Method for monitoring abnormality of fuel cell stack and system therefor

Abstract

The application discloses an abnormality monitoring method and system of a fuel cell stack. The operation process of the fuel cell stack is divided into an open circuit stage with no load connected, an activation polarization stage with low load started, and a normal operation stage with high load started. The abnormality monitoring method comprises the following steps: in the open circuit stage, the open circuit voltage of the battery is collected in real time, if the open circuit voltage is higher than a first set value for a continuous preset number of times, it is judged that the open circuit stage is abnormal, and a pre-warning prompt is triggered; in the activation polarization stage, the output power slope of the stack is calculated, if the power slope is less than a second set value within a specified time, it is judged that the activation polarization stage is abnormal, and a preset action is performed. The technical scheme of the application can monitor different stages of the fuel cell stack, improve the fault disposal efficiency, and increase the safety and stability of the stack operation.

Description

Anomaly Monitoring Methods and Systems for Fuel Cell Stacks Technical Field

[0001] This invention relates to the field of fuel cell testing technology, and specifically to a method and system for anomaly monitoring of fuel cell stacks. Background Technology

[0002] Fuel cell stacks, as devices that directly convert chemical energy into electrical energy, can be applied in fields such as vehicle-mounted power supplies, portable power sources, and distributed power generation. The operation of a fuel cell stack involves multiple stages from startup to stable power generation. Monitoring the operational status of the fuel cell stack is typically necessary. However, existing anomaly monitoring technologies for fuel cell stacks use uniform monitoring parameters, resulting in insufficient targeting for different stages. This can easily lead to the stack operating in an inefficient state for extended periods, accelerating catalyst degradation and increasing the risk of irreversible electrode damage. Furthermore, unaddressed power decay can further exacerbate energy consumption and increase the risk of failure. These problems collectively lead to delays in fault handling of fuel cell stacks, severely impacting their operational safety and stability. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an anomaly monitoring method for fuel cell stacks that can specifically monitor different stages of fuel cell stacks, improve fault handling efficiency, and increase the safety and stability of stack operation.

[0004] This application provides an abnormal monitoring method for a fuel cell stack, which pre-divides the operation process of the fuel cell stack into an open circuit stage with no load connection, an activation polarization stage with low load start-up, and a normal operation stage with high load start-up.

[0005] The anomaly monitoring method includes:

[0006] During the open circuit phase, the open circuit voltage of the battery is collected in real time. If the open circuit voltage is higher than a first set value after a preset number of consecutive cycles, the open circuit phase is judged to be abnormal and an early warning prompt is triggered.

[0007] During the activation polarization stage, the output power slope of the fuel cell is calculated. If the power slope is less than a second set value within a specified time, the activation polarization stage is judged to be abnormal, and processing is performed based on a preset action.

[0008] In one aspect, after the step of real-time acquisition of the battery's open-circuit voltage during the open-circuit phase, the method further includes:

[0009] The fluctuation range of the open-circuit voltage is monitored in real time. If the fluctuation range exceeds a preset fluctuation threshold for a preset number of consecutive preset times, an early warning is triggered.

[0010] In one aspect, the steps for calculating the output power slope of the fuel cell stack include:

[0011] The stack output power during the activation polarization stage is collected at preset time intervals;

[0012] The output power slope is obtained by calculating the ratio of the power difference between two adjacent data acquisitions to the time interval.

[0013] In one aspect, the preset action includes at least one of the following: load adjustment, gas purging, and shutdown protection.

[0014] In one aspect, the anomaly monitoring method further includes:

[0015] During the normal operation phase, the hydrogen pressure fed into the stack and the voltage of a single cell are collected in real time.

[0016] If the hydrogen pressure entering the stack is lower than the third set value, and the voltage of any single cell drops by more than the set threshold within a set time, it is determined to be a hydrogen starvation anomaly, triggering a hydrogen starvation warning.

[0017] Based on the aforementioned hydrogen hunger warning, the hydrogen supply will be adjusted accordingly.

[0018] In one aspect, the anomaly monitoring method further includes:

[0019] During the normal operation phase, the voltage of each individual battery cell is monitored in real time. If the voltage of any individual battery cell is lower than the fourth set value and the duration is greater than or equal to the set time, it is determined to be an reverse polarity abnormality, and preset processing actions are performed according to the voltage level.

