A fuel cell vehicle hydrogen leakage monitoring method, system, device and medium

Abstract

The application discloses a kind of fuel cell vehicle hydrogen leakage monitoring method, system, equipment and medium, it is related to hydrogen leakage monitoring field, method includes obtaining real-time sensor data on hydrogen pipeline;The sensor data includes pressure data, temperature data and humidity data;Utilize empirical mode decomposition method to the sensor data is preprocessed and first-order derivation, obtain the sensor rate of change numerical value of the position;The sensor rate of change numerical value includes pressure rate of change numerical value, temperature rate of change numerical value and humidity rate of change numerical value;According to the sensor rate of change numerical value of same time period, calculate the rate of change parameter value of the position of sensor;According to the rate of change parameter value of the position of sensor, preliminary positioning is carried out, and preliminary positioning result is obtained;According to the sensor rate of change numerical value, positioning is carried out using neural network model, and the positioning result of leakage point is obtained.The application improves the timeliness and reliability of hydrogen leakage monitoring.

Description

Technical Field

[0001] This invention relates to the field of hydrogen leakage monitoring, and in particular to a method, system, device, and medium for monitoring hydrogen leakage in fuel cell vehicles. Background Technology

[0002] Compared to electric vehicles powered by batteries, fuel cell vehicles have a series of advantages such as zero emissions, high efficiency, low noise, rapid refueling, and long driving range. As the main application carrier of hydrogen energy in the transportation field, fuel cell vehicles have become the next blue ocean for the automotive industry.

[0003] However, due to the high flammability and explosiveness of hydrogen, the technical requirements for the safety of fuel cell vehicles are extremely stringent. To ensure hydrogen safety in fuel cell vehicles, real-time monitoring of the hydrogen system's operating status is necessary. In the hydrogen system pipelines of fuel cell vehicles, complex operating conditions, high gas pressure within the pipelines, and hydrogen embrittlement can lead to high-pressure hydrogen leaks caused by wear or failure of internal vehicle components. If hydrogen leak diagnosis is inaccurate and leak points are not detected in a timely manner, it will not only result in resource waste but also pose serious safety hazards. Research on hydrogen leak diagnosis and leak point location technologies for fuel cell vehicles has extremely important practical significance for resource management, personnel safety, and equipment maintenance.

[0004] Existing methods for monitoring hydrogen leaks often employ high-precision hydrogen concentration sensors, typically deploying multiple sensors near the fuel cell engine, passenger cabin ceiling, and hydrogen storage tanks to obtain hydrogen concentration levels at the monitored locations. However, these methods are slow to react and have low monitoring efficiency. Another approach involves deploying pressure sensors in high-pressure hydrogen storage tanks to detect hydrogen leaks through pressure changes. However, in actual fuel cell operation, hydrogen pressure fluctuations are subject to randomness, leading to insufficient accuracy in the assessment. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, device, and medium for monitoring hydrogen leaks in fuel cell vehicles, which can improve the timeliness and reliability of hydrogen leak monitoring.

[0006] To achieve the above objectives, the present invention provides the following solution:

[0007] A method for monitoring hydrogen leakage in a fuel cell vehicle includes:

[0008] Acquire real-time sensor data from the hydrogen pipeline; the sensor data includes pressure data, temperature data, and humidity data;

[0009] The sensor data is preprocessed and its first derivative is calculated using the empirical mode decomposition method to obtain the sensor change rate values ​​at the sensor location; the sensor change rate values ​​include pressure change rate values, temperature change rate values, and humidity change rate values.

[0010] Calculate the rate of change parameter value of the sensor's location based on the sensor's rate of change value over the same time period;

[0011] Preliminary positioning is performed based on the rate of change parameter value of the sensor's location, resulting in a preliminary positioning result.

[0012] The location of the leak point is obtained by using a neural network model based on the sensor's rate of change value.

[0013] Optionally, the hydrogen pipeline is the pipeline between the hydrogen storage cylinder and the fuel cell of the fuel cell vehicle; the hydrogen pipeline is planned into three parts with the hydrogen storage cylinder, the first-stage pressure reducing valve, the second-stage pressure reducing valve and the fuel cell as nodes.

[0014] Optionally, the sensor data is preprocessed and its first derivative is calculated using the empirical mode decomposition method to obtain the sensor's rate of change value at the sensor's location, specifically including:

[0015] The sensor data is processed by removing random fluctuation signals using the empirical mode decomposition method to obtain preprocessed data.

