Method for establishing and diagnosing full-life-cycle fault library of fuel cell stack

Abstract

The invention discloses a method for establishing and diagnosing a full-life-cycle fault library of a fuel cell stack, and belongs to the technical field of hydrogen fuel cells. The method comprises the following steps: collecting stack data in an initial state of the stack to define an initial fault-free state of the stack; carrying out a pile fault sensitivity test, and collecting pile data in the test process to construct a pile initial state comprehensive fault library; carrying out an endurance test on the electric pile, and collecting electric pile data at fixed time intervals in the process to construct an electric pile process state comprehensive fault library; and when the pile endurance test is performed until the pile life is ended, collecting pile data at the moment so as to construct a pile life end state comprehensive fault library. According to the invention, the full life cycle state of the electric pile is collected, the corresponding fault database is established, convenience is provided for work such as electric pile working condition adjustment, recovery and repair according to the diagnosis result in the later period, and the service life of the electric pile is further prolonged.

Description

Technical Field

[0001] The present invention belongs to the technical field of hydrogen fuel cells. Specifically, the present invention relates to a method for establishing and diagnosing a full life cycle fault library of a fuel cell stack. Background Art

[0002] Proton exchange membrane fuel cells (PEMFCs), one of the most promising and popular fuel cell technologies, offer numerous advantages, including zero pollution, high energy conversion efficiency, short charging time, low operating temperature, and low noise. They are widely used in transportation vehicles, cogeneration, fixed base stations, mobile portable devices, drones, military equipment, and many other fields. Currently, the actual use of fuel cell products mainly relies on monitoring the voltage of individual cells to determine the health status of the fuel cell stack. While effective, this method can only determine the health status under macroscopic conditions and cannot be used for timely detection and corresponding adjustments. Furthermore, the factors that lead to voltage loss are numerous and coupled, such as membrane dryness, membrane wetness, water blockage, air shortage, and excessive temperature. Voltage acquisition alone cannot distinguish these related states. Therefore, alternative characterization methods are needed to determine fuel cell stack faults and optimize relevant parameters based on the fault state determination information to extend the fuel cell stack life.

[0003] Fuel cell stack fault diagnosis is a critical step in ensuring the efficient and safe operation of fuel cell systems. As a device that directly converts chemical energy into electrical energy, the performance and stability of fuel cells are crucial for practical applications. Therefore, regular fault diagnosis and maintenance of fuel cell stacks are essential. Diagnosis involves monitoring and analyzing various parameters of the fuel cell stack during use to determine the status and performance of the fuel cell. For example, parameters such as temperature, density, pressure, and current can provide information on the operating status of the fuel cell. The purpose of diagnosis is to detect faults in a timely manner and take appropriate measures to repair them, thereby ensuring the long-term stable operation of the fuel cell. Existing diagnostic methods are mostly single-minded, using only a few methods for fault analysis and diagnosis, which is not comprehensive, and fault diagnosis does not cover the entire life cycle.

[0004] To this end, the present invention proposes a method for establishing and diagnosing a fuel cell stack full life cycle fault library. Summary of the Invention

[0005] The present invention aims to overcome the shortcomings of the existing technology and proposes a method for establishing and diagnosing a full life cycle fault library of a fuel cell stack to achieve the following purposes: to collect the full life cycle status of the stack and establish a corresponding fault library, so as to facilitate the subsequent adjustment, recovery and repair of the stack working condition according to the diagnosis results, thereby improving the service life of the stack.

[0006] In order to achieve the above-mentioned object, the technical solution adopted by the present invention is: a method for establishing a fuel cell stack full life cycle fault library, the method comprising:

[0007] Step S1: collecting stack data in the initial state of the stack to define the initial fault-free state of the stack; performing a stack fault sensitivity test, and collecting stack data during the test to construct a comprehensive fault library for the initial state of the stack;

[0008] Step S2: Performing a durability test on the stack, collecting stack data at fixed intervals during the test, obtaining stack data that changes with time, and constructing a comprehensive fault library for the stack process status;

[0009] Step S3: When the stack durability test reaches the end of the stack life, the stack data at this time is collected to construct a comprehensive fault library of the stack end-of-life status.