[0020] In one aspect, the anomaly monitoring method includes:

[0021] Obtain the cumulative runtime of the fuel cell stack, current ambient temperature, and altitude;

[0022] Based on the dynamic correction model, the first set value, the second set value, the third set value, and the fourth set value are adapted to the influence of the stack aging degree, ambient temperature and humidity, and altitude and air pressure on the stack operation. The update cycle of the dynamic correction model is consistent with the parameter acquisition cycle.

[0023] In one aspect, the activation polarization phase further includes:

[0024] The catalyst activity correlation parameters of the fuel cell stack are obtained, including the fuel cell stack temperature change rate and anode hydrogen utilization rate.

[0025] If the power slope is less than the second set value within a specified time, and the rate of change of the stack temperature is greater than the set rate of change, and the hydrogen utilization rate is less than the set utilization rate, the activation polarization stage is judged to be abnormal.

[0026] In one aspect, the anomaly monitoring method further includes:

[0027] Based on the anomaly types and handling results of the preceding stages, adjust the monitoring parameters, load control strategies, and hydrogen pressure control parameters of the subsequent stages.

[0028] In addition, to solve the above problems, this application also provides an abnormal monitoring system for fuel cell stacks, which pre-divides the operation process of the fuel cell stack into an open circuit stage with no load access, an activation polarization stage with low load start-up, and a normal operation stage with high load start-up.

[0029] The anomaly monitoring system includes:

[0030] The first monitoring module is used to collect the open-circuit voltage of the battery in real time during the open-circuit stage. If the open-circuit voltage is higher than a first set value after a preset number of consecutive cycles, the open-circuit stage is judged to be abnormal and an early warning prompt is triggered.

[0031] The second monitoring module is used to calculate the output power slope of the fuel cell during the activation polarization stage. If the power slope is less than a second set value within a specified time, the activation polarization stage is judged to be abnormal, and processing is performed based on preset actions.

[0032] The beneficial effects of this invention are reflected in the following aspects: by clearly dividing the fuel cell stack operation process into an open-circuit stage, an activation polarization stage, and a normal operation stage, monitoring is performed specifically for the open-circuit and activation polarization stages. During the open-circuit stage, the open-circuit voltage under no-load conditions is monitored. By continuously judging the voltage a preset number of times, initial startup faults are quickly identified, preventing faults from entering subsequent stages. During the activation polarization stage, the output power slope is used as the main monitoring indicator to accurately capture anomalies. Pre-set actions are used for timely intervention to prevent long-term inefficient operation of the stack from exacerbating irreversible damage. This phased and precise monitoring mode achieves early identification and rapid response to abnormal faults, significantly improving fault handling efficiency. It forms full-process protection from the initial startup stage to the low-load stage, effectively avoiding cascading faults caused by anomalies in each stage, and improving the safety and stability of fuel cell stack operation. Attached Figure Description

[0033] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0034] Figure 1 is a schematic diagram of the process steps of the abnormal monitoring method for fuel cell stacks in this application;

[0035] Figure 2 is a schematic diagram of the process steps for calculating the output power slope of the fuel cell stack in the anomaly monitoring method of this application;

[0036] Figure 3 is a schematic diagram of the monitoring process steps during the normal operation phase of the anomaly monitoring method in this application;

[0037] Figure 4 is a schematic diagram of the process steps for dynamically adjusting the set value in the anomaly monitoring method of this application.

[0038] Figure 5 is a schematic diagram of the process steps for monitoring catalyst activity in the anomaly monitoring method of this application;

[0039] Figure 6 is a schematic diagram of the functional structure of the abnormal monitoring system of the fuel cell stack of this application. Detailed Implementation

[0040] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0041] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0042] As shown in Figure 1, this application provides an abnormal monitoring method for fuel cell stacks. The working principle of the fuel cell stack is to directly convert the chemical energy of hydrogen and oxygen into electrical energy through electrochemical oxidation-reduction reaction. The process does not involve combustion or mechanical movement, and has high energy conversion efficiency and clean emissions.

