation is within the allowable range of 1%, and the sensor arrangement is relatively reasonable (areas with large hydrogen concentrations and areas with large changes in hydrogen concentrations). This CFD model is used to simulate hydrogen leakage events in real scenes, and the Markov chain prediction model is trained using the simulation results. Taking the hydrogen fuel cell cabin as an example, the environmental data includes the length, width, height and other data of the fuel cell cabin, as well as the length, width, height and other data of the fuel cell stack. It can also include other geographical data required to establish the gas environment model. There are multiple sets of leakage conditions, which will be introduced in detail below. Through the gas environment model to simulate different leakage conditions, the current gas concentration and gas concentration change rate data under different conditions are obtained. The gas concentration change rate data can be calculated by the ratio of the difference in gas concentration within a certain period of time to the time. In this way, a database of hydrogen concentration changes under different leakage conditions is preliminarily constructed.

[0077] A calculation unit 20 is used to establish a state space according to the current gas concentration and the gas concentration change rate, and calculate an initial state transition probability matrix of the state space, and calculate a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents a plurality of time series after the current time series corresponding to the current gas concentration;

[0078] Specifically, a Markov chain prediction model is designed, and the current hydrogen concentration (current gas concentration) and the rate of change of hydrogen concentration (gas concentration change rate) are used as characteristic data. The risk can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the rate of change of hydrogen concentration exceed the 4% limit. According to the division, the initial state transfer probability matrix is ​​calculated, and then the basic calculation process and the matrix probability update are completed based on the initial state transfer probability matrix to obtain the state transfer probability matrix corresponding to the future time series. Finally, the predicted risk level is obtained, that is, the situation and risk level at the next moment can be predicted based on the current hydrogen concentration and the rate of change of hydrogen concentration, so as to carry out early warning and prevention and control. And due to the stable characteristics of the Markov chain state transfer matrix, multi-step prediction can be achieved, that is, the risk level corresponding to the future time series can be predicted, and the predicted data can be saved and written into the database, so as to adaptively update the existing database.

[0079] The output unit 30 is used to determine the risk level of the gas leakage according to the state transition probability matrix, and output the risk level to provide early warning and maintenance for the gas leakage.

[0080] Specifically, the Markov state prediction model obtained through training is finally embedded into the on-site hydrogen leakage monitoring, early warning and prevention and control system. When a hydrogen leakage occurs, the system will automatically judge sensor false alarms, check hydrogen leakage occurrence points, display hydrogen concentration in three dimensions in real time, predict hydrogen concentration in three dimensions in future event segments, provide hydrogen leakage repair suggestions, plan escape routes, and upload remote data to the laboratory end based on the training results. The laboratory end receives the remote data, and further corrects the CFD calculation results and graph neural network according to the deviation between the remote sensor measured data and the predicted results of the Markov chain prediction model, further improving the CFD calculation accuracy and model prediction accuracy.

[0081] Through this embodiment, environmental data can be obtained, a gas environment model can be established according to the environmental data, gas leakage conditions can be constructed, and under the gas leakage conditions, the current gas concentration and gas concentration change rate of the gas environment model can be obtained; a state space can be established according to the current gas concentration and gas concentration change rate, and the initial state transition probability matrix of the state space can be calculated, and the state transition probability matrix corresponding to the future time series can be calculated according to the initial state transfer matrix; the risk level of gas leakage can be determined according to the state transfer probability matrix, and the risk level can be output to warn and repair the gas leakage. Compared with the complex prediction device in the prior art, the model structure is simpler, easier to implement, and more efficient. At the same time, the requirements for the amount of data are lower, and only a small amount of feature data is required. Through subsequent continuous iteration and supplementation of measured data, the problems of less historical data, insufficient sequence integrity and low reliability can be solved, and it is also beneficial to reduce the time cost of system prediction and improve prediction efficiency. Therefore, the prediction efficiency of gas leakage can be improved, and the effect of timely prediction and prevention and control can be achieved.

