Fuel cell system water balance control methods, devices, vehicles and storage media

Abstract

This application relates to a method, apparatus, vehicle, and storage medium for water balance control in a fuel cell system. The method includes: acquiring a high-frequency resistance data sequence within a preset time window of the fuel cell stack, and calculating the rate of change and acceleration of the high-frequency resistance; generating a proton exchange membrane (PEM) water content prediction signal based on the rate of change and acceleration using a preset water content state prediction model; generating a gas path subsystem control command based on the prediction signal using a preset feedforward-feedback control strategy, and adjusting the preset gas path subsystem based on the command to maintain the PEM water content within a target range. Thus, by real-time monitoring of the rate of change of high-frequency resistance with load / time, the future trend of water content in the membrane electrode assembly (MEA) can be predicted, and gas path parameters can be adjusted before the water state in the stack deteriorates to an irreversible state. This solves the problems of delayed response and poor adaptability in water management in existing technologies, thereby achieving precise and stable control of water content.

Description

Technical Field

[0001] This application relates to the field of fuel cell technology, and in particular to a method, device, vehicle, and storage medium for water balance control of a fuel cell system. Background Technology

[0002] Proton exchange membrane fuel cells (PEMFCs), as efficient and clean energy conversion devices, show broad application prospects in transportation and stationary power generation. Their power generation performance and lifespan depend crucially on the precise management of the water content in the membrane electrode assembly (MEA): the proton exchange membrane must be kept sufficiently wet to achieve efficient proton conduction, but at the same time, it cannot be over-wetted to prevent liquid water from blocking the gas channels of the porous electrode. Maintaining this water balance is a key challenge in achieving high efficiency and long-term stable operation of fuel cells.

[0003] In related technologies, most mainstream water management solutions are passive response control based on instantaneous threshold values. This means that parameters such as the high-frequency resistance or voltage of the fuel cell are monitored in real time and compared with preset fixed thresholds. For example, when the high-frequency resistance is higher than a certain threshold, the membrane is determined to be dry, and the system starts humidification or lowers the operating temperature. When the high-frequency resistance is lower than another threshold and accompanied by voltage fluctuations, it is determined to be flooded, and the system alleviates the flooding by increasing airflow or purging.

[0004] However, this method relies entirely on adjusting when the monitored value exceeds a fixed threshold, at which point performance loss has already occurred, resulting in a response lag. Furthermore, the fixed threshold cannot adapt to dynamic load changes and state changes in the fuel cell, exhibiting poor adaptability, which urgently needs to be addressed. Summary of the Invention

[0005] This application provides a water balance control method, device, vehicle, and storage medium for a fuel cell system to solve problems such as delayed response and poor adaptability in water management in the prior art, thereby achieving precise and stable control of water content.

[0006] To achieve the above objectives, the first aspect of this application proposes a water balance control method for a fuel cell system, comprising the following steps: The high-frequency resistance data sequence of the fuel cell stack within a preset time window is obtained, and the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance are calculated based on the high-frequency resistance data sequence. Based on the rate of change and acceleration of the high-frequency resistance, a water content state prediction signal for the proton exchange membrane in the fuel cell stack is generated using a preset water content state prediction model. Based on the water content state prediction signal, a preset feedforward-feedback control strategy is used to generate a control command for the gas path subsystem in the fuel cell system. Based on the control command, the preset gas path subsystem is adjusted so that the water content of the proton exchange membrane is maintained within the target range.

[0007] According to one embodiment of this application, the step of generating a water content prediction signal for the proton exchange membrane in the fuel cell stack based on the rate of change and acceleration of the high-frequency resistance using a preset water content prediction model includes: Based on the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is greater than a first preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between a second preset threshold and a first preset safety margin. If the rate of change of the high-frequency resistance is greater than the first preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between the second preset threshold and the first preset safety margin, then a water content state prediction signal indicating that the proton exchange membrane is tending to dry out is generated.