[0043] A typical fuel cell stack consists of multiple individual cells connected in series. Each individual cell includes an anode, a cathode, and an electrolyte membrane. After hydrogen gas is introduced into the anode, it decomposes into protons under the action of a catalyst. ) and electrons ( Electrons pass through an external circuit to form an electric current, while protons migrate through the electrolyte membrane to the cathode. Simultaneously, oxygen (or air) is introduced into the cathode, where it combines with electrons and protons on the surface of the cathode catalyst to form water. This continuous reaction ensures a stable output of electrical energy. Throughout the process, the electrolyte membrane allows only protons to pass through, preventing gas leakage and electron passage, thus ensuring the orderly conduct of the reaction.

[0044] The operation of a fuel cell stack is pre-divided into three stages: an open-circuit stage with no load, an activation and polarization stage during low load startup, and a normal operation stage during high load startup. The operation of the fuel cell stack is divided into three continuous and progressive stages based on operating conditions, load status, and reaction characteristics. The open-circuit stage with no load is the initial preparation stage for stack startup. At this time, hydrogen and air have been introduced into the stack to form a stable electrochemical potential, but no external load is connected, and there is no current output. The initial state of the stack is detected through the no-load condition. The activation and polarization stage during low load startup is the stage where the stack transitions from no load to high load. At this time, a low load is connected, and a small current begins to be output. The electrochemical reaction gradually starts but at a slow rate. Charge transfer on the catalyst surface is limited, and the power increases slowly with the current. It is crucial to monitor catalyst activity and electrode reaction status. The normal operation stage during high load startup is the stage where the stack generates stable power. At this time, the load reaches the preset operating intensity, the current density is at a high level, the electrochemical reaction is efficient and stable, and the rated power is continuously output. It is crucial to prevent faults that affect operational stability, such as hydrogen starvation and reverse polarity.

[0045] Anomaly monitoring methods include:

[0046] Step S10: During the open-circuit phase, the open-circuit voltage of the battery is collected in real time. If the open-circuit voltage exceeds a first set value after a preset number of consecutive tests, an abnormality in the open-circuit phase is determined, triggering an early warning. At this stage, although hydrogen gas has been introduced into the stack to form an electrochemical potential with air, no external load is connected, and there is no current output. The risk is concentrated on initial faults during startup. By collecting the open-circuit voltage of individual cells in real time, a judgment is established for a preset number of consecutive tests, i.e., the preset number of tests (e.g., 3-5) must be met consecutively to avoid misjudgments caused by momentary electromagnetic interference or sensor fluctuations. When the open-circuit voltage exceeds a first set value (e.g., 0.9V~1.05V / cell) after a preset number of consecutive tests, it indicates that the stack may have problems such as membrane perforation or sealing failure. At this time, an early warning is triggered, allowing the stack startup process to be paused promptly, preventing the initial fault from entering the load phase and avoiding irreversible damage during startup from the source.

[0047] Step S20: During the activation polarization stage, the output power slope of the fuel cell stack is calculated. If the power slope is less than a second set value within a specified time, the activation polarization stage is deemed abnormal, and processing is performed based on preset actions. In this stage, the fuel cell stack is connected to a low load and begins to output a small current. The core risks are catalyst activity decay and gradual power decay caused by electrode structure deterioration, and such anomalies are difficult to identify directly using a single power value. This step uses the calculated fuel cell stack output power slope as a monitoring indicator. First, the fuel cell stack output power is continuously collected at preset time intervals, such as 10 seconds, during this stage. Then, the power slope is obtained by comparing the difference between two adjacent power values ​​with the time interval. If the power slope remains less than the second set value within the specified time, it indicates that the fuel cell stack may be experiencing a slowdown in reaction rate due to decreased catalyst activity, or that electrode structure deterioration may be hindering charge transfer. The specified time is 30-60 seconds, and the second set value is -0.02 to -0.01 kW / s. The negative slope represents power decay. At this time, the preset action is taken to alleviate the power decay trend in time, avoid the fuel cell stack from operating in an inefficient state for a long time, which will aggravate the irreversible damage to the catalyst and electrodes, and ensure the operational stability of the fuel cell stack during the transition from low load to high load.

