Fuel cell single cell voltage fault online diagnosis and recovery method, system and vehicle

Abstract

The embodiment of the present application relates to the technical field of fuel cell, and discloses a kind of fuel cell single voltage fault on-line diagnosis and recovery method, system and vehicle, comprising: the voltage signal of acquisition battery single;Identify whether battery single exists single voltage too low fault;When single voltage too low fault, execute emergency system shutdown, store fault information and prohibition start instruction;After detecting system low voltage power-on, carry out state detection, and when meeting preset detection condition, execute air, hydrogen gas pressure building operation;According to voltage signal determination pressure building result, if pressure building succeeds, then clear fault and prohibition start instruction, if pressure building fails, then offline maintenance is prompted;When pressure building result is pressure building success and receive start request, on-line diagnose stack is in membrane dry state or waterlogged state, and according to diagnostic result, execute corresponding performance recovery strategy.The present application can on-line, quickly distinguish single low fault root cause, and on-line recovery strategy is applied to recoverable fault.

Description

Technical Field

[0001] This invention relates to the field of fuel cell technology, and specifically to a method, system, and vehicle for online diagnosis and recovery of voltage faults in a fuel cell cell. Background Technology

[0002] Fuel cells, as a highly efficient and clean energy conversion device, directly convert the chemical energy of fuel into electrical energy, boasting significant advantages such as high energy conversion efficiency and zero environmental pollution. They demonstrate enormous application potential in fields such as transportation power and distributed power generation, and are one of the important technological pathways to achieving the "dual-carbon" strategic goals.

[0003] In real-world operating scenarios of automotive fuel cell systems, frequent low-voltage (single-cell low) faults in the fuel cell stack severely impact system output power, durability, and reliability. This fault is typically caused by two main factors:

[0004] First, there is irreversible physical damage, such as the proton exchange membrane being "pulled through" due to excessive local current or "burned through" due to overheating caused by cooling failure. Such damage is irreversible and requires shutdown and offline replacement of the membrane electrode.

[0005] Second, recoverable performance degradation or abnormal operating conditions, including:

[0006] 1) Membrane dryness: Low inlet humidity or excessive current density leads to dehydration of the proton exchange membrane, resulting in a decrease in proton conductivity;

[0007] 2) Water flooding: Liquid water accumulates in the catalyst layer or gas diffusion layer, hindering the transport of reactant gases;

[0008] 3) Temporary chemical poisoning or catalyst activity decay.

[0009] In theory, recoverable performance degradation or abnormal operating conditions can be mitigated or restored by adjusting operating conditions.

[0010] Currently, the common approach to handling single-low faults triggered during operation is to immediately perform an emergency shutdown and lock the system upon detection of a single low fault, regardless of the cause, prohibiting restart until offline maintenance. While this ensures safety, it may lead to many temporary faults that could be recovered online being misjudged as serious faults, resulting in unnecessary maintenance costs and vehicle downtime.

[0011] Therefore, it is necessary to develop a new method, system, and vehicle for online diagnosis and recovery of voltage faults in fuel cell cells. Summary of the Invention

[0012] In view of the shortcomings of the prior art, the purpose of this application is to provide a method, system and vehicle for online diagnosis and recovery of fuel cell cell voltage faults, which can quickly distinguish the root cause of single-cell voltage faults (including recoverable faults and unrecoverable faults) online, and apply a precise online recovery strategy for recoverable faults.

[0013] In a first aspect, embodiments of this application provide an online diagnosis and recovery method for voltage faults in a fuel cell cell, comprising the following steps:

[0014] Real-time acquisition of voltage signals from individual battery cells;

[0015] Based on the voltage signal, identify whether there is a fault of low cell voltage in a single battery cell;

[0016] When the low voltage fault of the single unit is detected, an emergency system shutdown is executed, cutting off the hydrogen, air, load and cooling cycles, and storing the fault information and the command to prevent power-on.

[0017] After detecting low-voltage power-on, the system status is checked, and when the preset detection conditions are met, air and hydrogen pressurization operations are performed.

[0018] The voltage signal of the battery cell determines the voltage building result. If the voltage building is successful, the fault is cleared and the power-on prohibition command is issued. If the voltage building fails, the failure result is stored and offline maintenance is prompted.

[0019] When the pressure build-up result is successful and a power-on request is received, the online activation program for the fuel cell stack is started to diagnose whether the fuel cell stack is in a dry membrane state or a flooded state, and the corresponding performance recovery strategy is executed according to the diagnosis result.

[0020] In the above technical solution, by real-time acquisition of cell voltage to identify faults, emergency shutdown to isolate risks, low-voltage power-on status detection, pressure build-up result judgment, and online diagnosis and targeted recovery of membrane dryness or water flooding, the online differentiation and intelligent handling of low cell voltage faults in fuel cells are realized. This effectively avoids the safety hazards caused by blind retrying and the availability loss caused by a one-size-fits-all shutdown, thereby improving the safety, reliability and operating efficiency of the system.