[0008] According to one embodiment of this application, the step of generating a water content prediction signal for the proton exchange membrane in the fuel cell stack based on the rate of change and acceleration of the high-frequency resistance using a preset water content prediction model includes: Based on the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is less than or equal to a third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin. If the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin, then it is determined whether the single cell voltage fluctuation variance of the fuel cell stack is greater than the preset voltage fluctuation variance threshold, whether the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference threshold, and whether the rate of change of the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference rate of change threshold. If the single-cell voltage fluctuation variance of the fuel cell stack is greater than the preset voltage fluctuation variance threshold, and / or the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference threshold, and / or the cathode voltage difference change rate of the fuel cell stack is greater than the preset cathode voltage difference change rate threshold, then it is determined that the cathode of the fuel cell stack has a preset flooding risk, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

[0009] According to one embodiment of this application, the step of generating a water content prediction signal for the proton exchange membrane in the fuel cell stack based on the rate of change and acceleration of the high-frequency resistance using a preset water content prediction model includes: Based on the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is less than or equal to a third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin. If the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin, then it is determined whether the target single cell voltage of the fuel cell stack is less than a preset voltage threshold, whether the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure threshold, and whether the rate of change of the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure change rate threshold. If the target single cell voltage of the fuel cell stack is less than the preset voltage threshold, and / or the anode differential pressure of the fuel cell stack is greater than the preset anode differential pressure threshold, and / or the anode differential pressure change rate of the fuel cell stack is greater than the preset anode differential pressure change rate threshold, then it is determined that the anode of the fuel cell stack has a preset flooding risk, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

[0010] According to one embodiment of this application, the water content state prediction signal indicates that the proton exchange membrane is tending to dry out, and the step of generating control commands for the gas path subsystem of the fuel cell system based on the water content state prediction signal using a preset feedforward-feedback control strategy includes: Based on the moisture content state prediction signal, a first instruction is generated to increase the outlet humidity of the humidifier in the gas path subsystem using a preset feedforward-feedback control strategy, and / or a second instruction is generated to reduce the air metering ratio of the cathode of the fuel cell stack.

[0011] According to one embodiment of this application, the water content state prediction signal indicates that the proton exchange membrane tends to be over-wet, and the step of generating control commands for the gas path subsystem of the fuel cell system based on the water content state prediction signal using a preset feedforward-feedback control strategy includes: If it is determined that there is a preset risk of water flooding at the cathode of the fuel cell stack, a third command is generated to increase the speed of the air compressor of the gas circuit subsystem, and / or a fourth command is generated to reduce the inlet humidity of the cathode of the fuel cell stack, based on the water content state prediction signal and using a preset feedforward-feedback control strategy. If it is determined that there is a preset risk of flooding at the anode of the fuel cell stack, a fifth instruction is generated based on the water content state prediction signal and using a preset feedforward-feedback control strategy to increase the purging frequency and purging duration of the anode of the fuel cell stack, and / or a sixth instruction is generated to increase the speed of the hydrogen circulation pump of the gas circuit subsystem.

[0012] According to the fuel cell system water balance control method proposed in this application, a high-frequency resistance data sequence within a preset time window of the fuel cell stack is obtained, and the rate of change and acceleration of change of the high-frequency resistance are calculated accordingly. Based on the rate of change and acceleration of change, a predicted signal of the proton exchange membrane water content is generated using a preset water content state prediction model. According to the predicted signal, a preset feedforward-feedback control strategy is used to generate a gas path subsystem control command, and the preset gas path subsystem is adjusted based on the command to maintain the proton exchange membrane water content within the target range. Thus, by monitoring the rate of change of high-frequency resistance with load / time in real time, the future trend of water content in the membrane electrode is predicted, and the gas path parameters are adjusted before the water state in the stack deteriorates to an irreversible state. This solves the problems of lag in water management response and poor adaptability in the prior art, thereby achieving precise and stable control of water content.