[0010] Preferably, in step S1, collecting stack data in the initial state of the stack to define an initial fault-free state of the stack includes:

[0011] Collect data on the entire stack in its initial state, including the stack voltage and total EIS. This data is defined as the stack being in a fault-free state.

[0012] In the initial state of the stack, single-cell level data is collected, including the average voltage of the single cell, the EIS data of each single cell, the gas distribution from the entire stack to the single cell, and the ohmic impedance, hydrogen permeation current, electrochemical surface area and other parameters of each single cell collected by the microcurrent excitation method. This data is defined as the single cell fault-free state;

[0013] In the initial state of the fuel cell stack, data at the single-chip partition level is collected, including the EIS data of the single-chip partition, the multi-point inspection voltage of the single-chip partition, and the distribution status of the internal medium of the fuel cell stack collected through in-situ testing. This data is defined as the fault-free state of the single-chip partition.

[0014] Preferably, the stack fault sensitivity test includes simulating faults on the stack, and the faults include gas deficiency, high temperature, high and low humidity.

[0015] Preferably, a stack fault sensitivity test is performed, and stack data during the test is collected to construct a comprehensive fault library of the stack initial state, including:

[0016] When conducting the stack fault sensitivity test, multi-point inspection voltage, whole stack voltage, single-chip average voltage, stack total EIS, single-chip EIS data, and single-chip partition EIS data are collected. After comparing and analyzing with the corresponding data under the fault-free state, relevant fault data is obtained and organized into the comprehensive fault library of the stack initial state.

[0017] Preferably, a stack fault sensitivity test is performed, and stack data during the test is collected to construct a comprehensive fault library of the stack initial state, including:

[0018] When conducting a stack fault sensitivity test, gas distribution detection equipment is used to collect gas distribution information from the entire stack to a single chip. At the same time, the EIS data of each single chip and the voltage at the end of the stack are collected. Based on the gas distribution information, single-chip EIS data, and stack end voltage, gas deficiency faults are graded and organized into a comprehensive fault library for the stack initial state.

[0019] Preferably, a stack fault sensitivity test is performed, and stack data during the test is collected to construct a comprehensive fault library of the stack initial state, including:

[0020] When conducting a stack fault sensitivity test, the fault-prone areas on a single chip partition are determined based on the internal medium distribution status of the stack collected through in-situ testing and combined with partition current detection;

[0021] Under different fault conditions, multi-point inspection is used to collect the multi-point inspection voltage of the fault-prone area as fault data and organize it into the comprehensive fault library of the initial state of the fuel cell stack;

[0022] Under different fault conditions, the EIS data of the single-chip partition of the fault-prone area is collected as fault data and organized into a comprehensive fault library of the initial state of the fuel cell stack.

[0023] Preferably, the step S2 includes:

[0024] During the endurance test, if the endurance loss is less than the preset loss threshold: when the endurance time is less than the preset time threshold, the stack data is collected at a first fixed time interval; when the endurance time is greater than or equal to the preset time threshold, the stack data is collected at a second fixed time interval, wherein the first fixed time interval is greater than the second fixed time interval;

[0025] If the durability loss is greater than or equal to the preset loss threshold: stack data is collected every third fixed time, and the third fixed time is less than the second fixed time.

[0026] Preferably, during the endurance test process of step S2, the status of the battery stack components is also detected, and the battery stack data corresponding to the status of the battery stack components when the battery stack cannot work is imported into the battery stack process status comprehensive fault library as fault data.

[0027] Preferably, the states of the components of the stack when the stack fails to work include cracks in the membrane electrode, pinholes in the membrane electrode, and shedding of the bipolar plate coating.

[0028] Preferably, the durability test road spectrum in step S2 includes an accelerated durability road spectrum and a real vehicle road spectrum.

