Hydrogen leakage detection method and device

Abstract

The invention discloses a hydrogen leakage detection method and device. According to the scheme, a hydrogen leakage detection request sent by a target fuel cell vehicle is received, the target fuel cell vehicle comprises a target fuel cell engine, and a plurality of preset hydrogen concentration sensors are installed in the target fuel cell engine; the hydrogen leakage detection request comprises a unique identifier corresponding to each preset hydrogen concentration sensor and a plurality of hydrogen concentration values; determining a hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to the plurality of hydrogen concentration values corresponding to each preset hydrogen concentration sensor; and when the hydrogen concentration change trend corresponding to one or more preset hydrogen concentration sensors is that the hydrogen leakage hidden danger exists and the hydrogen concentration change trend corresponding to the other preset hydrogen concentration sensors is that the hydrogen leakage does not exist, sending early warning information to target terminal equipment to prompt that the target fuel cell engine has the hydrogen leakage hidden danger. And timely maintenance is needed.

Description

Technical Field

[0001] The present application relates to the technical field of fuel cell vehicles, and in particular to a hydrogen leakage detection method and device. Background Art

[0002] In recent years, with the continuous development of society, people's living standards have been continuously improved, and people's demand for cars has also increased. Due to the increasingly serious energy shortage and environmental pollution caused by traditional cars, fuel cell vehicles using hydrogen as fuel have come into being. Fuel cell vehicles convert the chemical energy of hydrogen into electrical energy through fuel cell engines to provide power for the entire vehicle, with the advantages of high efficiency and environmental protection. However, hydrogen is a highly flammable and explosive gas, and the safety issues during the use of hydrogen cannot be ignored. Therefore, it is necessary to perform hydrogen leakage detection on fuel cell engines to improve the safety of fuel cell vehicles.

[0003] At present, the staff usually sets the corresponding hydrogen concentration value (i.e., the preset hydrogen concentration threshold) of the fuel cell engine based on experience when the fuel cell engine has a hydrogen leakage problem. During the operation of the fuel cell vehicle, the hydrogen concentration values ​​of various places in the fuel cell engine are collected by multiple preset hydrogen concentration sensors pre-installed in the fuel cell engine, and the collected hydrogen concentration values ​​are sent to the FCU (Fuel cell Control Unit, fuel cell engine main controller) in the fuel cell engine. When the hydrogen concentration value collected by a preset hydrogen concentration sensor is greater than the preset hydrogen concentration threshold, the FCU controls the fuel cell engine to stop working and sends an alarm message to the VCU (Vehicle Control Unit, vehicle controller) so that the VCU can alarm through the central control screen. Therefore, the existing hydrogen leakage detection method can only alarm when the fuel cell engine has a hydrogen leakage problem, so that to a certain extent, there are still safety hazards of combustion and explosion of fuel cell vehicles, and the safety of fuel cell vehicles cannot be effectively guaranteed. Summary of the invention

[0004] The present disclosure provides a hydrogen leakage detection method and device, the main purpose of which is to solve the problem that the existing hydrogen leakage detection method can only alarm when a hydrogen leakage problem occurs in a fuel cell engine, so that to a certain extent, there are still safety hazards of combustion and explosion of fuel cell vehicles.

[0005] A first aspect of the present disclosure provides a hydrogen leakage detection method, which is applied to a cloud platform. The method comprises:

[0006] Receiving a hydrogen leakage detection request sent by a target fuel cell vehicle, wherein the target fuel cell vehicle includes a target fuel cell engine, a plurality of preset hydrogen concentration sensors are installed in the target fuel cell engine, and the hydrogen leakage detection request includes a unique identifier corresponding to each preset hydrogen concentration sensor and a plurality of hydrogen concentration values, and the plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor are a plurality of hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensor;

[0007] Determine a hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to a plurality of hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor;

[0008] When the hydrogen concentration change trend corresponding to one or more preset hydrogen concentration sensors indicates that there is a risk of hydrogen leakage, and the hydrogen concentration change trend corresponding to the remaining preset hydrogen concentration sensors indicates that there is no hydrogen leakage, an early warning message is sent to the target terminal device to prompt that the target fuel cell engine has a risk of hydrogen leakage and needs to be repaired in time, wherein the early warning message includes a unique identifier corresponding to each preset hydrogen concentration sensor whose hydrogen concentration change trend indicates that there is a risk of hydrogen leakage.

[0009] In some embodiments, determining the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor includes:

[0010] A plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensors are input into the hydrogen leakage detection model so that the hydrogen leakage detection model outputs the hydrogen concentration variation trend corresponding to the preset hydrogen concentration sensors.

[0011] In some embodiments, the method further comprises:

[0012] Acquire a training sample set, wherein the training sample set includes a plurality of first training samples, a plurality of second training samples, and a plurality of third training samples. For any one of the first training samples, the first training sample includes a plurality of first sample hydrogen concentration values ​​and a first hydrogen concentration change trend. The plurality of first sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is no hydrogen leakage in the sample fuel cell engine, and the first hydrogen concentration change trend is no hydrogen leakage. For any one of the second training samples, the second training sample includes a plurality of second sample hydrogen concentration values ​​and a second hydrogen concentration change trend. The plurality of second sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is a potential hydrogen leakage in the sample fuel cell engine, and the second hydrogen concentration change trend is that there is a potential hydrogen leakage. For any one of the third training samples, the third training sample includes a plurality of third sample hydrogen concentration values ​​and a third hydrogen concentration change trend. The plurality of third sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is an imminent hydrogen leakage in the sample fuel cell engine, and the third hydrogen concentration change trend is that there is an imminent hydrogen leakage.

