A method and system for controlling a fuel cell air system

Abstract

This invention provides a control method and system for a fuel cell air system, relating to the field of air supply regulation technology for fuel cell stacks. The method first collects data on voltage, temperature, pressure, and oxygen concentration of the fuel cell stack. Based on the stack flow channel structure, the fuel cell stack is divided into multiple air supply zones. Air flow model parameters for each air supply zone are established, and an input data set for zone state estimation is constructed. The local oxygen concentration and humidity of each air supply zone are mathematically estimated using the flow model and electrochemical reaction model to obtain the zone operating state parameters. Combining voltage consistency, oxygen concentration distribution, humidity distribution, and historical operating characteristics, a health index reflecting the stack's health status is constructed. This invention can ensure stack reaction uniformity while considering stack health status, achieving adaptive optimization of air distribution and improving the performance stability of the fuel cell stack.

Description

Technical Field

[0001] This invention relates to the field of air supply technology for fuel cell stacks, and in particular to a control method and system for a fuel cell air system. Background Technology

[0002] During power generation, fuel cell stacks require a continuous supply of air to the cathode side to ensure the stable conduct of the electrochemical reaction. Traditional air systems typically employ a single air passage structure, supplying air to the entire fuel cell stack via an air compressor. The air supply is controlled in a closed loop based on factors such as stack current, stack temperature, and outlet oxygen concentration. However, this approach has the following drawbacks:

[0003] First, the airflow distribution inside a fuel cell stack exhibits significant regional differences. Due to variations in flow channel structure, pressure drop distribution, and local reaction intensity, even if the air compressor provides the same total air volume, the local oxygen concentration and humidity in different regions may still be significantly uneven, resulting in the coexistence of "oxygen-rich" and "oxygen-deficient" zones within the stack.

[0004] Secondly, the degradation process of fuel cell stacks is characterized by significant cumulative and regional variations. However, existing air control methods typically set fixed or semi-fixed air margins based on current load requirements, failing to incorporate the stack's health status into the air volume regulation mechanism. As operating time increases, even if early signs of degradation appear in localized areas of the stack, such as insufficient oxygen concentration, humidity deviations, or increased voltage fluctuations, the air system struggles to detect and compensate for them in a timely manner. This allows localized degradation to deepen further, impacting the overall lifespan of the fuel cell stack.

[0005] Therefore, we propose a control method and system for a fuel cell air system; the information disclosed in the background section is only for enhancing the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by providing a control method and system for a fuel cell air system, thereby solving the technical problems mentioned in the background section.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A control method for a fuel cell air system includes the following steps:

[0009] S1. Acquire real-time operating data of single cell voltage, stack temperature, stack pressure difference, air inlet pressure, and air outlet oxygen concentration of the fuel cell stack during operation; Based on the internal flow channel structure of the stack, divide the fuel cell stack into multiple air supply zones and establish a corresponding zone identifier for each air supply zone.

[0010] S2. Based on real-time operating data and partition identification, local oxygen concentration estimators and local humidity estimators for each air supply partition are constructed using air flow models and electrochemical reaction models. The partition operating status parameters of each air supply partition are output to characterize the gas distribution and reaction uniformity in the internal spatial dimension of the reactor.

[0011] S3. Based on voltage consistency, zoned operating status parameters, stack temperature fluctuations, and historical operating data, a health index is constructed to reflect the degradation level of the fuel cell stack; this index is used to describe the stack's performance maintenance capability and degradation trend throughout its entire life cycle.

[0012] S4. Based on the partition operation status parameters and health index of each air supply partition, construct a model for calculating the target air volume of each partition; determine the target air flow rate of each air supply partition according to the stack load requirements, so that the target air volume of each partition can simultaneously meet the overall stack reaction requirements, the stack internal voltage consistency requirements, and the stack health index maintenance requirements.

[0013] S5. Based on the target air volume for each air supply zone, control the opening of the throttle valve or flow guide mechanism corresponding to each air supply zone, and allocate the corresponding actual air flow to each air supply zone; maintain the coordination relationship between the total air supply of the air compressor and the air volume of each zone during the air allocation process to ensure the stability of the air system;

[0014] S6. Based on the air supply execution results of step S5, collect updated stack operation data in real time, and recalculate the partition operation status parameters and health index. When the health index decreases or the partition operation status deviates from the preset range, dynamically adjust the partition target air volume so that the air supply process can automatically correct the air margin allocation strategy in the early stage of fuel cell stack degradation.

[0015] S1 specifically includes:

[0016] The system collects basic operational data such as single cell voltage, stack temperature, stack pressure difference, air inlet pressure, and air outlet oxygen concentration during the operation of the fuel cell stack, and provides input conditions for subsequent zoning estimation.

[0017] Based on the flow channel geometry of the fuel cell stack, the stack is divided into multiple air supply zones, and a unique zone identifier is set for each air supply zone.

[0018] Establish corresponding airflow model parameters for each air supply zone, including flow channel resistance coefficient, local pressure loss characteristics, and air transmission path;

[0019] Construct input data sets for each air supply zone, associate the collected basic operational data with the zone identifiers, and form a unified data structure for spatial dimension estimation;

[0020] The structural information of each air supply zone is verified against the input dataset to ensure that subsequent estimation steps can be processed based on a consistent and accurate zone structure.

[0021] S2 specifically includes:

[0022] Based on a unified data structure, the air flow model is input to preliminarily estimate the gas flow distribution in each air supply zone;

[0023] The preliminary estimates were input into the regional reaction model related to the electrochemical reaction to calculate the local oxygen consumption and local water generation in each region.

[0024] A local oxygen concentration estimator is constructed to estimate the oxygen concentration of each air supply zone in real time, thereby obtaining the oxygen concentration distribution of each zone.

