Hydrogen leakage monitoring system and method for hydrogen energy train

Abstract

The application provides a hydrogen leakage monitoring system and method for a hydrogen energy train, and belongs to the technical field of safety monitoring of rail transit equipment, and comprises a near-field point sensing unit group, an environmental area sensing unit group, a data acquisition and preprocessing module, and a central controller.The data acquisition and preprocessing module synchronously acquires the in-bottle pressure, pipeline pressure and near-field point hydrogen concentration monitored by all the near-field point sensing units, the spatial environmental hydrogen concentration and environmental temperature monitored by the environmental area sensing unit, and calculates the pressure change rate.The central controller is internally provided with a hierarchical judgment engine, which is used for comparing the data of the data acquisition and preprocessing module with preset trigger conditions at each level, comparing the severity / danger level from high to low one by one, and generating corresponding alarm strategies and disposal instructions according to the judgment result.The application realizes early warning of leakage, accurate positioning of the leakage source, reduces the false alarm and missed alarm probability, and realizes intelligent hierarchical response.

Description

Technical Field

[0001] This invention belongs to the field of safety monitoring technology for rail transit equipment, and in particular relates to a hydrogen leakage monitoring system and method for hydrogen-powered trains. Background Technology

[0002] Currently, the main technical solution for monitoring hydrogen leaks in the high-pressure hydrogen system on hydrogen-powered trains is as follows: discrete hydrogen concentration sensors are deployed in the surrounding space of key areas such as the compartment containing the hydrogen storage tanks, pipeline connections, and valves. This system collects hydrogen concentration data at the sensor locations in real time and compares the monitored values ​​with preset fixed alarm thresholds. When the concentration monitored by any sensor exceeds the fixed threshold, an audible and visual alarm is triggered, alerting the driver and passengers to the risk of a hydrogen leak.

[0003] The above-mentioned scheme based on a single concentration sensor and fixed threshold alarm has the following drawbacks: First, there is a lag in early warning. Existing solutions rely solely on concentration as a single physical quantity for judgment, failing to detect early signs such as pressure drops and localized temperature rises in the initial stages of a leak. A hydrogen leak takes time to spread from its occurrence to the location of environmental sensors; during this period, the leak may have already progressed, leading to a delayed early warning response and missing the optimal window for early intervention.

[0004] Secondly, false alarms and missed alarms occur frequently. The fixed threshold judgment method has poor adaptability under the complex operating conditions of trains. Factors such as airflow disturbances and changes in cabin ventilation during train operation can easily cause fluctuations in concentration readings. On the one hand, the actual leakage may be diluted and fall below the alarm threshold, resulting in missed alarms. On the other hand, environmental interference may cause false alarms. The existing solution lacks a comprehensive analysis mechanism that verifies multiple parameters, and cannot effectively distinguish between actual leakage and environmental interference.

[0005] Third, the system lacks location and classification capabilities. Existing solutions only monitor environmental concentrations and cannot pinpoint the specific leak source. After an alarm is triggered, the system can only indicate that the concentration in a certain area exceeds the standard; drivers and passengers cannot determine which gas cylinder or pipeline is leaking, making precise isolation and rapid investigation difficult. Furthermore, existing solutions use a single, fixed alarm threshold, responding to all leaks in the same way, failing to implement differentiated measures based on the severity and development of the leak.

[0006] In summary, the existing technical solutions are essentially a passive, singular, and indiscriminate alarm mode, which is insufficient to meet the safety requirements of hydrogen-powered trains for early warning, precise location, and graded response to leaks. Summary of the Invention

[0007] This invention aims to solve the above-mentioned technical problems, thereby providing a hydrogen leakage monitoring system and method for hydrogen-powered trains. It breaks through the limitations of single concentration monitoring, integrates multi-source information such as pressure and temperature, and achieves earlier and more reliable detection of hydrogen leakage. Through collaborative analysis and comprehensive judgment of multi-sensor information, it reduces the probability of false alarms and missed alarms in complex operating environments, achieves accurate location of leakage sources, and dynamically generates a matching graded alarm strategy based on the severity and development trend of the leakage.