[0013] Iteratively train the preset deep learning model based on the training sample set;

[0014] After each round of training, determine whether the loss function of the preset deep learning model converges;

[0015] If the loss function converges, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0016] If the loss function does not converge, the model parameters of the preset deep learning model are optimized and adjusted according to the loss function, and the next round of training is entered based on the optimized and adjusted preset deep learning model.

[0017] In some embodiments, the hydrogen leakage detection request further includes the collection time corresponding to each hydrogen concentration value; determining the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor includes:

[0018] Calculate the hydrogen concentration change rate and the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor and the collection time corresponding to each hydrogen concentration value;

[0019] When the hydrogen concentration change rate is less than the first preset change rate threshold, and the average hydrogen concentration value is less than the historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is no hydrogen leakage;

[0020] When the hydrogen concentration change rate is greater than or equal to the first preset change rate threshold and the hydrogen concentration change rate is less than the second preset change rate threshold, or the average hydrogen concentration value is greater than or equal to the historical average hydrogen concentration value and the average hydrogen concentration value is less than the sum of the historical average hydrogen concentration value and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is a potential hydrogen leakage risk;

[0021] When the hydrogen concentration change rate is greater than the second preset change rate threshold, or the average hydrogen concentration value is greater than the sum of the historical average hydrogen concentration value and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is about to cause hydrogen leakage.

[0022] In some embodiments, the method further comprises:

[0023] When the hydrogen concentration change trend corresponding to any preset hydrogen concentration sensor indicates that a hydrogen leakage is about to occur, an alarm message is sent to the target fuel cell vehicle so that after receiving the alarm message, the target fuel cell vehicle controls the target fuel cell engine to stop working, and outputs and displays the alarm message through the central control screen, wherein the alarm message includes a unique identifier corresponding to the preset hydrogen concentration sensor whose hydrogen concentration change trend indicates that a hydrogen leakage is about to occur.

[0024] In some embodiments, the method further comprises:

[0025] If the loss function has not converged, determine whether the current cumulative iterative training time has reached the preset time threshold;

[0026] If the preset time threshold is reached, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0027] If the preset time threshold is not reached, the step of optimizing and adjusting the model parameters of the preset deep learning model according to the loss function is entered, and the next round of training is entered based on the optimized and adjusted preset deep learning model.

[0028] In some embodiments, the method further comprises:

[0029] If the loss function has not converged, determine whether the current cumulative number of iterative training times has reached the preset number threshold;

[0030] If the preset number threshold is reached, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0031] If the preset number of times is not reached, the step of optimizing and adjusting the model parameters of the preset deep learning model according to the loss function and entering the next round of training based on the optimized and adjusted preset deep learning model is entered.

[0032] A second aspect of the present disclosure provides a hydrogen leakage detection device, which is applied to a cloud platform, and includes:

[0033] A receiving unit, configured to receive a hydrogen leakage detection request sent by a target fuel cell vehicle, wherein the target fuel cell vehicle includes a target fuel cell engine, a plurality of preset hydrogen concentration sensors are installed in the target fuel cell engine, and the hydrogen leakage detection request includes a unique identifier corresponding to each preset hydrogen concentration sensor and a plurality of hydrogen concentration values, and the plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor are a plurality of hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensor;

[0034] a determination unit, configured to determine a hydrogen concentration variation trend corresponding to each preset hydrogen concentration sensor according to a plurality of hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor;

[0035] The first sending unit is used to send an early warning message to a target terminal device to prompt that a target fuel cell engine has a potential hydrogen leakage hazard and needs to be repaired in time when the hydrogen concentration change trend corresponding to one or more preset hydrogen concentration sensors indicates that there is a potential hydrogen leakage hazard and the hydrogen concentration change trend corresponding to the remaining preset hydrogen concentration sensors indicates that there is no hydrogen leakage, wherein the early warning message includes a unique identifier corresponding to each preset hydrogen concentration sensor whose hydrogen concentration change trend indicates that there is a potential hydrogen leakage hazard.

[0036] In some embodiments, the determination unit is specifically configured to input a plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensors into the hydrogen leakage detection model, so that the hydrogen leakage detection model outputs the hydrogen concentration variation trend corresponding to the preset hydrogen concentration sensors.

[0037] In some embodiments, the apparatus further comprises:

[0038] A training unit, used to obtain a training sample set, wherein the training sample set includes a plurality of first training samples, a plurality of second training samples, and a plurality of third training samples. For any first training sample, the first training sample includes a plurality of first sample hydrogen concentration values ​​and a first hydrogen concentration change trend. The plurality of first sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is no hydrogen leakage in the sample fuel cell engine, and the first hydrogen concentration change trend is no hydrogen leakage. For any second training sample, the second training sample includes a plurality of second sample hydrogen concentration values ​​and a second hydrogen concentration change trend. The plurality of second sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is a potential hydrogen leakage in the sample fuel cell engine, and the second hydrogen concentration change trend is that there is a potential hydrogen leakage. For any third training sample, the third training sample includes a plurality of third sample hydrogen concentration values ​​and a third hydrogen concentration change trend. The plurality of third sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is an imminent hydrogen leakage in the sample fuel cell engine, and the third hydrogen concentration change trend is that there is an imminent hydrogen leakage.

[0039] Iteratively train the preset deep learning model based on the training sample set;

[0040] After each round of training, determine whether the loss function of the preset deep learning model converges;

[0041] If the loss function converges, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0042] If the loss function does not converge, the model parameters of the preset deep learning model are optimized and adjusted according to the loss function, and the next round of training is entered based on the optimized and adjusted preset deep learning model.