[0082] In the specific implementation process, the calculation unit includes a first calculation module and a composition module. The first calculation module is used to calculate the transition probability from any state in the state space to all remaining states to obtain multiple state transition probabilities; the composition module is used to arrange the multiple state transition probabilities in a predetermined order to form the initial state transition probability matrix. The device calculates the initial state transition probability matrix through the above steps, so that the current state transition process can be represented.

[0083] Specifically, the risk can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the hydrogen concentration change rate exceed the 4% limit, forming four risk levels, and a state space containing four states is established according to the four risk levels: states =[1,2,3,4], where f states Indicates the risk state, 1, 2, 3, 4 correspond to the risk level. Then, according to the database data, the initial state transition probability matrix is ​​calculated as follows: Among them, T tm represents the initial state transition probability matrix, T ij represents the switching probability from state i to state j. States i and j correspond to the risk level states in the risk state space respectively, and satisfy the relationship

[0084] In some optional implementations, the calculation unit further includes a second calculation module and a third calculation module, the second calculation module is used to calculate the intermediate state transition probability matrix from the current time sequence to a time sequence before the future time sequence, and obtain multiple intermediate state transition probability matrices; the third calculation module is used to calculate the product of multiple intermediate state transition probability matrices, and obtain the state transition probability matrix corresponding to the future time sequence. The device calculates the state transition probability matrix corresponding to the future time sequence through the above steps to obtain the state corresponding to the future time sequence.

[0085] In the specific implementation process, after the preliminary state transition probability matrix is ​​established, the risk state of 2 to n time series can be predicted based on the state transition probability matrix. tm,n =T tm ·T tm,2 ·...·T tm,n-1 , where T tm,n is the state transition probability matrix of the nth time series. The risk level of the current time series is judged by the transition probability relationship of this probability matrix, so as to take corresponding early warning and prevention measures. At the same time, the actual state data after each step of prediction will be written into the database, and the state transition probability matrix T will continue to be dynamically updated. tm,n The Markov chain prediction model is trained repeatedly using the dynamic database of simulation results until the training is completed after reaching the allowable error range.

[0086] In some optional embodiments, the first acquisition unit includes an establishment module and a determination module, the establishment module is used to obtain the length, width and height dimensions of the environmental space where the gas is located, the leakage hole dimensions and the ventilation hole dimensions, and establish a grid model of the environmental space according to the length, width and height dimensions, the leakage hole dimensions and the ventilation hole dimensions; the determination module is used to encrypt each grid in the grid model, and determine the mass flow calculation formula and component transport equation of the gas to obtain the gas environment model. The device establishes a simulation model through the above steps, so that gas leakage, especially hydrogen leakage, in real situations can be simulated to predict future leakage.

[0087] Specifically, the environment space can be a fuel cell cabin, and a simulation model of the fuel cell cabin is established, such as Figure 3As shown in the figure, the fuel cell passenger ship is 30 meters long, 10 meters wide, and has 150 passenger seats. Its power source is mainly 40 120kW proton exchange membrane fuel cell groups, with a total installed power of 4.92MW (including 120kW reserve power). The length, width and height of a single stack are 1.07 meters, 0.762 meters and 1.99 meters respectively. It is placed in the fuel cell room on the deck. The fuel cell room, control cabin and cabin are all adjacent to each other on the second floor of the ship. A grid model of the fuel cell cabin is established according to the size of the fuel cell cabin, the exhaust port and the hatch. The entire calculation process is carried out using transient numerical calculation. In order to reduce the calculation time, save computing resources, and make the simulation easy to converge, the Icem tool (a professional CAE pre-processing software with powerful CAD model repair capabilities) is used to divide the structured grid, and the grids of the horizontal and longitudinal sections where the leakage holes and ventilation holes are located are encrypted, with a total of 358,949 grids. The leakage mode is continuous leakage caused by the rupture of the hydrogen supply pipeline in the fuel cell compartment. The hydrogen uses the ideal gas model, that is, PV = nRT. The outlet flow state is supersonic flow, so the hydrogen mass flow calculation formula is: Where Q is the mass flow rate of hydrogen leakage, P0 is the ambient pressure, P is the initial pressure, k is the gas adiabatic index, C d is the gas leakage coefficient, A is the area of ​​the leak hole, M is the molar mass of hydrogen, R is the gas constant, and T is the temperature. During the leakage and diffusion process of hydrogen, the gas flows continuously, satisfying the conservation of mass and energy, and each component satisfies the component transport equation. The Realize k-ε turbulence model is selected based on the characteristics of the leak hole. The Realize k-ε turbulence model is described by the following equation:

[0088] Where k is the turbulent kinetic energy, unit J, ε is the turbulent dissipation rate, ρ is the density, unit kg / m3, u j is the component of the velocity vector in the j direction, in m / s; x j is the transmission distance in the j direction, in m, μ, μ t are the dynamic viscosity coefficient and turbulent viscosity coefficient of the fluid, in kg / (m·s); C1, C2, C 1ε ,C 3ε is the transmission dissipation constant; σ k ,σ ε are the turbulent Prandtl numbers of turbulent kinetic energy and its dissipation rate, respectively; v is the correction term for turbulent dissipation rate; G k is the term for turbulent kinetic energy generated by the mean velocity gradient, G b is the term for turbulent kinetic energy generated by buoyancy, Y M =2ρεMa 2The dissipation term caused by wave expansion in compressible turbulence, Ma is the turbulent Mach number, S k , S ε It is a user-defined heat source item. The boundary conditions are set as the ambient pressure and temperature are set to standard atmospheric pressure (101.325 kPa) and 300 K. The influence of gravity is considered and the speed is set to 9.8 m / s 2 , vertically downward. The pressure inside the fractured tube is 1MPa, and the fracture surface is a circular hole with a diameter of 20mm.

[0089] In some optional embodiments, the first acquisition unit includes an acquisition module and a combination module, the acquisition module is used to acquire the opening and closing state of the environment space where the gas is located, the gas leakage duration and the ventilation intensity of the ventilation hole; the combination module is used to combine different values ​​of the opening and closing state, different values ​​of the gas leakage duration and different values ​​of the ventilation intensity to obtain multiple gas leakage conditions. The device constructs gas leakage conditions through the above steps, so that it can simulate real gas leakage scenarios, thereby acquiring gas leakage data under different scenarios.

[0090] Specifically, according to the opening and closing state of the hatch, the ventilation intensity of the vent and the leakage duration, a variety of leakage condition combinations are set, as shown in Table 1. In Fluent, it is set to pressurize the pressure outlet, and the gauge pressure is 0MPa. Because the front hatch leads to the cabin control room, in order to ensure that there is not too much hydrogen leakage to the control room and the cabin, the front hatch is half-open when it is opened, with a width and height of 0.3 meters and 2 meters respectively. It is set as a pressure outlet in Fluent, and the gauge pressure is 0MPa. The exhaust conditions of the vent can be set to natural ventilation and forced ventilation. The exhaust efficiency of forced ventilation is the mass flow rate of the gas, which is set as the mass flow outlet in Fluent. According to common ventilation conditions, boundary conditions are set, and gas monitoring points are inserted. In the transient numerical calculation process, the time step is fixed to 0.002s. According to the simulation results, a hydrogen leakage database is established, and the changes in hydrogen concentration over time at each hydrogen detection point under various conditions are established, and a hydrogen concentration change library under different leakage conditions is preliminarily constructed.

[0091] In some optional implementations, the device further includes a second acquisition unit and a determination unit, the second acquisition unit is used to acquire the actual state data corresponding to each of the future time series; the determination unit is used to verify the risk level according to the actual state data to obtain a verification error, and when the verification error is within the error tolerance range, the risk level is determined to be accurate. The device verifies the prediction results of the Markov state prediction model through the above steps, so that the accuracy of the model can be verified and further modified according to the verification results.