[0016] The first derivative of the preprocessed data is taken to obtain the sensor change rate value at the sensor location.

[0017] Optionally, the rate of change parameter value of the sensor's location is calculated based on the sensor's rate of change values ​​over the same time period, specifically including:

[0018] The change rate parameter value of the sensor location is calculated based on the sensor change rate value during the same time period and the weights under different operating conditions.

[0019] Optionally, the neural network model is an LSTM neural network.

[0020] The present invention also provides a hydrogen leakage monitoring system for fuel cell vehicles, comprising:

[0021] The acquisition module is used to acquire real-time sensor data on the hydrogen pipeline; the sensor data includes pressure data, temperature data, and humidity data.

[0022] The preprocessing and first-order derivative module is used to preprocess and perform first-order derivative calculation on the sensor data using the empirical mode decomposition method to obtain the sensor change rate value at the sensor location; the sensor change rate value includes the pressure change rate value, temperature change rate value, and humidity change rate value.

[0023] The calculation module is used to calculate the rate of change parameter value of the sensor's location based on the sensor's rate of change value over the same time period;

[0024] The preliminary positioning module is used to perform preliminary positioning based on the change rate parameter value of the sensor's location and obtain preliminary positioning results.

[0025] The final positioning module is used to locate the leak point by using a neural network model based on the sensor's rate of change value.

[0026] Optionally, the hydrogen pipeline is the pipeline between the hydrogen storage cylinder and the fuel cell of the fuel cell vehicle; the hydrogen pipeline is planned into three parts with the hydrogen storage cylinder, the first-stage pressure reducing valve, the second-stage pressure reducing valve and the fuel cell as nodes.

[0027] Optionally, the preprocessing and first-order derivative module specifically includes:

[0028] The preprocessing unit is used to remove random fluctuation signals from the sensor data using the empirical mode decomposition method to obtain preprocessed data.

[0029] The first-order derivative unit is used to perform first-order derivative on the preprocessed data to obtain the sensor change rate value at the sensor location.

[0030] The present invention also provides an electronic device, comprising:

[0031] One or more processors;

[0032] A storage device on which one or more programs are stored;

[0033] When the one or more programs are executed by the one or more processors, the one or more processors implement the method.

[0034] The present invention also provides a computer storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described thereon.

[0035] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0036] This invention first uses a threshold parameter to perform preliminary positioning based on the rate of change parameter value of the sensor's location, obtaining a preliminary positioning result and quickly performing preliminary positioning, thereby improving the response speed of leak detection. Then, based on the rate of change parameter value, a neural network model is used to perform positioning, obtaining the location result of the leak point, performing precise positioning, and improving the reliability of positioning. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 A schematic diagram of a hydrogen leakage monitoring method for fuel cell vehicles;

[0039] Figure 2 A schematic diagram of a hydrogen leakage monitoring system and sensor arrangement for a fuel cell vehicle.

[0040] Figure 3 A schematic diagram of the sensor pressure change rate when hydrogen leakage occurs in Region I;

[0041] Figure 4 The flowchart illustrates the hydrogen leakage monitoring method for fuel cell vehicles provided by this invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] The purpose of this invention is to provide a method, system, device, and medium for monitoring hydrogen leaks in fuel cell vehicles, which can improve the timeliness and reliability of hydrogen leak monitoring.

[0044] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] like Figure 1 and Figure 4 As shown, the present invention provides a method for monitoring hydrogen leakage in a fuel cell vehicle, comprising:

[0046] Step 101: Acquire real-time sensor data on the hydrogen pipeline; the sensor data includes pressure data, temperature data, and humidity data. The hydrogen pipeline is the pipeline between the hydrogen storage cylinder and the fuel cell in the fuel cell vehicle; the hydrogen pipeline is planned into three parts with the hydrogen storage cylinder, the first-stage pressure reducing valve, the second-stage pressure reducing valve, and the fuel cell as nodes.

