Automatic safety monitoring device of marine hydrogen fuel cell power generation system

Abstract

The invention relates to an automatic safety monitoring device of a marine hydrogen fuel cell power generation system, and belongs to the technical field of new energy power, the system comprises a monitoring module, a control analysis module and a cutting module; the monitoring module is used for acquiring state information of each part of a ship and sending the state information to the control analysis module; the state information comprises hydrogen concentration, flame smoke state and ventilation state; the control analysis module is used for analyzing the operation safety of the ship hydrogen fuel cell by taking the state information as an input feature and adopting a completely trained support vector machine classification model to obtain a safety state result, and if the safety state result is state abnormity, sending a state abnormity instruction to the cut-off module; the cut-off module is used for controlling a valve of a hydrogen supply channel to be closed according to the instruction of the abnormal state. According to the invention, the technical problem of low monitoring precision caused by scattered existing security monitoring schemes is solved.

Description

Technical Field

[0001] This invention relates to the field of new energy power technology, and in particular to an automated safety monitoring device for a marine hydrogen fuel cell power generation system. Background Technology

[0002] Fuel cells are a key device for hydrogen energy applications. They directly convert the chemical energy stored in the reaction gas into electrical energy, emitting only water. Therefore, fuel cells are a highly efficient, zero-emission conversion device with broad development prospects.

[0003] Currently, hydrogen fuel cells are most widely used in the automotive industry, and their application and promotion in the marine industry are also gradually increasing. Compared to automobiles, marine hydrogen fuel cells are generally installed in enclosed compartments of ships, where hydrogen is more likely to accumulate if a leak occurs, thus posing a greater safety concern. To address these issues, some professionals have conducted research and exploration. For example, the invention patent with patent application number CN202310776335.1, "A Safety Monitoring System and Method for Marine Hydrogen Fuel Cells," designs a safety monitoring system functionally independent of the hydrogen fuel cell system control unit, achieving real-time monitoring of hydrogen concentration in enclosed compartments. However, the reliability and accuracy of the ventilation status detection and judgment methods provided by this invention patent are not high enough, which may lead to misjudgments and affect the effectiveness of the safety monitoring system. Another example is the invention patent with patent application number CN202211021525.4, "A Hydrogen Energy Ship Safety System," which monitors hydrogen leaks, electrical and power fluctuations, and mechanical vibrations in real time, classifying accidents based on state parameters to ensure ship operation safety. It can be seen that this invention focuses on small ships and is not universally applicable. For example, the invention patent with patent application number CN202110786384.4, entitled "A Green Ship High-Safety Fuel Cell Power Generation System and Its Safe Operation Method", discloses the specific structure and operation method of the ship high-safety fuel cell power generation system. The control and judgment methods of this patent are simple, but the accuracy of the results depends on the accuracy of the sensors. The error or failure of the sensors will seriously affect the judgment results of the system.

[0004] Given that existing safety monitoring solutions are fragmented, each sensor outputs a single result based on data from its own sensors. For example, a high hydrogen concentration triggers a high hydrogen concentration alarm, and a flame detection alarm triggers a flame detection alarm. However, a high hydrogen concentration may indicate a hydrogen leak, which is a safety anomaly. Low air pressure, on the other hand, could be due to a fan malfunction or a problem with the cabin sealing causing air intake, and cannot be simply identified as a safety anomaly. Therefore, a comprehensive and accurate feedback system is needed to provide precise information about the hydrogen fuel cell and its operating status. Summary of the Invention

[0005] In view of this, it is necessary to provide an automated safety monitoring device for marine hydrogen fuel cell power generation systems to solve the problem that existing safety monitoring schemes are decentralized and obtain a result based on data from each monitoring sensor. For example, a high hydrogen concentration will output a high hydrogen concentration alarm, and a flame detection will output a flame detection alarm. However, a high hydrogen concentration may indicate a hydrogen leak, which is a safety anomaly. Low wind pressure may be due to a fan failure or a problem with the cabin sealing that causes air intake, resulting in low accuracy of decentralized monitoring.