[0092] Specifically, the Markov chain prediction model is repeatedly trained using the dynamic database of simulation results until the training is completed after the error tolerance is reached. The Markov state prediction model obtained through training is embedded into the on-site hydrogen leakage monitoring, early warning and prevention and control system. Figure 4 The system includes the stern, fuel cell compartment, control room, cabin and bow. The control room includes a central warning unit, and the fuel cell compartment includes a hydrogen fuel cell stack. Each of the above modules contains one or more hydrogen sensors. When a hydrogen leak occurs, the system will, based on the training results, have the functions of automatic sensor false alarm judgment, hydrogen leak point investigation, three-dimensional real-time display of hydrogen concentration, three-dimensional prediction of future hydrogen concentration events, hydrogen leak repair suggestions, escape route planning, and remote data upload to the laboratory. The laboratory receives the remote data and further corrects the CFD calculation results and graph neural network according to the deviation between the remote sensor measured data and the predicted results of the Markov chain prediction model, further improving the CFD calculation accuracy and model prediction accuracy.

[0093] In some other optional implementations, the first acquisition unit further includes a classification module for classifying the risk level according to the current gas concentration and the gas concentration change rate, obtaining the state corresponding to each risk level, and forming the state space with all the states. The device establishes the state space through the above steps, so that the transition probability between different states can be calculated, thereby predicting the state corresponding to a certain time sequence in the future, judging whether danger occurs, and performing risk prediction.

[0094] In the specific implementation process, the risks can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the hydrogen concentration change rate exceed the 4% limit, such as Figure 5 As shown, the horizontal axis a represents the current hydrogen concentration, and the vertical axis is the hydrogen concentration change rate Δa, which is expressed as Among them, f(a, Δa) represents the risk level, the independent variable a represents the current hydrogen concentration, and Δa represents the rate of change of hydrogen concentration, forming the above four risk levels, namely, danger level 1, danger level 2, danger level 3 and danger level 4, that is, the four values ​​of f(a, Δa) above, and a state space containing four states is established according to the four risk levels: f states =[1,2,3,4], where f states Indicates risk status, 1, 2, 3, 4 correspond to risk levels.

[0095] The gas leakage prediction device includes a processor and a memory. The first acquisition unit, the calculation unit, the output unit, etc. are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions. The modules are all located in the same processor; or, the modules are located in different processors in any combination.

[0096] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problem that the hydrogen leakage prediction device in the prior art cannot efficiently predict can be solved by adjusting the kernel parameters.

[0097] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0098] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the gas leakage prediction device.

[0099] Specifically, the gas leakage prediction method includes:

[0100] Step S201, obtaining environmental data, establishing a gas environment model according to the environmental data, constructing a gas leakage condition, and obtaining the current gas concentration and gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing the environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage;

[0101] Specifically, the present application uses hydrogen as gas for analysis, which is mainly divided into a laboratory end and a field end. The laboratory end consists of a reduced model test bench, CFD simulation (Computational Fluid Dynamics, CFD simulation for short), and a Markov chain prediction model training server. Its basic process is as follows: According to the characteristics of high diffusivity and large buoyancy of hydrogen in the air, the initial hydrogen leakage sensor is arranged. At the same time, in the CFD simulation, geometric modeling, meshing, calculation model setting, initial condition setting, numerical calculation, graphical display and other processes are performed. The CFD simulation results are corrected with the laboratory measured data, and the calculation model setting, meshing, initial conditions, and calculation methods are corrected, and then the CFD simulation model is used to optimize the laboratory sensor arrangement. This is repeated until the model calculation deviation is within the allowable range of 1%, and the sensor arrangement is relatively reasonable (areas with large hydrogen concentrations and areas with large changes in hydrogen concentrations). This CFD model is used to simulate hydrogen leakage events in real scenes, and the Markov chain prediction model is trained using the simulation results. Taking the hydrogen fuel cell cabin as an example, the environmental data includes the length, width, height and other data of the fuel cell cabin, as well as the length, width, height and other data of the fuel cell stack. It can also include other geographical data required to establish the gas environment model. There are multiple sets of leakage conditions, which will be introduced in detail below. Through the gas environment model to simulate different leakage conditions, the current gas concentration and gas concentration change rate data under different conditions are obtained. The gas concentration change rate data can be calculated by the ratio of the difference in gas concentration within a certain period of time to the time. In this way, a database of hydrogen concentration changes under different leakage conditions is preliminarily constructed.