[0048] In this embodiment, the operation of the fuel cell stack is clearly divided into an open-circuit stage, an activation polarization stage, and a normal operation stage, with monitoring specifically focused on the open-circuit and activation polarization stages. During the open-circuit stage, the open-circuit voltage under no-load conditions is monitored. By continuously checking the voltage a preset number of times, initial startup faults are quickly identified, preventing faults from entering subsequent stages. During the activation polarization stage, the output power slope is used as the primary monitoring indicator to accurately detect anomalies. Pre-set actions are implemented for timely intervention, preventing long-term inefficient operation of the stack from exacerbating irreversible damage. This phased and precise monitoring mode enables early identification and rapid response to abnormal faults, significantly improving fault handling efficiency. It provides end-to-end protection from the initial startup stage to the low-load stage, effectively avoiding cascading faults caused by anomalies at each stage, and increasing the safety and stability of the fuel cell stack operation.

[0049] In one embodiment of this application, after the step of real-time acquisition of the battery's open-circuit voltage during the open-circuit phase, the method further includes:

[0050] Step S110: Monitor the fluctuation amplitude of the open-circuit voltage in real time. If the fluctuation amplitude exceeds a preset fluctuation threshold after a preset number of consecutive cycles, an early warning is triggered. During this stage, the fuel cell stack is in a no-load state. Although there is no current output, the stability of the open-circuit voltage directly reflects the consistency of the initial state within the stack, such as membrane humidity uniformity, gas distribution balance, and single-cell consistency. Based on the real-time acquisition of the open-circuit voltage, the voltage fluctuation amplitude is further calculated, i.e., the difference between the maximum and minimum values ​​of the open-circuit voltage per unit time, such as the voltage change range within 3.5 seconds. A judgment design for continuous preset number of verifications is established. An early warning is not triggered by a single fluctuation exceeding the threshold, but rather by continuously meeting the preset number of times, such as 3-5 times. This eliminates false fluctuations caused by accidental factors such as instantaneous electromagnetic interference and sensor sampling errors, ensuring the accuracy of anomaly detection. When the voltage fluctuation amplitude exceeds the preset fluctuation threshold after a preset number of consecutive cycles, such as ±0.05V / 3.5s, it indicates that the fuel cell stack may have potential problems such as localized membrane drying, localized blockage of hydrogen / air flow channels, or poor single-cell consistency. At this time, an early warning prompt is triggered, which can promptly suspend the startup process and prevent such hidden problems from developing into more serious power anomalies or reverse polarity faults as the fuel cell stack enters the subsequent load stage, thus ensuring that the fuel cell stack has a stable electrochemical foundation from the initial startup stage.

[0051] As shown in Figure 2, the steps for calculating the output power slope of the fuel cell stack include:

[0052] Step S210: Collect the fuel cell stack output power during the activation polarization stage at preset time intervals. During this stage, the fuel cell stack is connected to a low load and begins to stably output a small current. Continuously collecting power data provides a basis for subsequent judgment of catalyst activity and electrode reaction status. First, the acquisition process for the preset time interval is clearly defined. Considering the slow reaction rate and gradual power change during the activation polarization stage, the preset time interval must balance data timeliness and stability. This avoids data redundancy and increased system computational burden due to excessively short intervals, while also preventing the loss of subtle power change trends due to excessively long intervals. During the acquisition process, the accuracy of data acquisition is ensured simultaneously. For example, a high-precision power sensor is used to capture the total output power of the fuel cell stack in real time, eliminating interference from instantaneous load fluctuations and slight voltage oscillations. By continuously collecting power data at preset time intervals, a continuous dataset of power changes during the activation polarization stage can be formed.

[0053] Step S220: Calculate the ratio of the power difference between two adjacent data acquisitions to the time interval to obtain the output power slope. The discrete power data acquired in step S210 is converted into a quantitative indicator reflecting the power change trend, namely the output power slope. The ratio is calculated using the power difference between two adjacent data acquisitions as the numerator and the time interval between the two acquisitions as the denominator. The result directly reflects the rate of change of the stack's output power per unit time. For example, if the power acquired in two adjacent data acquisitions is 120W and 118W respectively, and the preset time interval is 10s, then the power difference is -2W, and the output power slope is -0.2W / s. A negative slope indicates a power decay trend. By quantifying the power change trend, gradual anomalies in the activation polarization stage can be accurately captured, such as a slow power decrease caused by catalyst activity decay, which is difficult to identify directly from a single power value. Simultaneously, the accuracy of data matching must be ensured during the calculation process, that is, the power data from two adjacent data acquisitions must strictly correspond to the corresponding time interval to avoid slope calculation errors due to data misalignment.