[0043] In some embodiments, the hydrogen leakage detection request further includes a collection time corresponding to each hydrogen concentration value; a determination unit is specifically configured to calculate a hydrogen concentration change rate and an average hydrogen concentration value corresponding to a preset hydrogen concentration sensor according to a plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor and a collection time corresponding to each hydrogen concentration value;

[0044] When the hydrogen concentration change rate is less than the first preset change rate threshold, and the average hydrogen concentration value is less than the historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is no hydrogen leakage;

[0045] When the hydrogen concentration change rate is greater than or equal to the first preset change rate threshold and the hydrogen concentration change rate is less than the second preset change rate threshold, or the average hydrogen concentration value is greater than or equal to the historical average hydrogen concentration value and the average hydrogen concentration value is less than the sum of the historical average hydrogen concentration value and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is a potential hydrogen leakage risk;

[0046] When the hydrogen concentration change rate is greater than the second preset change rate threshold, or the average hydrogen concentration value is greater than the sum of the historical average hydrogen concentration value and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is about to cause hydrogen leakage.

[0047] In some embodiments, the apparatus further comprises:

[0048] The second sending unit is used to send an alarm message to the target fuel cell vehicle when the hydrogen concentration change trend corresponding to any preset hydrogen concentration sensor is that a hydrogen leakage is about to occur, so that the target fuel cell vehicle controls the target fuel cell engine to stop working after receiving the alarm message, and outputs and displays the alarm message through the central control screen, wherein the alarm message contains a unique identifier corresponding to the preset hydrogen concentration sensor whose hydrogen concentration change trend is that a hydrogen leakage is about to occur.

[0049] In some embodiments, the training unit is further used to: when the loss function has not converged, determine whether the current cumulative iterative training time reaches a preset time threshold;

[0050] When the preset time threshold is reached, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0051] When the preset time threshold is not reached, the steps of optimizing and adjusting the model parameters of the preset deep learning model according to the loss function and entering the next round of training based on the optimized and adjusted preset deep learning model are performed.

[0052] In some embodiments, the training unit is further used to: when the loss function has not converged, determine whether the current cumulative number of iterative training times reaches a preset number threshold;

[0053] When the preset number threshold is reached, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0054] When the preset number of times is not reached, the steps of optimizing and adjusting the model parameters of the preset deep learning model according to the loss function and entering the next round of training based on the optimized and adjusted preset deep learning model are performed.

[0055] A third aspect of the present disclosure provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above hydrogen leakage detection method is implemented.

[0056] A fourth aspect of the present disclosure provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the above hydrogen leakage detection method when executed by the processor.

[0057] By means of the above technical solution, the technical solution provided by the present disclosure has at least the following advantages:

[0058] The present disclosure provides a hydrogen leakage detection method and device. The present disclosure can determine the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor after a cloud platform receives a hydrogen leakage detection request sent by a target fuel cell vehicle and carries a unique identity identifier corresponding to each preset hydrogen concentration sensor and a plurality of hydrogen concentration values. When the hydrogen concentration change trend corresponding to one or several preset hydrogen concentration sensors indicates that there is a hydrogen leakage hazard, and the hydrogen concentration change trend corresponding to the remaining preset hydrogen concentration sensors indicates that there is no hydrogen leakage, the cloud platform sends an early warning message to the target terminal device to remind the owner of the target fuel cell vehicle that the target fuel cell engine has a hydrogen leakage hazard and needs to be repaired in time. In the present disclosure, since the cloud platform performs hydrogen leak detection based on multiple hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensor, it is possible to determine the changing trend of the hydrogen concentration near the preset hydrogen concentration sensor, thereby being able to determine whether the vicinity of the preset hydrogen concentration sensor is in a state of no hydrogen leakage, a state of a potential hydrogen leakage, or a state of an imminent hydrogen leakage; and since the warning information contains a unique identifier corresponding to each preset hydrogen concentration sensor with a potential hydrogen leakage, the owner of the target fuel cell vehicle and the vehicle maintenance personnel can quickly determine the location of the potential hydrogen leakage in the target fuel cell engine based on the warning information, thereby being able to effectively improve the efficiency of vehicle maintenance personnel in servicing the target fuel cell vehicle.

[0059] The above description is only an overview of the technical solution of the present disclosure. In order to more clearly understand the technical means of the present disclosure, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present disclosure more obvious and easy to understand, the specific implementation methods of the present disclosure are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0061] Figure 1 A flow chart of a hydrogen leakage detection method provided in an embodiment of the present disclosure;

[0062] Figure 2 A flow chart of another hydrogen leakage detection method provided in an embodiment of the present disclosure;

[0063] Figure 3 A block diagram of a hydrogen leakage detection device provided in an embodiment of the present disclosure;

[0064] Figure 4 A block diagram of another hydrogen leakage detection device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0065] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0066] In addition, the words “first”, “second” and the like used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different parts.

[0067] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in the present disclosure should have the common meanings understood by those skilled in the art to which the present disclosure belongs.

[0068] At present, staff usually set in advance based on experience the hydrogen concentration value (i.e., the preset hydrogen concentration threshold) corresponding to the fuel cell engine when a hydrogen leakage problem occurs in the fuel cell engine. During the operation of the fuel cell vehicle, the hydrogen concentration values ​​at various locations in the fuel cell engine are collected by multiple preset hydrogen concentration sensors pre-installed in the fuel cell engine, and the collected hydrogen concentration values ​​are sent to the FCU in the fuel cell engine. When the hydrogen concentration value collected by a preset hydrogen concentration sensor is greater than the preset hydrogen concentration threshold, the FCU controls the fuel cell engine to stop working and sends an alarm message to the VCU so that the VCU can sound an alarm through the central control screen.

[0069] Therefore, the existing hydrogen leakage detection method can only alarm when a hydrogen leakage problem occurs in the fuel cell engine, so that to a certain extent, there are still safety hazards of combustion and explosion of fuel cell vehicles, and the safety of fuel cell vehicles cannot be effectively guaranteed.