[0021] One possible implementation includes the following criteria for determining the low-voltage fault in a single cell:

[0022] The actual value of the lowest single-cell voltage is less than or equal to the fault threshold, and the duration of the fault is greater than or equal to the fault confirmation time.

[0023] In the above technical solution, by setting a minimum single-unit voltage threshold and a fault confirmation time as dual judgment conditions, the real fault can be accurately identified, avoiding false triggering caused by signal fluctuation interference, and improving the accuracy of fault detection and the stability of system operation.

[0024] One possible implementation includes the following preset detection conditions:

[0025] The system is in a low-voltage power-on state, has stored the aforementioned low-voltage fault, and the historical voltage build-up test results are not failures.

[0026] In the above technical solution, by limiting the detection conditions to low-voltage power-on, storage fault, and non-failure of historical voltage build-up results, the voltage build-up diagnostic process is ensured to be executed safely, avoiding repeated attempts on damaged fuel cells, achieving single-time intelligent diagnosis and effectively protecting the fuel cells.

[0027] One possible implementation includes the following specific steps for pressurizing by introducing air and hydrogen:

[0028] The system controls high pressure; the air system controls the air supply to the fuel cell stack at a first set pressure and a set flow rate, and opens the hydrogen supply system and nitrogen venting valve after the air flow rate reaches the target; the hydrogen system controls the hydrogen supply to the fuel cell stack at a second set pressure, and enters the cell voltage determination step after the hydrogen pressure reaches the target.

[0029] In the above technical solution, the step-by-step pressure building operation of first applying high pressure, then introducing air, and finally introducing hydrogen ensures that the pressure building process is standardized and controllable, providing a stable and reliable testing environment for fault diagnosis and improving the authenticity and credibility of the pressure building results.

[0030] One possible implementation method includes determining the pressure build-up result, specifically:

[0031] If the minimum unit voltage is greater than or equal to the voltage rating and remains within the specified time, the voltage build-up is considered successful, and the fault is cleared and the power-on prohibition command is issued; if the voltage does not meet the rating and the timeout occurs, the voltage build-up is considered a failure, the failure result is stored, and an offline maintenance prompt is sent.

[0032] In the above technical solution, the success or failure of voltage build-up is determined by the individual unit voltage and the time to reach the target, which can quickly distinguish between recoverable and unrecoverable faults, automatically perform fault clearing or offline maintenance prompts, simplify the fault handling logic and improve the accuracy of decision-making.

[0033] One possible implementation includes the online diagnosis of the fuel cell stack status and the corresponding recovery strategy, specifically comprising:

[0034] Real-time acquisition of fuel cell stack internal resistance, calculation of dry and wet state assessment values;

[0035] If the dry / wet state assessment value is greater than or equal to the preset value, it indicates that the fuel cell stack is in a dry state. In this case, the air flow rate is reduced and the coolant temperature is increased to optimize membrane hydration.

[0036] If the dry / wet condition assessment value is less than the preset value, it indicates that the fuel cell stack is in a flooded state. In this case, the air flow and pressure are increased, and a small power load is applied within the safety boundary to assist in drainage.

[0037] In the above technical solution, based on the calculation of dry and wet evaluation values ​​of the stack internal resistance, differentiated strategies are adopted, such as reducing air flow to increase water temperature, increasing air flow and pressure, and applying small power load, to achieve precise targeted recovery of membrane dryness and water flooding faults, improve recovery efficiency and not damage the stack.

[0038] One possible implementation method for calculating the wet / dry condition assessment value includes:

[0039] The dry and wet state assessment value is determined based on the stack internal resistance and the reference internal resistance value calibrated when the fuel cell stack is healthy and in good hydrothermal balance.

[0040] In the above technical solution, the dry and wet assessment values ​​are calculated using real-time internal resistance and healthy reference internal resistance, making the fuel cell stack condition diagnosis quantitative, objective, and unaffected by operating conditions. The algorithm is simple and easy to implement, making it suitable for real-time operation of on-board controllers.

[0041] One possible implementation involves calculating the in-reactor airflow, air pressure, coolant temperature, and power request value in real time during the online activation process, and using the average unit voltage as the control variable for closed-loop regulation.

[0042] In the above technical solution, the average unit voltage is used as the control variable for closed-loop regulation, and the air flow, pressure, coolant temperature and power request are optimized in real time to improve the adaptive capability and fault recovery effect of online activation.

[0043] Secondly, the online diagnosis and recovery system for fuel cell cell voltage faults according to the present invention includes:

[0044] The signal acquisition module is used to acquire the voltage signal of individual battery cells in real time.