[0013] To achieve the above objectives, a second aspect of this application provides a fuel cell system water balance control device, comprising: The calculation module is used to acquire the high-frequency resistance data sequence of the fuel cell stack within a preset time window, and to calculate the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance based on the high-frequency resistance data sequence. The generation module is used to generate a water content state prediction signal of the proton exchange membrane in the fuel cell stack based on the rate of change and acceleration of the change of the high-frequency resistance and using a preset water content state prediction model. The adjustment module is used to generate a control command for the gas path subsystem in the fuel cell system based on the water content state prediction signal and a preset feedforward-feedback control strategy, and adjust the preset gas path subsystem based on the control command so that the water content of the proton exchange membrane is maintained within the target range.

[0014] According to one embodiment of this application, the generation module is specifically used for: Based on the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is greater than a first preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between a second preset threshold and a first preset safety margin. If the rate of change of the high-frequency resistance is greater than the first preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between the second preset threshold and the first preset safety margin, then a water content state prediction signal indicating that the proton exchange membrane is tending to dry out is generated.

[0015] According to one embodiment of this application, the generation module is specifically used for: Based on the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is less than or equal to a third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin. If the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin, then it is determined whether the single cell voltage fluctuation variance of the fuel cell stack is greater than the preset voltage fluctuation variance threshold, whether the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference threshold, and whether the rate of change of the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference rate of change threshold. If the single-cell voltage fluctuation variance of the fuel cell stack is greater than the preset voltage fluctuation variance threshold, and / or the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference threshold, and / or the cathode voltage difference change rate of the fuel cell stack is greater than the preset cathode voltage difference change rate threshold, then it is determined that the cathode of the fuel cell stack has a preset flooding risk, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

[0016] According to one embodiment of this application, the generation module is specifically used for: Based on the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is less than or equal to a third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin. If the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin, then it is determined whether the target single cell voltage of the fuel cell stack is less than a preset voltage threshold, whether the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure threshold, and whether the rate of change of the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure change rate threshold. If the target single cell voltage of the fuel cell stack is less than the preset voltage threshold, and / or the anode differential pressure of the fuel cell stack is greater than the preset anode differential pressure threshold, and / or the anode differential pressure change rate of the fuel cell stack is greater than the preset anode differential pressure change rate threshold, then it is determined that the anode of the fuel cell stack has a preset flooding risk, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

[0017] According to one embodiment of this application, the water content state prediction signal indicates that the proton exchange membrane is tending to dry out, and the adjustment module is specifically used for: Based on the moisture content state prediction signal, a first instruction is generated to increase the outlet humidity of the humidifier in the gas path subsystem using a preset feedforward-feedback control strategy, and / or a second instruction is generated to reduce the air metering ratio of the cathode of the fuel cell stack.

[0018] According to one embodiment of this application, the moisture content state prediction signal indicates that the proton exchange membrane tends to be over-wet, and the adjustment module is specifically used for: If it is determined that there is a preset risk of water flooding at the cathode of the fuel cell stack, a third command is generated to increase the speed of the air compressor of the gas circuit subsystem, and / or a fourth command is generated to reduce the inlet humidity of the cathode of the fuel cell stack, based on the water content state prediction signal and using a preset feedforward-feedback control strategy. If it is determined that there is a preset risk of flooding at the anode of the fuel cell stack, a fifth instruction is generated based on the water content state prediction signal and using a preset feedforward-feedback control strategy to increase the purging frequency and purging duration of the anode of the fuel cell stack, and / or a sixth instruction is generated to increase the speed of the hydrogen circulation pump of the gas circuit subsystem.

[0019] According to the fuel cell system water balance control device proposed in this application, a high-frequency resistance data sequence within a preset time window of the fuel cell stack is acquired, and the rate of change and acceleration of change of the high-frequency resistance are calculated accordingly. Based on the rate of change and acceleration of change, a proton exchange membrane water content prediction signal is generated using a preset water content state prediction model. According to the prediction signal, a gas path subsystem control command is generated using a preset feedforward-feedback control strategy. Based on the command, the preset gas path subsystem is adjusted to maintain the proton exchange membrane water content within the target range. Thus, by real-time monitoring of the rate of change of high-frequency resistance with load / time, the future trend of water content in the membrane electrode is predicted, and the gas path parameters are adjusted before the water state in the stack deteriorates to an irreversible state. This solves the problems of delayed response and poor adaptability in water management in the prior art, thereby achieving precise and stable control of water content.