[0008] To achieve the above-mentioned objectives, this invention provides a hydrogen leak monitoring system for hydrogen-powered trains, comprising: The near-field point sensing unit group consists of near-field point sensing units deployed at the valves, pipelines and interfaces of each high-pressure hydrogen cylinder. Each near-field point sensing unit includes a pressure sensor and a point hydrogen concentration sensor. The pressure sensor is used to monitor the pressure inside the cylinder in real time, and the point hydrogen concentration sensor is used to measure the trace near-field point hydrogen concentration at the leakage source. The environmental area sensing unit group consists of environmental area sensing units deployed on the top of the cabin space where the hydrogen cylinder group is located. Each environmental area sensing unit includes a hydrogen concentration sensor and a temperature sensor. The hydrogen concentration sensor is used to measure the hydrogen concentration in the space environment, and the temperature sensor is used to measure the ambient temperature. The data acquisition and preprocessing module is used to simultaneously acquire the bottle pressure, pipeline pressure and near-field hydrogen concentration monitored by all near-field point sensing units, the spatial environment hydrogen concentration and ambient temperature monitored by the environmental area sensing unit, and calculate the pressure change rate. The central controller has a built-in hierarchical judgment engine, which compares the data from the data acquisition and preprocessing module, including pressure change rate, near-field hydrogen concentration, space environment hydrogen concentration and ambient temperature, with preset trigger conditions at each level. The severity / danger level is compared one by one from high to low, sequentially comparing the fourth-level warning, third-level warning, second-level warning and first-level warning, and generating corresponding alarm strategies and handling instructions based on the judgment results.

[0009] Furthermore, one near-field point sensing unit is arranged at the beginning and one at the end of each hydrogen cylinder, and is placed on the valve or pipeline.

[0010] Furthermore, one of the environmental area sensing units is arranged on the top of the cabin space on both sides of the front, middle and rear of the hydrogen cylinder group.

[0011] A method for monitoring hydrogen leakage in a hydrogen-powered train, using the aforementioned system, includes the following steps: Step 1: Data Synchronization Acquisition and Preprocessing: Synchronously acquire the bottle pressure and near-field hydrogen concentration monitored by all near-field point sensing units, and the spatial environment hydrogen concentration and ambient temperature of the environmental sensing units, and calculate the pressure change rate. Step 2: Priority-based collaborative judgment from high to low: The preprocessed data is compared with the preset trigger conditions at each level in turn. The severity / danger level is compared one by one from high to low. When the higher-level condition is met, the level is locked immediately and no further judgment is made. Step 3: Strategy Execution and Output: Based on the determined risk level, execute the corresponding alarm strategy and response action.

[0012] Furthermore, the judgment conditions for the Level 1 warning include: When there exists any gas cylinder i satisfying | P local_i | > Th dp1 And C local_i > Th c1 At that time, a Level 1 warning was issued; Among them, P local_i The i-th pressure sensor monitors the pressure value; Th dp1 The first-level threshold for the rate of change of pressure is defined as 10 Pa / s to 1000 Pa / s; C local_i The concentration value is monitored by the i-th concentration sensor; Th c1 This is the first-level threshold for near-field concentration.

[0013] Furthermore, the secondary early warning judgment: Based on meeting the Level 1 warning conditions, when the environmental concentration sensor reading C in the relevant area... env_j > Th ce2 And this state lasts for more than t hold At the specified time, a Level 2 warning is activated; Among them, C env_j For the j-th concentration sensor to monitor the concentration value, Th ce2 The environmental concentration secondary threshold, t hold This is the duration threshold.

[0014] Furthermore, the three-level early warning judgment: When any environmental area sensing unit simultaneously satisfies C env_j > Th ce3 And T env_j > Th t At that time, a level three warning was issued; Among them, C env_j For the j-th concentration sensor to monitor the concentration value, Th ce3 T represents the third-level threshold for environmental concentration. env_jFor the j-th temperature sensor to monitor the temperature value, Th t This is the environmental temperature risk threshold.

[0015] Furthermore, the hierarchical collaborative judgment also includes four levels of early warning judgment: Level 4 alert will be activated when the following conditions are met: Any environmental area sensing unit simultaneously satisfies C env_j > Th ce4 And T env_j > 400 ℃, where Th ce4 The threshold is set at level four for environmental concentration.

[0016] Furthermore, the pressure change rate P local_i This represents the change in pressure per unit time.

[0017] Furthermore, the data acquisition module synchronously acquires all sensor data at a sampling frequency of not less than 100 Hz.