[0045] The fault identification module is used to receive the voltage signal transmitted by the signal acquisition module, identify whether there is a low voltage fault in a single battery cell based on the voltage signal, and send a fault trigger signal when the low voltage fault is detected.

[0046] The emergency control module is used to receive the fault trigger signal, execute the emergency system shutdown, cut off the hydrogen, air, load and cooling cycles, and store fault information and power-on prohibition instructions.

[0047] The power-on detection and pressure building module is used to detect whether the system is in a low-voltage power-on state. After detecting that the system is in a low-voltage power-on state, it performs system status detection and sends a pressure building start signal when the preset detection conditions are met. Upon receiving the pressure building start signal, it performs air and hydrogen pressurization operations.

[0048] The voltage build-up determination module is used to receive the voltage signal collected by the signal acquisition module, determine the voltage build-up result based on the voltage signal, and clear the fault and prohibit the power-on command if the voltage build-up is successful; if the voltage build-up fails, the failure result is stored and offline maintenance is prompted.

[0049] The status diagnosis and performance recovery module is used to start the fuel cell stack online activation program when the pressure build-up determination module determines that the pressure build-up is successful and receives the power-on request, to diagnose whether the fuel cell stack is in a membrane dry state or a water flooded state online, and to execute the corresponding performance recovery strategy according to the diagnosis results.

[0050] In the above technical solution, the complete fault diagnosis and recovery functions are realized in hardware through the coordinated work of signal acquisition, fault identification, emergency control, power-on detection and voltage build-up, voltage build-up determination, status diagnosis and performance recovery modules. It is convenient to deploy, highly versatile, and easy to apply in engineering and system integration.

[0051] Thirdly, embodiments of this application provide a vehicle that employs the online diagnosis and recovery system for fuel cell cell voltage faults as described in this invention. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application will be described below.

[0053] Figure 1 This is a block diagram of a vehicle disclosed in an embodiment of this application;

[0054] Figure 2 This is one of the flowcharts for the online diagnosis and recovery method for fuel cell cell voltage faults disclosed in the embodiments of this application;

[0055] Figure 3 This is the second flowchart of the online diagnosis and recovery method for fuel cell cell voltage faults disclosed in the embodiments of this application;

[0056] Figure 4 This is a block diagram of the online diagnosis and recovery system for fuel cell cell voltage faults disclosed in an embodiment of this application;

[0057] Explanation of reference numerals in the attached figures:

[0058] 1. Vehicle; 2. Online diagnosis and recovery system for fuel cell cell voltage faults; 21. Signal acquisition module; 22. Battery pack; 22. Fault identification module; 23. Emergency control module; 24. Power-on detection and pressure build-up module; 25. Pressure build-up determination module; 26. Status diagnosis and performance recovery module. Detailed Implementation

[0059] The embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The detailed description of the following embodiments and the accompanying drawings are used to illustrate the principles of this application by way of example, but should not be used to limit the scope of this application, that is, this application is not limited to the described embodiments.

[0060] Please see Figure 1 , Figure 1 This is a schematic diagram of a vehicle disclosed in an embodiment of this application. Vehicle 1 is a fuel cell vehicle; as an example, it can be, but is not limited to, a hydrogen fuel cell vehicle. A vehicle includes a fuel cell stack and an online diagnosis and recovery system for fuel cell cell voltage faults as described in this application.

[0061] Please see Figure 2 , Figure 2 This is one of the flowcharts for an online diagnosis and recovery method for voltage faults in a fuel cell cell disclosed in this application. The online diagnosis and recovery method for voltage faults in a fuel cell cell includes the following steps:

[0062] Real-time acquisition of voltage signals from individual battery cells.

[0063] The system identifies whether a battery cell has a low voltage fault based on voltage signals.

[0064] When a low cell voltage fault is detected, an emergency system shutdown is executed, cutting off hydrogen, air, load, and cooling cycles, and storing fault information and a power-on prohibition command.

[0065] After detecting low-voltage power-on, the system status is checked, and when the preset detection conditions are met, air and hydrogen pressurization operations are performed.

[0066] The voltage signal of each battery cell determines the voltage build-up result. If the voltage build-up is successful, the fault is cleared and the power-on prohibition command is issued. If the voltage build-up fails, the failure result is stored and an offline maintenance prompt is sent.

[0067] When the pressure build-up result is successful and a power-on request is received, the online activation program for the fuel cell stack is started to diagnose whether the fuel cell stack is in a dry or flooded state, and the corresponding performance recovery strategy is executed according to the diagnosis result.

[0068] This method achieves online differentiation and intelligent handling of fuel cell cell voltage low-voltage faults by real-time acquisition of cell voltage for fault identification, emergency shutdown to isolate risks, low-voltage power-on status detection, pressure build-up result judgment, and online diagnosis and targeted recovery of membrane dryness or water flooding. It effectively avoids the safety hazards caused by blind retrying and the availability loss caused by indiscriminate shutdown, thereby improving system safety, reliability and operating efficiency.