[0020] To achieve the above objectives, a third aspect of this application provides a vehicle comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the fuel cell system water balance control method as described in the above embodiments.

[0021] To achieve the above objectives, a fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the fuel cell system water balance control method as described in the above embodiments.

[0022] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0023] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a water balance control method for a fuel cell system according to an embodiment of this application; Figure 2 This is a block diagram of a fuel cell system water balance control device provided according to an embodiment of this application; Figure 3 This is a structural schematic diagram of a vehicle provided according to an embodiment of this application. Detailed Implementation

[0024] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0025] The following describes, with reference to the accompanying drawings, a method, apparatus, vehicle, and storage medium for controlling the water balance of a fuel cell system according to embodiments of this application. First, the method for controlling the water balance of a fuel cell system according to embodiments of this application will be described with reference to the accompanying drawings.

[0026] Figure 1 This is a flowchart of a fuel cell system water balance control method according to an embodiment of this application.

[0027] For example, such as Figure 1 As shown, the water balance control method for this fuel cell system includes the following steps: In step S101, the high-frequency resistance data sequence of the fuel cell stack within a preset time window is obtained, and the rate of change of high-frequency resistance and the acceleration of change of high-frequency resistance are calculated based on the high-frequency resistance data sequence.

[0028] Specifically, the system uses a sensing unit (i.e., a sensor) integrated into the controller to continuously acquire the high-frequency resistance (HFR) of the fuel cell stack at a fixed sampling period (e.g., 100ms). HFR Parameters such as load current I, voltage V, cathode inlet and outlet pressure difference ΔP, inlet / outlet relative humidity RH, flow rate, air excess ratio, air compressor speed, hydrogen circulation pump speed, and temperature, etc., are all high-frequency resistance values ​​collected in chronological order within a preset time window (e.g., the past 2 seconds including the current moment, corresponding to 20 data points). These parameters together constitute the high-frequency resistance data sequence used for subsequent analysis of the water content change trend of the proton exchange membrane electrode. The system, through a processing unit (such as an embedded MCU), uses the least squares method to fit the optimal straight line of resistance change over time within the preset time window based on the high-frequency resistance data sequence. The slope of this straight line represents the rate of change of the high-frequency resistance. This rate of change quantifies the speed and direction of change in the water content of the proton exchange membrane electrode (positive value indicates drying, negative value indicates wetting). To further assess the urgency of the trend, the system can perform a linear regression on the series of continuous rate of change values ​​calculated above. The slope obtained from this regression is the acceleration of the high-frequency resistance change. This acceleration can reveal whether the process of drying or wetting the proton exchange membrane is accelerating or decelerating, thus providing a key criterion for the confidence level of the early warning.

[0029] It should be noted that, to ensure the accuracy and reliability of subsequent trend predictions, the system can first perform preprocessing operations such as filtering, noise reduction, and calibration on the raw sensor data read from the sensing unit through the processing unit to generate stable and reliable high-quality data. Only after obtaining this high-quality data will the system call algorithms (such as linear regression) to calculate the rate of change (first derivative) and acceleration (second derivative) of the high-frequency resistance.

[0030] In step S102, based on the rate of change and acceleration of the high-frequency resistance, a water content state prediction signal for the proton exchange membrane in the fuel cell stack is generated using a preset water content state prediction model.

[0031] Specifically, after obtaining the rate of change and acceleration of the high-frequency resistance, a pre-set water content state prediction model can be used to output a water content state prediction signal for the proton exchange membrane based on the trend characteristics of the high-frequency resistance (i.e., the rate of change and acceleration of the high-frequency resistance).

[0032] To facilitate understanding, the following details how to generate a water content prediction signal for the proton exchange membrane in a fuel cell stack based on the rate of change and acceleration of the high-frequency resistance, using a pre-defined water content state prediction model.