[0102] Step S202, establishing a state space according to the current gas concentration and the gas concentration change rate, and calculating an initial state transition probability matrix of the state space, and calculating a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents a plurality of time series after the current time series corresponding to the current gas concentration;

[0103] Specifically, a Markov chain prediction model is designed, and the current hydrogen concentration (current gas concentration) and the rate of change of hydrogen concentration (gas concentration change rate) are used as characteristic data. The risk can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the rate of change of hydrogen concentration exceed the 4% limit. According to the division, the initial state transfer probability matrix is ​​calculated, and then the basic calculation process and the matrix probability update are completed based on the initial state transfer probability matrix to obtain the state transfer probability matrix corresponding to the future time series. Finally, the predicted risk level is obtained, that is, the situation and risk level at the next moment can be predicted based on the current hydrogen concentration and the rate of change of hydrogen concentration, so as to carry out early warning and prevention and control. And due to the stable characteristics of the Markov chain state transfer matrix, multi-step prediction can be achieved, that is, the risk level corresponding to the future time series can be predicted, and the predicted data can be saved and written into the database, so as to adaptively update the existing database.

[0104] Step S203, determining the risk level of gas leakage according to the state transition probability matrix, and outputting the risk level to provide early warning and maintenance for gas leakage.

[0105] Specifically, the Markov state prediction model obtained through training is finally embedded into the on-site hydrogen leakage monitoring, early warning and prevention and control system. When a hydrogen leakage occurs, the system will automatically judge sensor false alarms, check hydrogen leakage occurrence points, display hydrogen concentration in three dimensions in real time, predict hydrogen concentration in three dimensions in future event segments, provide hydrogen leakage repair suggestions, plan escape routes, and upload remote data to the laboratory end based on the training results. The laboratory end receives the remote data, and further corrects the CFD calculation results and graph neural network according to the deviation between the remote sensor measured data and the predicted results of the Markov chain prediction model, further improving the CFD calculation accuracy and model prediction accuracy.

[0106] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, at least the following steps are implemented:

[0107] Step S201, obtaining environmental data, establishing a gas environment model according to the environmental data, constructing a gas leakage condition, and obtaining the current gas concentration and gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing the environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage;

[0108] Specifically, the present application uses hydrogen as gas for analysis, which is mainly divided into a laboratory end and a field end. The laboratory end consists of a reduced model test bench, CFD simulation (Computational Fluid Dynamics, CFD simulation for short), and a Markov chain prediction model training server. Its basic process is as follows: According to the characteristics of high diffusivity and large buoyancy of hydrogen in the air, the initial hydrogen leakage sensor is arranged. At the same time, in the CFD simulation, geometric modeling, meshing, calculation model setting, initial condition setting, numerical calculation, graphical display and other processes are performed. The CFD simulation results are corrected with the laboratory measured data, and the calculation model setting, meshing, initial conditions, and calculation methods are corrected, and then the CFD simulation model is used to optimize the laboratory sensor arrangement. This is repeated until the model calculation deviation is within the allowable range of 1%, and the sensor arrangement is relatively reasonable (areas with large hydrogen concentrations and areas with large changes in hydrogen concentrations). This CFD model is used to simulate hydrogen leakage events in real scenes, and the Markov chain prediction model is trained using the simulation results. Taking the hydrogen fuel cell cabin as an example, the environmental data includes the length, width, height and other data of the fuel cell cabin, as well as the length, width, height and other data of the fuel cell stack. It can also include other geographical data required to establish the gas environment model. There are multiple sets of leakage conditions, which will be introduced in detail below. Through the gas environment model to simulate different leakage conditions, the current gas concentration and gas concentration change rate data under different conditions are obtained. The gas concentration change rate data can be calculated by the ratio of the difference in gas concentration within a certain period of time to the time. In this way, a database of hydrogen concentration changes under different leakage conditions is preliminarily constructed.