[0025] A local humidity estimator is constructed, which combines local oxygen concentration and water generation to estimate the local humidity distribution of each air supply zone, forming zone operating status parameters, including zone oxygen concentration and zone humidity, etc.

[0026] The consistency of the zone operation status parameters is verified to ensure that the zone operation status parameters of each air supply zone can be used as the input data source for the subsequent construction of the health index.

[0027] S3 specifically includes:

[0028] Input the operating status parameters of the partition and extract the feature indicators that reflect the local operating consistency, including partition voltage consistency, partition oxygen concentration deviation and partition humidity deviation;

[0029] By combining historical operating data of the heap (cumulative operating time, cumulative load changes, historical temperature fluctuations, etc.), a multi-dimensional set of input features for constructing a health index is formed.

[0030] Input features are fed into the degradation modeling module to fit the performance degradation trend of the heap, and intermediate degradation parameters are obtained to reflect the rate of heap degradation.

[0031] Based on intermediate degradation parameters, a health index is constructed so that the health index can reflect the current degree of degradation and future degradation trend of fuel cell stacks.

[0032] The health index is checked to see if it falls within the preset health status range, and the health index is used as the input variable for the subsequent model for calculating the target air volume for each zone.

[0033] S4 specifically includes:

[0034] Input the partition operation status parameters to determine whether the current oxygen concentration and humidity levels of each air supply partition meet the current load requirements of the stack.

[0035] The comparison results of health index and operating status are input into the air volume optimization module to construct a comprehensive optimization target, including overall reactor reaction requirements, reactor internal consistency requirements, and health index maintenance requirements.

[0036] Based on the optimization objective, the target air volume for each air supply zone is calculated using an air allocation algorithm, forming a zone air volume allocation list;

[0037] Perform a feasibility constraint check on the zone air volume allocation list to ensure that the sum of the target air volume for all zones does not exceed the total air volume currently available from the air compressor.

[0038] The target air volume for each zone that passes the constraint test will be used as the control input for subsequent air supply execution steps.

[0039] S5 specifically includes:

[0040] Input the target air volume for each zone, and generate adjustment commands for each throttle valve or flow guide mechanism based on the target value for each air supply zone.

[0041] According to the adjustment command, control the actual opening degree of the throttle valve or flow guide mechanism of each air supply zone so that each air supply zone obtains the corresponding actual air flow.

[0042] Monitor the total air supply of the air compressor and compare it with the actual air flow of each air supply zone to ensure the overall air supply coordination of the air system.

[0043] Based on the comparison results, the air compressor output and throttle valve opening are adjusted synchronously to avoid mismatch between the total air flow and the zone air flow.

[0044] The actual airflow of each air supply zone is used as the air supply execution result and provided to subsequent dynamic adjustment steps.

[0045] S6 specifically includes:

[0046] Based on the air supply execution results, the operating data of the fuel cell stack is re-collected, and the basic operating data set is updated;

[0047] The updated basic operational data is input into the estimation process to recalculate the partition operational state parameters, so that the spatial dimension state estimation remains real-time.

[0048] The partition's running status parameters are input into the health index construction process to obtain the updated health index, ensuring that the health status estimate is dynamically updated as the running status changes.

[0049] Compare the updated health index with the preset health range. When the health index drops or the zone's operating status deviates from the preset range, trigger the recalculation of the zone's target air volume.

[0050] Based on the triggering results, the target air volume of the partition is dynamically adjusted so that the air supply process can be adaptively corrected in the early stage of reactor degradation, thereby achieving comprehensive optimization of reactor internal reaction uniformity and reactor lifetime performance.

[0051] A control system for a fuel cell air system includes:

[0052] The data acquisition module is used to collect operating data of the fuel cell stack, including single cell voltage, stack temperature, stack pressure difference, air inlet pressure, air outlet pressure, and oxygen concentration information.

[0053] The partition modeling module is used to divide the fuel cell stack into multiple air supply partitions based on the flow channel structure of the fuel cell stack, and to establish corresponding air flow model parameters for each air supply partition.

[0054] The zone status estimation module is used to calculate the local oxygen concentration and local humidity of each air supply zone based on the zone input data set and air flow model parameters, and obtain the zone operating status parameters.

[0055] The health index construction module is used to calculate the intermediate degradation parameters of the fuel cell stack based on the partition operation status parameters and historical operation characteristics, and to construct the health index of the fuel cell stack based on the intermediate degradation parameters.

[0056] The target air volume calculation module is used to construct the air demand index of each air supply zone based on the zone operation status parameters and health index, and to calculate the target air flow of each air supply zone.

[0057] The execution control module is used to generate corresponding throttle valve opening commands based on the target air flow rate of the zone, and control the throttle valves or flow guiding mechanisms of the air supply zone to perform opening adjustment;

[0058] The closed-loop update module is used to re-collect the operating data of the fuel cell stack after the control is executed, update the operating status parameters and health index of the zone, and dynamically adjust the target airflow of the zone when the health index or the operating status of the zone deviates from the preset range.

[0059] The beneficial effects of this invention are as follows:

[0060] This invention expands air supply from traditional "overall quantity control" to "regional quantity control" by dividing the air supply into zones according to the flow channel structure of the fuel cell stack and calculating the local oxygen concentration and humidity of each air supply zone based on an air flow model. This method can differentiate the allocation of air volume according to the local needs of different areas, thereby significantly reducing the deviation of oxygen concentration and humidity within the stack and improving the consistency and stability of the fuel cell stack operation.

[0061] This invention constructs a health index for fuel cell stacks by extracting voltage consistency, regional oxygen concentration deviation, regional humidity deviation, and historical operating characteristics. This health index is then used as a weighting correction factor in the target air volume calculation process, enabling the air control strategy to proactively adjust according to changes in the stack's health status, rather than simply being oriented towards the current load. This technology can automatically improve local air supply capacity in the early stages of stack degradation, mitigating regional decline and extending the lifespan of the fuel cell stack.