[0006] To achieve the above objectives, in a first aspect, the present invention provides an automated safety monitoring device for a marine hydrogen fuel cell power generation system, comprising: a monitoring module, a control and analysis module, and a cut-off module; The monitoring module is used to acquire status information of various parts of the ship and send the status information to the control and analysis module; the status information includes hydrogen concentration, flame smoke status, and ventilation status; The control analysis module is used to analyze the operational safety of the ship's hydrogen fuel cell by using the state information as input features and employing a fully trained support vector machine classification model to obtain a safety state result. If the safety state result is an abnormal state, an abnormal state command is sent to the cut-off module. The cut-off module is used to control the valve of the hydrogen supply channel to close according to the abnormal status command.

[0007] In one possible implementation, the monitoring module includes a leak detection unit, a fire alarm detection unit, and a ventilation detection unit; The leakage monitoring unit is deployed in various hydrogen-related areas of the ship to acquire hydrogen concentration information in each hydrogen-related area and send the hydrogen concentration information to the control and analysis module. The fire detection unit is used to detect the flame and smoke status of the ship's cabin environment and send the flame and smoke status to the control and analysis module. The ventilation detection unit is used to detect the wind pressure value in each ventilation area of ​​the ship and send the wind pressure value to the control and analysis module.

[0008] In one possible implementation, the control analysis module includes an input unit, an analysis and prediction unit, a decision-making unit, and an instruction issuing unit; The input unit is used to receive the status information and send the status information to the analysis and prediction module; The analysis and prediction unit is used to analyze the operational safety of ship hydrogen fuel cells using the state information as input features and a fully trained support vector machine classification model to obtain the safety status result. The decision unit is used to receive the security status result and determine whether the security status result is a normal security result or an abnormal status. If the status is abnormal, the abnormal status instruction is sent to the instruction issuing unit. The instruction issuing unit is used to send the instruction indicating an abnormal status to the cut-off module.

[0009] In one possible implementation, the analysis and prediction unit includes a support vector machine subunit; The support vector machine subunit is used to predict the operational safety of ship hydrogen fuel cells by using the state information as input features and employing a fully trained support vector machine classification model to obtain the safety status result.

[0010] In one possible implementation, the ventilation detection unit for each ship ventilation area includes a first ventilation fan and a second ventilation fan, with either the first ventilation fan or the second ventilation fan participating in ventilation at any given time.

[0011] In one possible implementation, the ventilation detection unit further includes a fan starter; The fan starter is used to control the start and stop of the ventilation fan and to feed back the operating status of the ventilation fan to the decision unit.

[0012] In one possible implementation, the hydrogen concentration information being greater than a hydrogen concentration threshold is used as the first judgment condition; The presence of a fire, indicated by the flame and smoke conditions, is used as the second judgment condition. The third judgment condition is that the wind pressure value in the ventilation state is lower than the wind pressure threshold. The decision unit is used to determine that the security status result is abnormal when either the first judgment condition or the second judgment condition is met. The decision unit is also used to send a wind turbine start command to the wind turbine starter when the third judgment condition is met, and to determine the safety status result as an abnormal status based on the abnormal wind turbine operation feedback from the wind turbine starter.

[0013] In one possible implementation, the decision-making unit includes a first decision-making subunit, a timing determination unit, and a second decision-making subunit; When the third judgment condition is met, the first decision subunit is used to determine whether there is a ventilation fan running. If the first ventilation fan is running, a first fan running command will be generated and sent to the timing unit. The timing determination unit is used to determine whether the third determination condition is met after waiting for a set time. If it is met, the first ventilation fan is determined to be a failed fan, and the result that the first ventilation fan is a failed fan is sent to the first decision subunit. The first decision subunit is also configured to, upon receiving the result that the first ventilation fan is a failed fan, send a second ventilation fan start command to the fan starter to start the second ventilation fan, and send a first ventilation fan stop command to the fan starter to stop the first ventilation fan; The timing determination unit is used to determine whether the third determination condition is met after waiting for a set time. If it is met, the second ventilation fan is determined to be a failed fan, and the result that the second ventilation fan is a failed fan is sent to the first decision subunit. The first decision subunit is also used to send the result that the first ventilation fan is a failed fan and the result that the second ventilation fan is a failed fan to the second decision subunit; The second decision subunit is used to determine the safety status result as an abnormal state based on the result that the first ventilation fan is a failed fan and the result that the second ventilation fan is a failed fan.

[0014] In one possible implementation, the decision-making unit further includes an optimization decision-making subunit; When the third judgment condition is met, the first decision subunit is used to determine whether there is a ventilation fan running. If there is no ventilation fan participating in the ventilation work, the optimization decision subunit is used to obtain the total running time of the ventilation fan, and prioritize the start-up order of the ventilation fans from low to high based on the total running time, and send the priority ranking result to the first decision subunit. The first decision subunit is also used to send ventilation fan start commands according to the priority order.