[0109] Step S202, establishing a state space according to the current gas concentration and the gas concentration change rate, and calculating an initial state transition probability matrix of the state space, and calculating a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents a plurality of time series after the current time series corresponding to the current gas concentration;

[0110] Specifically, a Markov chain prediction model is designed, and the current hydrogen concentration (current gas concentration) and the rate of change of hydrogen concentration (gas concentration change rate) are used as characteristic data. The risk can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the rate of change of hydrogen concentration exceed the 4% limit. According to the division, the initial state transfer probability matrix is ​​calculated, and then the basic calculation process and the matrix probability update are completed based on the initial state transfer probability matrix to obtain the state transfer probability matrix corresponding to the future time series. Finally, the predicted risk level is obtained, that is, the situation and risk level at the next moment can be predicted based on the current hydrogen concentration and the rate of change of hydrogen concentration, so as to carry out early warning and prevention and control. And due to the stable characteristics of the Markov chain state transfer matrix, multi-step prediction can be achieved, that is, the risk level corresponding to the future time series can be predicted, and the predicted data can be saved and written into the database, so as to adaptively update the existing database.

[0111] Step S203, determining the risk level of gas leakage according to the state transition probability matrix, and outputting the risk level to provide early warning and maintenance for gas leakage.

[0112] Specifically, the Markov state prediction model obtained through training is finally embedded into the on-site hydrogen leakage monitoring, early warning and prevention and control system. When a hydrogen leakage occurs, the system will automatically judge sensor false alarms, check hydrogen leakage occurrence points, display hydrogen concentration in three dimensions in real time, predict hydrogen concentration in three dimensions in future event segments, provide hydrogen leakage repair suggestions, plan escape routes, and upload remote data to the laboratory end based on the training results. The laboratory end receives the remote data, and further corrects the CFD calculation results and graph neural network according to the deviation between the remote sensor measured data and the predicted results of the Markov chain prediction model, further improving the CFD calculation accuracy and model prediction accuracy.

[0113] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0114] The present application also provides a computer program product, including a computer program. When the computer program is executed by a processor, the steps of the above method in each embodiment of the present application are implemented:

[0115] Step S201, obtaining environmental data, establishing a gas environment model according to the environmental data, constructing a gas leakage condition, and obtaining the current gas concentration and gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing the environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage;

[0116] Specifically, the present application uses hydrogen as gas for analysis, which is mainly divided into a laboratory end and a field end. The laboratory end consists of a reduced model test bench, CFD simulation (Computational Fluid Dynamics, CFD simulation for short), and a Markov chain prediction model training server. Its basic process is as follows: According to the characteristics of high diffusivity and large buoyancy of hydrogen in the air, the initial hydrogen leakage sensor is arranged. At the same time, in the CFD simulation, geometric modeling, meshing, calculation model setting, initial condition setting, numerical calculation, graphical display and other processes are performed. The CFD simulation results are corrected with the laboratory measured data, and the calculation model setting, meshing, initial conditions, and calculation methods are corrected, and then the CFD simulation model is used to optimize the laboratory sensor arrangement. This is repeated until the model calculation deviation is within the allowable range of 1%, and the sensor arrangement is relatively reasonable (areas with large hydrogen concentrations and areas with large changes in hydrogen concentrations). This CFD model is used to simulate hydrogen leakage events in real scenes, and the Markov chain prediction model is trained using the simulation results. Taking the hydrogen fuel cell cabin as an example, the environmental data includes the length, width, height and other data of the fuel cell cabin, as well as the length, width, height and other data of the fuel cell stack. It can also include other geographical data required to establish the gas environment model. There are multiple sets of leakage conditions, which will be introduced in detail below. Through the gas environment model to simulate different leakage conditions, the current gas concentration and gas concentration change rate data under different conditions are obtained. The gas concentration change rate data can be calculated by the ratio of the difference in gas concentration within a certain period of time to the time. In this way, a database of hydrogen concentration changes under different leakage conditions is preliminarily constructed.

