Device and method for managing failures in hydrogen fuel cell based mobility

Abstract

An apparatus and method for managing a failure of a hydrogen fuel cell-based mobility are disclosed. The failure management apparatus includes a collection unit that collects mobility-related status information and communication network-related status information when the hydrogen fuel cell-based mobility is turned on; a diagnosis unit that applies the collected status information to a diagnosis model that has undergone learning related to failure diagnosis to diagnose whether there is an abnormality in the mobility, and if it is diagnosed that there is an abnormality in the mobility, determines at least one of the possibility of explosion and the risk of damage based on the location of the mobility and the cause of the diagnosed abnormality; and a drive control unit that controls the operation of the mobility based on the determined result.

Description

Technology Field

[0001] The present invention relates to a fault management device, and more specifically, to a device and method for managing faults in a hydrogen fuel cell-based mobility that diagnoses whether there is an abnormality in the hydrogen fuel cell-based mobility and supports driving control of the said mobility according to the diagnosis result. Background Technology

[0002] Generally, a hydrogen fuel cell is a power generation device that utilizes the reverse reaction of electrolysis to produce electricity and heat by supplying hydrogen extracted from sources such as petroleum gas as fuel and reacting it with oxygen from the air. Compared to turbine power generation methods using fossil fuels, hydrogen fuel cells are an eco-friendly energy source with higher energy efficiency and lower greenhouse gas emissions, and they are currently being used as a power source for transportation mobility.

[0003] Meanwhile, compared to conventional fossil fuel-based mobility, safety issues such as fire and explosion are becoming prominent in hydrogen fuel cell-based mobility; consequently, research on fault diagnosis to prevent safety accidents in advance is continuously underway. Prior art literature

[0004] Published Patent Application No. 10-2023-0032443 (March 7, 2023) The problem to be solved

[0005] The problem that the present invention aims to solve is to provide an apparatus and method for managing failures in hydrogen fuel cell-based mobility, which diagnoses abnormalities in hydrogen fuel cell-based mobility from various angles using artificial intelligence, and controls the operation of the mobility based on the location of the mobility when an abnormality occurs in the diagnosis results. means of solving the problem

[0006] To solve the above problem, the fault management device according to the present invention includes: a collection unit that collects status information related to the mobility and status information related to the communication network when the ignition of the hydrogen fuel cell-based mobility is turned on; a diagnosis unit that applies the collected status information to a diagnosis model in which fault diagnosis-related learning is performed to diagnose whether there is an abnormality in the mobility, and if it is diagnosed that there is an abnormality in the mobility, determines at least one of the possibility of explosion and the risk of damage based on the location of the mobility and the cause of the diagnosed abnormality; and a drive control unit that controls the operation of the mobility based on the determined result.

[0007] In addition, the above-mentioned collection unit is characterized by collecting at least one state information among hydrogen leakage, stack voltage, stack current, stack temperature, stack hydrogen flow rate, stack vibration, power pack hydrogen pressure, power pack temperature, power pack hydrogen flow rate, power pack current, battery voltage, battery current, battery temperature, battery charging speed, battery electrolyte leakage, mobility body coolant temperature, mobility body oil pressure, mobility body hydraulic pressure, mobility body hydraulic temperature, mobility body pump status, mobility internal communication / electronic control status, supercapacitor temperature, supercapacitor capacitance, supercapacitor self-discharge, supercapacitor internal resistance, supercapacitor voltage, external communication signal strength, external communication speed, external communication latency, external communication packet loss rate, and external communication reconnection frequency.

[0008] In addition, the diagnostic unit is characterized by classifying the state of the mobility using the diagnostic model into normal, hydrogen leakage, stack abnormality, abnormality of at least one of the power pack and battery, communication abnormality, pump abnormality, and supercapacitor abnormality, and diagnosing whether there is an abnormality.

[0009] In addition, the diagnostic unit is characterized by performing transfer learning of the diagnostic model in real-time or at preset intervals by linking with an external cloud server that derives an answer relatively closer to the correct answer than the diagnostic model.

[0010] In addition, the diagnostic unit provides at least one of diagnostic information diagnosed by the diagnostic model and driving control information based on the diagnostic information to the cloud server, and performs the transfer learning using at least one of the diagnostic information and driving control information derived from the cloud server based on the provided information.

[0011] In addition, the diagnostic unit is characterized by determining that there is a possibility of explosion when the mobility is located in a sealed space, and determining that there is relatively less possibility of explosion when the mobility is located in a non-sealed space compared to when it is located in a sealed space.

[0012] In addition, the diagnostic unit determines the risk of damage to the mobility when there is relatively no possibility of explosion, and determines that there is a risk of damage to the mobility if the cause of the abnormality of the mobility corresponds to at least one of a stack abnormality, a power pack and battery abnormality, and a pump abnormality.