[0062] This invention establishes zoned operating state parameters through mathematical models such as oxygen consumption rate, water generation rate, local oxygen concentration estimation, and local humidity estimation. This allows air demand to be calculated comprehensively based on air flow characteristics and electrochemical reaction mechanisms, rather than relying on a single sensor. This approach enhances the accuracy of air control judgments, reduces dependence on sensor layout, and improves the responsiveness to changes in air supply demand under complex operating conditions.

[0063] This invention constructs an air demand index that includes oxygen concentration deviation and humidity deviation, and uses a health index to correct this index. The target air volume can meet the reactor reaction requirements while avoiding problems such as excessive compressor operation or insufficient air supply. This method makes the air supply more in line with the actual reaction requirements, which can improve system energy consumption performance, increase hydrogen utilization, and reduce the mechanical load on the air compressor.

[0064] This invention smooths the throttle valve opening by using rate-of-change constraints and a trimming function, preventing sudden local airflow changes from impacting the fuel cell stack. Furthermore, it synchronously adjusts the air compressor speed based on changes in zoned air volume, ensuring the air compressor output flow matches the zoned air demand. This coordinated control improves the response speed and stability of the air supply system, helping to maintain stable stack performance under dynamic load conditions.

[0065] This invention establishes a closed-loop air supply regulation mechanism covering the entire operating cycle by re-collecting operational data, updating partition operating status parameters, updating health indices, and dynamically adjusting the target air volume for each partition when the health status deviates or local air anomalies occur. This mechanism allows the air control strategy to be continuously updated according to the reactor status, possessing self-learning and adaptive characteristics, which helps maintain reactor performance stability during long-term operation and effectively suppresses the spread of local degradation. Attached Figure Description

[0066] Figure 1 This is a schematic diagram of a control method for a fuel cell air system according to the present invention;

[0067] Figure 2 This is a schematic diagram of the control system framework for a fuel cell air system according to the present invention. Detailed Implementation

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] Example 1: As Figure 1 As shown, this embodiment provides a control method for a fuel cell air system, including the following steps:

[0070] S1. Steps for acquiring stack operation data and establishing a partition structure: Acquire real-time operation data of the fuel cell stack during operation, such as single cell voltage, stack temperature, stack pressure difference, air inlet pressure, and air outlet oxygen concentration; and based on the internal flow channel structure of the stack, divide the fuel cell stack into multiple air supply partitions and establish a corresponding partition identifier for each air supply partition.

[0071] S2. Estimation steps for zoned operation status: Based on the real-time operation data and zone identifiers from step S1, local oxygen concentration estimators and local humidity estimators for each air supply zone are constructed using air flow models and electrochemical reaction models; and the zoned operation status parameters for each air supply zone are output to characterize the gas distribution and reaction uniformity in the internal spatial dimension of the reactor.

[0072] S3. Construction steps of the stack health index: Based on the voltage consistency, zoned operating status parameters, stack temperature fluctuation and historical operating data obtained in steps S1 and S2, a health index is constructed to reflect the degradation level of the fuel cell stack; the health index is used to describe the stack's performance maintenance capability and degradation trend throughout its entire life cycle.

[0073] S4. Steps for determining the target air volume for each air supply zone: Based on the zone operation status parameters of each air supply zone obtained in step S2 and the health index obtained in step S3, construct a model for calculating the target air volume for each zone; determine the target air flow rate of each air supply zone according to the current stack load requirements, so that the target air volume of each zone simultaneously meets the overall stack reaction requirements, the stack internal voltage consistency requirements, and the requirements for maintaining the stack health index.

[0074] S5. Zonal Air Supply Execution Steps: Based on the zonal target air volume in step S4, control the opening of the throttle valve or flow guiding mechanism corresponding to each air supply zone, and allocate the corresponding actual air flow to each air supply zone; and maintain the coordination relationship between the total air supply of the air compressor and the zonal air volume during the air allocation process to ensure the overall stability of the air system.

[0075] S6. Dynamic adjustment steps of partition-health coupling: Based on the air supply execution results of step S5, the updated stack operation data is collected in real time, and the partition operation status parameters of step S2 and the health index of step S3 are recalculated; when the health index decreases or the partition operation status deviates from the preset range, the partition target air volume of step S4 is dynamically adjusted, so that the air supply process can automatically correct the air margin allocation strategy in the early stage of fuel cell stack degradation, thereby achieving comprehensive optimization of the uniformity of internal reaction and stack lifetime performance.

[0076] S1 specifically includes the following sub-steps:

[0077] S110. Acquisition of Basic Operating Data: Acquire basic operating data of the fuel cell stack during operation, including: single cell voltage, stack temperature, stack pressure difference, and air inlet pressure. With air outlet pressure Oxygen concentration at the air outlet, etc.

[0078] Among them, air inlet pressure The air outlet pressure is used to characterize the pressure of air before it enters the fuel cell stack. This data is used to characterize the pressure of air flowing through the fuel cell stack towards the outlet. The baseline operating data serves as input for subsequent air supply zoning state estimation.

[0079] S120. Division of air supply zones: Based on the flow channel geometry, flow channel arrangement and local pressure drop variation of the fuel cell stack, the fuel cell stack is divided into multiple air supply zones.

[0080] The air supply zones are divided using a "pressure drop similarity criterion," meaning that adjacent channels i and j that are assigned to the same air supply zone must meet the following conditions:

[0081]

[0082] in This represents the pressure drop of air in flow channel i. This indicates the pressure drop of air in flow channel j. This indicates the preset pressure drop threshold. The above criteria ensure that airflow channels falling into the same air supply zone have consistent air transmission resistance.

[0083] Set a unique zone identifier for each air supply zone. This is used for subsequent data association and model calculation.