[0029] The present application also proposes a fuel cell stack full life cycle fault diagnosis method, using a full life cycle fault library constructed according to the above-mentioned fuel cell stack full life cycle fault library establishment method, the method comprising:

[0030] When a fault occurs in the stack, first determine whether the current stack operating condition belongs to the preset normal operating condition. If not, refer to the full life cycle fault library to correct the current operating condition; if it does, proceed to the next step;

[0031] Determine whether the EIS data of the current fuel cell stack is abnormal. If so, refer to the full life cycle fault database to query the corresponding fault problem and correct it. If not, proceed to the next step.

[0032] Determine whether the current stack multi-point inspection voltage is abnormal. If so, refer to the full life cycle fault database to query the corresponding fault problem and correct it. If not, proceed to the next step.

[0033] The microcurrent excitation method is used to collect parameters such as the ohmic impedance, hydrogen permeation current, and electrochemical surface area of ​​each cell in the current stack. It is determined whether the parameters are abnormal. If so, the corresponding fault problem is searched and corrected by referring to the full life cycle fault database. If not, proceed to the next step.

[0034] The fault problem was further identified by disassembling the battery stack for characterization analysis.

[0035] The technical effect of the present invention is as follows: the present invention constructs a full life cycle fault library through multi-dimensional stack data collection, divided into initial, process, and end-of-life states. Among them, the introduction of fault data such as gas shortage fault classification, positioning of fault-prone areas, and component failure state detection ensures that the fault library is comprehensive and accurate. In addition, based on the constructed high-reliability full life cycle fault library, the diagnostic method of the present application can realize fault diagnosis through basic operating conditions, impedance, multi-point inspection voltage, and micro-current excitation values, achieving efficient and accurate diagnosis, providing strong support for fuel cell stack full life cycle fault management, and improving the reliability and service life of the stack. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flow chart of a method for establishing a full life cycle fault library for a fuel cell stack provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The following is a further detailed description of the specific implementation of the present invention through the description of the embodiments with reference to the accompanying drawings, with the aim of helping those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention and to facilitate its implementation. It should be noted that the terms "first" and "second" described in this application are only used to facilitate the description of the technical solution to distinguish components. The corresponding component configurations may be the same or different, and are not intended to limit this application. In order to make the technical solution of the present invention clearer, the present invention is explained through the following embodiments.

[0038] This embodiment provides a method for establishing a fuel cell stack full life cycle fault library. Figure 1 Steps S1 to S3 of this method are described in detail.

[0039] Step S1: collecting stack data in the initial state of the stack to define the initial fault-free state of the stack; performing a stack fault sensitivity test, and collecting stack data during the test to construct a comprehensive fault library for the initial state of the stack.

[0040] The initial state of the stack is the initial stage of the stack's entire life cycle. During this stage, the stack is stable, so the stack data in the initial state is collected to define the stack's initial fault-free state, which serves as a reference benchmark for subsequent stack fault analysis.

[0041] The data collection method used in this embodiment includes:

[0042] (1) EIS (Electrochemical Impedance Spectroscopy): Analyze the membrane state based on the distribution relaxation time method (DRT) and collect the corresponding EIS data (impedance data).

[0043] (2) Multi-point inspection: By collecting voltage data at different locations in the fuel cell stack, we can understand whether the voltage distribution inside the fuel cell stack is uniform. The voltage changes differently under different operating conditions, which helps to discover possible local performance differences or potential faults inside the fuel cell stack.

[0044] (3) Gas distribution status: used to detect the distribution of reaction gases (such as hydrogen, oxygen, etc.) in the fuel cell stack.

[0045] (4) Microcurrent excitation: used to collect parameters such as the stack ohmic impedance, hydrogen permeation current, and electrochemical surface area.

[0046] (5) In-situ detection: Collect the distribution status of the internal medium of the battery stack, and combine the partition current detection to collect the status of the single chip.

[0047] (6) Voltage detection: detect the voltage of the entire stack and the voltage of a single chip.

[0048] At the same time, this embodiment collects data at three levels, including whole stack level data (i.e., complete stack product), single chip level data (i.e., a power generation unit, not necessarily a single cell structure in the physical sense), and single chip partition level data (i.e., cathode or anode surface partition of the power generation unit). Specifically, the stack data is collected in the initial state of the stack to define the initial fault-free state of the stack, including:

[0049] Collect data on the entire stack in its initial state, including the stack voltage and total EIS. This data is defined as the stack being in a fault-free state.