[0070] In order to ensure the safety of fuel cell vehicles, the present disclosure provides a hydrogen leakage detection method, which is applied to a cloud platform, such as Figure 1 As shown, the method includes:

[0071] 101. Receive a hydrogen leak detection request sent by a target fuel cell vehicle.

[0072] The target fuel cell vehicle includes a target fuel cell engine, which includes an FCU, a fuel cell, a TBOX (Telematics Box, an on-board terminal), a plurality of preset hydrogen concentration sensors, etc., wherein the installation positions of the plurality of preset hydrogen concentration sensors may be, but are not limited to, at the tail exhaust outlet of the fuel cell engine, at the purge port of the fuel cell stack box, directly above the fuel cell engine assembly, etc.; the hydrogen leakage detection request includes a unique identifier and a plurality of hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor, and for any preset hydrogen concentration sensor, the plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor are a plurality of hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensor.

[0073] During the operation of the target fuel cell vehicle, each preset hydrogen concentration sensor in the target fuel cell engine collects nearby hydrogen concentration values ​​in real time, and sends the unique identity, the collected hydrogen concentration value and the collection time to the FCU via the CAN bus. After receiving the preset number of multiple hydrogen concentration values ​​sent by each preset hydrogen concentration sensor, the FCU can send the unique identity corresponding to each preset hydrogen concentration sensor, the multiple hydrogen concentration values ​​and the collection time corresponding to each hydrogen concentration value to the TBOX via the CAN bus (wherein the preset number can be but is not limited to 5, 10, 15, etc., and the present disclosure does not make specific limitations on this). Then, the TBOX sends a hydrogen leakage detection request carrying the unique identity corresponding to each preset hydrogen concentration sensor and the multiple hydrogen concentration values ​​to the cloud platform via the wireless network. At this time, the cloud platform can receive the hydrogen leakage detection request sent by the target fuel cell vehicle.

[0074] 102. Determine a hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to a plurality of hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor.

[0075] For any preset hydrogen concentration sensor, the corresponding hydrogen concentration variation trend is one of: no hydrogen leakage, potential hydrogen leakage and imminent hydrogen leakage.

[0076] After receiving the hydrogen leakage detection request sent by the target fuel cell vehicle and carrying the unique identification and multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor, the cloud platform can determine the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor. Since the cloud platform performs hydrogen leakage detection based on the multiple hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensor, it is possible to determine the change trend of the hydrogen concentration near the preset hydrogen concentration sensor, and thus determine whether the preset hydrogen concentration sensor is in a state of no hydrogen leakage, a state of potential hydrogen leakage, or a state of imminent hydrogen leakage.

[0077] 103. When the hydrogen concentration change trend corresponding to one or more preset hydrogen concentration sensors indicates that there is a risk of hydrogen leakage, and the hydrogen concentration change trend corresponding to the remaining preset hydrogen concentration sensors indicates that there is no hydrogen leakage, a warning message is sent to the target terminal device.

[0078] Among them, the target terminal device is the terminal device of the target fuel cell vehicle owner, which can be but is not limited to: a smart phone, a tablet computer, etc.; wherein the warning information includes a unique identifier corresponding to each preset hydrogen concentration sensor with a hydrogen leakage hazard, and a hydrogen concentration change trend.

[0079] After the cloud platform determines the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor, when the hydrogen concentration change trend corresponding to one or several preset hydrogen concentration sensors indicates that there is a risk of hydrogen leakage, and the hydrogen concentration change trends corresponding to the remaining preset hydrogen concentration sensors indicate that there is no hydrogen leakage, the cloud platform needs to send an early warning message to the target terminal device to remind the owner of the target fuel cell vehicle that the target fuel cell engine has a risk of hydrogen leakage and needs to be repaired in time.

[0080] It should be noted that since the warning information contains the unique identifier corresponding to each preset hydrogen concentration sensor with a hydrogen leakage risk, the owner of the target fuel cell vehicle and the vehicle maintenance personnel can quickly determine the location of the hydrogen leakage risk in the target fuel cell engine based on the warning information, thereby effectively improving the efficiency of vehicle maintenance personnel in servicing the target fuel cell vehicle.

[0081] The above is the hydrogen leakage detection method provided by the embodiment of the present disclosure. In the present disclosure, since the cloud platform performs hydrogen leakage detection based on multiple hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensor, it is possible to determine the changing trend of the hydrogen concentration near the preset hydrogen concentration sensor, thereby being able to determine whether the vicinity of the preset hydrogen concentration sensor is in a state of no hydrogen leakage, a state of a potential hydrogen leakage, or a state of an imminent hydrogen leakage; and since the warning information contains a unique identifier corresponding to each preset hydrogen concentration sensor with a potential hydrogen leakage, the owner of the target fuel cell vehicle and the vehicle maintenance personnel can quickly determine the location of the potential hydrogen leakage in the target fuel cell engine based on the warning information, thereby being able to effectively improve the efficiency of vehicle maintenance personnel in servicing the target fuel cell vehicle.

[0082] To explain in more detail below, the present disclosure provides another method for detecting hydrogen leakage, as follows: Figure 2 As shown, the method includes:

[0083] 201. Iteratively train the preset deep learning model based on the training sample set to obtain a hydrogen leak detection model.

[0084] In order to ensure that the cloud platform can accurately determine the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor based on the multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor after receiving the hydrogen leakage detection request sent by the target fuel cell vehicle, the cloud platform needs to pre-train the hydrogen leakage detection model. The following will explain in detail how to train the hydrogen leakage detection model.

[0085] (1) Obtain a training sample set.