[0084] in and The data can be obtained by: using CFD fluid simulation software (such as ANSYS Fluent) to simulate the airflow distribution of a single cell channel during the fuel cell stack design phase, and obtaining the pressure distribution data of each channel under rated operating conditions; or by using fluid dynamics formulas to perform theoretical calculations based on the channel geometry (length, hydraulic diameter). This indicates the preset pressure drop threshold, which is usually set to 5%-10% of the average pressure drop of the flow channel, or set to a specific value (e.g., 50Pa~100Pa) according to the control accuracy requirements, so as to ensure that the flow channels divided into the same zone have similarity in fluid dynamics.

[0085] S130. Establishment of airflow model parameters for each air supply zone; Based on the geometry of each air supply zone and the corresponding gas transport path, establish the airflow model:

[0086]

[0087] in: The pressure drop of air within air supply zone k; Airflow rate for air supply zone k; The drag coefficient of air supply zone k (obtained through CFD simulation or experimental calibration); The linear loss coefficient for air supply zone k (obtained through CFD simulation or experimental calibration).

[0088] drag coefficient With linear loss coefficient All must meet the physical feasibility conditions:

[0089]

[0090] The methods for obtaining model parameters include:

[0091] The airflow within the flow channel was simulated on a CFD simulation platform to obtain the pressure drop-flow rate relationship for each zone. Airflow rate and pressure drop curves were measured on an experimental bench. This could be achieved by blocking the flow channel inlets of non-test zones or by individually ventilating a specific flow channel group in a dummy stack test to collect pressure drop data at different flow rates. Parameters were then obtained through least squares fitting. and .

[0092] S140. Construction of the partition input data set: Integrating basic operational data with air supply partition identifiers. Perform associations to build a structured data set:

[0093]

[0094] in: The unique identifier for the air supply zone; This represents the set of single-cell voltages corresponding to air supply zone k. The set of temperatures corresponding to partition k; This refers to the oxygen concentration component in the outlet oxygen concentration corresponding to zone k. When the fuel cell stack outlet manifold is equipped with multiple oxygen sensors, each corresponding to a different zone location... The values ​​are the actual measured values ​​from the corresponding sensors; when the system is equipped with only a single oxygen sensor at the main outlet, The total outlet oxygen concentration is measured at the initial moment, or the total outlet oxygen concentration is mapped to each zone according to the distance weight between the sensor installation location and the outlet of each zone, and used as the observation input or boundary condition for subsequent state estimation.

[0095] Data set D is used to ensure that the subsequent partition operation status estimation model can directly access various basic operation data.

[0096] S150. Consistency Verification of Partition Structure and Data Set: Perform consistency verification on the structural information of the air supply partition and the partition input data set D. The verification content includes:

[0097] Partition identifier Does it correspond one-to-one with the corresponding flow channel group?

[0098] Model parameters Does it match the geometry of the corresponding air supply zone?

[0099] Does dataset D contain all the data required to construct the partition runtime status parameters?

[0100] Model parameters Does it meet the physical feasibility requirements?

[0101]

[0102] When the verification fails, the system triggers a re-division of the air supply zones or a recalibration of the air flow model parameters at the software logic level of the control system, ensuring that subsequent estimation steps can be performed based on complete, accurate and physically feasible data.

[0103] S2 specifically includes the following sub-steps:

[0104] S210. Preliminary estimation of gas flow distribution within the air supply zone: Based on the data set D obtained in step S150, the zone air flow rate is...

[0105] As input to the airflow model, substitute it into the airflow model equations:

[0106]

[0107] in: : Pressure drop of air within air supply zone k; Air flow rate in air supply zone k; : The drag coefficient of air supply zone k; Linear loss coefficient for air supply zone k;

[0108] Calculate the zone pressure drop for each air supply zone. This constitutes the basic flow distribution, used to further reflect the smoothness of air flow within each zone.

[0109] S220, Calculation of local oxygen consumption and local water generation in air supply zones: based on zone airflow. And the set of single-cell voltages related to the partition Calculate the oxygen consumption rate of the air supply zone.

[0110] With water generation rate .

[0111] The following mathematical expression is used:

[0112] Oxygen consumption rate of air supply zones:

[0113]

[0114] in Oxygen consumption rate of air supply zone k; : Oxygen consumption coefficient of air supply zone k (derived from pile current density and stoichiometry); : Air flow rate of air supply zone k.

[0115] Water generation rate:

[0116]

[0117] in: : Water generation rate in air supply zone k; : Water generation coefficient of air supply zone k (determined based on electrochemical reaction equation); this step yields oxygen consumption and water generation, characterizing the intensity of the local reaction.

[0118] S230. Local oxygen concentration estimation for air supply zones: based on the oxygen consumption rate calculated in step S220. Estimate the local oxygen concentration in the air supply zone. The formula for estimating virtual oxygen concentration is as follows:

[0119]

[0120] in: Local oxygen concentration in air supply zone k; Oxygen concentration at the air inlet (measured from inlet gas parameters); Oxygen consumption rate of air supply zone k; : Airflow rate in air supply zone k. This formula reflects the law that oxygen concentration decreases as oxygen consumption increases, enabling real-time mathematical estimation of local oxygen concentration.

[0121] S240. Local humidity estimation of the air supply zone: based on the water generation rate obtained in step S220. A formula for estimating local humidity in air supply zones is constructed.

[0122] Local humidity of air supply zones It can be represented as:

[0123]

[0124] in Local humidity in air supply zone k; Initial humidity of the air inlet; : Water generation rate in air supply zone k; : Airflow rate of air supply zone k. This formula describes the trend of increasing local humidity within the air supply zone as the amount of water produced by the electrochemical reaction increases.

[0125] Ultimately, the local oxygen concentration will be reduced. With local humidity Together, they serve as the zoning operation status parameters for the air supply zone.

[0126] S250. Consistency verification of partition operation status parameters: Based on the partition operation status parameters in steps S230 and S240, establish consistency verification criteria.