[0050] In the initial state of the stack, single-cell level data is collected, including the average voltage of the single cell, the EIS data of each single cell, the gas distribution from the entire stack to the single cell, and the ohmic impedance, hydrogen permeation current, electrochemical surface area and other parameters of each single cell collected by the microcurrent excitation method. This data is defined as the single cell fault-free state;

[0051] In the initial state of the fuel cell stack, data at the single-chip partition level is collected, including the EIS data of the single-chip partition, the multi-point inspection voltage of the single-chip partition, and the distribution status of the internal medium of the fuel cell stack collected through in-situ testing. This data is defined as the fault-free state of the single-chip partition.

[0052] To establish a comprehensive fault library for the initial state of the stack, this embodiment conducts a stack fault sensitivity test to simulate various faults, including gas deficiency, high temperature, and high and low humidity. Simultaneously, stack data under corresponding fault conditions is collected and analyzed to obtain fault data for use in constructing the comprehensive fault library for the initial state of the stack.

[0053] During the stack fault sensitivity test, multi-point inspection voltage, overall stack voltage, single-chip average voltage, stack total EIS, single-chip EIS data, and single-chip partition EIS data are collected and compared with the corresponding data in the fault-free state to obtain relevant fault data and organize them into a comprehensive fault library for the stack initial state. Under various fault simulations in the stack fault sensitivity test, various stack data will vary relative to the fault-free state. This embodiment records these data changes under various fault simulations and stores them in the comprehensive fault library for the stack initial state to facilitate subsequent fault diagnosis.

[0054] Furthermore, in order to improve the reference value of the comprehensive fault library of the initial state of the fuel cell stack, the present embodiment also deeply analyzes the relationship between the data and sets up a gas shortage fault classification based on multiple data. Specifically, when conducting a fuel cell stack fault sensitivity test, the present embodiment uses gas distribution detection equipment to collect gas distribution conditions from the entire stack to a single chip, and at the same time collects EIS data of each single chip and the voltage at the end of the fuel cell stack. According to the gas distribution conditions, single chip EIS data, and fuel cell stack end voltage, the gas shortage fault classification is performed and organized into the comprehensive fault library of the initial state of the fuel cell stack. Among them, the present embodiment divides the gas shortage fault into three levels: mild gas shortage, moderate gas shortage and severe gas shortage:

[0055] Mild air deficiency: first flow threshold > gas flow ≥ second flow threshold, first increment threshold < EIS impedance increment ≤ second increment threshold, first reduction threshold < terminal voltage reduction ≤ second reduction threshold;

[0056] Moderate gas deficiency: second flow rate threshold > gas flow rate ≥ third flow rate threshold, second increment threshold < EIS impedance increment ≤ third increment threshold, second reduction threshold < terminal voltage reduction ≤ third reduction threshold;

[0057] Severe gas shortage: gas flow rate < third flow rate threshold, EIS impedance increment > third increment threshold, terminal voltage drop > third drop threshold. The above thresholds can be flexibly selected according to actual conditions during implementation.

[0058] Based on the above-described gas-deficient fault classification, the stack data after the stack fault sensitivity test can be processed to obtain corresponding fault data and stored in the comprehensive fault library of the stack initial state. Compared with a simple comparison with the fault-free state, this fault analysis is more accurate and reliable. The gas-deficient fault classification provided in this embodiment is only an illustrative reference. During specific implementation, the associated data of other faults can also be processed in accordance with this embodiment.

[0059] In addition, the construction of the comprehensive fault library of the initial state of the fuel cell stack described in this embodiment also includes:

[0060] When conducting a stack fault sensitivity test, the fault-prone areas on a single-chip partition are determined based on the internal medium distribution status of the stack collected through in-situ testing and combined with partition current detection.