[0033] In one possible implementation, in some embodiments, a water content state prediction signal for the proton exchange membrane in the fuel cell stack is generated using a preset water content state prediction model based on the rate of change and acceleration of the high-frequency resistance. This includes: determining whether the rate of change of the high-frequency resistance is greater than a first preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between a second preset threshold and a first preset safety margin, based on the rate of change and acceleration of the high-frequency resistance. If the rate of change of the high-frequency resistance is greater than the first preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between the second preset threshold and the first preset safety margin, then a water content state prediction signal indicating that the proton exchange membrane is tending to dry out is generated.

[0034] It should be noted that the preset water content prediction model here refers to a mathematical model based on preset rules and thresholds. After receiving the rate of change and acceleration of high-frequency resistance, as well as other pre-processed sensor data, it can predict the future trend of water content changes in the proton exchange membrane electrode based on its built-in judgment logic based on electrochemical and fluid dynamics principles. The output of this model is a warning signal representing a specific risk (such as a drought warning or a flood warning), rather than a precise water content value.

[0035] Specifically, based on the rate of change of high-frequency resistance and the acceleration of high frequency resistance change It can perform dynamic trend judgment, that is, judge the rate of change of high-frequency resistance. Is it greater than the first preset threshold? , where the first preset threshold A positive threshold is used to filter noise interference. This value is not fixed but determined based on the signal-to-noise ratio of the high-frequency resistance measurement signal. It is set by analyzing the standard deviation of the rate of change of the high-frequency resistance of the fuel cell stack under stable operating conditions. This ensures that the system only responds to significant, non-noise-induced drying trends. Furthermore, a safety margin judgment can be performed, i.e., judging the current high-frequency resistance value in the high-frequency resistance data sequence. Is it greater than the second preset threshold? With the first preset safety margin The difference between them, where the first preset safety margin This represents the high-frequency resistance changing from the warning value (i.e., within the system response delay time Δt) during the system response time delay Δt. The change has reached a dangerous threshold requiring emergency intervention. The estimated change (which can be dynamically adjusted according to operating conditions) can be dynamically adjusted based on the response speed of the gas circuit subsystem in the fuel cell system and the severity of load changes, so as to ensure that the warning time of the control system is earlier than the time required for the water state of the proton exchange membrane electrode to deteriorate to a dangerous level.

[0036] When both of the above conditions are met, that is, (This indicates that the proton exchange membrane is continuously drying out), and (This indicates that the proton exchange membrane is no longer wet and is approaching the danger zone, but has not yet reached the threshold requiring emergency intervention.) The preset water content state prediction model can conclude that the proton exchange membrane electrode not only exhibits clear drying kinetic characteristics, but its current state has also entered the warning zone before danger. Therefore, the system can generate a water content state prediction signal indicating that the proton exchange membrane is tending to dry out, thereby achieving early warning and the purpose of pre-intervention.

[0037] It should be noted that, even when both of the above conditions are met, if the acceleration of the high-frequency resistance change is also determined... This indicates that the desiccation of the proton exchange membrane is accelerating, further enhancing the reliability of the prediction.

[0038] In one possible implementation, in other embodiments, a water content state prediction signal for the proton exchange membrane in the fuel cell stack is generated using a preset water content state prediction model based on the rate of change and acceleration of the high-frequency resistance. This includes: determining whether the rate of change of the high-frequency resistance is less than or equal to a third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin; if the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin. The sum of the values ​​between the values ​​determines whether the single-cell voltage fluctuation variance of the fuel cell stack is greater than a preset voltage fluctuation variance threshold, whether the cathode voltage difference of the fuel cell stack is greater than a preset cathode voltage difference threshold, and whether the cathode voltage difference change rate of the fuel cell stack is greater than a preset cathode voltage difference change rate threshold. If the single-cell voltage fluctuation variance of the fuel cell stack is greater than the preset voltage fluctuation variance threshold, and / or the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference threshold, and / or the cathode voltage difference change rate of the fuel cell stack is greater than the preset cathode voltage difference change rate threshold, then it is determined that the cathode of the fuel cell stack has a preset flooding risk, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