[0127] The verification content includes: physical feasibility verification: Ensure that the estimated values ​​of local oxygen concentration and local humidity are free from physical errors.

[0128] Traffic reasonableness verification: >0; Ensure that air supply zone k is in a state of airflow during actual operation.

[0129] Zone consistency check: Determine whether the local oxygen concentration in all sub-channels within the same air supply zone meets the following requirements: ;in: , : Local oxygen concentration in sub-channels i and j of the air supply zone; δ: Oxygen concentration deviation threshold.

[0130] If any of the above conditions are not met, the parameter update process will be triggered, including adjusting model parameters or recalculating partition running status parameters, to ensure that subsequent health index construction steps have reliable input.

[0131] S3 specifically includes the following sub-steps:

[0132] S310. Extraction of Partition Operational Status Parameter Features: Based on the air supply partition operational status parameters obtained in step S250, mathematical features used to characterize the internal consistency of the fuel cell stack are extracted, including:

[0133] Characteristics of oxygen concentration deviation in different zones:

[0134]

[0135] in: : Standard deviation of oxygen concentration in each air supply zone, used to characterize the uniformity of oxygen concentration distribution; N: Number of air supply zones; Local oxygen concentration in air supply zone k; : The average oxygen concentration across all air supply zones.

[0136] Zonal humidity deviation characteristics:

[0137]

[0138] in: Standard deviation of humidity in air supply zones; Local humidity in air supply zone k; : The average local humidity of all air supply zones.

[0139] Single-cell voltage consistency characteristics:

[0140]

[0141] in: M: Standard deviation of single cell voltage consistency; M: Number of single cells in the fuel cell stack; : The voltage of the i-th single cell; : The average voltage of all individual cells.

[0142] Three deviation features extracted , , Used to reflect the consistency of the heap in spatial dimensions.

[0143] S320. Construction of Historical Operating Data and Degradation-Related Features: Constructing historical operating features related to reactor aging and degradation, including:

[0144] Cumulative running time characteristics This indicates the cumulative operating time of the fuel cell stack from its activation to the current moment.

[0145] Cumulative load impact characteristics :

[0146]

[0147] in: J: Cumulative load impact of the fuel cell stack; J: Number of time points for load sampling; The pile current at time j. : The pile current at time j-1 (or the previous sampling time).

[0148] Historical temperature fluctuation characteristics :

[0149]

[0150] in: K: Standard deviation of stack temperature fluctuation; K: Number of temperature sampling points; : The pile temperature at time m; Historical average reactor temperature. These characteristics reflect the reactor's long-term service life and degradation trends.

[0151] S330. Calculation of degradation parameters: Based on the various features obtained in steps S310 and S320, intermediate degradation parameters representing the degradation rate are constructed in a weighted manner. :

[0152]

[0153] in: Intermediate degradation parameters of fuel cell stacks; : Weighting coefficients, used to reflect the contribution of each feature to the degradation rate (obtained through calibration or training). This step unifies spatial consistency features and historical degradation features into the same mathematical framework, forming a continuous and cumulative degradation measurement variable.

[0154] S340, Construction of the Health Index: Based on the intermediate degradation parameters of step S330 Constructing a health index for fuel cell stacks :

[0155]

[0156] in: Fuel cell stack health index, value range A higher value indicates a higher level of health. : An intermediate degradation parameter reflecting the trend of reactor degradation. This formula maps the degradation parameter to health through an exponential relationship, so that the health index naturally decreases as the degree of degradation increases.

[0157] S350. Health Index Availability Verification and Output: Perform availability verification on the health index HI obtained in step S340, including:

[0158] Range verification: If the range is exceeded, the weighting coefficients will be readjusted. .

[0159] Change smoothness check: ;in:

[0160] Health index from the previous period; Threshold for changes in health index.

[0161] Anomaly Trigger Mechanism: If the health index verification fails, it triggers a recalculation of partition running status parameters or an update of historical characteristics. Upon successful verification, the health index is updated. This serves as input for subsequent optimization of target air volume in different zones.

[0162] S4 specifically includes the following sub-steps:

[0163] S410. Calculation of demand deviation based on zoned operating status parameters: Based on the air supply zoned operating status parameters (oxygen concentration) obtained in step S250. ,humidity The oxygen concentration and humidity deviations of each air supply zone are calculated to measure whether the current air volume meets the reactor reaction requirements.

[0164] Zone oxygen concentration deviation:

[0165]

[0166] in: Oxygen concentration deviation in air supply zone k; Target oxygen concentration (preset by reactor load requirements); : Local oxygen concentration in air supply zone k.

[0167] Zone humidity deviation:

[0168]

[0169] in: Humidity deviation in air supply zone k; Target humidity level (set according to the water storage management strategy); The local humidity of air supply zone k. Oxygen concentration deviation and humidity deviation are used to describe whether the air supply to the zone is too much or too little.

[0170] S420, Correction of the weighting of air demand based on the health index: Based on the health index obtained in step S340 The air flow demand of each air supply zone is weighted and adjusted to achieve dynamic adjustment based on "health priority" or "performance priority".

[0171] Construct a health correction factor:

[0172]

[0173] in: Health correction coefficient for air supply zone k; Health decline influencing factors (obtained through calibration); Fuel cell stack health index.

[0174] Explanation: Health index declined ( (Shrinks) → Increased airflow → The control system is more inclined to increase the air volume in this zone to suppress degradation. High health index → ≈1→ Air volume decision maintains standard strategy.

[0175] S430. Construction of the zoned air demand function: Combining the zoned operating status deviations (oxygen concentration deviation and humidity deviation) with health correction coefficients, a demand function for solving the target air volume is obtained:

[0176]

[0177] in: Air demand index for air supply zone k; Oxygen concentration deviation weight; Humidity deviation weight; Oxygen concentration deviation in air supply zone k; Humidity deviation of air supply zone k.