[0013] In addition, the drive control unit is characterized by controlling the ignition of the mobility to be turned off and outputting an explosion-related warning message when it is determined that there is a possibility of explosion, or that there is a relative possibility of explosion and a risk of damage, and controlling the ignition of the mobility to be kept on and outputting an abnormal condition notification message when it is determined that there is a relative possibility of explosion.

[0014] A fault management method performed by a fault management device for managing the state of a hydrogen fuel cell-based mobility according to the present invention comprises: a step of collecting status information related to the mobility and status information related to a communication network when the mobility is turned on; a step of diagnosing whether there is an abnormality in the mobility by applying the collected status information to a diagnostic model that has undergone fault diagnosis-related learning; a step of determining at least one of the possibility of explosion and the risk of damage based on the location of the mobility and the cause of the diagnosed abnormality when it is diagnosed that there is an abnormality in the mobility; and a step of controlling the operation of the mobility based on the determined result. Effects of the invention

[0015] According to an embodiment of the present invention, an accurate diagnosis of abnormalities in hydrogen fuel cell-based mobility can be performed by using artificial intelligence to subdivide them into hydrogen leakage, stack abnormalities, power pack / battery abnormalities, communication abnormalities, pump abnormalities, and supercapacitor abnormalities.

[0016] In addition, it can support optimal drive control by determining the possibility of explosion based on the location of the mobility. Brief explanation of the drawing

[0017] FIG. 1 is a configuration diagram for explaining a fault management system according to an embodiment of the present invention. FIG. 2 is a block diagram illustrating a fault management device according to an embodiment of the present invention. FIG. 3 is a block diagram for explaining a control unit according to an embodiment of the present invention. FIG. 4 is a diagram illustrating the structure of a diagnostic model according to an embodiment of the present invention. FIG. 5 is a drawing for explaining the criteria for determining the possibility of explosion according to an embodiment of the present invention. FIG. 6 is a diagram illustrating transfer learning of a diagnostic model according to an embodiment of the present invention. FIG. 7 is a flowchart illustrating a fault management method according to an embodiment of the present invention. FIG. 8 is a block diagram illustrating a computing device according to an embodiment of the present invention. Specific details for implementing the invention

[0018] Embodiments of the present invention are described below with reference to the attached drawings so that those skilled in the art can easily implement them. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification are denoted by similar reference numerals.

[0019] In this specification and drawings (hereinafter referred to as the 'this specification'), redundant descriptions of identical components are omitted.

[0020] Furthermore, when a component is described in this specification as being 'connected' or 'connected' to another component, it should be understood that it may be directly connected to or connected to the other component, or that there may be other components in between. On the other hand, when a component is described in this specification as being 'directly connected' or 'directly connected' to another component, it should be understood that there are no other components in between.

[0021] Furthermore, the terms used in this specification are used merely to describe specific embodiments and are not intended to limit the invention.

[0022] Additionally, in this specification, singular expressions may include plural expressions unless the context clearly indicates otherwise.

[0023] Furthermore, in this specification, terms such as 'comprising' or 'having' are intended merely to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not excluding in advance the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0024] Additionally, in this specification, the term "and / or" includes a combination of the plurality of described items or any of the plurality of described items. In this specification, "A or B" may include "A," "B," or "both A and B."

[0025] In addition, detailed descriptions of known functions and configurations that may obscure the essence of the invention will be omitted in this specification.

[0027] FIG. 1 is a configuration diagram for explaining a fault management system according to an embodiment of the present invention.

[0028] Referring to FIG. 1, the fault management system (400) uses artificial intelligence to diagnose whether there is an abnormality in the hydrogen fuel cell-based mobility from various angles, and if an abnormality occurs in the diagnosis results, controls the operation of the mobility based on the location of the mobility. The fault management system (400) includes a fault management device (100), a cloud server (200), and a user terminal (300).

[0029] A fault management device (100) is installed in the mobility (M) and controls the operation according to whether the mobility (M) is faulty. Here, the mobility (M) is a means of transportation that operates based on a hydrogen fuel cell. That is, the mobility (M) may have the same overall structure as a conventional fossil fuel-based mobility, except that the engine block is replaced by a power pack (not shown). The power pack may be a power supply unit that integrates a fuel cell stack, an air supply device (air compressor, humidifier, etc.), a hydrogen supply module, a thermal management system, and a power conversion device (DC / DC converter, inverter) into a single module.

[0030] The fault management device (100) can diagnose faults using a diagnostic model to which artificial intelligence (AI) technology is applied. The diagnostic model is a model that has completed learning to diagnose faults based on the status information of the mobility (M). The fault management device (100) can automatically control the driving state of the mobility (M) based on the abnormality, the possibility of explosion, and the risk of damage of the mobility (M) as diagnosed results. Through this, the fault management device (100) can prevent safety accidents in advance and provide safety to the driver. In addition, the fault management device (100) generates monitoring information including the status information of the mobility (M), diagnostic-related information, and driving-state-related information. The fault management device (100) can output the monitoring information or transmit it to a user terminal (300).