[0039] Specifically, in order to accurately identify the early risk of cathode flooding in a fuel cell stack, the system can first perform a basic state assessment, namely, assessing the rate of change of high-frequency resistance. Is it less than or equal to the third preset threshold? (Same as the first preset threshold), and also perform a safety margin judgment, that is, determine whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the fourth preset threshold. With the second preset safety margin The sum between, where the fourth preset threshold The second preset safety margin is the reference value for excessive humidity. It is a positive value. If ≤ Established, due to The positive threshold already covers the rate of change of high-frequency resistance. In the case of ≤0 (i.e., the high-frequency resistance remains unchanged or continues to decrease), this negative rate of change indicates that the water content of the proton exchange membrane electrode has not decreased, but rather remains at a high level or continues to increase, which is the basic trend of water flooding; and The establishment of this condition indicates that the water content of the proton exchange membrane electrode has entered the excessively wet range, deviating from the ideal operating conditions, but has not yet reached the level that would cause serious performance degradation.

[0040] If both of the above conditions are met, the system can assume that the fuel cell stack has reached the basic state of being flooded. Further targeted judgments will then be made to avoid misjudgment when the proton exchange membrane is in a normal humid state.

[0041] Specifically, determining the variance of single-cell voltage fluctuation in a fuel cell stack. Is it greater than the preset voltage fluctuation variance threshold? Among them, the preset voltage fluctuation variance threshold This value can be used to determine the severity of voltage fluctuations. It can be determined based on a baseline value for voltage fluctuations in a healthy state of the fuel cell stack and a preset safety margin. If... If this condition is confirmed, it indicates that the uneven transport of reactants caused by flooding leads to drastic fluctuations in the performance of some individual cells. This is direct evidence of performance perturbation; the cathode flow channel is randomly and unevenly blocked by liquid water, resulting in significant differences in the amount of reactant gas obtained by different individual cells, thus causing drastic voltage fluctuations throughout the fuel cell stack. It can also be used to determine the cathode voltage difference in the fuel cell stack. Is it greater than the preset cathode differential pressure threshold? Among them, cathode pressure difference It is the difference between the cathode inlet pressure and the cathode outlet pressure, and it is a positive value. The target values ​​for cathode voltage difference under different power levels. This is the cathode differential pressure fluctuation threshold, which can be set based on the cathode differential pressure fluctuation of the fuel cell stack under normal operating conditions through continuous monitoring and dynamic adjustment mechanisms. If... If this condition is met, it indicates that the cathode pressure difference is higher than the normal range under the same operating conditions, which is evidence of static blockage in the flow channel. Accumulated liquid water occupies the space in the gas flow channel, leading to increased fluid resistance, manifested as an inlet and outlet pressure difference that is abnormally higher than the normal target value under this operating condition. Furthermore, the rate of change of the cathode pressure difference in the fuel cell stack can also be determined. Is it greater than the preset threshold for the rate of change of cathode differential pressure? Among them, the preset threshold for the rate of change of cathode pressure difference It can be used to determine the upward trend of the cathode pressure difference. This value can be dynamically determined based on the measurement noise level of the cathode pressure sensor and the pressure difference fluctuation range of the fuel cell stack under normal operating conditions. If If the cathode pressure difference continues to rise, it indicates that liquid water is accumulating in the gas flow channel, causing blockage. This is a very direct characteristic of flooding.

[0042] Among them, the single-cell voltage fluctuation variance of the fuel cell stack The calculation method is as follows:

[0043] in, For single cell voltage, The average voltage of the fuel cell stack. This represents the number of cells in the fuel cell stack.

[0044] Cathode voltage drop rate of fuel cell stack The calculation method is as follows:

[0045] in, The pressure difference between the cathode inlet and outlet is... Sampling time, For time intervals.

[0046] If at least one of the above three conditions is met, and combined with the two conditions that are simultaneously met in the basic over-wet state judgment, the preset moisture content state prediction model can conclude that the cathode of the battery stack is at risk of flooding (entering the warning zone before danger). Therefore, the system can generate a moisture content state prediction signal indicating that the proton exchange membrane is tending to be over-wet, thereby achieving early warning and the purpose of pre-intervention.