[0086] The training sample set includes a plurality of first training samples, a plurality of second training samples and a plurality of third training samples. For any first training sample, the first training sample includes a plurality of first sample hydrogen concentration values ​​and a first hydrogen concentration change trend. The plurality of first sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected by a hydrogen concentration sensor when there is no hydrogen leakage in the sample fuel cell engine, and the first hydrogen concentration change trend is no hydrogen leakage. For any second training sample, the second training sample includes a plurality of second sample hydrogen concentration values ​​and a second hydrogen concentration change trend. The plurality of second sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected by a hydrogen concentration sensor when there is a hidden danger of hydrogen leakage in the sample fuel cell engine, and the second hydrogen concentration change trend is that there is a hidden danger of hydrogen leakage. For any third training sample, the third training sample includes a plurality of third sample hydrogen concentration values ​​and a third hydrogen concentration change trend. The plurality of third sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected by a hydrogen concentration sensor when there is about to be a hydrogen leakage in the sample fuel cell engine, and the third hydrogen concentration change trend is that a hydrogen leakage is about to occur.

[0087] (2) Iteratively train the preset deep learning model based on the training sample set.

[0088] Among them, the preset deep learning model is a deep learning model established according to any existing deep learning algorithm, and the present disclosure does not make any specific limitations on this.

[0089] The preset deep learning model is iteratively trained based on a training sample set including multiple first training samples, multiple second training samples and multiple third training samples; wherein, after each round of training, it is determined whether the loss function of the preset deep learning model converges; if the loss function converges, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model; if the loss function does not converge, the model parameters of the preset deep learning model are optimized and adjusted according to the loss function, and the next round of training is entered based on the optimized and adjusted preset deep learning model.

[0090] That is, after performing multiple rounds of iterative training on the preset deep learning model based on the training sample set, when the loss function of the preset deep learning model converges, it is determined that the preset deep learning model at this time is a hydrogen leak detection model.

[0091] In some specific cases, even after a large number of iterative trainings, the loss function of the preset deep learning model will not converge. Therefore, in order to avoid the endless iterative training of the preset deep learning model, when it is determined that the loss function of the preset deep learning model obtained after the current round of training has not converged, the following two methods can be used for processing, but are not limited to:

[0092] (1) If the loss function of the preset deep learning model has not converged, determine whether the current cumulative iterative training time of the preset deep learning model based on the training sample set iterative training reaches a preset time threshold.

[0093] If the current cumulative iterative training time reaches the preset time threshold, it means that the iterative training time has met the requirement. At this time, the iterative training can be stopped, and the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model.

[0094] If the current cumulative iterative training time does not reach the preset time threshold, the model parameters of the preset deep learning model can be optimized and adjusted according to the loss function of the preset deep learning model, and the next round of training can be entered based on the optimized and adjusted preset deep learning model.

[0095] (2) If the loss function of the preset deep learning model has not converged, determine whether the current cumulative number of iterative training of the preset deep learning model based on the training sample set has reached a preset number threshold.

[0096] If the current cumulative number of iterative training times reaches the preset threshold, it means that the number of iterative training times has met the requirement. At this time, the iterative training can be stopped, and the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model.

[0097] If the current cumulative number of iterative training times does not reach the preset threshold, the model parameters of the preset deep learning model can be optimized and adjusted according to the loss function of the preset deep learning model, and the next round of training can be entered based on the optimized and adjusted preset deep learning model.

[0098] 202. Receive a hydrogen leak detection request sent by a target fuel cell vehicle.

[0099] Regarding step 202, receiving a hydrogen leak detection request sent by a target fuel cell vehicle, please refer to Figure 1 The description of the corresponding parts will not be repeated here in the embodiments of the present disclosure.

[0100] 203. Determine a hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to a plurality of hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor.

[0101] After receiving a hydrogen leakage detection request sent by the target fuel cell vehicle and carrying a unique identification code and multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor, the cloud platform can determine the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor.

[0102] Specifically, in this step, the cloud platform may determine the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor in any of the following two ways:

[0103] (1) For any preset hydrogen concentration sensor, multiple hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor are input into the hydrogen leakage detection model, so that the hydrogen leakage detection model outputs the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor.

[0104] (2) The hydrogen leakage detection request sent by TBOX to the cloud platform through the wireless network also includes the collection time corresponding to each hydrogen concentration value; for any preset hydrogen concentration sensor, the hydrogen concentration change rate and the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor are calculated according to the multiple hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor and the collection time corresponding to each hydrogen concentration value; when the hydrogen concentration change rate corresponding to the preset hydrogen concentration sensor is less than the first preset change rate threshold, and the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor is less than the historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is no hydrogen leakage; when the hydrogen concentration change rate corresponding to the preset hydrogen concentration sensor is greater than or equal to the first preset change rate threshold and the preset hydrogen concentration sensor is less than the historical average hydrogen concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is no hydrogen leakage. When the hydrogen concentration change rate corresponding to the hydrogen concentration sensor is less than the second preset change rate threshold, or the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor is greater than or equal to the historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor and the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor is less than the sum of the historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is that there is a hidden danger of hydrogen leakage; when the hydrogen concentration change rate corresponding to the preset hydrogen concentration sensor is greater than the second preset change rate threshold, or the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor is greater than the sum of the historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is that a hydrogen leakage is about to occur.

[0105] The historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor is obtained by averaging multiple historical hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor carried in multiple historical hydrogen leakage detection requests previously sent by the target fuel cell vehicle.

[0106] Among them, according to the multiple hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor and the collection time corresponding to each hydrogen concentration value, the specific process of calculating the hydrogen concentration change rate and the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor is:

[0107] First, the average value of multiple hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor is calculated to obtain the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor; secondly, among the multiple hydrogen concentration values, the hydrogen concentration value with the smallest collection time is determined as the first hydrogen concentration value, and the hydrogen concentration value with the largest collection time is determined as the second hydrogen concentration value, the first hydrogen concentration value is subtracted from the second hydrogen concentration value to obtain the hydrogen concentration difference, and the collection time corresponding to the first hydrogen concentration value is subtracted from the collection time corresponding to the second hydrogen concentration value to obtain the collection time interval, and the hydrogen concentration difference is divided by the collection time interval to obtain the hydrogen concentration change rate corresponding to the preset hydrogen concentration sensor.