[0047] Multiple sensors are installed on the hydrogen pipeline between the high-pressure hydrogen storage tank and the fuel cell in a fuel cell vehicle to acquire pressure, temperature, and humidity data signals at different locations in the pipeline, monitoring the real-time pressure, temperature, and humidity within the pipeline. Based on the ultra-high pressure state in the hydrogen storage tank, the hydrogen gas needs to pass through several pressure reducing valves before finally entering the fuel cell. A two-stage pressure reducing valve system, typical of fuel cell vehicle hydrogen systems, is adopted. The hydrogen pipeline is divided into three parts, defined as Region I, Region II, and Region III, with the hydrogen storage tank, first-stage pressure reducing valve, second-stage pressure reducing valve, and fuel cell as nodes. Two sensors are installed in each region of the hydrogen pipeline. The sensors should be located where the hydrogen flow rate is stable, and the distance between the two sensors in each pipeline segment should be as large as possible. A fuel cell vehicle hydrogen leakage monitoring system and sensor arrangement method are described below. Figure 2 As shown, the temperature, pressure, and humidity data collected by each sensor are denoted as Ti, Pi, and Hi, i = 1, 2, 3, ..., 6, and the sensor numbers are 1 to 6.

[0048] Step 102: Preprocess the sensor data and perform first-order differentiation using the empirical mode decomposition method to obtain the sensor change rate values ​​at the sensor location; the sensor change rate values ​​include pressure change rate values, temperature change rate values, and humidity change rate values.

[0049] Step 102 specifically includes: removing random fluctuation signals from the sensor data using the empirical mode decomposition method to obtain preprocessed data; and taking the first derivative of the preprocessed data to obtain the sensor change rate value at the sensor location.

[0050] The pressure, temperature, and humidity signals acquired by the sensor are preprocessed using the Empirical Mode Decomposition (EDM) method. The first-order derivative of the preprocessed data is then performed to obtain the rates of change of pressure, temperature, and humidity at the sensor location. The data preprocessing method involves extracting high-frequency and low-frequency components from the sensor signals using EDM. Through iterative filtering, the non-stationary signal is decomposed into a residual sequence (RES) and a series of intrinsic mode functions (IMFs), thus removing random fluctuations. The first-order derivative of the preprocessed data yields the rates of change of pressure, temperature, and humidity at the sensor location, denoted as... i = 1, 2, 3, ..., 6.

[0051] Step 103: Calculate the rate of change parameter value of the sensor location based on the sensor's rate of change values ​​over the same time period.

[0052] Step 103 specifically includes: calculating the rate of change parameter value of the sensor location based on the sensor change rate value in the same time period and the weights under different operating conditions.

[0053] Observe the trends of the rates of change of pressure, temperature, and humidity over time. For example... Figure 3 As shown, under normal circumstances, the pressure change rate curves of the sensors are basically stable with small fluctuations. When the pressure change rates of sensors 1 and 2 change abruptly and gradually decrease over time, it can be preliminarily determined that there is a hydrogen leak in area I of the fuel cell hydrogen pipeline. To improve the accuracy of the judgment, the combined effects of multiple parameters such as pressure, temperature, and humidity are considered.

[0054] Constructor

[0055] in, Let i represent the rate of change of pressure, temperature, and humidity at each sensor after the first derivative, i = 1, 2, 3, ..., 6. Different weights C = {c1, c2, c3} are assigned to these parameters under different operating conditions.

[0056] Step 104: Perform preliminary positioning based on the rate of change parameter value of the sensor's location to obtain preliminary positioning results.

[0057] The rate of change parameters of each sensor location are calculated based on the pressure, temperature and humidity change rates within the same time period. Whether the rate of change parameter value exceeds the threshold is used to determine whether a leak has occurred and to achieve preliminary location of the leak area. The hydrogen leak area of ​​the fuel cell vehicle is determined by comprehensively considering the range of pressure, temperature and humidity change rates and the sensor location.

[0058] The threshold can be defined as 1.1U' i , where U' i The change rate parameter values ​​are for normal operation of the fuel cell without any leakage, i = 1, 2, 3, ..., 6.

[0059] When the rate of change parameter exceeds the threshold, it is preliminarily determined that there is a leak in the hydrogen system of the fuel cell vehicle. The vehicle issues a hydrogen leak fault warning message. By observing the value of i, it is determined that the leak occurs in the corresponding sensor location area. That is, when i = 1, 2, the hydrogen leak occurs in area I; when i = 3, 4, the hydrogen leak occurs in area II; and when i = 5, 6, the hydrogen leak occurs in area III.

[0060] Step 105: Based on the sensor's rate of change value, a neural network model is used to locate the leak point. The neural network model is an LSTM neural network.

[0061] By inputting the sensor pressure, temperature, and humidity change rates within the leak area into the LSTM neural network model, the specific location of the leak point can be obtained, improving the speed and accuracy of hydrogen leak diagnosis.