[0047] Therefore, a gradual decrease in the high-frequency resistance value, trending towards a low steady state, usually indicates that the proton exchange membrane is in an over-wetted state. Simultaneously, if significant voltage fluctuations are detected, it suggests that the electrochemical reaction process may be hindered. Furthermore, if this is accompanied by a continuous increase in the cathode-side voltage difference, it often indicates a risk of blockage in the fluid channels. When multiple conditions occur simultaneously, it can be determined that the cathode portion of the fuel cell stack has a potential risk of flooding, requiring timely and appropriate control measures.

[0048] In one possible implementation, in other embodiments, a water content state prediction signal for the proton exchange membrane in the fuel cell stack is generated using a preset water content state prediction model based on the rate of change and acceleration of the high-frequency resistance. This includes: determining whether the rate of change of the high-frequency resistance is less than or equal to a third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin; if the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin; The sum of two preset safety margins is used to determine whether the target single cell voltage of the fuel cell stack is less than a preset voltage threshold, whether the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure threshold, and whether the rate of change of the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure rate threshold. If the target single cell voltage of the fuel cell stack is less than the preset voltage threshold, and / or the anode differential pressure of the fuel cell stack is greater than the preset anode differential pressure threshold, and / or the rate of change of the anode differential pressure of the fuel cell stack is greater than the preset anode differential pressure rate threshold, then it is determined that the anode of the fuel cell stack has a preset risk of flooding, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

[0049] Specifically, in order to accurately identify the early risk of anode flooding in a fuel cell stack, the system can also first perform a basic state assessment, namely, assessing the rate of change of high-frequency resistance. Is it less than or equal to the third preset threshold? Furthermore, it is determined whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the fourth preset threshold. With the second preset safety margin The sum of the two conditions. When both conditions are met, the system can consider the fuel cell stack to be in a basic over-wet state that could lead to flooding. Further targeted judgments will then be made to avoid misjudgment when the proton exchange membrane is in a normal wet state.

[0050] Specifically, determining the target single-cell voltage of the fuel cell stack. Is it less than the preset voltage threshold? Among them, the target single cell voltage This is the minimum voltage among all individual cells in the fuel cell stack. This is the minimum limit for the voltage of a single cell. It is a positive value. To approximate the minimum limit range of a single cell voltage, the setting can be adjusted based on the actual conditions of the fuel cell stack. If The presence of a positive voltage reading indicates that the minimum single-cell voltage in the fuel cell stack is close to the set minimum limit. When the anode channel or diffusion layer is blocked by water, the uneven transport of reactants due to flooding leads to insufficient hydrogen supply in certain areas, causing a significant drop in the single-cell voltage in those regions. The positive voltage reading can also be used to determine the anode pressure difference in the fuel cell stack. Is it greater than the preset anode differential pressure threshold? ,in, The target values ​​for different power levels corresponding to the anode pressure difference, and the anode pressure difference fluctuation threshold. The value can be dynamically determined based on the differential pressure fluctuation range of the fuel cell stack under normal operating conditions. If If this is true, it indicates that the pressure difference between the anode inlet and outlet is higher than the target value under this operating condition. A continuous increase in the anode pressure difference suggests that liquid water is accumulating in the gas flow channel, causing blockage and increasing the resistance to hydrogen flow. This manifests as an abnormally high anode inlet and outlet pressure difference compared to the normal target value under this operating condition. Furthermore, it can also be used to determine the rate of change of the anode pressure difference in the fuel cell stack. Is it greater than the preset threshold for the rate of change of anode pressure? Among them, the threshold of the rate of change of anode pressure difference It can be used to determine the upward trend of the anode pressure differential. This value can be dynamically determined based on the measurement noise level of the anode pressure sensor and the pressure differential fluctuation range of the fuel cell stack under normal operating conditions. If The establishment of this condition indicates that the anode pressure differential is rising rapidly and continuously, liquid water is accumulating at an accelerated rate, and the blockage is worsening, which is a very direct characteristic of flooding.

[0051] If at least one of the above three conditions is met, and combined with the two conditions that are simultaneously met in the basic over-wet state judgment, the preset moisture content state prediction model can conclude that the anode of the battery stack is at risk of flooding (entering the warning zone before danger). Therefore, the system can generate a moisture content state prediction signal indicating that the proton exchange membrane is tending to be over-wet, thereby achieving early warning and the purpose of pre-intervention.