[0178] Subsequently, health corrections were introduced: ;in: Air demand index adjusted for health factors; Health correction coefficient. This step enables joint modeling of partition status and health status.

[0179] S440, Calculation of Zoned Target Air Volume: Based on air demand index Calculate the target air volume for air supply zone k. .

[0180] The following linear mapping model is adopted:

[0181]

[0182] in: Target airflow rate for air supply zone k; Base airflow (set by the current load on the heap); Airflow sensitivity coefficient; Air demand index after health index correction.

[0183] If the oxygen concentration or humidity in a zone is low → Increase → Increase; if the health index is low → Increase → Increase → Automatically increase air volume to slow down degradation.

[0184] S450, Feasibility constraint of zone target air volume: Sum the target air volumes of all zones and sum them with the maximum air supply volume that the air compressor can currently output. Comparison to ensure the feasibility of meeting air supply requirements:

[0185]

[0186] in: : The maximum output air flow rate of the air compressor under the current operating conditions; N: The number of air supply zones.

[0187] If the above constraints are not met, the target air volume for each zone will be scaled proportionally:

[0188]

[0189] in: : The feasible target airflow for constrained air supply zone k. This step ensures that the target airflow meets both stack performance and air compressor capacity limits.

[0190] S5 specifically includes the following sub-steps:

[0191] S510, Generation of throttle valve / guide mechanism control commands: based on the zone target airflow obtained in step S450. (or after constraint) Using the throttle valve flow model of the air supply zone, a corresponding control opening command is generated for each air supply zone.

[0192] The flow rate model for throttle valves uses the following mathematical relationship:

[0193]

[0194] in: : Actual airflow rate in air supply zone k; : The throttle valve flow coefficient of air supply zone k; : The throttle valve opening of air supply zone k; Pressure drop in air supply zone k (already calculated in S210).

[0195] Solve for the target opening :

[0196]

[0197] in: : Target throttle valve opening for air supply zone k. If constrained airflow is used, then... Alternative .

[0198] S520, Adjustment of the actual opening of the throttle valve / guide mechanism: To avoid airflow fluctuations caused by excessively rapid changes in the throttle valve opening, an opening change limit ("slope constraint") is introduced:

[0199]

[0200] in: : The actual opening degree of the throttle valve in air supply zone k at the current time t; The opening degree of the throttle valve at the previous moment; : Maximum rate of change of throttle valve opening (control smoothness parameter).

[0201] The final openness is:

[0202]

[0203] in: Limit the amount of change to to Interval clipping function.

[0204] S530, Consistency verification of total air supply from air compressor and air volume between zones: Based on the actual air flow rate of all air supply zones. Calculate the current total air consumption:

[0205]

[0206] in: : Total airflow across all air supply zones at the current moment; N: Number of air supply zones. This is then compared with the actual output flow rate of the air compressor. Compare:

[0207] like If so, it means the gas supply capacity meets the demand; if This needs to be compensated by adjusting the air compressor speed.

[0208] The air compressor flow model is as follows:

[0209]

[0210] in: Air compressor output flow rate; Air compressor speed; : Air compressor flow gain coefficient.

[0211] S540, Synchronous adjustment of air compressor speed: When insufficient air supply occurs ( When this occurs, the air supply is compensated by increasing the air compressor speed:

[0212]

[0213] in: : Current air compressor speed; : The air compressor speed at the previous moment; : Air compressor speed adjustment gain (used to control the adjustment range).

[0214] Simultaneously apply safety constraints:

[0215]

[0216] in: The minimum permissible speed of the air compressor; : The maximum permissible speed of the air compressor.

[0217] S550. Generation of air supply execution results: Based on the adjustment results of steps S520 and S540, the following is obtained:

[0218] The throttle valve's final operating opening degree Actual airflow in each air supply zone Actual output flow rate of air compressor The above is taken as the "air supply execution result" and fed back to step S610 (dynamic adjustment step) for closed-loop updating of the partition operation status parameters and health index.

[0219] S6 specifically includes the following sub-steps:

[0220] S610. Update of operational data for air supply execution results: Based on the air supply execution results obtained in step S550, including:

[0221] Throttling valve operating opening ;

[0222] Actual airflow in each air supply zone ;

[0223] Actual output flow rate of air compressor ;

[0224] Reacquire the operating data set of the fuel cell stack. ,in:

[0225]

[0226] in: : The set of single-cell voltages corresponding to air supply zone k; : The set of temperatures corresponding to air supply zone k; , : The current air inlet and outlet pressures; The oxygen concentration component corresponding to air supply zone k at the current moment. (Data set) Used to re-estimate the partition's running status.

[0227] After executing the air conditioning command in step S550, a gas path response waiting time (e.g., 0.5s-2s) needs to be set. Once the pipeline pressure and gas distribution reach a new steady state, the data acquisition in step S610 can be executed. This waiting time is related to the fuel cell stack's inlet pipeline volume and the air compressor's response characteristics, and is used to eliminate the impact of dynamic airflow fluctuations on the accuracy of state estimation.

[0228] S620. Recalculation of partition running status parameters: using the data set updated in step S610. Recalculate the following partition running status parameters:

[0229] Re-estimation of local oxygen concentration:

[0230]

[0231] in: : Local oxygen concentration in air supply zone k at the current moment; Oxygen concentration at the air inlet; Oxygen consumption rate of air supply zone k at the current moment; : Air flow rate of air supply zone k at the current moment.

[0232] Re-estimation of local humidity:

[0233]

[0234] in: : Local humidity of air supply zone k at the current moment; : Air inlet humidity; : Water generation rate in air supply zone k at the current moment; : Air flow rate in air supply zone k at the current moment. These state parameters serve as input for the next health index update.