[0015] In one possible implementation, an alarm module is also included, which is communicatively connected to the control and analysis module; The control analysis module is also used to send alarm commands; The alarm module is used to issue a security anomaly alarm based on the alarm command.

[0016] The beneficial effects of this invention are as follows: To address the problem of dispersed monitoring devices in marine hydrogen fuel cell power generation systems, where each parameter is analyzed independently and the low correlation between parameters leads to a single parameter failing to accurately represent the working environment of the hydrogen fuel cell, this invention integrates monitoring of hydrogen concentration, flame smoke status, and ventilation status through a monitoring module. Using these parameters as inputs, a machine learning model in the control analysis module predicts and analyzes the working environment status of the hydrogen fuel cell, determining its safety status. If an anomaly is detected, the hydrogen supply channel is cut off via a cutoff module to prevent a hydrogen explosion. Furthermore, the high reliability of the results from the machine learning model's prediction of real-time status information further improves the monitoring accuracy of the monitoring device. Attached Figure Description

[0017] Figure 1 A system architecture diagram of an embodiment of the automated safety monitoring device for a marine hydrogen fuel cell power generation system provided by the present invention; Figure 2 The system architecture diagram of a decision unit in the automated safety monitoring device for a marine hydrogen fuel cell power generation system provided by the present invention is shown. Detailed Implementation

[0018] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0019] Before demonstrating the embodiments, the following terms will be explained.

[0020] A hydrogen fuel cell is a device that directly converts the chemical energy of hydrogen and oxygen into electrical energy; it is a type of energy conversion equipment.

[0021] Safety analysis is a process of systematically assessing the risks and hazards that may occur during the operation of a hydrogen fuel cell system, equipment, or process. Its purpose is to identify potential safety hazards, take corresponding measures to reduce risks, and ensure the safe operation of the system.

[0022] A specific embodiment of the present invention discloses an automated safety monitoring device for a marine hydrogen fuel cell power generation system. Please refer to [link / reference]. Figure 1 It includes: monitoring module 1, control and analysis module 2, and cut-off module 3; The monitoring module is used to acquire status information of various parts of the ship and send the status information to the control and analysis module; the status information includes hydrogen concentration, flame smoke status, and ventilation status; It should be noted that hydrogen is a flammable and explosive gas, and its leakage may lead to serious safety accidents. By monitoring the hydrogen concentration in real time through a monitoring module, the system can immediately issue an alarm and take measures (such as cutting off the hydrogen supply and activating the ventilation system) once the concentration exceeds the safety threshold, effectively avoiding the risk of explosion or fire caused by hydrogen accumulation.

[0023] Flame and Smoke Detection: In a marine environment, fire is a serious safety hazard. Flame and smoke detection can quickly detect the generation of ignition sources or smoke, and promptly identify early signs of a fire. Combined with hydrogen concentration monitoring, the cause of the fire can be more accurately determined (such as whether it was caused by a hydrogen leak), allowing for targeted firefighting measures to be taken and reducing fire damage.

[0024] Ventilation status monitoring: Good ventilation is a key measure to reduce the risk of hydrogen accumulation. By monitoring the ventilation status and ensuring the normal operation of the ventilation system, even in the event of a hydrogen leak, forced ventilation can quickly dilute the hydrogen concentration and reduce the risk of explosion.

[0025] In some embodiments of the present invention, please refer to Figure 1 The monitoring module includes a leakage monitoring unit 11, a fire alarm detection unit 12, and a ventilation detection unit 13; The leakage monitoring unit is deployed in various hydrogen-related areas of the ship to acquire hydrogen concentration information in each hydrogen-related area and send the hydrogen concentration information to the control and analysis module. The fire detection unit is used to detect the flame and smoke status of the ship's cabin environment and send the flame and smoke status to the control and analysis module. The ventilation detection unit is used to detect the wind pressure value in each ventilation area of ​​the ship and send the wind pressure value to the control and analysis module.

[0026] It should be noted that the leak monitoring unit is a hydrogen concentration sensor. The hydrogen concentration sensor is usually installed in the location where hydrogen may accumulate and at the ventilation outlet. The number of hydrogen concentration sensors is determined based on the size, layout and ventilation conditions of the compartment.