[0031] The cloud server (200) communicates with the fault management device (100). The cloud server (200) may be a cloud that derives the optimal answer for fault diagnosis and drive control of the hydrogen fuel cell-based mobility (M), and its performance may derive an answer that is relatively closer to the correct answer than the diagnostic model of the fault management device (100). That is, the cloud server (200) may include at least one server computer and be equipped with high-performance computing capabilities. Preferably, the cloud server (200) includes an artificial intelligence model linked to the diagnostic model of the fault management device (100), and can perform transfer learning in real time or at preset intervals using the said artificial intelligence model. Here, the artificial intelligence model may be a model with higher diagnostic accuracy than the diagnostic model of the fault management device (100).

[0032] The user terminal (300) is a terminal used by the user and performs communication with the fault management device (100). The user terminal (300) receives monitoring information from the fault management device (100) and outputs the received monitoring information. Through this, the user terminal (300) supports the user in checking information related to mobility (M) in real time from a remote location. That is, when the user confirms that an abnormality has occurred in the mobility (M), the user terminal (300) helps the user to quickly respond to the accident even from a remote location.

[0033] Meanwhile, the fault management system (400) supports communication between the fault management device (100), the cloud server (200), and the user terminal (300) by establishing a communication network (450). The communication network (450) may be composed of a backbone network and a subscriber network. The backbone network may be composed of one or more integrated networks among an X.25 network, a Frame Relay network, an ATM network, an MPLS (Multi-Protocol Label Switching) network, and a GMPLS (Generalized Multi-Protocol Label Switching) network. The subscriber network may be FTTH (Fiber To The Home), ADSL (Asymmetric Digital Subscriber Line), cable network, Zigbee, Bluetooth, Wireless LAN (IEEE 802.11b, IEEE 802.11a, IEEE 802.11g, IEEE 802.11n), Wireless Hart (ISO / IEC 62591-1), ISA 100.11a (ISO / IEC 62734), CoAP (Constrained Application Protocol), MQTT (Message Queuing Telemetry Transport), WIBro (Wireless Broadband), WiMAX, 3G, HSDPA (High Speed ​​Downlink Packet Access), 4G, 5G, and 6G, etc. In some embodiments, the communication network (450) may be an internet network and a mobile communication network. Additionally, the communication network (450) may include any other widely known or future-developed wireless or wired communication methods.

[0035] FIG. 2 is a block diagram illustrating a fault management device according to an embodiment of the present invention.

[0036] Referring to FIGS. 1 and 2, the fault management device (100) includes a communication unit (10), a sensor unit (20), a camera unit (30), a control unit (40), an output unit (50), and a storage unit (60).

[0037] The communication unit (10) performs communication with the cloud server (200), the user terminal (300), and the mobility (M). The communication unit (10) transmits at least one of diagnostic information and driving control information to the cloud server (200) and receives the diagnostic information and driving control information derived from the cloud server (200). Additionally, the communication unit (10) transmits monitoring information related to the mobility (M) to the user terminal (300) and transmits the driving control signal of the mobility (M) to the ECU (Electronic Control Unit) (not shown) of the mobility (M).

[0038] The sensor unit (20) measures status information related to the mobility (M) and may include a hydrogen detection sensor, a current sensor, a voltage sensor, a temperature sensor, a pressure sensor, etc. Specifically, the sensor unit (20) measures hydrogen leakage of the mobility (M). The sensor unit (20) measures voltage, temperature, hydrogen flow rate, vibration, etc. related to the stack (not shown) of the mobility (M). The sensor unit (20) measures hydrogen pressure, temperature, hydrogen flow rate, current, etc. related to the power pack of the mobility (M). The sensor unit (20) measures voltage, current, temperature, charging speed, etc. related to the battery (not shown) of the mobility (M). The sensor unit (20) measures coolant temperature, oil pressure, hydraulic pressure, hydraulic temperature, pump status, etc. related to the vehicle body of the mobility (M). The sensor unit (20) measures temperature, capacitance, internal resistance, voltage, etc. related to the super capacitor (not shown) of the mobility (M). The sensor unit (20) transmits the measured state information to the control unit (40).

[0039] The camera unit (30) includes at least one camera and generates image information by capturing images to determine the location of the mobility (M). Preferably, the camera unit (30) may be installed on the roof of the mobility (M) to capture the surrounding environment of the mobility (M) (e.g., terrain, buildings, obstacles, etc.). Here, the image information may serve as a criterion for determining whether the mobility (M) is located in a closed space. Here, the camera unit (30) normally maintains an idle state, but is activated to capture images when a wake-up signal from the control unit (40) is detected.