[0069] The implementation of this method requires the following hardware and software conditions:

[0070] 1. Build a fuel cell engine with control functions for various subsystems, including a fuel cell stack (referred to as a stack), an air supply subsystem, a hydrogen supply subsystem, a water and heat management subsystem, a DC / DC converter, and a fuel cell engine system controller (Fuel Cell Control Unit, FCCU), etc.

[0071] 2. Configure the Controller Area Network (CAN) communication protocol between the DC / DC converter and the FCCU to ensure that the DC / DC converter can receive the DC-DC input current request signal from the FCCU, and that the FCCU can receive the DC-DC output current and voltage actual value signals from the DC / DC converter. The FCCU software calculates the product of the DC-DC output current and voltage to obtain the actual net power signal of the system.

[0072] 3. Configure the CAN communication protocol between the Cell Voltage Monitor (CVM) module and the FCCU so that the FCCU can receive the minimum and average cell voltage data sent by the CVM.

[0073] 4. Connect the hydrogen inlet solenoid valve inlet pipe of the fuel cell engine to the hydrogen supply pipe of the test system, and connect the main water circuit and auxiliary water circuit to the cooling water circuit of the test system to ensure a continuous supply of hydrogen and cooling water circulation during the operation of the fuel cell engine.

[0074] 5. As described in this application, complete the application layer software model of the corresponding algorithm in Matlab / Simulink and integrate it into the FCCU software architecture.

[0075] 6. Compile the FCCU software model that integrates this method into C code, and use a USB CAN analyzer to flash the generated S19 file (i.e., the program burning file that is common to microcontrollers or controllers) into the actual physical controller.

[0076] Once the aforementioned hardware and software conditions are ready, high and low voltage power is supplied to the fuel cell engine through the test bench, and the hydrogen supply circuit and cooling water supply circuit of the test bench are turned on. The fuel cell engine is then requested to start up (i.e., the vehicle requests a command) through the Integration Calibration and Acquisition (INCA) software.

[0077] Please see Figure 3 , Figure 3This is a second flowchart illustrating an online diagnosis and recovery method for voltage faults in a fuel cell unit, as disclosed in this application. After the fuel cell engine is started, the online diagnosis and recovery method for voltage faults in a fuel cell unit can be implemented according to the following steps:

[0078] S1. During the operation of the fuel cell system, the voltage signal of each battery cell is collected in real time by the CVM to identify low voltage faults. The criteria for determining a low voltage fault are as follows:

[0079] If the actual value of the lowest single-cell voltage is less than or equal to the fault threshold (initial value of the fault threshold: 0.4V, calibrable), and the duration is greater than or equal to the fault confirmation time (initial value of the fault confirmation time: 0.01s, calibrable), then proceed to S2; otherwise, remain in S1.

[0080] S2. Execute emergency system shutdown, automatically cutting off hydrogen supply, air supply, load circuit, and cooling circulation (lowering high pressure) to ensure immediate isolation of the fault condition and prevent further damage. It also stores the individual cell low voltage fault and system startup prohibition command, then proceeds to S3. The storage module is a functional module in the FCCU's underlying software; the FCCU application layer software can retrieve the stored information in real time during operation.

[0081] S3. After the system is powered on at low voltage, it automatically performs system status detection and, when preset detection conditions are met, performs air and hydrogen pressurization operations. The preset detection conditions include:

[0082] 1) The system is currently in a low-voltage power-on state;

[0083] 2) A single-unit low voltage fault is stored;

[0084] 3) Historical pressure testing results were not considered failures;

[0085] If all the above conditions are met, proceed to S4; otherwise, remain in S3. To avoid scenarios where repeated attempts to establish open-circuit voltage may worsen the fault due to irreversible internal damage to the fuel cell stack, historical voltage build-up detection results are incorporated. This ensures that only one intelligent diagnostic is performed, avoiding repeated attempts to maximize stack protection.

[0086] S4, pressurization operation by introducing air and hydrogen, specifically includes:

[0087] The system controls high pressure; the air system controls the air supply to the fuel cell stack at a set pressure and flow rate, and opens the hydrogen supply system and nitrogen venting valve after the air flow rate reaches the target; the hydrogen system controls the hydrogen supply to the fuel cell stack at a set pressure, and enters the cell voltage determination step after the hydrogen pressure reaches the target.

[0088] For example, S4 is as follows:

[0089] S41. Control the DC / DC converter to enter standby mode, and control the DC / DC converter output relay to close (to bring high voltage to the system). Control the air system to operate, control the air pressure to the first set pressure (the calibration value, e.g., 0.06 bar), and control the air flow to the set flow rate (the calibration value, e.g., 25 g / s) (to supply air to the fuel cell stack). After completion, proceed to S42.