[0018] Advantages of this invention compared to existing technologies: 1. Achieve early warning of leaks: This invention, by deploying near-field point-type sensing units at each gas cylinder valve, can capture trace amounts of hydrogen gas and pressure changes before the leak source has diffused, enabling extremely early identification of leak events. Compared to existing technologies that rely solely on environmental concentration monitoring, this invention can significantly advance the warning time, buying valuable time for emergency response.

[0019] 2. Achieve precise location of the leak source: This invention equips each gas cylinder with an independent near-field sensing unit and identifies them with cylinder numbers. When a near-field unit triggers an alarm, the system can directly lock onto the specific cylinder number, achieving precise location of the leak source. This overcomes the shortcomings of existing technologies that can only indicate "concentration exceeds the standard in a certain area" but cannot pinpoint the specific leak source.

[0020] 3. Significantly reduces the probability of false alarms and false negatives: This invention universally employs multi-condition AND logic in its alarm rules, such as using simultaneous exceedances of pressure and concentration, along with confirmation of the duration, as trigger conditions. This effectively avoids false alarms caused by instantaneous noise from a single sensor, normal train vibration, or brief environmental fluctuations. Simultaneously, the near-field unit deploys one monitoring point at each end of each gas cylinder, while the environmental unit deploys six monitoring points at the beginning, middle, and end of the cabin roof, forming blind-spot-free monitoring coverage and ensuring no missed alarms from a system architecture perspective.

[0021] 4. Achieve intelligent hierarchical response: Based on the leakage evolution law, this invention formulates a graded alarm rule from level one to four, adopting a priority judgment logic from high to low. Each level is triggered by a combination of multi-parameter conditions and bound with differentiated automated handling instructions. The system can intelligently distinguish four different risk states: micro-leakage, diffusion, fire precursor, and emergency danger, and automatically execute precise responses from initiating ventilation to emergency braking, achieving a leap from fixed threshold alarms to intelligent graded decision-making. Attached Figure Description

[0022] Figure 1 This is a block diagram of the hydrogen leakage control logic of the present invention.

[0023] Figure 2 This is a schematic diagram showing the arrangement of the various sensors in this invention.

[0024] Figure 3 This is a flowchart of the method for generating a multi-sensor collaborative sensing and alarm strategy for hydrogen leakage in hydrogen-powered trains according to the present invention. Detailed Implementation

[0025] To better understand the purpose, structure, and function of this invention, the following detailed description of a hydrogen leakage monitoring system and method for hydrogen-powered trains, in conjunction with the accompanying drawings, is provided.

[0026] I. System Setup Method Reference Figure 1 The hydrogen leak monitoring system for hydrogen-powered trains mainly includes the following hardware components: near-field point sensing unit group, environmental area sensing unit group, data acquisition and preprocessing module, central controller, actuator, and human-machine interface in the driver's cab.

[0027] (a) Deployment of near-field point sensing units: Reference Figure 2 For each high-pressure hydrogen cylinder, near-field point sensing units are deployed at the valves and interfaces most prone to leakage. Each near-field point sensing unit consists of a pressure sensor and a point-type hydrogen concentration sensor. The pressure sensor is installed at the cylinder valve outlet and on the pipeline to monitor the cylinder pressure P in real time. local_i And pipeline pressure; a point-type hydrogen concentration sensor is installed within 5 cm of a potential leak point, including valves and interfaces, to capture trace amounts of hydrogen that have not yet diffused, measuring the hydrogen concentration at a trace near-field point at the leak source. The measured value is C. local_i .

[0028] Because hydrogen cylinders are most prone to leakage at the head and tail, a near-field point-type sensing unit is placed at both ends of each cylinder, located on valves or pipelines. A single trainset of hydrogen-powered trains contains 11 or 21 hydrogen cylinders. Taking a 21-cylinder trainset as an example, a total of 42 near-field point-type sensing units are required. Each monitoring point is equipped with one pressure sensor and one point-type concentration sensor. All near-field units are identified by cylinder number "i" to achieve precise location of the leak source.

[0029] (II) Deployment of environmental area sensing units: Reference Figure 2 An environmental sensing unit is placed at the top of the compartment where hydrogen cylinders are located, in an area where hydrogen tends to accumulate. Because hydrogen is much less dense than air, leaked hydrogen will rise rapidly and accumulate at the top of the compartment. Therefore, placing the environmental sensing unit at the top of the compartment allows for the most effective monitoring of spatial concentration.