[0054] In one embodiment of this application, the preset actions include at least one of the following: load regulation, gas purging, and shutdown protection. These preset actions are tiered and targeted intervention measures designed to address abnormal power slopes during the activation polarization stage. They cover at least three types of actions: load regulation, gas purging, and shutdown protection, and can be used individually or in combination depending on the severity of the anomaly. Load regulation reduces the stack load ramp-up rate, decreases the stack reaction intensity, alleviates the working pressure on the catalyst and electrodes, and creates conditions for anomaly recovery. Gas purging specifically refers to initiating the anode hydrogen purging procedure, which can remove reaction byproducts or residual air that may accumulate on the electrode surface, improve hydrogen distribution uniformity, and restore the contact efficiency between the catalyst active sites and the gas. Shutdown protection is a fallback measure for extreme anomaly scenarios. When the calculated power slope is still less than the second set value after two consecutive calculations, it indicates that the anomaly cannot be alleviated by conventional intervention. In this case, cutting off the hydrogen supply and maintaining air purging can prevent irreversible catalyst deactivation, electrode corrosion, and other serious damage, minimizing the risk of fault escalation.

[0055] As shown in Figure 3, the anomaly monitoring method also includes:

[0056] Step S310: During normal operation, the feed hydrogen pressure and individual cell voltage are collected in real time. During normal operation, the stack needs to continuously output power to meet demand. One risk is hydrogen starvation due to a mismatch between hydrogen supply and reaction consumption. Feed hydrogen pressure and individual cell voltage are key parameters directly reflecting the hydrogen supply status and electrochemical reaction balance. By synchronously acquiring these two parameters in real time, a high-precision hydrogen pressure sensor monitors the feed hydrogen pressure to ensure the capture of changes in hydrogen supply caused by fluctuations in hydrogen source pressure and pipeline leaks. Simultaneously, a voltage acquisition module collects the voltage of each individual cell, accurately reflecting the electrochemical reaction status of each cell. This synchronous acquisition of two parameters ensures data real-time performance and avoids misjudgments caused by environmental interference or sensor errors when monitoring only one parameter.

[0057] Step S320: If the hydrogen pressure fed into the stack is lower than the third set value, and the voltage of any single cell drops beyond a set threshold within a set time, it is determined to be a hydrogen starvation anomaly, triggering a hydrogen starvation warning. A comprehensive judgment logic based on hydrogen supply status parameters and reaction status parameters ensures the accuracy of anomaly identification. First, the judgment conditions are clearly defined: 1) The hydrogen pressure fed into the stack is lower than the third set value, such as 35 kPa, indicating that the hydrogen supply can no longer meet the current reaction requirements of the stack, which is a prerequisite for hydrogen starvation; 2) The voltage of any single cell drops beyond a set threshold within a set time, such as 1-3 seconds, and the set threshold is greater than or equal to 5%, which is a direct manifestation of hydrogen starvation limiting the electrochemical reaction. Simultaneously, the judgment rule requires both conditions to be met simultaneously and continuously verified, excluding false anomaly scenarios such as instantaneous hydrogen pressure drops and random deviations in single-cell voltage. When all the above conditions are met, a hydrogen starvation anomaly can be accurately determined and a warning can be triggered, ensuring timely identification in the early stages of hydrogen starvation and preventing serious faults such as reverse polarity and catalyst failure.

[0058] Step S330: Based on the hydrogen hunger warning, the hydrogen supply is adjusted in a coordinated manner. Upon triggering the hydrogen hunger warning, the linkage control between the hydrogen supply system and the warning signal is immediately activated. Electrical signals are used to control the hydrogen release module to increase the hydrogen supply flow rate or increase the hydrogen supply pressure, rapidly restoring the infeed hydrogen pressure to the normal range (e.g., 40-60 kPa). This ensures that the anode hydrogen supply rate matches the current reaction consumption rate of the fuel cell. During the coordinated adjustment, both response speed and stability are considered to avoid a sudden increase in hydrogen supply leading to excessively high or low hydrogen pressure. Closed-loop control provides real-time feedback of the infeed hydrogen pressure and individual cell voltage data until the parameters stabilize, thus quickly terminating the hydrogen hunger state. This prevents irreversible damage to the fuel cell due to prolonged hydrogen shortage, such as power drop or reverse polarity, ensuring stable power generation during normal operation.