[0090] S42. Based on the actual airflow value collected by the airflow sensor, determine whether the actual airflow value is greater than or equal to the calibrated airflow value (initial value: 24g / s). If the actual airflow value is greater than or equal to the calibrated airflow value, proceed to S43; otherwise, determine whether the unmet maintenance time is less than or equal to the first time calibration value (initial value: 5s). If so, maintain S42; otherwise, return to S3.

[0091] S43. Control the operation of the hydrogen storage system (open the hydrogen cylinder, etc.), control the operation of the hydrogen supply system, control the hydrogen pressure request to the second set pressure (e.g., 0.2 bar) (to supply hydrogen to the fuel cell stack), control the opening of the nitrogen venting valve, and proceed to S44 after completion.

[0092] S44. Using the actual hydrogen pressure value collected by the hydrogen pressure sensor, determine whether the actual hydrogen pressure value is greater than or equal to the second pressure calibration value (initial value: 0.19 bar). If the actual hydrogen pressure value is greater than or equal to the hydrogen pressure calibration value, proceed to S5; otherwise, determine whether the unmet maintenance time is less than or equal to the second time calibration value (initial value: 5 s). If yes, maintain S44; otherwise, return to S3.

[0093] S5. Determine the voltage build-up result based on the voltage signal of each battery cell. If the voltage build-up is successful, clear the fault and issue a power-off prohibition command. If the voltage build-up fails, store the failure result and prompt for offline maintenance. Specifically, determining the voltage build-up result includes:

[0094] If the minimum unit voltage is greater than or equal to the voltage rating and remains within the specified time, the voltage build-up is considered successful, and the fault is cleared and the power-on prohibition command is issued; if the voltage does not meet the rating and the timeout occurs, the voltage build-up is considered a failure, the failure result is stored, and an offline maintenance prompt is sent.

[0095] For example, S5 is as follows:

[0096] S51. Collect the actual value of the individual cell voltage through CVM, and determine whether the lowest actual value of the individual cell voltage is greater than or equal to the third voltage calibration value (initial value: 0.92V), and whether the maintenance time is greater than or equal to the third time calibration value (initial value: 3s). If so, the arbitration voltage build-up detection result is that the voltage build-up is successful and outputs the result of successful voltage build-up to the storage module. The storage module will then arbitrate the historical voltage build-up detection result as successful, clear the individual cell voltage too low fault and the system power-on prohibition command, and enter S52. Otherwise, determine whether the maintenance time that is not satisfied is less than or equal to the third time calibration value (initial value: 5s). If so, maintain S51; otherwise, the arbitration voltage build-up detection result is that the voltage build-up fails and outputs the result of voltage build-up failure to the storage module. The storage module will then arbitrate the historical voltage build-up detection result as failed and send it to the vehicle controller. The vehicle controller will then send a display request to the vehicle display screen: "Cell stack fault, please perform offline repair", and enter S52.

[0097] S52. Stop the operation of each subsystem, for example: control the hydrogen supply system to stop working and request the hydrogen pressure to be 0 bar; control the air system to stop working, request the air flow rate to be 0 g / s and the air pressure to be 0 bar; control the nitrogen venting valve to close; control the DCF output relay to disconnect (i.e., reduce the high pressure) and jump to S53.

[0098] S53. The fuel cell system is in a low-voltage power-on state. If a vehicle start-up request is received and no "disable system start-up" command is stored, proceed to S6; otherwise, maintain S53. The vehicle start-up request is sent by the vehicle controller (Electronic Control Unit, ECU), while in the bench test system, the operator requests the command through the INCA software.

[0099] S6. The fuel cell system is powered on and reads the historical pressure build-up test results from the storage module. If the pressure build-up is successful, the online activation program of the fuel cell stack is started to restore the online performance of the fuel cell stack. After completion, the historical pressure build-up test results are output to the storage module as "invalid" and the process proceeds to S1. If the pressure build-up fails, the process proceeds to S1.

[0100] In one possible embodiment, online diagnosis of the fuel cell stack status and corresponding recovery strategy includes: real-time acquisition of the stack internal resistance and calculation of a dry / wet state assessment value. If the dry / wet state assessment value is greater than or equal to a preset value, indicating that the fuel cell stack is in a dry state, the airflow is reduced and the coolant temperature is increased to optimize membrane hydration. If the dry / wet state assessment value is less than the preset value, indicating that the fuel cell stack is in a flooded state, the airflow and pressure are increased, and a small power load within the safety boundary is applied to assist in drainage.

[0101] In one possible embodiment, the method for calculating the dry / wet condition assessment value includes:

[0102] The dry and wet state assessment value is determined based on the stack internal resistance and the reference internal resistance value calibrated when the fuel cell stack is healthy and in good hydrothermal balance.