[0030] Each environmental sensing unit consists of a high-precision hydrogen concentration sensor and a temperature sensor. Since the hydrogen cylinders are placed adjacent to each other, one environmental sensing unit is located on the top of each of the three compartments on either side of the hydrogen cylinder group, totaling six environmental monitoring points. Each monitoring point is labeled with the number j, and the high-precision hydrogen concentration sensor measures the ambient concentration C. env_j Temperature sensor measures ambient temperature T env_j .

[0031] In the above deployment scheme, the near-field point sensing unit undertakes the functions of early identification and precise positioning, while the environmental area sensing unit undertakes the function of overall assessment. The two form a full-coverage monitoring system in terms of physical distribution.

[0032] (III) Data Acquisition and Preprocessing Module: After the system is powered on, it enters continuous monitoring mode. The data acquisition module synchronously acquires sensor data from all near-field and environmental units at a sampling frequency of no less than 100 Hz.

[0033] For pressure signals, the current pressure value P of each gas cylinder is collected. local_i The pressure change rate P is calculated in real time by the central controller. local_i The method for calculating the rate of pressure change is as follows: the duration can be taken as 1 s to 10 s, and the rate of pressure change P... local_i P is the change in pressure per unit time. local_i = d Plocal_i / dt. When P local_i A positive value indicates a decrease in pressure, and the larger the absolute value, the higher the leakage rate.

[0034] For concentration signals, C is collected from each near-field point-type concentration sensor. local_iAnd the Cenv_j of the concentration sensors for each environmental area. All concentration sensors must be calibrated under standard hydrogen concentration before installation to ensure measurement accuracy.

[0035] For temperature signals, T values ​​from temperature sensors in each environmental area are collected. env_j .

[0036] Data preprocessing includes: using median filtering or low-pass filtering algorithms to denoise the raw sensor data, eliminating noise signals introduced by train operation vibration, electromagnetic interference, etc.; and smoothing the pressure change rate calculation results to avoid misjudgment caused by instantaneous fluctuations.

[0037] (iv) The central controller has a built-in hierarchical judgment engine, which is used to perform hierarchical collaborative judgment on the pre-processed multi-source sensor data according to the priority order of level four, level three, level two and level one, and generate corresponding alarm strategies and handling instructions based on the judgment results.

[0038] (v) The actuator, including the audible and visual alarm, valve controller, ventilation controller, brake controller and fire controller, is used to execute the alarm strategy and handling instructions output by the central controller.

[0039] (vi) The human-machine interface in the driver's cab is used to display alarm level and fault location information.

[0040] II. Parameters and Early Warnings The concentration sensor monitors concentration, the temperature sensor monitors temperature, and the pressure sensor monitors pressure. Different warning values ​​are set for these sensors, as shown in Table 1. Table 1 Parameter Comparison Table Note: In this context, the subscript i represents the near-field unit corresponding to the i-th gas cylinder, and j represents the j-th environmental monitoring point.

[0041] III. Specific process of hierarchical collaborative judgment logic: like Figure 3 As shown, the central controller performs hierarchical collaborative judgments according to priority from high to low. The specific process is as follows: Step 1: Data Synchronization Acquisition and Preprocessing: Synchronously acquire the bottle pressure and near-field hydrogen concentration monitored by all near-field point sensing units, and the spatial environment hydrogen concentration and ambient temperature of the environmental sensing units, and calculate the pressure change rate. Step 2: Priority-based collaborative judgment from high to low: The preprocessed data is compared with the preset trigger conditions at each level in turn. The severity / danger level is compared one by one from high to low. When the higher-level condition is met, the level is locked immediately and no further judgment is made. (1) Level 4 Warning Judgment: First, determine whether the triggering conditions for a Level 4 warning are met. A Level 4 warning is the highest risk level, and it is triggered when any of the following conditions are met: Condition 1 is the combustion condition: C of the sensing unit in any environmental area env_j > Th ce4 And T env_j > 400 ℃, both conditions must be met simultaneously. Among them, Th ce4 The environmental concentration level four threshold can be set to 40000×10. -6 .

[0042] Condition 4-2 is the open flame condition: any flame detector detects a flame signal.