[0059] In one embodiment of this application, the anomaly monitoring method further includes:

[0060] Step S340: During normal operation, the voltage of each individual cell is monitored in real time. If the voltage of any individual cell falls below the fourth preset value and the duration is greater than or equal to the preset time, it is determined to be a reverse polarity anomaly, and preset processing actions are executed according to voltage levels. During this stage, the stack is in a high-load, stable power generation state. Reverse polarity anomalies are often caused by insufficient hydrogen supply to a single cell, severe catalyst activity degradation, or single-cell structural deterioration. If not handled promptly, they can quickly lead to irreversible damage to a single cell and even spread to the entire stack. By monitoring the output voltage of each individual cell in real time, a dual judgment based on voltage threshold and duration is established. First, the fourth preset value is defined as the critical threshold for reverse polarity anomalies, and a minimum duration is set. For example, the fourth preset value is 0.4V, and the minimum duration is 2 seconds. This avoids false judgments caused by instantaneous load fluctuations or sensor malfunctions, ensuring the accuracy of anomaly identification. When the voltage of any single cell falls below the fourth set value for a duration that meets the set requirement, it is determined to be a reverse polarity anomaly. At this time, preset handling actions are executed according to voltage levels: for mild reverse polarity, mild interventions such as load reduction and hydrogen purging are taken; for moderate reverse polarity, deep load reduction and emergency hydrogen supply measures are taken; and for severe reverse polarity, shutdown protection is immediately triggered. Through tiered handling, the possibility of stack operation is preserved to the greatest extent, and damage can be stopped quickly in the event of a serious anomaly, so as to prevent the failure from spreading to the entire stack.

[0061] As shown in Figure 4, the anomaly monitoring methods include:

[0062] Step S410: Obtain the cumulative runtime of the fuel cell stack, the current ambient temperature, and the altitude; obtain key environmental and self-state parameters that affect the fuel cell stack's operating status and the rationality of monitoring thresholds. The cumulative runtime is accurately recorded by the timing module of the fuel cell stack control system, reflecting the degree of fuel cell stack aging; the current ambient temperature is collected in real time by environmental sensors, capturing the impact of temperature on electrochemical reaction rates and gas solubility; and the altitude is obtained by an altitude sensor, indirectly reflecting changes in air pressure. The simultaneous acquisition of these three types of parameters ensures the comprehensiveness and real-time nature of the data, providing accurate input for subsequent dynamic correction models and avoiding rigid threshold settings due to neglecting fuel cell stack aging and environmental changes.

[0063] Step S420, based on the dynamic correction model, adjusts the first, second, third, and fourth setpoints according to the influence of stack aging, ambient temperature and humidity, and altitude and air pressure on stack operation. The update cycle of the dynamic correction model is consistent with the parameter acquisition cycle. Step S420 aims to address the problem that fixed thresholds cannot adapt to stack aging and dynamic environmental changes, improving the accuracy of anomaly monitoring. Based on the preset dynamic correction model, the correlation between cumulative stack operating time and aging level, the influence of ambient temperature / humidity on electrochemical reactions, and the correlation between altitude and oxygen content are first analyzed. Then, the first to fourth setpoints are adjusted accordingly based on these patterns. Simultaneously, the update cycle of the dynamic correction model is consistent with the parameter acquisition cycle, ensuring that the thresholds adjust in real time to follow parameter changes, avoiding misjudgments or missed judgments caused by threshold lag. This ensures that anomaly monitoring at each stage is always adapted to the real-time state of the stack and environmental conditions, further improving the adaptability and reliability of the monitoring system.

[0064] As shown in Figure 5, the activation polarization stage also includes:

[0065] Step S201 involves acquiring catalyst activity-related parameters for the fuel cell stack, including the stack temperature change rate and anode hydrogen utilization rate. Catalyst activity during the activation and polarization phase directly determines the electrochemical reaction efficiency, while the stack temperature change rate and anode hydrogen utilization rate are key parameters that indirectly reflect catalyst activity. This step involves real-time acquisition of stack temperature data using a temperature sensor and calculation of the temperature change rate to capture changes in catalyst reaction intensity. The anode hydrogen utilization rate is calculated using hydrogen supply flow sensor data and stack output power data, reflecting the conversion efficiency of hydrogen on the catalyst surface. The simultaneous acquisition of these two types of parameters provides supplementary information on catalyst activity for subsequent anomaly detection based on power slope, avoiding misjudgments caused by relying solely on a single power indicator.