[0108] 204a. When the hydrogen concentration change trend corresponding to one or more preset hydrogen concentration sensors indicates that there is a risk of hydrogen leakage, and the hydrogen concentration change trend corresponding to the remaining preset hydrogen concentration sensors indicates that there is no hydrogen leakage, a warning message is sent to the target terminal device.

[0109] Among them, regarding step 204a, when the hydrogen concentration change trend corresponding to one or more preset hydrogen concentration sensors indicates that there is a risk of hydrogen leakage, and the hydrogen concentration change trend corresponding to the remaining preset hydrogen concentration sensors indicates that there is no hydrogen leakage, sending an early warning message to the target terminal device, you can refer to Figure 1 The description of the corresponding parts will not be repeated here in the embodiments of the present disclosure.

[0110] 204b. When the hydrogen concentration change trend corresponding to any preset hydrogen concentration sensor indicates that hydrogen leakage is about to occur, an alarm message is sent to the target fuel cell vehicle.

[0111] The alarm information includes a unique identifier corresponding to a preset hydrogen concentration sensor whose hydrogen concentration change trend indicates that hydrogen leakage is about to occur.

[0112] After the cloud platform determines the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor, when the hydrogen concentration change trend corresponding to any preset hydrogen concentration sensor is about to cause hydrogen leakage, the cloud platform needs to send an alarm message carrying the unique identifier corresponding to the preset hydrogen concentration sensor whose hydrogen concentration change trend is about to cause hydrogen leakage to the target fuel cell vehicle, that is, it is necessary to send the alarm message to the TBOX in the target fuel cell vehicle through the wireless network. After receiving the alarm message, the TBOX will send the alarm message to the FCU through the CAN bus. After receiving the alarm message, the FCU will control the target fuel cell engine to stop working to prevent the target fuel cell engine from continuing to work and causing combustion and explosion, and will send the alarm message to the VCU through the CAN bus, and the VCU will output and display the alarm message through the central control screen.

[0113] It should be noted that, since the alarm information contains the alarm information of the unique identifier corresponding to the preset hydrogen concentration sensor whose hydrogen concentration change trend is about to cause hydrogen leakage, the owner of the target fuel cell vehicle and the vehicle maintenance personnel can quickly determine the location where the hydrogen leakage is about to occur in the target fuel cell engine based on the alarm information, thereby effectively improving the efficiency of vehicle maintenance personnel in servicing the target fuel cell vehicle.

[0114] It should be noted that, in the present disclosure, when hydrogen leakage is about to occur in the target fuel cell engine, the target fuel cell engine is controlled to stop working and an alarm is sounded, thus being able to avoid combustion and explosion of the target fuel cell engine to a certain extent.

[0115] As the above Figure 1 and Figure 2 In order to realize the method shown in the figure, the embodiment of the present disclosure provides a hydrogen leakage detection device. This device embodiment corresponds to the aforementioned method embodiment. For the sake of ease of reading, this device embodiment will not repeat the details of the aforementioned method embodiment one by one, but it should be clear that the hydrogen leakage detection device in this device embodiment can correspond to all the contents of the aforementioned method embodiment. Specifically, Figure 3 As shown, the device comprises:

[0116] The receiving unit 31 is used to receive a hydrogen leakage detection request sent by a target fuel cell vehicle, wherein the target fuel cell vehicle includes a target fuel cell engine, and a plurality of preset hydrogen concentration sensors are installed in the target fuel cell engine. The hydrogen leakage detection request includes a unique identifier corresponding to each preset hydrogen concentration sensor and a plurality of hydrogen concentration values, and the plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor are a plurality of hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensor.

[0117] A determining unit 32, configured to determine a hydrogen concentration variation trend corresponding to each preset hydrogen concentration sensor according to a plurality of hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor received by the receiving unit 31;

[0118] The first sending unit 33 is used to send an early warning message to the target terminal device to prompt that the target fuel cell engine has a hydrogen leakage hazard and needs to be repaired in time when the determination unit 32 determines that the hydrogen concentration change trend corresponding to one or more preset hydrogen concentration sensors is that there is a hydrogen leakage hazard, and the hydrogen concentration change trend corresponding to the remaining preset hydrogen concentration sensors is that there is no hydrogen leakage, wherein the early warning message includes a unique identifier corresponding to each preset hydrogen concentration sensor whose hydrogen concentration change trend is that there is a hydrogen leakage hazard.

[0119] In some embodiments, Figure 4As shown, the determination unit 32 is specifically used to input multiple hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensors into the hydrogen leakage detection model, so that the hydrogen leakage detection model outputs the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensors.

[0120] In some embodiments, Figure 4 As shown, the device also includes:

[0121] The training unit 34 is used to obtain a training sample set, wherein the training sample set includes a plurality of first training samples, a plurality of second training samples, and a plurality of third training samples. For any first training sample, the first training sample includes a plurality of first sample hydrogen concentration values ​​and a first hydrogen concentration change trend. The plurality of first sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is no hydrogen leakage in the sample fuel cell engine, and the first hydrogen concentration change trend is no hydrogen leakage. For any second training sample, the second training sample includes a plurality of second sample hydrogen concentration values ​​and a second hydrogen concentration change trend. The plurality of second sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is a potential hydrogen leakage in the sample fuel cell engine, and the second hydrogen concentration change trend is that there is a potential hydrogen leakage. For any third training sample, the third training sample includes a plurality of third sample hydrogen concentration values ​​and a third hydrogen concentration change trend. The plurality of third sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is about to be a hydrogen leakage in the sample fuel cell engine, and the third hydrogen concentration change trend is that there is about to be a hydrogen leakage.