[0043] If any of the above conditions are met, a Level 4 warning will be immediately activated.

[0044] (2) Level 3 Early Warning Judgment: If the Level 4 conditions are not met, then determine whether the Level 3 early warning triggering conditions are met. The Level 3 early warning is a fire risk warning, and the triggering condition is: any environmental area sensing unit simultaneously meets C env_j > Th ce3 And T env_j >Th t Among them, Th ce3 The environmental concentration level three threshold can be set to 10000×10. -6 ;Th t The environmental temperature risk threshold can be set to 200 ℃.

[0045] If the above conditions are met, a Level 3 warning will be activated.

[0046] (3) Level II Warning Judgment: If the Level III conditions are not met, then determine whether the Level II warning triggering conditions are met. A Level II warning is a confirmation of leakage and diffusion, and triggering it requires the simultaneous fulfillment of the following two conditions: Condition 2-1: The first-level warning condition has been met, that is, there exists any gas cylinder i that satisfies | P local_i | > Th dp1 And C local_i > Th c1 ; Condition 2-2: Environmental concentration sensor reading C in the relevant area env_j > Th ce2 And this state lasts for more than t hold Seconds. Among them, Th ce2 The secondary threshold for environmental concentration can be set to 5000 × 10⁻⁶. -6 ;t hold The duration threshold can be set to 60 seconds.

[0047] If all of the above conditions are met, a Level II warning will be issued.

[0048] (4) Level 1 Warning Judgment: If the Level 2 conditions are not met, then determine whether the Level 1 warning triggering conditions are met. The Level 1 warning is the initial warning, and the triggering condition is: there exists any gas cylinder i that satisfies |P local_i | > Th dp1 And C local_i > Th c1 Among them, Th dp1 The first-level threshold for the rate of change of pressure can be set from 10 Pa / s to 1000 Pa / s; Th c1 The first-order threshold for near-field concentration can be set to 100×10. -6 .

[0049] If the above conditions are met, a Level 1 warning will be issued.

[0050] Step 3: Strategy Execution and Output: Based on the determined risk level, execute the corresponding alarm strategy and response action.

[0051] The above order of judgment ensures that the highest risk is dealt with first, which is in line with the basic principle of safety first.

[0052] IV. Specific Implementation of Tiered Alarm Strategy and Response Actions: Based on the risk levels identified in steps S2 to S5, the central controller generates and executes corresponding alarm strategies and handling instructions.

[0053] (a) Level 1 Early Warning Response Strategy: When a Level 1 warning is triggered, the central controller performs the following actions: Precise location display: "Suspected micro-leakage of gas cylinder unit i" is highlighted on the human-machine interface of the cab control console, where i is the gas cylinder number corresponding to the near-field unit that triggered the first-level warning, thus achieving precise location of the leak source.

[0054] Automatic ventilation activation: Sends a start command to the ventilation controller in the cylinder group area to automatically activate enhanced ventilation, accelerating the dilution and discharge of leaked hydrogen.

[0055] Log recording: Records the fault unit identifier and trigger time, stores them in the system log, and prompts subsequent maintenance checks.

[0056] Train operation status: The train can continue to operate normally without slowing down or stopping.

[0057] (II) Level II Early Warning Response Strategy: When a Level 2 warning is triggered, the central controller performs the following additional operations on top of the Level 1 warning response measures: Audible and visual alarm: Triggering the audible and visual alarm device in the driver's cab will sound a buzzer alarm and flash a yellow warning light, alerting the driver and passengers to a confirmed risk of leakage and spread.

[0058] Active isolation: A closing command is sent to the solenoid valve at the mouth of gas cylinder i, automatically closing the hydrogen supply valve of that gas cylinder and cutting off the source of leakage.

[0059] Reduced speed operation: Send a reduced power operation command to the train control system to notify the power system to reduce the output power as appropriate and reduce the demand for hydrogen.

[0060] (III) Three-level early warning response strategy: When a Level 3 alert is triggered, the central controller performs the following actions: Emergency Alarm: Triggers the emergency audible and visual alarm in the driver's cab, the buzzer emits a high-frequency intermittent alarm, and the console displays the words "Fire Warning" in red.

[0061] Extended isolation: Send a closing command to all bottle neck solenoid valves of the relevant bottle group module to isolate the entire bottle group.