[0066] Step S202: If the power slope is less than the second set value within a specified time, and the stack temperature change rate is greater than the set change rate, and the hydrogen utilization rate is less than the set utilization rate, then an activation polarization stage anomaly is determined. Step S202 is the precise determination step for activation polarization stage anomalies. It adopts a multi-parameter comprehensive judgment using core and auxiliary indicators to solve the problem of easy interference from the single power slope indicator. The core judgment condition is that the power slope is less than the second set value within a specified time, while simultaneously superimposed with the constraints of two catalyst activity-related parameters: the stack temperature change rate is greater than the set change rate, and the anode hydrogen utilization rate is less than the set utilization rate. Only when all three conditions are met simultaneously is it determined that the activation polarization stage is abnormal. This not only eliminates false anomalies caused by fluctuations in a single parameter, but also accurately identifies real anomalies caused by catalyst activity decay or electrode structure deterioration, ensuring the reliability of anomaly determination and providing an accurate basis for subsequent targeted treatment.

[0067] In one embodiment of this application, the anomaly monitoring method further includes:

[0068] Step S50: Based on the anomaly types and handling results of the preceding stages, adjust the monitoring parameters, load control strategies, and hydrogen pressure control parameters of the subsequent stages. First, summarize the anomaly types and corresponding handling results of the preceding stages. Then, based on the preset linkage logic, specifically adjust the key control parameters of the subsequent stages. Regarding monitoring parameters, extend the monitoring time of anomaly-related indicators or increase the sampling frequency; regarding load control strategies, reduce the load ramp-up rate or limit the maximum output power; regarding hydrogen pressure control parameters, adjust the initial value of the hydrogen pressure fed into the stack or the warning threshold. Through this linkage adjustment from preceding anomalies to subsequent adaptation, a full-cycle protection closed loop is formed, preventing unresolved anomalies from escalating in the preceding stages from expanding in subsequent stages, further improving the continuity and safety of stack operation.

[0069] As shown in Figure 6, this application also provides an abnormal monitoring system for a fuel cell stack, which pre-divides the operation process of the fuel cell stack into an open circuit stage with no load access, an activation and polarization stage with low load start-up, and a normal operation stage with high load start-up.

[0070] The anomaly monitoring system includes: a first monitoring module 10 and a second monitoring module 20.

[0071] The first monitoring module 10 is used to collect the open-circuit voltage of the battery in real time during the open-circuit phase. If the open-circuit voltage exceeds a first set value after a preset number of consecutive tests, an abnormality in the open-circuit phase is determined, and an early warning is triggered. The first monitoring module 10 is a dedicated monitoring unit designed for the open-circuit phase of the fuel cell stack. It identifies initial faults in the initial startup phase by accurately monitoring the open-circuit voltage. When the stack is in the open-circuit phase without load connection, the first monitoring module 10 captures the open-circuit voltage data of individual cells in real time through voltage acquisition components. At the same time, it has built-in verification logic for a preset number of consecutive tests to avoid misjudgments caused by instantaneous interference. When the open-circuit voltage collected for a preset number of consecutive tests is higher than the first set value, the first monitoring module 10 immediately determines that there is an abnormality in the open-circuit phase and triggers an early warning simultaneously, promptly suspending the stack startup process. This prevents the risk of faulty operation from the source and ensures the safe entry of the stack into the load phase.

[0072] The second monitoring module 20 is used to calculate the output power slope of the fuel cell stack during the activation polarization stage. If the power slope is less than a second set value within a specified time, an anomaly is determined in the activation polarization stage, and processing is performed based on preset actions. The second monitoring module 20 is the core monitoring and processing unit in the activation polarization stage. During the activation polarization stage when the fuel cell stack starts up under low load, it first collects the fuel cell stack output power data at preset time intervals and obtains the output power slope through built-in calculation logic. Then, it compares the power slope with the second set value in real time. If the power slope is continuously less than the set value within a specified time, an anomaly is determined in the activation polarization stage. Subsequently, the second monitoring module 20 triggers preset actions according to the degree of anomaly to promptly alleviate the power decay trend, prevent the fuel cell stack from operating inefficiently for a long time and exacerbating irreversible damage, and ensure the stability of the fuel cell stack transitioning from low load to high load.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for anomaly monitoring of a fuel cell stack, characterized in that, The operation of the fuel cell stack is pre-divided into an open-circuit phase with no load, an activation and polarization phase with low load startup, and a normal operation phase with high load startup. The abnormality monitoring method includes: in the open-circuit phase, the open-circuit voltage of the battery is collected in real time. If the open-circuit voltage is higher than a first set value for a preset number of consecutive cycles, the open-circuit phase is judged to be abnormal, and an early warning is triggered. In the activation and polarization phase, the output power slope of the stack is calculated. If the power slope is less than a second set value within a specified time, the activation and polarization phase is judged to be abnormal, and processing is performed based on preset actions. The preset actions include at least one of the following: load adjustment, gas purging, and shutdown protection.