[0062] 1000 simulated leak points were randomly generated in regions I, II, and III using the Monte Carlo sampling method. The input features of the LSTM neural network were extracted, including the pressure, temperature, and humidity change rate data of the sensors in the corresponding region when each leak point leaked. The output features of the LSTM neural network were the specific location coordinates of the leak point.

[0063] The LSTM neural network model was trained using 70% of the dataset as the training set. The input features were the pressure, temperature, and humidity change rates of the corresponding sensors in the leak area at each leak point, and the output features were the specific location coordinates of the leak point. The test set was used with 30% of the dataset, and the deviation between the predicted and actual leak point location coordinates was used as the evaluation metric to demonstrate the effectiveness of the LSTM neural network model.

[0064] The pressure, temperature, and humidity change rate data from the sensors in the leak area are input into a trained LSTM neural network to obtain the predicted coordinates of the leak point and display the leak point location to the vehicle owner.

[0065] This invention aims to optimize the arrangement of sensors in hydrogen pipelines. By processing sensor signals, it can comprehensively determine the area where hydrogen leaks occur based on multiple parameters such as pressure, temperature, and humidity. Furthermore, by combining a neural network model, it can further achieve precise location of hydrogen leak points, thereby improving the speed and accuracy of hydrogen leak monitoring in fuel cell vehicles and ensuring that passengers can take timely safety measures.

[0066] The present invention also provides a hydrogen leakage monitoring system for fuel cell vehicles, comprising:

[0067] The acquisition module is used to acquire real-time sensor data on the hydrogen pipeline; the sensor data includes pressure data, temperature data, and humidity data.

[0068] The preprocessing and first-order derivative module is used to preprocess and perform first-order derivative calculation on the sensor data using the empirical mode decomposition method to obtain the sensor change rate value at the sensor location; the sensor change rate value includes the pressure change rate value, temperature change rate value, and humidity change rate value.

[0069] The calculation module is used to calculate the rate of change parameter value of the sensor's location based on the sensor's rate of change value over the same time period.

[0070] The preliminary positioning module is used to perform preliminary positioning based on the rate of change parameter value of the sensor's location, and obtain preliminary positioning results.

[0071] The final positioning module is used to locate the leak point by using a neural network model based on the sensor's rate of change value.

[0072] As an optional implementation, the hydrogen pipeline is the pipeline between the hydrogen storage cylinder and the fuel cell of the fuel cell vehicle; the hydrogen pipeline is planned into three parts with the hydrogen storage cylinder, the first-stage pressure reducing valve, the second-stage pressure reducing valve and the fuel cell as nodes.

[0073] As an optional implementation, the preprocessing and first-order derivative module specifically includes:

[0074] The preprocessing unit is used to remove random fluctuation signals from the sensor data using the empirical mode decomposition method to obtain preprocessed data.

[0075] The first-order derivative unit is used to perform first-order derivative on the preprocessed data to obtain the sensor change rate value at the sensor location.

[0076] The present invention also provides an electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon; wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described herein.

[0077] The present invention also provides a computer storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described thereon.

[0078] This invention employs a two-layer system for hydrogen leak monitoring: First, by using a combination of multiple parameters—pressure, temperature, and humidity—to reduce the impact of sudden changes in any single parameter, the rate of change can quickly determine whether a leak has occurred and preliminarily locate the leak area, improving the leak monitoring response speed. This facilitates alerting users and enabling them to take emergency evasive action, ensuring the safety of users and vehicles, and demonstrating high timeliness. Second, through a further LSTM neural network model, the specific location of the leak point can be pinpointed. This improved accuracy is beneficial for determining the cause of accidents in fuel cell vehicles and for maintenance personnel, resulting in a system with high precision and reliability.

[0079] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0080] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for monitoring hydrogen leakage in a fuel cell vehicle, characterized in that, include: Acquire real-time sensor data from the hydrogen pipeline; the sensor data includes pressure data, temperature data, and humidity data; The sensor data is preprocessed and its first derivative is calculated using the empirical mode decomposition method to obtain the sensor change rate values ​​at the sensor location; the sensor change rate values ​​include pressure change rate values, temperature change rate values, and humidity change rate values. The calculation of the rate of change parameter value of the sensor location based on the sensor's rate of change values ​​over the same time period specifically includes: calculating the rate of change parameter value of the sensor location based on the sensor's rate of change values ​​over the same time period and the weights of different operating parameters; the rate of change parameter value is calculated using the following formula: ； Among them, T i P i H i The data represents the temperature, pressure, and humidity collected from each sensor, where i is the sensor number. Weights for different operating condition parameters, These represent the rate of change of pressure, temperature, and humidity at each sensor after the first derivative; This refers to the value of the rate of change parameter. Preliminary location is determined based on the rate of change parameter value of the sensor location, and a preliminary location result is obtained. Specifically, when the rate of change parameter value exceeds the threshold, it is preliminarily determined that a hydrogen leak has occurred. The value of i is observed and the leak is recorded in the corresponding sensor location area, thereby determining the hydrogen leak area. The location of the leak point is obtained by using a neural network model based on the sensor's rate of change value. Specifically, this includes inputting the sensor's pressure change rate, temperature change rate, and humidity change rate within the hydrogen leak area into the neural network model to obtain the specific location coordinates of the leak point.