[0040] The control unit (40) performs overall control of the fault management device (100). When the mobility (M) is turned on, the control unit (40) collects status information related to the mobility (M) and status information related to the communication network (450). The control unit (40) applies the collected status information to a diagnostic model that has undergone fault diagnosis training to diagnose whether there is an abnormality in the mobility (M). Here, the diagnostic model is a model that incorporates artificial intelligence technology and can be implemented as an Artificial Neural Network (ANN) or a Multilayer Perceptron (MLP), but is not limited thereto. If the control unit (40) diagnoses that there is no abnormality in the mobility (M), it maintains the current operation without any additional operation. Additionally, if the control unit (40) diagnoses that there is an abnormality in the mobility (M), it activates the camera unit (30) to check the location of the mobility (M) and determines the possibility of an explosion based on the image information generated from the camera unit (30). Additionally, the control unit (40) determines whether there is a risk of damage to the mobility (M) due to a diagnosed abnormal cause, even though there is no possibility of explosion in the mobility (M). The control unit (40) controls the operation of the mobility (M) based on the determined possibility of explosion and risk of damage. That is, the control unit (30) generates a drive control signal to control the operation of the mobility (M) and can control the generated drive control signal to be transmitted to the ECU of the mobility (M). The control unit (40) generates monitoring information including status information of the mobility (M), diagnosis-related information, driving status-related information, etc. The control unit (40) can control the monitoring information to be output or transmitted to a user terminal (300).

[0041] The output unit (50) outputs monitoring information generated from the control unit (40). The output unit (50) is installed on a dashboard provided inside the mobility (M) and can visually output the information (e.g., display, head-up, etc.). In addition, the output unit (50) can audibly output an alarm message or a notification message (e.g., alarm sound, notification sound, etc.).

[0042] The storage unit (60) stores a program or algorithm for operating the fault management device (100). The storage unit (60) stores information collected, generated, diagnosed, and determined during the process of managing mobility (M)-related faults. For example, the storage unit (60) may store status information, diagnostic information, drive control information, monitoring information, etc. The storage unit (60) may include a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, a magnetic disk, and an optical disk, etc.

[0044] FIG. 3 is a block diagram for explaining a control unit according to an embodiment of the present invention, FIG. 4 is a diagram for explaining the structure of a diagnostic model according to an embodiment of the present invention, FIG. 5 is a diagram for explaining a criterion for determining the possibility of explosion according to an embodiment of the present invention, and FIG. 6 is a diagram for explaining transfer learning of a diagnostic model according to an embodiment of the present invention.

[0045] Referring to FIGS. 1 to 6, the control unit (40) includes a collection unit (41), a diagnosis unit (42), and a drive control unit (43).

[0046] The collection unit (41) collects status information related to the mobility (M) and status information related to the communication network (450) when the mobility (M) is turned on. The collection unit (41) collects the status of the stack, power pack, battery, body, and super capacitor of the mobility (M). For example, the collection unit (41) collects at least one of hydrogen leakage, voltage of the stack, temperature of the stack, hydrogen flow rate of the stack, vibration of the stack, hydrogen pressure of the power pack, temperature of the power pack, hydrogen flow rate of the power pack, current of the power pack, voltage of the battery, current of the battery, temperature of the battery, charging speed of the battery, coolant temperature of the mobility (M) vehicle body, oil pressure of the mobility (M) vehicle body, hydraulic pressure of the mobility (M) vehicle body, hydraulic temperature of the mobility (M) vehicle body, pump status of the mobility (M) vehicle body, internal communication (e.g., CAN communication) of the mobility (M), electronic control (e.g., ECU) status, temperature of the super capacitor, capacitance of the super capacitor, internal resistance of the super capacitor, and voltage of the super capacitor. In addition, the collection unit (41) collects the status of communication related to the communication network (450). That is, the collection unit (41) collects information related to the status of communication performed with an external device. For example, the collection unit (41) collects at least one of the signal strength of the external communication, the speed of the external communication, the delay time of the external communication, the packet loss rate of the external communication, and the reconnection frequency of the external communication.

[0047] The diagnostic unit (42) applies the collected status information to a diagnostic model (DM) that has undergone fault diagnosis-related learning to diagnose whether there is an abnormality in the mobility (M). That is, the diagnostic unit (42) uses the diagnostic model (DM) to classify the state of the mobility (M) into hydrogen leakage (C1), stack abnormality (C2), abnormality of at least one of the power pack and battery (C3), communication abnormality (C4), pump abnormality (C5), supercapacitor abnormality (C6), and normal (C7) to diagnose whether there is an abnormality. Preferably, the diagnostic unit (42) can use the diagnostic model (DM) to perform an optimal diagnosis close to the correct answer for the situation occurring in the actual mobility (M).