[0027] The fire detection unit consists of a flame detector and a smoke detector.

[0028] The ventilation detection unit uses a wind pressure sensor to detect the working effect of the ventilation fan.

[0029] It should be noted that if sensors used for safety status detection, such as hydrogen concentration sensors, flame detectors, smoke detectors, and wind pressure sensors, do not have self-testing capabilities, two units should be installed at the same location for mutual verification. If the difference in the detection data from the two sensors exceeds a set range, a fault is determined. If the sensors have self-testing capabilities, the presence of a fault is determined by the output self-test signal.

[0030] The control analysis module is used to analyze the operational safety of the ship's hydrogen fuel cell by using the state information as input features and employing a fully trained support vector machine classification model to obtain a safety state result. If the safety state result is an abnormal state, an abnormal state command is sent to the cut-off module. It should be noted that machine learning models can use multi-dimensional state information such as hydrogen concentration, flame smoke state, and ventilation status as input features to conduct a comprehensive analysis of the operational safety of ship hydrogen fuel cells in a data-driven manner. Compared with traditional threshold judgment methods, machine learning models can capture more complex safety state patterns, thereby more accurately identifying potential safety risks.

[0031] In some embodiments of the present invention, please refer to Figure 1 The control analysis module includes an input unit 21, an analysis and prediction unit 22, a decision-making unit 23, and an instruction issuing unit 24; The input unit is used to receive the status information and send the status information to the analysis and prediction module; The analysis and prediction unit is used to analyze the operational safety of ship hydrogen fuel cells using the state information as input features and a fully trained support vector machine classification model to obtain the safety status result. The decision unit is used to receive the security status result and determine whether the security status result is a normal security result or an abnormal status. If the status is abnormal, the abnormal status instruction is sent to the instruction issuing unit. The instruction issuing unit is used to send the instruction indicating an abnormal status to the cut-off module.

[0032] It should be noted that by clearly defining the functional units of input, analysis, decision-making, and command issuance, the control and analysis module can efficiently process the status information received from the monitoring module. The input unit ensures the accurate reception and transmission of data, while the analysis and prediction unit uses machine learning models to perform in-depth analysis of the data. This hierarchical processing approach improves the overall efficiency of the system.

[0033] In some embodiments of the present invention, the analysis and prediction unit includes a support vector machine subunit; The support vector machine subunit is used to predict the operational safety of ship hydrogen fuel cells using the state information as input features, and obtain the safety state result.

[0034] It should be noted that Support Vector Machine (SVM) is a machine learning method based on statistical learning theory and structural risk minimization. SVM transforms the problem of finding the optimal classification surface that maximizes the classification margin into solving a convex quadratic optimization problem. The optimal classification function of SVM is expressed as:

[0035] In the above formula The number of support vectors, For kernel functions, commonly used kernel functions include linear kernel functions, polynomial kernel functions, radial basis function kernel functions, and sigmoid kernel functions.

[0036] The optimal SVM classification function described above is a hyperplane composed of support vectors from 1 to l. This hyperplane can classify the data in the feature space according to their classes. In the formula... Let be the weight of the i-th support vector. Let b be the classification result of the i-th support vector, and b be the intercept.

[0037] Specifically, the feature values ​​composed of signal data monitored by hydrogen concentration sensors, flame detectors, smoke detectors, wind pressure sensors, etc., are used as the input to the SVM model, as follows:

[0038] In the above formula, X is the input of the SVM model, and x is the collected signal data, which consists of p data points.

[0039] The safety state of the fuel cell system is used as the output of the SVM model, and normal and abnormal conditions are marked as 1 and -1, respectively, and represented by y.

[0040] Based on the above method, multiple sets of state data under different working conditions were collected, and the above sample data were divided into two parts: a training set of m groups and a test set of n groups, which were used for training and testing of the SVM model, respectively.

[0041] Then the i-th group of sample data is represented as:

[0042] The training sample set is represented as:

[0043] The test sample set is represented as follows:

[0044] Furthermore, to avoid some eigenvalues ​​having excessively large ranges while others have excessively small ranges, and to prevent numerical computation difficulties caused by calculating the inner product during training to compute the kernel function, for eigenvalues... The normalization calculation method is as follows:

[0045] In the above formula, Eigenvalues The maximum value, Eigenvalues The minimum value.