[0122] Iteratively train the preset deep learning model based on the training sample set;

[0123] After each round of training, determine whether the loss function of the preset deep learning model converges;

[0124] If the loss function converges, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0125] If the loss function does not converge, the model parameters of the preset deep learning model are optimized and adjusted according to the loss function, and the next round of training is entered based on the optimized and adjusted preset deep learning model.

[0126] In some embodiments, Figure 4 As shown, the hydrogen leakage detection request also includes the collection time corresponding to each hydrogen concentration value; the determination unit 32 is specifically used to calculate the hydrogen concentration change rate and the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor and the collection time corresponding to each hydrogen concentration value;

[0127] When the hydrogen concentration change rate is less than the first preset change rate threshold, and the average hydrogen concentration value is less than the historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is no hydrogen leakage;

[0128] When the hydrogen concentration change rate is greater than or equal to the first preset change rate threshold and the hydrogen concentration change rate is less than the second preset change rate threshold, or the average hydrogen concentration value is greater than or equal to the historical average hydrogen concentration value and the average hydrogen concentration value is less than the sum of the historical average hydrogen concentration value and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is a potential hydrogen leakage risk;

[0129] When the hydrogen concentration change rate is greater than the second preset change rate threshold, or the average hydrogen concentration value is greater than the sum of the historical average hydrogen concentration value and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is about to cause hydrogen leakage.

[0130] In some embodiments, Figure 4 As shown, the device also includes:

[0131] The second sending unit 35 is used to send an alarm message to the target fuel cell vehicle when the determination unit 32 determines that the hydrogen concentration change trend corresponding to any preset hydrogen concentration sensor is about to cause a hydrogen leakage, so that the target fuel cell vehicle controls the target fuel cell engine to stop working after receiving the alarm message, and outputs and displays the alarm message through the central control screen, wherein the alarm message includes a unique identifier corresponding to the preset hydrogen concentration sensor whose hydrogen concentration change trend is about to cause a hydrogen leakage.

[0132] In some embodiments, Figure 4 As shown, the training unit 34 is also used to: when the loss function has not converged, determine whether the current cumulative iterative training time reaches a preset time threshold;

[0133] When the preset time threshold is reached, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0134] When the preset time threshold is not reached, the steps of optimizing and adjusting the model parameters of the preset deep learning model according to the loss function and entering the next round of training based on the optimized and adjusted preset deep learning model are performed.

[0135] In some embodiments, Figure 4 As shown, the training unit 34 is also used to: when the loss function has not converged, determine whether the current cumulative number of iterative training times reaches a preset number threshold;

[0136] When the preset number threshold is reached, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model;

[0137] When the preset number of times is not reached, the steps of optimizing and adjusting the model parameters of the preset deep learning model according to the loss function and entering the next round of training based on the optimized and adjusted preset deep learning model are performed.

[0138] As described above, the feeding device provided in the embodiment of the present disclosure includes a processor and a memory. The above-mentioned receiving unit 31, determining unit 32 and first sending unit 33 are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0139] An embodiment of the present disclosure provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above hydrogen leakage detection method is implemented.

[0140] An embodiment of the present disclosure provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the above hydrogen leakage detection method is implemented when the processor executes the computer program.

[0141] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0142] In a typical configuration, the device includes one or more processors (CPU), memory and bus. The device may also include input / output interface, network interface and the like.

[0143] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.

[0144] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0145] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0146] It will be appreciated by those skilled in the art that the embodiments of the present disclosure may be provided as methods, systems or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] The above are only embodiments of the present disclosure and are not intended to limit the present disclosure. For those skilled in the art, the present disclosure may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the scope of the claims of the present disclosure.

[0148] All terms used in the present disclosure have the same meanings as those understood by those of ordinary skill in the art to which the present disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in general dictionaries, for example, should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0149] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0150] So far, various embodiments of the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0151] Although some specific embodiments of the present disclosure have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified or some technical features may be replaced by equivalents without departing from the scope and spirit of the present disclosure. In particular, the various technical features mentioned in the various embodiments may be combined in any manner as long as there is no structural conflict.

Claims

1. A method for detecting hydrogen leakage, characterized in that: The method is applied to a cloud platform, and the method comprises: Receiving a hydrogen leakage detection request sent by a target fuel cell vehicle, wherein the target fuel cell vehicle includes a target fuel cell engine, a plurality of preset hydrogen concentration sensors are installed in the target fuel cell engine, the hydrogen leakage detection request includes a unique identifier and a plurality of hydrogen concentration values ​​corresponding to each of the preset hydrogen concentration sensors, and the plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensors are a plurality of hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensors; Determine a hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to a plurality of hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor; When the hydrogen concentration change trend corresponding to one or more of the preset hydrogen concentration sensors indicates that there is a risk of hydrogen leakage, and the hydrogen concentration change trend corresponding to the other preset hydrogen concentration sensors indicates that there is no hydrogen leakage, an early warning message is sent to the target terminal device to prompt that the target fuel cell engine has a risk of hydrogen leakage and needs to be repaired in time, wherein the early warning message includes a unique identifier corresponding to each of the preset hydrogen concentration sensors whose hydrogen concentration change trend indicates that there is a risk of hydrogen leakage.

2. The hydrogen leakage detection method according to claim 1, characterized in that: The determining the hydrogen concentration change trend corresponding to each preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to each preset hydrogen concentration sensor includes: A plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor are input into a hydrogen leakage detection model so that the hydrogen leakage detection model outputs a hydrogen concentration variation trend corresponding to the preset hydrogen concentration sensor.