[0103] In one possible embodiment, during the online activation process, the infeed air flow rate, air pressure, coolant temperature, and power request value are calculated in real time, and closed-loop regulation is performed using the average unit voltage as the control variable.

[0104] For example, during the online performance recovery process of the fuel cell stack, the FCCU will calculate the stack feed airflow, pressure, coolant temperature (representing the stack temperature), and power request value according to the following model, and send them to the subsystem for execution. The principle is an active closed-loop regulation system that uses the stack average as the process control variable and the key controllable parameters of the system as setpoints. The goal is to diagnose and restore performance by adjusting the system input and observing the voltage response.

[0105] Introduce dry and wet condition assessment values: ;

[0106] in, These are the dry and wet condition assessment values ​​for the fuel cell stack. ≥0 indicates that the resistance is higher than the reference value, and the condition is too dry; A value less than 0 indicates that the resistance is below the reference value and the condition is too wet. This is the reference internal resistance value calibrated when the fuel cell stack is in a healthy state and in good hydrothermal balance, in Ω; This is the internal resistance of the fuel cell stack, obtained through impedance hardware on the system (such as the impedance function module built into the DC-DC converter injecting a small alternating excitation current into the fuel cell stack), and the unit is Ω.

[0107] Example airflow request value:

[0108] in, The airflow rate is the real-time airflow value under normal operating conditions, in g / s. This is the requested real-time airflow rate, in g / s. This is the flow rate adjustment coefficient, with a recommended range of 0.1 to 0.5.

[0109] Calibration method: In known "dry" and "wet" fault bench tests, adjust the flow rate changes to produce a clear voltage recovery trend. This represents the standard average cell voltage value corresponding to the real-time current under normal operating conditions. This represents the current actual average single-cell voltage value.

[0110] Air pressure request value:

[0111] in, This represents the air pressure value under normal operating conditions in real time, in bar. This is a real-time air pressure request value, in bar. This is the pressure regulation coefficient, and its value is determined in the same way as the calibration method. .

[0112] Coolant temperature request:

[0113] in, This represents the coolant inlet water temperature under normal operating conditions in real time, in °C. γ is the real-time air pressure request value, in °C; γ is the temperature adjustment coefficient, recommended range: 2.0~5.0, in °C / V. Calibration method: In a slightly dry state, adjust γ so that the increase in water temperature can effectively promote voltage recovery (e.g., a recovery of >5mV per minute).

[0114] Power request value:

[0115] in, This is the real-time power request value, in watts (W). This is the total voltage of the fuel cell stack, expressed in volts (V). It is more suitable for calculating power than using the voltage of individual fuel cells. The instantaneous safe current limit, measured in amperes (A), is used to ensure that all operations do not damage the fuel cell stack. It is calculated in real-time based on the current highest stack temperature, coolant temperature, and gas pressure (formula provided by the fuel cell stack manufacturer). This is the power activation factor. Recommended range: 0.02~0.10, used to ensure that the calculated power request is well below the determined absolute maximum safe power. The voltage deviation influence constant is recommended to be in the range of 0.03 to 0.08V. This determines the rate at which the voltage deviation affects the power request intensity. The activation function determines the load loading status based on the dry and wet state of the fuel cell stack.

[0116] The principles for establishing the power load model are: 1. No additional load should be applied when the membrane is dry to prevent accelerated dehydration. 2. Only when flooding is confirmed is a small positive load allowed for auxiliary heat generation and drainage. 3. Power requests must be strictly limited within real-time safety boundaries. This achieves differentiated safety control.

[0117] The core principle of the above online activation method is based on real-time monitoring of the stack's internal resistance, online diagnosis of whether the stack is in a "membrane dry" or "flooded" state, and automatic switching to different recovery strategies accordingly. When "membrane dry" is determined, the system will actively reduce airflow and increase coolant temperature to optimize membrane hydration operating conditions; when "flooded" is determined, it will synergistically increase airflow and pressure, and apply a small power load limited by strict safety boundaries to drain water through a combination of physical purging and electrochemical heat generation.

[0118] This method realizes a paradigm shift from fault detection-emergency shutdown to intelligent diagnosis-targeted recovery. By distinguishing between unrecoverable and recoverable faults online, it completely avoids the security risks of "blind retries" and the availability losses of "one-size-fits-all" shutdowns in traditional solutions.