2. The method for monitoring hydrogen leakage in a fuel cell vehicle according to claim 1, characterized in that, The hydrogen pipeline is the pipeline between the hydrogen storage cylinder and the fuel cell in the fuel cell vehicle; the hydrogen pipeline is planned into three parts with the hydrogen storage cylinder, the first-stage pressure reducing valve, the second-stage pressure reducing valve and the fuel cell as nodes.

3. The method for monitoring hydrogen leakage in a fuel cell vehicle according to claim 1, characterized in that, The sensor data is preprocessed and its first derivative is calculated using the Empirical Mode Decomposition (EMD) method to obtain the sensor's rate of change at its location. Specifically, this includes: The sensor data is processed by removing random fluctuation signals using the empirical mode decomposition method to obtain preprocessed data. The first derivative of the preprocessed data is taken to obtain the sensor change rate value at the sensor location.

4. The method for monitoring hydrogen leakage in a fuel cell vehicle according to claim 1, characterized in that, The neural network model is an LSTM neural network.

5. A hydrogen leakage monitoring system for fuel cell vehicles, characterized in that, include: The acquisition module is used to acquire real-time sensor data on the hydrogen pipeline; the sensor data includes pressure data, temperature data, and humidity data. The preprocessing and first-order derivative module is used to preprocess and perform first-order derivative calculation on the sensor data using the empirical mode decomposition method to obtain the sensor change rate value at the sensor location; the sensor change rate value includes the pressure change rate value, temperature change rate value, and humidity change rate value. The calculation module is used to calculate the rate of change parameter value of the sensor location based on the sensor's rate of change values ​​over the same time period. Specifically, it includes: calculating the rate of change parameter value of the sensor location based on the sensor's rate of change values ​​over the same time period and the weights of different operating parameters; the rate of change parameter value is calculated using the following formula: ； Among them, T i P i H i The data represents the temperature, pressure, and humidity collected from each sensor, where i is the sensor number. Weights for different operating condition parameters, These represent the rate of change of pressure, temperature, and humidity at each sensor after the first derivative; This refers to the value of the rate of change parameter. The preliminary positioning module is used to perform preliminary positioning based on the change rate parameter value of the sensor location and obtain preliminary positioning results. Specifically, it includes: when the change rate parameter value exceeds the threshold, it is preliminarily determined that a hydrogen leak has occurred, the i value is observed, and the leak occurs in the corresponding sensor location area, thereby determining the hydrogen leak area. The final positioning module is used to locate the leak point using a neural network model based on the sensor change rate values. Specifically, it includes inputting the sensor pressure change rate, temperature change rate, and humidity change rate within the hydrogen leak area into the neural network model to obtain the specific location coordinates of the leak point.

6. The hydrogen leakage monitoring system for fuel cell vehicles according to claim 5, characterized in that, The hydrogen pipeline is the pipeline between the hydrogen storage cylinder and the fuel cell in the fuel cell vehicle; the hydrogen pipeline is planned into three parts with the hydrogen storage cylinder, the first-stage pressure reducing valve, the second-stage pressure reducing valve and the fuel cell as nodes.

7. The hydrogen leakage monitoring system for fuel cell vehicles according to claim 5, characterized in that, The preprocessing and first-order derivative module specifically includes: The preprocessing unit is used to remove random fluctuation signals from the sensor data using the empirical mode decomposition method to obtain preprocessed data. The first-order derivative unit is used to perform first-order derivative on the preprocessed data to obtain the sensor change rate value at the sensor location.

8. An electronic device, characterized in that, include: One or more processors; A storage device on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method as described in any one of claims 1 to 4.

9. A computer storage medium, characterized in that, It stores a computer program thereon, wherein the computer program, when executed by a processor, implements the method as described in any one of claims 1 to 4.

Images