[0117] Step S202, establishing a state space according to the current gas concentration and the gas concentration change rate, and calculating an initial state transition probability matrix of the state space, and calculating a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents a plurality of time series after the current time series corresponding to the current gas concentration;

[0118] Specifically, a Markov chain prediction model is designed, and the current hydrogen concentration (current gas concentration) and the rate of change of hydrogen concentration (gas concentration change rate) are used as characteristic data. The risk can be divided into four main areas and risk levels based on whether the current hydrogen concentration and the rate of change of hydrogen concentration exceed the 4% limit. According to the division, the initial state transfer probability matrix is ​​calculated, and then the basic calculation process and the matrix probability update are completed based on the initial state transfer probability matrix to obtain the state transfer probability matrix corresponding to the future time series. Finally, the predicted risk level is obtained, that is, the situation and risk level at the next moment can be predicted based on the current hydrogen concentration and the rate of change of hydrogen concentration, so as to carry out early warning and prevention and control. And due to the stable characteristics of the Markov chain state transfer matrix, multi-step prediction can be achieved, that is, the risk level corresponding to the future time series can be predicted, and the predicted data can be saved and written into the database, so as to adaptively update the existing database.

[0119] Step S203, determining the risk level of gas leakage according to the state transition probability matrix, and outputting the risk level to provide early warning and maintenance for gas leakage.

[0120] Specifically, the Markov state prediction model obtained through training is finally embedded into the on-site hydrogen leakage monitoring, early warning and prevention and control system. When a hydrogen leakage occurs, the system will automatically judge sensor false alarms, check hydrogen leakage occurrence points, display hydrogen concentration in three dimensions in real time, predict hydrogen concentration in three dimensions in future event segments, provide hydrogen leakage repair suggestions, plan escape routes, and upload remote data to the laboratory end based on the training results. The laboratory end receives the remote data, and further corrects the CFD calculation results and graph neural network according to the deviation between the remote sensor measured data and the predicted results of the Markov chain prediction model, further improving the CFD calculation accuracy and model prediction accuracy.

[0121] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0122] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0123] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0124] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0126] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0127] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0128] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0129] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0130] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0131] 1) In the prediction method of gas leakage of the present application, environmental data can be obtained, a gas environment model can be established according to the environmental data, gas leakage conditions can be constructed, and under the gas leakage conditions, the current gas concentration and gas concentration change rate of the gas environment model can be obtained; a state space can be established according to the current gas concentration and gas concentration change rate, and the initial state transition probability matrix of the state space can be calculated, and the state transition probability matrix corresponding to the future time series can be calculated according to the initial state transfer matrix; the risk level of gas leakage can be determined according to the state transfer probability matrix, and the risk level can be output to warn and repair the gas leakage. Compared with the complex prediction method in the prior art, the model structure is simpler, easier to implement, and more efficient. At the same time, the requirements for the amount of data are lower, and only a small amount of feature data is required. Through subsequent continuous iteration and supplementation of measured data, the problems of less historical data, insufficient sequence integrity and low reliability can be solved, and it is also beneficial to reduce the time cost of system prediction and improve prediction efficiency. Therefore, the prediction efficiency of gas leakage can be improved, and the effect of timely prediction and prevention and control can be achieved.

[0132] 2) In the prediction device for gas leakage of the present application, environmental data can be obtained, a gas environment model can be established according to the environmental data, gas leakage conditions can be constructed, and under the gas leakage conditions, the current gas concentration and gas concentration change rate of the gas environment model can be obtained; a state space can be established according to the current gas concentration and the gas concentration change rate, and the initial state transition probability matrix of the state space can be calculated, and the state transition probability matrix corresponding to the future time series can be calculated according to the initial state transfer matrix; the risk level of gas leakage can be determined according to the state transfer probability matrix, and the risk level can be output to warn and repair the gas leakage. Compared with the complex prediction device in the prior art, the model structure is simpler, easier to implement, and more efficient. At the same time, the requirements for the amount of data are lower, and only a small amount of feature data is required. Through subsequent continuous iteration and supplementation of measured data, the problems of less historical data, insufficient sequence integrity and low reliability can be solved, and it is also beneficial to reduce the time cost of system prediction and improve prediction efficiency. Therefore, the prediction efficiency of gas leakage can be improved, and the effect of timely prediction and prevention and control can be achieved.