[0061] For example, locations such as gas inlets and outlets, where media easily accumulate and current density drops sharply, can be considered prone to failure. In-situ testing methods include but are not limited to partitioned current collectors, partitioned flow field plates, electromagnetic induction, partitioned MEA manufacturing, spatially resolved capillary mass spectrometers, optically transparent single cells, X-ray imaging, synchrotron X-ray imaging, neutron imaging, nuclear magnetic resonance, pressure-sensitive blankets, thin-film thermocouples, thin-film thermal resistors, distributed fiber optic sensors, infrared imagers, fluorescence attenuation thermometry, and Bragg gratings.

[0062] Under different fault conditions, multi-point inspections are used to collect multi-point inspection voltages in the fault-prone areas as fault data and organize them into a comprehensive fault library for the initial state of the stack. Under different fault conditions, EIS data from single-chip partitions in the fault-prone areas is collected as fault data and organized into the comprehensive fault library for the initial state of the stack. Fault analysis in fault-prone areas makes it easier to locate fault data, and the collected fault data is more accurate and reliable, thereby improving the reference value of the constructed comprehensive fault library for the initial state of the stack.

[0063] Step S2: Performing a durability test on the fuel cell stack. During the test, fuel cell stack data is collected at fixed intervals to obtain fuel cell stack data that changes with the time axis, so as to construct a comprehensive fault library of the fuel cell stack process status.

[0064] In order to solve the failure problems that may occur in the fuel cell stack after long-term use during its entire life cycle, this embodiment conducts a durability test on the fuel cell stack to obtain relevant data for building a comprehensive fault library of the fuel cell stack process status.

[0065] For the durability test, the selection of the durability test road spectrum is the key to the test design. The durability test road spectrum of this embodiment includes the accelerated durability road spectrum and the real vehicle road spectrum. Among them, under normal circumstances, the real vehicle road spectrum can be used. The real vehicle road spectrum collects real road driving data (such as vehicle speed, load, road conditions, etc.) through on-board equipment, directly reproduces the actual operation scene of the vehicle, is closer to the actual battery stack usage environment, and has high data reliability. When necessary, the accelerated durability road spectrum can also be used. Compared with the real vehicle road spectrum, by strengthening the key degradation factors (such as high load cycle, temperature fluctuation, start-stop frequency, etc.), the test time is shortened and the efficiency is improved. However, during the calculation process of the accelerated durability road spectrum, it is necessary to fit the corresponding durability time of the battery stack according to the acceleration factor to ensure the accuracy and reliability of the data.

[0066] Specifically, during the endurance test, if the endurance loss is less than a preset loss threshold: when the endurance time is less than the preset time threshold, the stack data is collected at a first fixed time interval; when the endurance time is greater than or equal to the preset time threshold, the stack data is collected at a second fixed time interval, wherein the first fixed time interval is greater than the second fixed time interval;

[0067] If the durability loss is greater than or equal to the preset loss threshold: stack data is collected every third fixed time, where the third fixed time is less than the second fixed time. The thresholds can be flexibly selected according to actual conditions during implementation.

[0068] For example, in this embodiment, when the durability loss is less than 8%, the stack data is collected every 200 hours. After 5000 hours of durability, the collection frequency is changed to every 100 hours. When the durability loss is greater than 8%, the data is collected every 50 hours, thereby obtaining the stack data that changes with the time axis. Whether the subsequent frequency needs to be changed is confirmed based on the collected data and the stack status.

[0069] During the endurance test, premature component failure may occur, causing the stack to become inoperable. This may include cracks in the membrane electrode, pinholes in the membrane electrode, and shedding of the bipolar plate coating. Such failures are fatal to the stack and need to be stored in the comprehensive fault library of the stack process status. Therefore, this embodiment also detects the status of the stack components during the endurance test, and imports the stack data corresponding to the status of the stack components when the stack is inoperable as fault data (for example, hydrogen permeation current, short-circuit resistance, ECSA, plate contact resistance, coating status, whole stack airtightness, etc.) into the comprehensive fault library of the stack process status.

[0070] Step S3: When the stack endurance test reaches the end of its life, the stack data at this time is collected to construct a comprehensive fault library for the end of the stack life state. Finally, the comprehensive fault library for the initial state of the stack, the comprehensive fault library for the process state of the stack, and the comprehensive fault library for the end of the stack life state are aggregated to obtain a fuel cell stack full life cycle library.