[0046] Furthermore, the optimal parameters c (penalty factor) and g (kernel function parameter) are selected by using cross-validation and grid traversal principles; To train an SVM model, an SVM model is established using parameters c and g. The SVM model is then trained using the training set of sample data. A radial basis function kernel is selected, and its expression is as follows.

[0047] SVM model testing utilizes a sample data test set to validate and test the established SVM model.

[0048] The cut-off module is used to control the valve of the hydrogen supply channel to close according to the abnormal status command.

[0049] It should be noted that once a safety anomaly is detected, an alarm is immediately triggered and the shutdown module is notified to ensure that measures are taken in time before a dangerous situation occurs, minimizing the probability of an accident.

[0050] In this embodiment, to address the problem of dispersed monitoring devices and independent analysis of each parameter in marine hydrogen fuel cell power generation systems, resulting in low correlation between parameters and the inability of a single parameter to accurately represent the hydrogen fuel cell's operating environment, this invention integrates monitoring of hydrogen concentration, flame smoke status, and ventilation status through a monitoring module. Using these parameters as inputs, a machine learning model in the control analysis module predicts and analyzes the operating environment status of the hydrogen fuel cell to determine its safety status. If an anomaly is detected, the hydrogen supply channel is cut off via a cutoff module to prevent a hydrogen explosion. Furthermore, the high reliability of the results from the machine learning model's prediction of real-time status information further improves the monitoring accuracy of the monitoring device.

[0051] In some embodiments of the present invention, the ventilation detection unit of each ship ventilation area includes a first ventilation fan and a second ventilation fan, and at any given time, either the first ventilation fan or the second ventilation fan participates in the ventilation operation.

[0052] It should be noted that by setting up at least two ventilation fans in the same location, the redundancy of the fans in terms of risk resistance and fault tolerance can be increased.

[0053] Furthermore, by setting up ventilation fans, misjudgments caused by other factors are avoided. When the cabin ventilation is normal, negative pressure is generated, and the wind pressure sensor directly detects this negative pressure, eliminating interference from other irrelevant factors. Therefore, the solution of this invention provides more direct detection of ventilation status and a more reasonable evaluation of ventilation effectiveness, preventing misjudgments that could affect the effectiveness of the safety monitoring system.

[0054] In some embodiments of the present invention, the ventilation detection unit further includes a fan starter; The fan starter is used to control the start and stop of the ventilation fan and to feed back the operating status of the ventilation fan to the decision unit.

[0055] In this embodiment, the fan starter directly controls the start and stop of the ventilation fan, and two fans are set up at the same ventilation location to achieve one in use and one as a backup. When the control and analysis module detects a low air pressure alarm and neither fan is running, it sends a fan start command to the fan starter. When it detects a fan failure, it sends a fan switching command to start another fan, and at the same time receives the status feedback signal from the fan starter.

[0056] In some embodiments of the present invention, the hydrogen concentration information being greater than a hydrogen concentration threshold is used as a first judgment condition; The presence of a fire, indicated by the flame and smoke conditions, is used as the second judgment condition. The third judgment condition is that the wind pressure value in the ventilation state is lower than the wind pressure threshold. The decision unit is used to determine that the security status result is abnormal when either the first judgment condition or the second judgment condition is met. The decision unit is also used to send a wind turbine start command to the wind turbine starter when the third judgment condition is met, and to determine the safety status result as an abnormal status based on the abnormal wind turbine operation feedback from the wind turbine starter.

[0057] In this embodiment, a high hydrogen concentration alarm is output when the hydrogen concentration remains below a first preset value, and a very high hydrogen concentration alarm is output when the hydrogen concentration remains below a second preset value. The first preset value is lower than the second preset value. The working effect of the ventilation fan is detected by the wind pressure sensor, and a low wind pressure alarm is output when the ventilation wind pressure remains below a preset value.

[0058] Furthermore, when the third judgment condition is met, in order to avoid misjudging the hydrogen fuel operating environment due to ambiguity in the positioning of the ventilation status, this embodiment performs further analysis when the third judgment condition is met, thereby improving the accuracy and effectiveness of the monitoring system.