[0235] S630. Dynamic Update of Health Index: Based on the partition operating status parameters obtained in step S620, recalculate the relevant characteristics of the health index.

[0236] Dynamic updates of oxygen concentration deviation:

[0237]

[0238] in: Standard deviation of oxygen concentration at the current moment; : The average oxygen concentration of all air supply zones at the current moment.

[0239] Dynamic updates of humidity deviation:

[0240]

[0241] in: : Standard deviation of humidity at the current moment; : The average humidity of all zones at the current moment.

[0242] Dynamic updates of single-cell voltage consistency:

[0243]

[0244] in: : Standard deviation of single-cell voltage at the current moment; : The voltage of the i-th single cell at the current moment; : The average value of all single cell voltages at the current moment.

[0245] Update of dynamic intermediate degradation parameters:

[0246]

[0247] All weights and variable definitions remain consistent with those in the previous text.

[0248] Recalculation of health index:

[0249]

[0250] in: : Current health index. This index is used to dynamically adjust the target air volume.

[0251] S640, Triggering judgment of health status deviation: Determine whether the change in health index exceeds the safe range:

[0252]

[0253] in: Threshold for changes in health index; Health index from the previous period.

[0254] When the above conditions are met or any of the following situations occur:

[0255] Oxygen concentration deviation exceeds the standard in any air supply zone:

[0256]

[0257] Humidity deviation exceeds the standard in any air supply zone:

[0258]

[0259] This triggers the process of recalculating the air volume.

[0260] in: The minimum permissible oxygen concentration; : Maximum permissible humidity. To avoid frequent fluctuations in the health index due to sensor noise or instantaneous fluctuations, the trigger determination must meet the "continuity constraint condition": that is, the above trigger condition must be met continuously for N consecutive sampling periods (e.g., 3-5 consecutive periods) before the dynamic adjustment process of S650 can be confirmed.

[0261] S650, Dynamic adjustment of zone target air volume: If the triggering condition is met, the zone target air volume is recalculated.

[0262]

[0263] in: : The target airflow rate for air supply zone k at the current moment; Base airflow rate under current load; Airflow sensitivity coefficient; The current air demand index after health correction. The throttle valve opening is then updated, triggering the next round of the S510–S550 algorithm to form continuous closed-loop control.

[0264] Simultaneously, an adjustment gain saturation limit is set: if the oxygen concentration deviation in the zone is not eliminated or the health index does not recover after M consecutive dynamic adjustments, and the calculated target airflow has reached the physical limit of the flow channel or caused the air compressor to surge, then the increase in air volume will be stopped, and a "zone fault warning" signal will be output to indicate that shutdown maintenance or drainage operations are required to prevent the system energy consumption from running out of control due to excessive air volume requests.

[0265] Example 2: Figure 2 As shown, this embodiment provides a control system for a fuel cell air system, including:

[0266] The data acquisition module is used to collect operating data of the fuel cell stack, including single cell voltage, stack temperature, stack pressure difference, air inlet pressure, air outlet pressure and oxygen concentration information, and associate the operating data with the partition identifier of the air supply partition to form a partition input data set;

[0267] The partition modeling module is used to divide the fuel cell stack into multiple air supply partitions based on the flow channel structure of the fuel cell stack, and to establish corresponding air flow model parameters for each air supply partition. The air flow model parameters include the drag coefficient and linear loss coefficient of the air supply partition, which are used to characterize the relationship between air pressure drop and air flow rate in each air supply partition.

[0268] The zone status estimation module is used to calculate the local oxygen concentration and local humidity of each air supply zone based on the zone input data set and air flow model parameters, and obtain the zone operating status parameters.

[0269] The health index construction module is used to calculate the intermediate degradation parameters of the fuel cell stack based on the partition operation status parameters and historical operation characteristics, and to construct the health index of the fuel cell stack based on the intermediate degradation parameters, which is used to characterize the health status of the fuel cell stack.

[0270] The target air volume calculation module is used to construct the air demand index of each air supply zone based on the zone operation status parameters and health index, calculate the target air flow of each air supply zone, and perform air supply capacity constraint processing on the target air flow to obtain an executable zone target air flow.

[0271] The execution control module is used to generate corresponding throttle valve opening commands based on the target air flow of the zone, and control the throttle valve or flow guide mechanism of the air supply zone to perform opening adjustment; it is also used to adjust the air compressor speed based on the changes in the air flow of the zone, so that the air compressor output air volume matches the air flow of the zone.

[0272] The closed-loop update module is used to re-collect the operating data of the fuel cell stack after the control is executed, update the partition operating status parameters and health index, and dynamically adjust the partition target air flow when the health index or partition operating status deviates from the preset range, so as to realize the closed-loop regulation of the fuel cell air supply.