[0071] This application collects EIS, gas distribution, operating parameters, multi-point inspection and other information in the initial state of the stack, and combines the stack fault sensitivity test to collect the initial fault information and baseline value of the stack. During the durability of the stack, relevant parameters are collected at fixed intervals, and relevant parameters are collected according to fixed time periods to build a fault library that changes with the time axis until the stack fails due to durability. The above method can collect parameters related to the initial state, process state, and end-of-life state of the stack, so as to build a fault library for the entire life cycle of the stack.

[0072] This embodiment also proposes a fuel cell stack full life cycle fault diagnosis method, using a full life cycle fault library constructed according to the above-mentioned fuel cell stack full life cycle fault library establishment method, the method comprising:

[0073] When a fault occurs in the battery stack, first determine whether the current battery stack operating condition belongs to the preset normal operating condition (temperature, humidity and other environmental conditions are normal). If not, refer to the full life cycle fault library to correct the current operating condition; if it does, proceed to the next step;

[0074] Determine whether the EIS data of the current fuel cell stack is abnormal. If so, refer to the full life cycle fault database to query the corresponding fault problem and correct it. If not, proceed to the next step.

[0075] Determine whether the current stack multi-point inspection voltage is abnormal. If so, refer to the full life cycle fault database to query the corresponding fault problem and correct it. If not, proceed to the next step.

[0076] The microcurrent excitation method is used to collect parameters such as the ohmic impedance, hydrogen permeation current, and electrochemical surface area of ​​each cell in the current stack. It is determined whether the parameters are abnormal. If so, the corresponding fault problem is searched and corrected by referring to the full life cycle fault database. If not, proceed to the next step.

[0077] The fault problem was further identified by disassembling the battery stack for characterization analysis.

[0078] Based on the highly reliable full-life cycle fault library constructed in this embodiment, the diagnostic method of this embodiment can realize fault diagnosis through basic working conditions, impedance, multi-point inspection voltage, and micro-current excitation values, thereby improving diagnostic efficiency and accuracy.

[0079] The present invention has been described above with reference to the accompanying drawings. Obviously, the specific implementation of the present invention is not limited to the above-described method. Any non-substantial improvements made using the method concepts and technical solutions of the present invention, or any direct application of the above-described concepts and technical solutions to other situations without modification, fall within the scope of protection of the present invention.

Claims

1. A method for establishing a fuel cell stack full life cycle fault library, characterized by: The method comprises: Step S1: collecting stack data in the initial state of the stack to define the initial fault-free state of the stack; performing a stack fault sensitivity test, and collecting stack data during the test to construct a comprehensive fault library for the initial state of the stack; Step S2: Performing a durability test on the stack, collecting stack data at fixed intervals during the test, obtaining stack data that changes with time, and constructing a comprehensive fault library for the stack process status; Step S3: When the stack durability test reaches the end of the stack life, the stack data at this time is collected to construct a comprehensive fault library of the stack end-of-life status.

2. The method for establishing a fuel cell stack full life cycle fault library according to claim 1, characterized in that: In step S1, collecting stack data in the initial state of the stack to define an initial fault-free state of the stack includes: Collect data on the entire stack in its initial state, including the stack voltage and total EIS. This data is defined as the stack being in a fault-free state. In the initial state of the stack, single-cell level data is collected, including the average voltage of the single cell, the EIS data of each single cell, the gas distribution from the entire stack to the single cell, and the ohmic impedance, hydrogen permeation current, electrochemical surface area and other parameters of each single cell collected by the microcurrent excitation method. This data is defined as the single cell fault-free state; In the initial state of the fuel cell stack, data at the single-chip partition level is collected, including the EIS data of the single-chip partition, the multi-point inspection voltage of the single-chip partition, and the distribution status of the internal medium of the fuel cell stack collected through in-situ testing. This data is defined as the fault-free state of the single-chip partition.

3. The method for establishing a fuel cell stack full life cycle fault library according to claim 1, characterized in that: The stack fault sensitivity test includes simulating faults on the stack, and the faults include gas deficiency, high temperature, and high and low humidity.