[0059] In one specific embodiment of the present invention, please refer to Figure 1 and Figure 2 The decision-making unit 23 includes a first decision-making subunit 231, a timing determination unit 232, and a second decision-making subunit 233; When the third judgment condition is met, the first decision subunit is used to determine whether there is a ventilation fan running. If the first ventilation fan is running, a first fan running command will be generated and sent to the timing unit. The timing determination unit is used to determine whether the third determination condition is met after waiting for a set time. If it is met, the first ventilation fan is determined to be a failed fan, and the result that the first ventilation fan is a failed fan is sent to the first decision subunit. The first decision subunit is also configured to, upon receiving the result that the first ventilation fan is a failed fan, send a second ventilation fan start command to the fan starter to start the second ventilation fan, and send a first ventilation fan stop command to the fan starter to stop the first ventilation fan; The timing determination unit is used to determine whether the third determination condition is met after waiting for a set time. If it is met, the second ventilation fan is determined to be a failed fan, and the result that the second ventilation fan is a failed fan is sent to the first decision subunit. The first decision subunit is also used to send the result that the first ventilation fan is a failed fan and the result that the second ventilation fan is a failed fan to the second decision subunit; The second decision subunit is used to determine the safety status result as an abnormal state based on the result that the first ventilation fan is a failed fan and the result that the second ventilation fan is a failed fan.

[0060] In this embodiment, if the third judgment condition is met, the operating status of the existing ventilation fans is first analyzed. If they are not operating, the ventilation fans are turned on to reduce the hydrogen concentration and thus eliminate the safety threat to the operation of the hydrogen fuel cell. If the third judgment condition is still met after each ventilation fan has been turned on for a period of time, then all ventilation fans have failed, indicating a potential risk to the hydrogen fuel cell operating environment, and the hydrogen supply channel is then suspended.

[0061] In a specific embodiment: S201, determine if there is a low wind pressure alarm. If yes, execute S202; otherwise, continue executing S201. S202, determine whether fan 1 is running; if yes, execute S203; otherwise, execute S208. S203, after a delay, determine whether the low wind pressure alarm still exists. If yes, execute S204; otherwise, execute S201. S204, determine that fan 1 has failed, output a fan 1 failure alarm, and execute S205; S205, determine whether the current state of fan 2 is failure. If yes, execute S206; otherwise, execute S207. S206, it is determined that both fan 1 and fan 2 have failed, and an alarm for all fans has failed is output. S207, start fan 2, stop fan 1, execute S201; S208, determine whether fan 2 is running; if yes, execute S209; otherwise, execute S213. S209, after a delay, determine whether the low wind pressure alarm still exists. If yes, execute S210; otherwise, execute S201. S210, determine that fan 2 has failed, output a fan 2 failure alarm, and execute S211; S211, determine whether the current state of fan 1 is failure. If yes, execute S206; otherwise, execute S212. S212, start fan 1, stop fan 2, execute S201; S213, neither fan 1 nor fan 2 is running. Compare the total running time of the two fans and start the fan with the shorter total running time, then execute S201.

[0062] In a more optimized solution, the decision-making unit further includes an optimization decision-making subunit; When the third judgment condition is met, the first decision subunit is used to determine whether there is a ventilation fan running. If there is no ventilation fan participating in the ventilation work, the optimization decision subunit is used to obtain the total running time of the ventilation fan, and prioritize the start-up order of the ventilation fans from low to high based on the total running time, and send the priority ranking result to the first decision subunit. The first decision subunit is also used to send ventilation fan start commands according to the priority order.

[0063] In some embodiments of the present invention, please refer to Figure 1 It also includes an alarm module 4, which is communicatively connected to the control and analysis module; The control analysis module is also used to send alarm commands; The alarm module is used to issue a security anomaly alarm based on the alarm command.

[0064] In this embodiment, by setting up an alarm module, a response and decision can be made as quickly as possible to further reduce risks.

[0065] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An automated safety monitoring device for a marine hydrogen fuel cell power generation system, characterized in that, include: Monitoring module, control and analysis module, and cut-off module; The monitoring module is used to acquire status information of various parts of the ship and send the status information to the control and analysis module; the status information includes hydrogen concentration, flame smoke status, and ventilation status; The control analysis module is used to analyze the operational safety of the ship's hydrogen fuel cell by using the state information as input features and employing a fully trained support vector machine classification model to obtain a safety state result. If the safety state result is an abnormal state, an abnormal state command is sent to the cut-off module. The cut-off module is used to control the valve of the hydrogen supply channel to close according to the abnormal status command.

2. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 1, characterized in that, The monitoring module includes a leak detection unit, a fire alarm detection unit, and a ventilation detection unit; The leakage monitoring unit is deployed in various hydrogen-related areas of the ship to acquire hydrogen concentration information in each hydrogen-related area and send the hydrogen concentration information to the control and analysis module. The fire detection unit is used to detect the flame and smoke status of the ship's cabin environment and send the flame and smoke status to the control and analysis module. The ventilation detection unit is used to detect the wind pressure value in each ventilation area of ​​the ship and send the wind pressure value to the control and analysis module.

3. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 2, characterized in that, The control analysis module includes an input unit, an analysis and prediction unit, a decision-making unit, and an instruction issuing unit; The input unit is used to receive the status information and send the status information to the analysis and prediction module; The analysis and prediction unit is used to analyze the operational safety of ship hydrogen fuel cells using the state information as input features and a fully trained support vector machine classification model to obtain the safety status result. The decision unit is used to receive the security status result and determine whether the security status result is a normal security result or an abnormal status. If the status is abnormal, the abnormal status instruction is sent to the instruction issuing unit. The instruction issuing unit is used to send the instruction indicating an abnormal status to the cut-off module.

4. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 3, characterized in that, The analysis and prediction unit includes a support vector machine subunit; The support vector machine subunit is used to predict the operational safety of ship hydrogen fuel cells by using the state information as input features and employing a fully trained support vector machine classification model to obtain the safety status result.

5. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 3, characterized in that, Each ventilation detection unit in the ventilation area of ​​the ship includes a first ventilation fan and a second ventilation fan, and at any given time, either the first ventilation fan or the second ventilation fan is engaged in ventilation work.

6. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 5, characterized in that, The ventilation detection unit also includes a fan starter; The fan starter is used to control the start and stop of the ventilation fan and to feed back the operating status of the ventilation fan to the decision unit.

7. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 6, characterized in that, The hydrogen concentration information being greater than the hydrogen concentration threshold is used as the first judgment condition; The presence of a fire, indicated by the flame and smoke conditions, is used as the second judgment condition. The third judgment condition is that the wind pressure value in the ventilation state is lower than the wind pressure threshold. The decision unit is used to determine that the security status result is abnormal when either the first judgment condition or the second judgment condition is met. The decision unit is also used to send a wind turbine start command to the wind turbine starter when the third judgment condition is met, and to determine the safety status result as an abnormal status based on the abnormal wind turbine operation feedback from the wind turbine starter.

8. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 7, characterized in that, The decision-making unit includes a first decision-making subunit, a timing determination unit, and a second decision-making subunit; When the third judgment condition is met, the first decision subunit is used to determine whether there is a ventilation fan running. If the first ventilation fan is running, a first fan running command will be generated and sent to the timing unit. The timing determination unit is used to determine whether the third determination condition is met after waiting for a set time. If it is met, the first ventilation fan is determined to be a failed fan, and the result that the first ventilation fan is a failed fan is sent to the first decision subunit. The first decision subunit is also configured to, upon receiving the result that the first ventilation fan is a failed fan, send a second ventilation fan start command to the fan starter to start the second ventilation fan, and send a first ventilation fan stop command to the fan starter to stop the first ventilation fan; The timing determination unit is used to determine whether the third determination condition is met after waiting for a set time. If it is met, the second ventilation fan is determined to be a failed fan, and the result that the second ventilation fan is a failed fan is sent to the first decision subunit. The first decision subunit is also used to send the result that the first ventilation fan is a failed fan and the result that the second ventilation fan is a failed fan to the second decision subunit; The second decision subunit is used to determine the safety status result as an abnormal state based on the result that the first ventilation fan is a failed fan and the result that the second ventilation fan is a failed fan.

9. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 8, characterized in that, The decision-making unit also includes an optimization decision-making subunit; When the third judgment condition is met, the first decision subunit is used to determine whether there is a ventilation fan running. If there is no ventilation fan participating in the ventilation work, the optimization decision subunit is used to obtain the total running time of the ventilation fan, and prioritize the start-up order of the ventilation fans from low to high based on the total running time, and send the priority ranking result to the first decision subunit. The first decision subunit is also used to send ventilation fan start commands according to the priority order.

10. The automated safety monitoring device for a marine hydrogen fuel cell power generation system according to claim 1, characterized in that, It also includes an alarm module, which is communicatively connected to the control and analysis module; The control analysis module is also used to send alarm commands; The alarm module is used to issue a security anomaly alarm based on the alarm command.

Images