[0048] The diagnostic model (DM) may be a multilayer perceptron and includes an input layer (IL), a plurality of hidden layers (HL, HL1 to HLk), and an output layer (OL). Each of the plurality of layers (IL, HL, OL) includes a plurality of nodes. For example, the input layer (IL) may include n input nodes (i1 to in), and the output layer (OL) may include 7 output nodes (o1 to o7). Additionally, among the hidden layers (HL), the first hidden layer (HL1) may include a number of nodes (h11 to h1a), the second hidden layer (HL2) may include b number of nodes (h21 to h2b), and the k-th hidden layer (HLK) may include c number of nodes (hk1 to hkc).

[0049] All multiple nodes in each layer have operations. In particular, multiple nodes in different layers are connected by weighted channels. Here, the result of an operation by a single node becomes the input to the next layer node to which the weight is applied.

[0050] In this way, a node corresponding to a layer of the diagnostic model (DM) receives a value with weights applied to the input from a node of the previous layer, sums it to apply an activation function, and transmits the result as an input to the next layer. Accordingly, when state information is input to the input layer (IL) of the diagnostic model (DM), the diagnostic model (DM) performs multiple operations with weights applied from multiple layers (IL, HL, OL) to the state information and outputs a diagnostic result, which is a predicted diagnostic value. At this time, the diagnostic result may be classified and output as hydrogen leakage (C1), stack abnormality (C2), abnormality of at least one of the power pack and battery (C3), communication abnormality (C4), pump abnormality (C5), supercapacitor abnormality (C6), and normal (C7).

[0051] Here, the diagnostic model (DM) can diagnose a hydrogen leak (C1) by using status information related to a drop in pressure of the hydrogen tank (not shown) of the mobility (M), and detection of hydrogen leaks in the hydrogen tank, input / output line, power pack, etc. of the mobility (M). The diagnostic model (DM) can diagnose a stack abnormality (C2) by using at least one of status information related to a drop in the preset voltage of the stack, a current amount lower than the amount of current applied to the stack, a rise in the temperature of the stack, a decrease in the hydrogen flow rate of the stack, analysis by spectroscopy, and vibration analysis regarding impact on the stack. The diagnostic model (DM) can diagnose an abnormality (C3) of at least one of the power pack and the battery by using at least one of status information related to a drop in hydrogen pressure of the power pack, a rise in the temperature of the power pack, a decrease in the hydrogen flow rate of the power pack, overcurrent / undercurrent of the power pack, overvoltage / undervoltage of the battery, overcurrent / undercurrent of the battery, a rise in the temperature of the battery, a decrease in the charging speed of the battery, and electrolyte leakage of the battery. The diagnostic model (DM) can diagnose a communication abnormality (C4) by using at least one of the status information related to a decrease in the signal strength of external communication, a decrease in the speed of external communication, an increase in the latency of external communication, an increase in the packet loss rate of external communication, and an increase in the frequency of reconnection of external communication. The diagnostic model (DM) can diagnose a pump abnormality (C5) indicating an abnormal hydraulic fluid pressure by using at least one of the status information related to an increase in the coolant temperature of the mobility body, an increase in the oil pressure of the mobility body, an increase in the hydraulic pressure of the mobility body, an increase in the hydraulic temperature of the mobility body, the pump status of the mobility body, and the internal communication / electronic control status of the mobility. The diagnostic model (DM) can diagnose a supercapacitor abnormality (C6) by using at least one of the following status information: a rise in temperature of the supercapacitor (fast charging / discharging, cell failure), a decrease in capacitance of the supercapacitor (decrease in stored charge), an increase in internal resistance of the supercapacitor (cell degradation, contact failure), an increase in self-discharge of the supercapacitor (insulation breakdown, internal short circuit), and an overvoltage / undervoltage of the supercapacitor (balancing circuit abnormality).Finally, the diagnostic module (DM) can diagnose it as normal (C7) if it determines that there are no abnormalities in the collected status information.

[0052] When the diagnostic unit (42) diagnoses that there is an abnormality in the mobility (M), it determines the possibility of explosion based on the location of the mobility (M). At this time, the diagnostic unit (42) activates the camera unit (30) to generate image information about the surroundings of the mobility (M) and determines the possibility of explosion based on the generated image information. Using the image information, the diagnostic unit (42) determines that there is a possibility of explosion if the mobility (M) is located in a sealed space, and determines that there is relatively less possibility of explosion if the mobility (M) is located in a non-sealed space compared to when it is located in a sealed space. That is, the diagnostic unit (42) determines that if the mobility (M) malfunctions in a sealed space, there is a high possibility of explosion because hydrogen can accumulate in that space, and if the mobility (M) malfunctions in a non-sealed space, there is a low possibility of explosion because hydrogen can escape into the air. For example, the sealed space may be a parking lot located indoors, and the non-sealed space may be an open space located outdoors, but is not limited thereto. Meanwhile, the diagnostic unit (42) can determine that there is relatively no possibility of explosion even if the hydrogen leak is minor and the unit is located in a sealed space.