[0119] Please see Figure 4 , Figure 4 This is a block diagram of an online diagnosis and recovery system for fuel cell cell voltage faults disclosed in an embodiment of this application. The system includes a signal acquisition module 21, a fault identification module 22, an emergency control module 23, a power-on detection and pressure build-up module 24, a pressure build-up determination module 25, and a status diagnosis and performance recovery module 26. The signal acquisition module 21 is used to acquire the voltage signal of the fuel cell cell in real time. The fault identification module 22 is used to receive the voltage signal transmitted by the signal acquisition module 21, identify whether the fuel cell cell has a low voltage fault based on the voltage signal, and send a fault trigger signal when the low voltage fault is detected. The emergency control module 23 is used to receive the fault trigger signal, execute an emergency system shutdown, cut off hydrogen, air, load, and cooling circulation, and store fault information and a power-on prohibition command. The power-on detection and pressure build-up module 24 is used to detect whether the system is in a low-voltage power-on state, perform system status detection after detecting low-voltage power-on, send a pressure build-up start signal when preset detection conditions are met, and execute air and hydrogen pressurization operations upon receiving the pressure build-up start signal. The voltage build-up determination module 25 receives the voltage signal acquired by the signal acquisition module 21, determines the voltage build-up result based on the voltage signal, and clears the fault and prohibits power-on if the voltage build-up is successful; if the voltage build-up fails, it stores the failure result and prompts for offline maintenance. The status diagnosis and performance recovery module 26, when the voltage build-up determination module 25 determines that the voltage build-up is successful and receives a power-on request, initiates the fuel cell stack online activation program, diagnoses whether the fuel cell stack is in a membrane dry state or a water-flooded state, and executes the corresponding performance recovery strategy based on the diagnosis result.

[0120] This system achieves complete fault diagnosis and recovery functions through the coordinated operation of 26 modules, including signal acquisition, fault identification, emergency control, power-on detection and pressure build-up, pressure build-up determination, status diagnosis and performance recovery. It is easy to deploy, highly versatile, and convenient for engineering applications and system integration.

[0121] This application significantly improves the operational safety and reliability of fuel cell systems. Existing systems, when experiencing low cell voltage, either blindly restart, potentially exacerbating irreversible damage such as membrane perforation, or directly shut down the system, failing to differentiate the fault type. This application, through online open-circuit voltage diagnostics, accurately identifies the root cause of the fault. For irreparable hardware faults such as membrane perforation, it directly locks the system and prompts for repair, avoiding the safety risks associated with operating with faults or repeated attempts. Compared to traditional experience-based approaches, this application implements control based on diagnostic results, resulting in a more rigorous logic and significantly reducing the possibility of secondary stack damage and safety accidents due to improper handling.

[0122] This application significantly improves system availability and reduces usage and maintenance costs. For recoverable faults such as membrane dryness and flooding, which are common in actual operation, this application avoids a blanket shutdown approach, preventing unnecessary vehicle downtime. The system can automatically execute targeted activation procedures without disassembling the unit or performing maintenance, quickly restoring fuel cell stack performance and ensuring normal vehicle operation. This significantly reduces towing, on-site inspections, and unplanned maintenance due to temporary faults, improving vehicle uptime efficiency and user experience, while reducing maintenance manpower and time costs.

[0123] This application achieves more intelligent and precise fault management capabilities. Compared to traditional systems with only simple "detection-protection" logic, this application upgrades fault handling to a complete intelligent closed loop of "diagnosis-classification-disposal." Through multi-level control logic and control algorithms tailored to the characteristics of the fuel cell stack, the system can not only identify the occurrence of faults but also determine the fault type and autonomously execute corresponding handling strategies. This improvement represents a significant breakthrough in the intelligentization of fuel cell systems, providing crucial support for subsequent health status management, remote diagnostics, and lifespan optimization, and contributing to enhanced product technology and market competitiveness.

[0124] This application also provides a feasible path for the early identification and intervention of fuel cell stack performance degradation. During the diagnosis and repair of recoverable faults, the system essentially performs a refined online condition monitoring process. The voltage build-up process, activation parameter adjustments, and voltage response changes all accurately reflect the internal health status of the fuel cell stack. Based on continuous monitoring and analysis of this data, the system can not only cope with sudden low-voltage faults but also identify the trend of slow performance decline in advance, enabling earlier and more gentle preventative control.

[0125] The examples are not limited to those described above. Those skilled in the art can make modifications or alterations based on the above description, and all such modifications and alterations should fall within the scope of protection of the appended claims. Those skilled in the art can understand that implementing all or part of the processes of the above embodiments and making equivalent changes according to the claims of this application still fall within the scope of this application.

Claims

1. A method for online diagnosis and recovery of voltage faults in a fuel cell cell, characterized in that, Includes the following steps: Real-time acquisition of voltage signals from individual battery cells; Based on the voltage signal, identify whether there is a fault of low cell voltage in a single battery cell; When the low voltage fault of the single unit is detected, an emergency system shutdown is executed, cutting off the hydrogen, air, load and cooling cycles, and storing the fault information and the command to prevent power-on. After detecting low-voltage power-on, the system status is checked, and when the preset detection conditions are met, air and hydrogen pressurization operations are performed. The voltage signal of the battery cell determines the voltage building result. If the voltage building is successful, the fault and the power-on prohibition command are cleared. If the voltage building fails, the failure result is stored and offline maintenance is prompted. When the pressure build-up result is successful and a power-on request is received, the online activation program for the fuel cell stack is started to diagnose whether the fuel cell stack is in a dry membrane state or a flooded state, and the corresponding performance recovery strategy is executed according to the diagnosis result.