4. A method for establishing a fuel cell stack full life cycle fault library according to claim 1 or 3, characterized in that: Conduct a stack fault sensitivity test and collect stack data during the test to build a comprehensive stack initial state fault library, including: When conducting the stack fault sensitivity test, multi-point inspection voltage, whole stack voltage, single-chip average voltage, stack total EIS, single-chip EIS data, and single-chip partition EIS data are collected. After comparing and analyzing with the corresponding data under the fault-free state, relevant fault data is obtained and organized into the comprehensive fault library of the stack initial state. When conducting a stack fault sensitivity test, gas distribution detection equipment is used to collect gas distribution information from the entire stack to a single chip. At the same time, the EIS data of each single chip and the voltage at the end of the stack are collected. Based on the gas distribution information, single-chip EIS data, and stack end voltage, gas deficiency faults are graded and organized into a comprehensive fault library for the stack initial state.

5. The method for establishing a fuel cell stack full life cycle fault library according to claim 1 or 3, characterized in that: Conduct a stack fault sensitivity test and collect stack data during the test to build a comprehensive stack initial state fault library, including: When conducting a stack fault sensitivity test, the fault-prone areas on a single chip partition are determined based on the internal medium distribution status of the stack collected through in-situ testing and combined with partition current detection; Under different fault conditions, multi-point inspection is used to collect the multi-point inspection voltage of the fault-prone area as fault data and organize it into the comprehensive fault library of the initial state of the fuel cell stack; Under different fault conditions, the EIS data of the single-chip partition of the fault-prone area is collected as fault data and organized into a comprehensive fault library of the initial state of the fuel cell stack.

6. The method for establishing a fuel cell stack full life cycle fault library according to claim 1, characterized in that: The step S2 comprises: During the endurance test, if the endurance loss is less than the preset loss threshold: when the endurance time is less than the preset time threshold, the stack data is collected at a first fixed time interval; when the endurance time is greater than or equal to the preset time threshold, the stack data is collected at a second fixed time interval, wherein the first fixed time interval is greater than the second fixed time interval; If the durability loss is greater than or equal to the preset loss threshold: stack data is collected every third fixed time, and the third fixed time is less than the second fixed time.

7. The method for establishing a fuel cell stack full life cycle fault library according to claim 1, characterized in that: During the endurance test process of step S2, the status of the battery stack components is also detected, and the battery stack data corresponding to the status of the battery stack components when the battery stack cannot work is imported into the battery stack process status comprehensive fault library as fault data.

8. The method for establishing a fuel cell stack full life cycle fault library according to claim 7, characterized in that: The states of the components of the battery stack when the battery stack cannot work include cracks in the membrane electrode, pinholes in the membrane electrode, and shedding of the bipolar plate coating.

9. The method for establishing a fuel cell stack full life cycle fault library according to claim 1, characterized in that: The durability test road spectrum in step S2 includes an accelerated durability road spectrum and a real vehicle road spectrum.

10. A fuel cell stack full life cycle fault diagnosis method, using a fuel cell stack full life cycle fault library constructed according to the fuel cell stack full life cycle fault library establishment method according to any one of claims 1 to 9, characterized in that: The method comprises: When a fault occurs in the stack, first determine whether the current stack operating condition belongs to the preset normal operating condition. If not, refer to the full life cycle fault library to correct the current operating condition; if it does, proceed to the next step; Determine whether the EIS data of the current fuel cell stack is abnormal. If so, refer to the full life cycle fault database to query the corresponding fault problem and correct it. If not, proceed to the next step. Determine whether the current stack multi-point inspection voltage is abnormal. If so, refer to the full life cycle fault database to query the corresponding fault problem and correct it; if not, proceed to the next step; The microcurrent excitation method is used to collect parameters such as the ohmic impedance, hydrogen permeation current, and electrochemical surface area of ​​each cell in the current stack. It is determined whether the parameters are abnormal. If so, the corresponding fault problem is searched and corrected by referring to the full life cycle fault database. If not, proceed to the next step. The fault problem was further identified by disassembling the battery stack for characterization analysis.

Images