[0062] Emergency preparedness: Activate the fire protection system pre-pressurization (i.e., pre-pressurize the fire protection pipelines) to prepare for possible fire extinguishing operations; at the same time, suggest on the driver's cab human-machine interface that "the train will stop at the next station for inspection".

[0063] (iv) Level 4 Early Warning Response Strategy: When a Level 4 alert is triggered, the central controller executes the highest level of emergency response: Disaster Alarm: Triggers the highest level of alarm for the entire train. The driver's cab's audible and visual alarm device emits the strongest alarm signal, and at the same time, the disaster alarm is broadcast to the entire train via the train's public address system.

[0064] Emergency braking: Immediately send an emergency braking command to the train braking control system to execute an emergency stop.

[0065] Active protection: Automatically releases explosion suppression or fire extinguishing media in the compartment.

[0066] Global notification: Sends a disaster alarm signal to the dispatch center, reporting the train's location and the accident status.

[0067] (v) Alarm status maintenance and reset monitoring: After an alarm is triggered, the system maintains the current alarm status and continuously monitors the relevant sensor data. Only when all triggering conditions are no longer met and the preset safety observation period is exceeded will the system automatically reset and return to step 1 to continue monitoring.

[0068] Advantages of this invention: This invention constructs a hierarchical collaborative sensing network by deploying near-field point sensing units at each gas cylinder valve and environmental area sensing units at key locations on the top of the compartment. This enables the near-field units to capture trace amounts of hydrogen and pressure changes before the leak source has diffused, achieving very early identification of leak events and precise location of faulty gas cylinders. The environmental units are responsible for monitoring the spatial concentration accumulation and environmental thermal state after the leak has diffused, assessing the overall risk. This addresses the fundamental shortcomings of existing technologies, such as delayed early warning and inability to locate leaks.

[0069] Based on the leakage evolution law, this invention formulates a graded alarm rule from level one to four, employing a priority judgment logic from high to low. Each level is triggered by a combination of multi-parameter conditions and bound to differentiated automated handling instructions. This enables the system to intelligently distinguish between four different risk states: micro-leakage, diffusion, fire precursor, and emergency danger, and automatically execute precise responses, significantly improving the scientific nature and effectiveness of emergency response.

[0070] This invention universally employs multi-condition AND logic and duration confirmation as trigger conditions in various alarm rules. This effectively avoids false alarms caused by instantaneous noise from a single sensor, normal train vibration, or brief environmental fluctuations, significantly improving the seriousness and accuracy of alarms. At the same time, the blind-spot-free deployment of the near-field unit and the redundant monitoring of the environmental unit ensure no missed alarms from the system architecture level.

[0071] In summary, this invention achieves blind-spot-free monitoring coverage from the leak source to the spatial environment by constructing a hierarchical sensing network that combines master-slave collaboration and point-to-surface integration; it completes intelligent decision-making from micro-leak identification to emergency response by designing a four-level hierarchical response strategy based on priority arbitration; and it fundamentally improves the system's anti-interference capability and alarm reliability by employing multi-condition AND logic and a duration judgment mechanism. The synergistic effect of these technologies constitutes a proactive safety protection system with advanced early warning, precise positioning, intelligent hierarchical classification, and reliable operation, comprehensively solving the core problems of existing technologies such as delayed early warning, frequent false alarms and missed alarms, and inability to locate and classify leaks.

Claims

1. A hydrogen leak monitoring system for a hydrogen-powered train, characterized in that, include: The near-field point sensing unit group consists of near-field point sensing units deployed at valves, pipelines and interfaces of each high-pressure hydrogen cylinder. Each near-field point sensing unit includes a pressure sensor and a point hydrogen concentration sensor. The pressure sensor is used to monitor the pressure inside the hydrogen cylinder and pipeline in real time, and the point hydrogen concentration sensor is used to measure the trace near-field point hydrogen concentration at the leak source. The environmental area sensing unit group consists of environmental area sensing units deployed on the top of the cabin space where the hydrogen cylinder group is located. Each environmental area sensing unit includes a hydrogen concentration sensor and a temperature sensor. The hydrogen concentration sensor is used to measure the hydrogen concentration in the space environment, and the temperature sensor is used to measure the ambient temperature. The data acquisition and preprocessing module is used to simultaneously acquire the bottle pressure, pipeline pressure and near-field hydrogen concentration monitored by all near-field point sensing units, the spatial environment hydrogen concentration and ambient temperature monitored by the environmental area sensing unit, and calculate the pressure change rate. The central controller has a built-in hierarchical judgment engine, which compares the data from the data acquisition and preprocessing module, including pressure change rate, near-field hydrogen concentration, space environment hydrogen concentration and ambient temperature, with preset trigger conditions at each level. The severity / danger level is compared one by one from high to low, and corresponding alarm strategies and handling instructions are generated based on the judgment results.