[0053] Additionally, the diagnostic unit (42) determines the risk of damage to the mobility (M) based on the diagnosed abnormal cause, if there is relatively no possibility of explosion in the mobility (M). The diagnostic unit (42) may determine that there is a possibility of damage to the mobility (M) body if at least one of the diagnosed abnormal cause is a stack abnormality (C2), a power pack / battery abnormality (C3), and a pump abnormality (C5).

[0054] Here, the diagnostic unit (42) can perform transfer learning of the diagnostic model (DM) in real time or at preset intervals by linking with an external cloud server (200) that derives an answer that is relatively closer to the correct answer than the diagnostic model (DM).

[0055] For example, transfer learning can be performed through the following process. The diagnostic unit (42) performs a fault diagnosis of the mobility (M) based on the diagnostic model (DM). If no abnormalities occur in the diagnosed results, the diagnostic unit (42) transmits the diagnostic information to the cloud server (200). If abnormalities occur in the diagnosed results, the diagnostic unit (42) diagnoses the possibility of explosion and the risk of damage based on the image information, and transmits the diagnosed diagnostic information and the possibility of explosion and the risk of damage to the cloud server (200). The cloud server (200) independently derives the fault diagnosis and the possibility of explosion and the risk of damage based on the received information. At this time, the cloud server (200) can derive an answer that is relatively closer to the correct answer than the diagnostic model (DM), and as a result, it can predict fault-related diagnostic information and the possibility of explosion and the risk of damage more accurately than the diagnostic model (DM). The cloud server (200) transmits the derived diagnostic information and the possibility of explosion and the risk of damage to the diagnostic model (DM). The diagnostic model (DM) performs transfer learning based on information received from the cloud server (200), and can further improve diagnostic accuracy through transfer learning. That is, when diagnosing a fault in the future, the diagnostic model (DM) can perform fault diagnosis using the transferred-learned diagnostic model.

[0056] The drive control unit (43) controls the operation of the mobility (M) based on the determined possibility of explosion and risk of damage. If the drive control unit (43) determines that there is a possibility of explosion of the mobility (M), it controls the ignition of the mobility (M) to be turned off and outputs an explosion-related warning message. The explosion-related warning message is a message regarding the possibility of explosion of the mobility (M) and countermeasures, and may be output to display at least one of visual and auditory effects, and power for outputting the message may be provided from a stored battery. If the drive control unit (43) determines that there is relatively no possibility of explosion of the mobility (M) but there is a risk of damage to the mobility (M), it controls the ignition of the mobility (M) to be turned off and outputs an explosion-related warning message. Additionally, if the drive control unit (43) determines that there is relatively no possibility of explosion of the mobility (M) and there is no risk of damage to the mobility (M), it controls the ignition of the mobility (M) to be kept on and outputs an abnormal condition notification message. An abnormal condition notification message is a message regarding the part where an abnormality occurred in the mobility (M) and the corresponding measures, and can be output such that at least one of a visual effect and an auditory effect is present, and power for outputting the message can be supplied from a stored battery.

[0057] The drive control unit (43) can control the mobility (M) to ensure the safety of the user (e.g., driver, passenger) riding the mobility (M) through the control described above, while ensuring optimal response measures are performed. That is, if there is a risk of explosion or damage to the mobility (M), the user can ensure safety by automatically turning off the engine and simultaneously evacuate quickly according to the warning message. In addition, if there is relatively no risk of explosion or damage to the mobility (M), the engine is kept running, allowing the user to secure time to minimize damage by moving to a place (e.g., shoulder, empty lot, etc.) that will not cause damage to the surroundings even if the mobility (M) explodes or is damaged, while simultaneously taking optimal action according to the notification message.

[0058] Additionally, the drive control unit (43) generates monitoring information including status information of the mobility (M), diagnostic information, driving status information, etc. The drive control unit (43) can control the monitoring information to be output or transmitted to a user terminal (300). Here, the monitoring information may include messages delivered to the user, such as warning messages and notification messages.

[0060] FIG. 7 is a flowchart illustrating a fault management method according to an embodiment of the present invention.

[0061] Referring to FIGS. 1 and 7, the fault management method can accurately diagnose abnormalities in a hydrogen fuel cell-based mobility (M) by using artificial intelligence to subdivide them into hydrogen leakage, stack abnormalities, power pack / battery abnormalities, communication abnormalities, pump abnormalities, and supercapacitor abnormalities. In addition, the fault management method can support optimal drive control by determining the possibility of explosion based on the location of the mobility (M).

[0062] In step S110, the hydrogen fuel cell-based mobility (M) is turned on. The mobility (M) is turned on by a user.