3. The hydrogen leakage detection method according to claim 2, characterized in that: The method further comprises: Acquire a training sample set, wherein the training sample set includes a plurality of first training samples, a plurality of second training samples, and a plurality of third training samples. For any one of the first training samples, the first training sample includes a plurality of first sample hydrogen concentration values ​​and a first hydrogen concentration change trend. The plurality of first sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is no hydrogen leakage in the sample fuel cell engine, and the first hydrogen concentration change trend is no hydrogen leakage. For any one of the second training samples, the second training sample includes a plurality of second sample hydrogen concentration values ​​and a second hydrogen concentration change trend. The plurality of second sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is a potential hydrogen leakage in the sample fuel cell engine, and the second hydrogen concentration change trend is that there is a potential hydrogen leakage. For any one of the third training samples, the third training sample includes a plurality of third sample hydrogen concentration values ​​and a third hydrogen concentration change trend. The plurality of third sample hydrogen concentration values ​​are a plurality of hydrogen concentration values ​​collected when there is about to be a hydrogen leakage in the sample fuel cell engine, and the third hydrogen concentration change trend is that there is about to be a hydrogen leakage. Iterative training is performed on the preset deep learning model based on the training sample set; wherein, After each round of training, determining whether the loss function of the preset deep learning model converges; If the loss function converges, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model; If the loss function does not converge, the model parameters of the preset deep learning model are optimized and adjusted according to the loss function, and the next round of training is entered based on the preset deep learning model after optimization and adjustment.

4. The hydrogen leakage detection method according to claim 1, characterized in that: The hydrogen leakage detection request also includes the collection time corresponding to each of the hydrogen concentration values; the determining the hydrogen concentration change trend corresponding to each of the preset hydrogen concentration sensors according to the multiple hydrogen concentration values ​​corresponding to each of the preset hydrogen concentration sensors includes: Calculate the hydrogen concentration change rate and the average hydrogen concentration value corresponding to the preset hydrogen concentration sensor according to the multiple hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensor and the collection time corresponding to each of the hydrogen concentration values; When the hydrogen concentration change rate is less than a first preset change rate threshold, and the average hydrogen concentration value is less than the historical average hydrogen concentration value corresponding to the preset hydrogen concentration sensor, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is no hydrogen leakage; When the hydrogen concentration change rate is greater than or equal to the first preset change rate threshold and the hydrogen concentration change rate is less than the second preset change rate threshold, or the average hydrogen concentration value is greater than or equal to the historical average hydrogen concentration value and the average hydrogen concentration value is less than the sum of the historical average hydrogen concentration value and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is a potential hydrogen leakage hazard; When the hydrogen concentration change rate is greater than the second preset change rate threshold, or the average hydrogen concentration value is greater than the sum of the historical average hydrogen concentration value and the target concentration value, it is determined that the hydrogen concentration change trend corresponding to the preset hydrogen concentration sensor is about to cause hydrogen leakage.

5. The hydrogen leakage detection method according to claim 1, characterized in that: The method further comprises: When the hydrogen concentration change trend corresponding to any one of the preset hydrogen concentration sensors indicates that a hydrogen leakage is about to occur, an alarm message is sent to the target fuel cell vehicle so that after receiving the alarm message, the target fuel cell vehicle controls the target fuel cell engine to stop working, and outputs and displays the alarm message through the central control screen, wherein the alarm message contains a unique identifier corresponding to the preset hydrogen concentration sensor whose hydrogen concentration change trend indicates that a hydrogen leakage is about to occur.

6. The hydrogen leakage detection method according to claim 3, characterized in that: The method further comprises: If the loss function has not converged, determining whether the current cumulative iterative training time has reached a preset time threshold; If the preset time threshold is reached, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model; If the preset time threshold is not reached, the step of optimizing and adjusting the model parameters of the preset deep learning model according to the loss function, and entering the next round of training based on the optimized and adjusted preset deep learning model is performed.

7. The hydrogen leakage detection method according to claim 3, characterized in that: The method further comprises: If the loss function has not converged, determining whether the current cumulative number of iterative training times reaches a preset number threshold; If the preset number threshold is reached, the preset deep learning model obtained after this round of training is determined as the hydrogen leak detection model; If the preset number threshold is not reached, the step of optimizing and adjusting the model parameters of the preset deep learning model according to the loss function, and entering the next round of training based on the preset deep learning model after optimization and adjustment is performed.

8. A hydrogen leakage detection device, characterized in that: The device is applied to a cloud platform, and comprises: A receiving unit, configured to receive a hydrogen leakage detection request sent by a target fuel cell vehicle, wherein the target fuel cell vehicle comprises a target fuel cell engine, a plurality of preset hydrogen concentration sensors are installed in the target fuel cell engine, the hydrogen leakage detection request comprises a unique identifier and a plurality of hydrogen concentration values ​​corresponding to each of the preset hydrogen concentration sensors, and the plurality of hydrogen concentration values ​​corresponding to the preset hydrogen concentration sensors are a plurality of hydrogen concentration values ​​continuously collected by the preset hydrogen concentration sensors; a determination unit, configured to determine a hydrogen concentration variation trend corresponding to each of the preset hydrogen concentration sensors according to a plurality of hydrogen concentration values ​​corresponding to each of the preset hydrogen concentration sensors; The first sending unit is used to send an early warning message to a target terminal device when the hydrogen concentration change trend corresponding to one or more of the preset hydrogen concentration sensors indicates that there is a potential hydrogen leakage risk and the hydrogen concentration change trend corresponding to the other preset hydrogen concentration sensors indicates that there is no hydrogen leakage, so as to prompt that the target fuel cell engine has a potential hydrogen leakage risk and needs to be repaired in time, wherein the early warning message includes a unique identifier corresponding to each of the preset hydrogen concentration sensors whose hydrogen concentration change trend indicates that there is a potential hydrogen leakage risk.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the hydrogen leakage detection method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the hydrogen leakage detection method according to any one of claims 1 to 7.

Images