2. The hydrogen leakage monitoring system for a hydrogen-powered train according to claim 1, characterized in that: One near-field point sensing unit is arranged at the beginning and one at the end of each hydrogen cylinder, and is placed on the valve or pipeline.

3. The hydrogen leakage monitoring system for a hydrogen-powered train according to claim 1, characterized in that: One environmental area sensing unit is arranged on the top of each of the cabin spaces on both sides of the front, middle and rear of the hydrogen cylinder group.

4. A method for monitoring hydrogen leakage in a hydrogen-powered train, employing the system as described in any one of claims 1 to 3, characterized in that, Includes the following steps: Step 1: Data Synchronization Acquisition and Preprocessing: Synchronously acquire the bottle pressure and near-field hydrogen concentration monitored by all near-field point sensing units, and the spatial environment hydrogen concentration and ambient temperature of the environmental sensing units, and calculate the pressure change rate. Step 2: Priority-based collaborative judgment from high to low: The preprocessed data is compared with the preset trigger conditions at each level in turn. The severity / danger level is compared one by one from high to low. The comparison is carried out in turn for the fourth-level warning, the third-level warning, the second-level warning and the first-level warning. When the higher-level condition is met, the level is locked immediately and no further judgment is made. Step 3: Strategy Execution and Output: Based on the determined risk level, execute the corresponding alarm strategy and response action.

5. The method for monitoring hydrogen leakage in a hydrogen-powered train according to claim 4, characterized in that, The judgment conditions for the first-level warning include: when there exists any gas cylinder i that satisfies |P local_i | > Th dp1 And C local_i > Th c1 At that time, a Level 1 warning was issued; Among them, P local_i The i-th pressure sensor monitors the pressure value; Th dp1 The first-level threshold for the rate of change of pressure is defined as 10 Pa / s to 1000 Pa / s; C local_i The concentration value is monitored by the i-th concentration sensor; Th c1 This is the first-level threshold for near-field concentration.

6. The method for monitoring hydrogen leakage in a hydrogen-powered train according to claim 5, characterized in that, The aforementioned level-two early warning judgment: Based on meeting the Level 1 warning conditions, when the environmental concentration sensor reading C in the relevant area... env_j > Th ce2 And it continues to exceed t hold At the specified time, a Level 2 warning is activated; Among them, C env_j For the j-th concentration sensor to monitor the concentration value, Th ce2 The environmental concentration secondary threshold, t hold This is the duration threshold.

7. The method for monitoring hydrogen leakage in a hydrogen-powered train according to claim 4, characterized in that, The three-level early warning judgment: When any environmental area sensing unit simultaneously satisfies C env_j > Th ce3 And T env_j > Th t At that time, a level three warning was issued; Among them, C env_j For the concentration value monitored by the j-th concentration sensor, Th ce3 T represents the third-level threshold for environmental concentration. env_j For the j-th temperature sensor to monitor the temperature value, Th t This is the environmental temperature risk threshold.

8. The method for monitoring hydrogen leakage in a hydrogen-powered train according to claim 4, characterized in that, The hierarchical collaborative judgment also includes four levels of early warning judgment: Level 4 alert will be activated when the following conditions are met: Any environmental area sensing unit simultaneously satisfies C env_j > Th ce4 And T env_j > 400 ℃, where Th ce4 The threshold is set at level four for environmental concentration.

9. The method for monitoring hydrogen leakage in a hydrogen-powered train according to any one of claims 4 to 8, characterized in that, The pressure change rate P local_i This represents the change in pressure per unit time.

10. The method for monitoring hydrogen leakage in a hydrogen-powered train according to any one of claims 4 to 8, characterized in that, The data acquisition module synchronously acquires all sensor data at a sampling frequency of not less than 100 Hz.

Images