2. The method for online diagnosis and recovery of fuel cell cell voltage faults according to claim 1, characterized in that, The criteria for determining the low voltage fault of a single cell include: The actual value of the lowest single-cell voltage is less than or equal to the fault threshold, and the duration of the fault is greater than or equal to the fault confirmation time.

3. The method for online diagnosis and recovery of fuel cell cell voltage faults according to claim 1, characterized in that, The preset detection conditions include: The system is in a low-voltage power-on state, has stored the aforementioned low-voltage fault, and the historical voltage build-up test results are not failures.

4. The method for online diagnosis and recovery of fuel cell cell voltage faults according to claim 1, characterized in that, The pressurization operations involving air and hydrogen supply specifically include: High voltage on the control system; The air control system supplies air to the fuel cell stack at a first set pressure and a set flow rate. Once the air flow rate reaches the target, the hydrogen supply system and the nitrogen venting valve are activated. The hydrogen system is controlled to supply hydrogen to the fuel cell stack at a second set pressure. Once the hydrogen pressure reaches the target, the cell voltage determination step begins.

5. The method for online diagnosis and recovery of fuel cell cell voltage faults according to claim 1, characterized in that, The determination of the pressure build-up result specifically includes: If the minimum unit voltage is greater than or equal to the voltage rating and remains within the specified time, the voltage build-up is considered successful, and the fault is cleared and the power-on prohibition command is issued; if the voltage does not meet the rating and the timeout occurs, the voltage build-up is considered a failure, the failure result is stored, and an offline maintenance prompt is sent.

6. The method for online diagnosis and recovery of fuel cell cell voltage faults according to claim 1, characterized in that, The online diagnostics system determines whether the fuel cell stack is in a dry or flooded state and executes corresponding performance recovery strategies based on the diagnostic results, specifically including: Real-time acquisition of fuel cell stack internal resistance, calculation of dry and wet state assessment values; If the dry / wet state assessment value is greater than or equal to the preset value, it indicates that the fuel cell stack is in a dry state. In this case, the air flow rate is reduced and the coolant temperature is increased to optimize membrane hydration. If the dry / wet condition assessment value is less than the preset value, it indicates that the fuel cell stack is in a flooded state. In this case, the air flow and pressure are increased, and a small power load is applied within the safety boundary to assist in drainage.

7. The method for online diagnosis and recovery of fuel cell cell voltage faults according to claim 6, characterized in that, The method for calculating the dry / wet condition assessment value includes: The dry and wet state assessment value is determined based on the stack internal resistance and the reference internal resistance value calibrated when the fuel cell stack is healthy and in good hydrothermal balance.

8. The method for online diagnosis and recovery of fuel cell cell voltage faults according to claim 1, characterized in that, During the online activation process, the infeed air flow rate, air pressure, coolant temperature and power request value are calculated in real time, and closed-loop regulation is performed using the average unit voltage as the control variable.

9. A system for online diagnosis and recovery of voltage faults in a fuel cell cell, characterized in that, include: The signal acquisition module (21) is used to acquire the voltage signal of the battery cell in real time; The fault identification module (22) is used to receive the voltage signal transmitted by the signal acquisition module (21), identify whether the battery cell has a low cell voltage fault based on the voltage signal, and send a fault trigger signal when the low cell voltage fault is identified. The emergency control module (23) is used to receive the fault trigger signal, execute the emergency system shutdown, cut off the hydrogen, air, load and cooling cycle, and store fault information and power-on prohibition instructions; The power-on detection and pressure building module (24) is used to detect whether the system is in a low-voltage power-on state. After detecting that the system is in a low-voltage power-on state, it performs system status detection and sends a pressure building start signal when the preset detection conditions are met. Upon receiving the pressure building start signal, it performs air and hydrogen pressurization operations. The voltage build-up determination module (25) is used to receive the voltage signal collected by the signal acquisition module (21), determine the voltage build-up result based on the voltage signal, clear the fault and prohibit the power-on command if the voltage build-up is successful, and store the failure result and prompt offline maintenance if the voltage build-up fails. The status diagnosis and performance recovery module (26) is used to start the online activation program of the fuel cell stack when the pressure build-up determination module (25) determines that the pressure build-up is successful and receives the power-on request, to diagnose online whether the fuel cell stack is in a membrane dry state or a water flooded state, and to execute the corresponding performance recovery strategy according to the diagnosis results.

10. A vehicle, characterized in that, The online diagnosis and recovery system for fuel cell cell voltage faults as described in claim 9 is adopted.

Images