[0063] In step S120, the fault management device (100) collects information related to mobility (M). The fault management device (100) collects status information related to mobility (M) and status information related to the communication network (450). The status information related to mobility (M) may include information related to hydrogen leakage, stack abnormality, power pack / battery abnormality, operating fluid pressure (pump abnormality) of mobility (M), supercapacitor abnormality, etc., and the status information related to the communication network (450) may include information related to the communication status between external devices, etc.

[0064] In step S130, the fault management device (100) diagnoses an abnormality in the mobility (M). The fault management device (100) determines the abnormality in the mobility (M) using a learned diagnostic model related to fault diagnosis. The fault management device (100) can diagnose whether there is an abnormality by classifying the state of the mobility (M) using the diagnostic model into normal, hydrogen leakage, stack abnormality, abnormality of at least one of the power pack and battery, communication abnormality, pump abnormality, and supercapacitor abnormality.

[0065] In step S140, the fault management device (100) determines whether there is a diagnosis that there is a problem with the mobility (M). If the fault management device (100) determines that there is a problem with the mobility (M), it performs step S150, and if it determines that there is no problem with the mobility (M), it performs step S120 again.

[0066] In step S150, the fault management device (100) determines the possibility of explosion of the mobility (M). If the mobility (M) is located in a sealed space, the fault management device (100) determines that there is a possibility of explosion and performs step S160, and if the mobility (M) is located in a non-sealed space, it determines that there is relatively less possibility of explosion compared to when it is located in a sealed space and performs step S170.

[0067] In step S160, the fault management device (100) changes the ignition of the mobility (M) to an off state and outputs a warning message. The fault management device (100) turns off the ignition to prevent an explosion and outputs an explosion-related warning message to help the user take quick action.

[0068] In step S170, the fault management device (100) determines whether there is a risk of damage to the mobility (M). The fault management device (100) may determine that there is a risk of damage to the mobility (M) body if the cause of the abnormality of the mobility (M) is at least one of a stack abnormality (C2), a power pack / battery abnormality (C3), and a pump abnormality (C5). If there is a risk of damage, the fault management device (100) performs step S160, and if there is no risk of damage, performs step S180.

[0069] In step S180, the fault management device (100) outputs a notification message while keeping the mobility (M) ignition on. The fault management device (100) outputs a notification message regarding an abnormal condition while keeping the mobility (M) ignition on so that it can move to an area where it can avoid causing surrounding damage even if an explosion occurs, thereby helping the user to take action regarding the fault later.

[0071] FIG. 8 is a block diagram illustrating a computing device according to an embodiment of the present invention.

[0072] Referring to FIG. 8, the computing device (TN100) may be a device described in the present specification (e.g., a fault management device, a cloud server, a user terminal, etc.).

[0073] The computing device (TN100) may include at least one processor (TN110), a transceiver (TN120), and a memory (TN130). Additionally, the computing device (TN100) may further include a storage device (TN140), an input interface device (TN150), an output interface device (TN160), etc. The components included in the computing device (TN100) may be connected by a bus (TN170) to communicate with each other.

[0074] The processor (TN110) can execute a program command stored in at least one of the memory (TN130) and the storage device (TN140). The processor (TN110) may refer to a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to embodiments of the present invention are performed. The processor (TN110) may be configured to implement the procedures, functions, and methods described in relation to embodiments of the present invention. The processor (TN110) can control each component of the computing device (TN100).

[0075] Each of the memory (TN130) and the storage device (TN140) can store various information related to the operation of the processor (TN110). Each of the memory (TN130) and the storage device (TN140) may be composed of at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory (TN130) may be composed of at least one of read-only memory (ROM) and random access memory (RAM).

[0076] The transmitting and receiving device (TN120) can transmit or receive wired or wireless signals. The transmitting and receiving device (TN120) can be connected to a network to perform communication.

[0078] Meanwhile, embodiments of the present invention are not limited to being implemented only through the apparatus and / or methods described so far, but may also be implemented through a program that realizes a function corresponding to the configuration of the embodiments of the present invention or a recording medium on which such program is recorded, and such implementation can be easily achieved by a person skilled in the art to which the present invention belongs based on the description of the embodiments described above.

[0080] Although embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements made by a person skilled in the art using the basic concept of the present invention as defined in the following claims also fall within the scope of the present invention. Explanation of the symbols

[0081] 10: Communications Department 20: Sensor section 30: Camera Department 40: Control unit 41: Collection Department 42: Diagnostic Department 43: Drive control unit 50: Output section 60: Storage section 100: Fault Management Device 200: Cloud Server 300: User terminal 400: Fault Management System 450: Communication network