2. The anomaly monitoring method according to claim 1, characterized in that, After the step of acquiring the open-circuit voltage of the battery in real time during the open-circuit phase, the method further includes: monitoring the fluctuation range of the open-circuit voltage in real time; if the fluctuation range exceeds a preset fluctuation threshold after a preset number of consecutive preset times, an early warning prompt is triggered.

3. The anomaly monitoring method according to claim 1, characterized in that, The steps for calculating the output power slope of the fuel cell stack include: collecting the output power of the fuel cell stack during the activation polarization stage at preset time intervals; calculating the ratio of the power difference between two adjacent collections to the time interval to obtain the output power slope.

4. The anomaly monitoring method according to claim 1, characterized in that, The abnormality monitoring method further includes: during the normal operation phase, real-time acquisition of the hydrogen pressure fed into the stack and the voltage of a single cell; if the hydrogen pressure fed into the stack is lower than a third set value, and the voltage of any single cell drops by more than a set threshold within a set time, it is determined to be a hydrogen starvation abnormality, triggering a hydrogen starvation warning; based on the hydrogen starvation warning, the hydrogen supply is adjusted accordingly.

5. The anomaly monitoring method according to claim 4, characterized in that, The abnormality monitoring method further includes: during the normal operation phase, real-time monitoring of the voltage of each individual battery cell; if the voltage of any individual battery cell is lower than the fourth set value and the duration is greater than or equal to the set time, it is determined to be an reverse polarity abnormality, and preset processing actions are performed according to voltage levels.

6. The anomaly monitoring method according to claim 5, characterized in that, The anomaly monitoring method includes: acquiring the cumulative operating time of the fuel cell stack, the current ambient temperature, and the altitude; based on a dynamic correction model, adjusting the first set value, the second set value, the third set value, and the fourth set value according to the influence of fuel cell stack aging degree, ambient temperature and humidity, and altitude and air pressure on fuel cell stack operation, and the update cycle of the dynamic correction model is consistent with the parameter acquisition cycle.

7. The anomaly monitoring method according to claim 1, characterized in that, The activation polarization stage further includes: acquiring catalyst activity correlation parameters of the fuel cell stack, including the fuel cell stack temperature change rate and anode hydrogen utilization rate; if the power slope is less than a second set value within a specified time, and the fuel cell stack temperature change rate is greater than a set change rate, and the hydrogen utilization rate is less than a set utilization rate, the activation polarization stage is judged to be abnormal.

8. The anomaly monitoring method according to claim 1, characterized in that, The anomaly monitoring method further includes: adjusting the monitoring parameters, load control strategy, and hydrogen pressure control parameters of the subsequent stage based on the anomaly type and processing results of the preceding stage.

9. An anomaly monitoring system for a fuel cell stack, characterized in that, The operation of the fuel cell stack is pre-divided into an open-circuit phase with no load, an activation and polarization phase with low load startup, and a normal operation phase with high load startup. The abnormality monitoring system includes: a first monitoring module, used to collect the open-circuit voltage of the battery in real time during the open-circuit phase; if the open-circuit voltage is higher than a first set value after a preset number of consecutive cycles, an abnormality is determined in the open-circuit phase, and an early warning is triggered; a second monitoring module, used to calculate the output power slope of the stack during the activation and polarization phase; if the power slope is less than a second set value within a specified time, an abnormality is determined in the activation and polarization phase, and processing is performed based on preset actions, wherein the preset actions include at least one of the following: load adjustment, gas purging, and shutdown protection.

Images