[0273] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0274] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0275] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A control method for a fuel cell air system, characterized in that, Includes the following steps: S1. Acquire real-time operating data of single cell voltage, stack temperature, stack pressure difference, air inlet pressure, and air outlet oxygen concentration of the fuel cell stack during operation; Based on the internal flow channel structure of the stack, divide the fuel cell stack into multiple air supply zones and establish a corresponding zone identifier for each air supply zone. S2. Based on real-time operating data and partition identification, local oxygen concentration estimators and local humidity estimators for each air supply partition are constructed using air flow models and electrochemical reaction models. The partition operating status parameters of each air supply partition are output to characterize the gas distribution and reaction uniformity in the internal spatial dimension of the reactor. S3. Based on the voltage consistency, zoned operation status parameters, stack temperature fluctuations and historical operation data obtained in steps S1 and S2, construct a health index to reflect the degradation level of the fuel cell stack. Used to describe the heap's performance maintenance capability and degradation trend throughout its entire life cycle; S4. Based on the zone operation status parameters and health index of each air supply zone, construct a model for calculating the target air volume of each zone; Determine the target airflow for each air supply zone based on the stack load requirements, so that the target airflow for each zone simultaneously meets the overall stack reaction requirements, the internal stack voltage consistency requirements, and the stack health index maintenance requirements. S5. Based on the target air volume for each zone, control the opening of the throttle valve or flow guide mechanism corresponding to each air supply zone, and allocate the corresponding actual air flow to each air supply zone. During the air distribution process, maintain a coordinated relationship between the total air supply of the air compressor and the air volume of each zone to ensure the stability of the air system; S6. Based on the air supply execution result of step S5, collect the updated stack operation data in real time, and recalculate the partition operation status parameters and the health index; when the health index decreases or the partition operation status deviates from the preset range, dynamically adjust the partition target air volume so that the air supply process can automatically correct the air margin allocation strategy in the early stage of fuel cell stack degradation. S5 specifically includes: Input the target air volume for each zone, and generate adjustment commands for each throttle valve or flow guide mechanism based on the target value for each air supply zone; According to the adjustment command, control the actual opening degree of the throttle valve or flow guide mechanism of each air supply zone so that each air supply zone obtains the corresponding actual air flow. Monitor the total air supply of the air compressor and compare it with the actual air flow of each air supply zone to ensure the overall air supply coordination of the air system. Based on the comparison results, the air compressor output and throttle valve opening are adjusted synchronously to avoid mismatch between the total air flow and the zone air flow. The actual airflow of each air supply zone is used as the air supply execution result and provided to subsequent dynamic adjustment steps.

2. The control method for a fuel cell air system according to claim 1, characterized in that, S1 specifically includes: The system collects basic operating data on single cell voltage, stack temperature, stack pressure difference, air inlet pressure, and air outlet oxygen concentration during the operation of the fuel cell stack, and provides input conditions for subsequent zoning estimation. Based on the flow channel geometry of the fuel cell stack, the stack is divided into multiple air supply zones, and a unique zone identifier is set for each air supply zone. Establish corresponding airflow model parameters for each air supply zone, including flow channel resistance coefficient, local pressure loss characteristics, and air transmission path; Construct input data sets for each air supply zone, associate the collected basic operational data with the zone identifiers, and form a unified data structure for spatial dimension estimation; The structural information of each air supply zone is verified against the input dataset to ensure that subsequent estimation steps can be processed based on a consistent and accurate zone structure.

3. The control method for a fuel cell air system according to claim 1, characterized in that, S2 specifically includes: Based on a unified data structure, the air flow model is input to preliminarily estimate the gas flow distribution in each air supply zone; The preliminary estimates were input into the regional reaction model related to the electrochemical reaction to calculate the local oxygen consumption and local water generation in each region. A local oxygen concentration estimator is constructed to estimate the oxygen concentration of each air supply zone in real time, thereby obtaining the oxygen concentration distribution of each zone. A local humidity estimator is constructed, which combines local oxygen concentration and water generation to estimate the local humidity distribution of each air supply zone, forming zone operating status parameters, including zone oxygen concentration and zone humidity. The consistency of the zone operation status parameters is verified to ensure that the zone operation status parameters of each air supply zone can be used as the input data source for the subsequent construction of the health index.

4. The control method for a fuel cell air system according to claim 1, characterized in that, S3 specifically includes: Input the operating status parameters of the partition and extract the feature indicators that reflect the local operating consistency, including partition voltage consistency, partition oxygen concentration deviation and partition humidity deviation; By combining historical operating data of the heap, a multi-dimensional set of input features is formed to construct the health index; Input features are fed into the degradation modeling module to fit the performance degradation trend of the heap, and intermediate degradation parameters are obtained to reflect the rate of heap degradation. Based on intermediate degradation parameters, a health index is constructed so that the health index can reflect the current degree of degradation and future degradation trend of fuel cell stacks. The health index is checked to see if it falls within the preset health status range, and the health index is used as the input variable for the subsequent model for calculating the target air volume for each zone.

5. The control method for a fuel cell air system according to claim 1, characterized in that, S4 specifically includes: Input the partition operation status parameters to determine whether the current oxygen concentration and humidity levels of each air supply partition meet the current load requirements of the stack. The comparison results of health index and operating status are input into the air volume optimization module to construct a comprehensive optimization target, including overall reactor reaction requirements, reactor internal consistency requirements, and health index maintenance requirements. Based on the optimization objective, the target air volume for each air supply zone is calculated using an air allocation algorithm, forming a zone air volume allocation list; Perform a feasibility constraint check on the zone air volume allocation list to ensure that the sum of the target air volume for all zones does not exceed the total air volume currently available from the air compressor. The target air volume for each zone that passes the constraint test will be used as the control input for subsequent air supply execution steps.

6. The control method for a fuel cell air system according to claim 1, characterized in that, S6 specifically includes: Based on the air supply execution results, the operating data of the fuel cell stack is re-collected, and the basic operating data set is updated; The updated basic operational data is input into the estimation process to recalculate the partition operational state parameters, so that the spatial dimension state estimation remains real-time. The partition's operating status parameters are input into the health index construction process to obtain the updated health index, ensuring that the health status estimate is dynamically updated as the operation changes.

7. The control method for a fuel cell air system according to claim 6, characterized in that, S6 also includes: comparing the updated health index with the preset health range, and triggering the recalculation of the target air volume for the zone when the health index drops or the zone's operating status deviates from the preset range; Based on the triggering results, the target air volume of the partition is dynamically adjusted so that the air supply process can be adaptively corrected in the early stage of reactor degradation, thereby achieving comprehensive optimization of reactor internal reaction uniformity and reactor lifetime performance.

8. A control system for a fuel cell air system, employing the control method for a fuel cell air system according to any one of claims 1-7, characterized in that, include: The system includes a data acquisition module, a partition modeling module, a partition state estimation module, a health index construction module, a target air volume calculation module, an execution control module, and a closed-loop update module.

Images