[0133] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for predicting gas leakage, characterized in that: include: Acquire environmental data, establish a gas environment model according to the environmental data, construct a gas leakage condition, and acquire the current gas concentration and the gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing the environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage; Establishing a state space according to the current gas concentration and the gas concentration change rate, and calculating an initial state transition probability matrix of the state space, and calculating a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents a plurality of time series after a current time series corresponding to the current gas concentration; The risk level of the gas leakage is determined according to the state transition probability matrix, and the risk level is output to provide early warning and maintenance for the gas leakage.

2. The prediction method according to claim 1, characterized in that: Calculating the initial state transition probability matrix of the state space includes: Calculating the transition probability from any state in the state space to all remaining states to obtain multiple state transition probabilities; Arrange the plurality of state transition probabilities in a predetermined order to form the initial state transition probability matrix.

3. The prediction method according to claim 1, characterized in that: Calculating the state transition probability matrix corresponding to the future time series according to the initial state transition probability matrix includes: Calculating an intermediate state transition probability matrix from the current time sequence to a time sequence before the future time sequence, and obtaining a plurality of the intermediate state transition probability matrices; The product of a plurality of the intermediate state transition probability matrices is calculated to obtain the state transition probability matrix corresponding to the future time sequence.

4. The prediction method according to claim 1, characterized in that: Acquiring environmental data and establishing a gas environment model according to the environmental data includes: Obtaining the length, width and height dimensions, leakage hole dimensions and ventilation hole dimensions of the environment space where the gas is located, and establishing a grid model of the environment space according to the length, width and height dimensions, the leakage hole dimensions and the ventilation hole dimensions; Each grid in the grid model is encrypted, and the mass flow calculation formula and component transport equation of the gas are determined to obtain the gas environment model.

5. The prediction method according to claim 1, characterized in that: Construct gas leak conditions, including: Obtaining the open and closed state of the environment space where the gas is located, the duration of gas leakage, and the ventilation intensity of the ventilation holes; A plurality of gas leakage conditions are obtained by combining different values ​​of the opening and closing state, different values ​​of the gas leakage duration, and different values ​​of the ventilation intensity.

6. The prediction method according to claim 1, characterized in that: After determining the risk level of gas leakage according to the state transition probability matrix, the method further includes: Acquire actual state data corresponding to each future time series; The risk level is verified according to the actual status data to obtain a verification error, and when the verification error is within an allowable error range, it is determined that the risk level is accurate.

7. The prediction method according to claim 1, characterized in that: Establishing a state space according to the current gas concentration and the gas concentration change rate includes: Risk levels are divided according to the current gas concentration and the gas concentration change rate to obtain states corresponding to each risk level, and all the states constitute the state space.

8. A gas leakage prediction device, characterized in that: include: A first acquisition unit is used to acquire environmental data, establish a gas environment model according to the environmental data, construct a gas leakage condition, and acquire a current gas concentration and a gas concentration change rate of the gas environment model under the gas leakage condition, wherein the environmental data is geographic data characterizing an environmental space where the gas is located, and the gas leakage condition is a condition characterizing the intensity of the gas leakage; a calculation unit, configured to establish a state space according to the current gas concentration and the gas concentration change rate, calculate an initial state transition probability matrix of the state space, and calculate a state transition probability matrix corresponding to a future time series according to the initial state transition probability matrix, wherein the future time series represents a plurality of time series after a current time series corresponding to the current gas concentration; The output unit is used to determine the risk level of gas leakage according to the state transition probability matrix, and output the risk level to provide early warning and maintenance for gas leakage.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the gas leakage prediction method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the gas leakage prediction method according to any one of claims 1 to 7.

Citation Information

Cited By

  • Temperature measurement method and system for underground mining electrical equipment based on infrared technology

    CN119756594A

  • Generator hydrogen leakage monitoring method, device and equipment based on CFD simulation and medium

    CN119830810A

  • Intelligent fire-fighting big data analysis method based on cloud service

    CN119939192A

  • A cloud service-based intelligent fire-fighting big data analysis method

    CN119939192B

  • Intelligent early warning method and system based on hydrogen leakage danger level in industrial production

    CN120148200A