Claims

Claim 1 A collection unit that collects status information related to the mobility and status information related to the communication network when the hydrogen fuel cell-based mobility is turned on; a diagnosis unit that applies the collected status information to a diagnosis model that has undergone fault diagnosis learning to diagnose whether there is an abnormality in the mobility, and if it is diagnosed that there is an abnormality in the mobility, determines at least one of the possibility of explosion and the risk of damage based on the location of the mobility and the cause of the diagnosed abnormality; and a drive control unit that controls the operation of the mobility based on the determined result.The diagnostic unit includes, wherein the diagnostic unit classifies the state of the mobility into normal, hydrogen leakage, stack abnormality, power pack abnormality, battery abnormality, communication abnormality, pump abnormality, and supercapacitor abnormality using the diagnostic model to diagnose whether there is an abnormality, and provides at least one of the diagnostic information diagnosed by the diagnostic model and the drive control information based on the diagnostic information to an external cloud server that derives an answer relatively closer to the correct answer than the diagnostic model, and performs transfer learning of the diagnostic model in real time or at preset intervals using at least one of the diagnostic information and drive control information derived from the cloud server based on the provided information, and performs fault diagnosis using the transferred-learned diagnostic model when a fault is diagnosed thereafter, and if there is an abnormality in the diagnosed result, activates at least one camera to generate video information about the surroundings of the mobility, and determines whether the mobility is located in a sealed space based on the generated video information, and if the result determined that it is located in a sealed space, determines that there is a possibility of explosion, and if it is located in an open space, determines that there is relatively less possibility of explosion than when it is located in a sealed space, and if there is relatively less possibility of explosion, determines the risk of damage to the mobility, wherein the cause of the abnormality is a stack abnormality, power pack A fault management device characterized by determining that there is a risk of damage to the mobility if at least one of an abnormality, a battery abnormality, and a pump abnormality is present, and the drive control unit controls the mobility to switch to an off state and output an explosion-related warning message if it is determined that there is a possibility of explosion or if there is a relatively low possibility of explosion and there is a risk of damage, and controls the mobility to keep the engine on and output an abnormality-related notification message if it is determined that there is a relatively low possibility of explosion and there is no risk of damage. Claim 2 A fault management device according to claim 1, wherein the collection unit collects at least one state information among hydrogen leakage, stack voltage, stack current, stack temperature, stack hydrogen flow rate, stack vibration, power pack hydrogen pressure, power pack temperature, power pack hydrogen flow rate, power pack current, battery voltage, battery current, battery temperature, battery charging speed, battery electrolyte leakage, mobility body coolant temperature, mobility body oil pressure, mobility body hydraulic pressure, mobility body hydraulic temperature, mobility body pump status, mobility internal communication / electronic control status, supercapacitor temperature, supercapacitor capacitance, supercapacitor self-discharge, supercapacitor internal resistance, supercapacitor voltage, external communication signal strength, external communication speed, external communication delay time, external communication packet loss rate, and external communication reconnection frequency. Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete Claim 7 delete Claim 8 delete Claim 9 A fault management method performed by a fault management device that manages the state of a hydrogen fuel cell-based mobility, comprising: a step of collecting status information related to the mobility and status information related to a communication network when the ignition of the mobility is turned on; a step of diagnosing whether there is an abnormality in the mobility by applying the collected status information to a diagnostic model that has undergone fault diagnosis-related learning; a step of determining at least one of the possibility of explosion and the risk of damage based on the location of the mobility and the cause of the diagnosed abnormality when it is diagnosed that there is an abnormality in the mobility; and a step of controlling the operation of the mobility based on the determined result.The method includes, wherein the diagnosing step classifies the state of the mobility into normal, hydrogen leakage, stack abnormality, power pack abnormality, battery abnormality, communication abnormality, pump abnormality, and supercapacitor abnormality using the diagnostic model to diagnose whether there is an abnormality, provided at least one of the diagnostic information diagnosed by the diagnostic model and the drive control information based on the diagnostic information to an external cloud server that derives an answer relatively closer to the correct answer than the diagnostic model, and performs transfer learning of the diagnostic model in real-time or at preset intervals using at least one of the diagnostic information and drive control information derived from the cloud server based on the provided information, and performs a fault diagnosis using the transferred-learned diagnostic model when a fault diagnosis occurs thereafter, and the judging step activates at least one camera to generate image information regarding the surroundings of the mobility, and determines whether the mobility is located in a sealed space based on the generated image information, and if the result of the judgment is that it is located in a sealed space, determines that there is a possibility of explosion, and if it is located in an open space, determines that there is relatively less possibility of explosion than when it is located in a sealed space, and if there is relatively less possibility of explosion, determines the risk of damage to the mobility, wherein the cause of the abnormality is a stack abnormality, A fault management method characterized by determining that there is a risk of damage to the mobility if at least one of a power pack abnormality, a battery abnormality, and a pump abnormality is present, and the controlling step is characterized by controlling the mobility to switch to an off state and output an explosion-related warning message if it is determined that there is a possibility of explosion or if there is a relative possibility of explosion and there is a risk of damage, and controlling the mobility to maintain an on state and output an abnormal state notification message if it is determined that there is a relative possibility of explosion and there is no risk of damage.

Images