[0052] Therefore, when the high-frequency resistance gradually approaches a low steady-state, which is typically characterized by high surface humidity of the proton exchange membrane; simultaneously, if the minimum single-cell voltage drops rapidly and gradually approaches the system's preset danger threshold, and the anode voltage difference begins to rise, it indicates that there may be partial blockage in the flow channel. Considering these multiple phenomena, the system can determine that the anode of the fuel cell stack is at risk of flooding, and appropriate measures need to be taken promptly to avoid potential performance degradation or damage.

[0053] In step S103, based on the water content state prediction signal, a preset feedforward-feedback control strategy is used to generate a control command for the gas path subsystem in the fuel cell system. Based on the control command, the preset gas path subsystem is adjusted so that the water content of the proton exchange membrane is maintained within the target range.

[0054] In other words, based on real-time monitoring and state prediction signals of the current water content of the proton exchange membrane within the fuel cell system (i.e., water content state prediction signal), a comprehensive control strategy combining preset feedforward and feedback control (i.e., preset feedforward-feedback control strategy) can generate precise control commands for the gas path subsystem. This feedforward control module can respond in advance to dynamic changes in the system, while the feedback control dynamically corrects the deviation between the actual output and the target value, effectively improving control response speed and stability. Based on the generated control commands, the system adjusts key parameters of the gas path subsystem in real time to ensure that the water content of the proton exchange membrane remains within the preset target range, thereby guaranteeing the efficient, stable, and safe operation of the fuel cell.

[0055] As one possible implementation, in some embodiments, the water content state prediction signal indicates that the proton exchange membrane is tending to dry out. Based on the water content state prediction signal, a preset feedforward-feedback control strategy is used to generate control instructions for the gas path subsystem in the fuel cell system. This includes: based on the water content state prediction signal, using the preset feedforward-feedback control strategy, generating a first instruction to increase the outlet humidity of the humidifier in the gas path subsystem, and / or generating a second instruction to reduce the air metering ratio of the cathode in the fuel cell stack.

[0056] Specifically, upon receiving a desiccation warning signal (i.e., a moisture content prediction signal indicating that the proton exchange membrane is approaching desiccation), the system controller can generate a first instruction to increase the outlet humidity of the humidifier in the gas path subsystem through a feedforward control strategy. Based on this first instruction, the humidifier's humidification capacity can be actively increased. If the system does not have a humidifier, the controller can automatically skip the adjustment operation for the cathode inlet humidity, avoiding ineffective control. Furthermore, the controller can also generate a second instruction to reduce the air metering ratio at the cathode of the fuel cell stack, based on actual conditions. This second instruction can moderately reduce the cathode air metering ratio to reduce the purging effect of excessive dry air on the cell interior, thereby mitigating the risk of water loss from the membrane electrode assembly.

[0057] It should be noted that while performing the above operations, the system will also continuously monitor the high-frequency resistance of the fuel cell stack and its rate of change in real time. When the rate of change of the high-frequency resistance... Less than or equal to the first preset threshold And the current high-frequency resistance value Less than or equal to If this state can be stably maintained for a period of time t, the system can determine that the risk of desiccation has been largely eliminated. Subsequently, the controller will gradually exit feedforward compensation control and rely mainly on feedback control loops to achieve more precise and stable humidity regulation. The time parameter t can be flexibly set according to the specific operating characteristics of the fuel cell stack, historical operating data, and actual operating conditions to ensure the adaptability and reliability of the control.

[0058] The calculation for reducing the cathode air metering ratio control is as follows: Feedforward term: ; Feedback items: .

[0059] in, For air metering ratio, , and To control the gain coefficient, its specific value can be comprehensively debugged and tuned according to the working characteristics, operating environment and control objectives of the fuel cell stack system in actual application, so as to ensure that the system response performance and control accuracy meet the design requirements.

[0060] feedforward term Its core predictive function can be reflected in the dynamic response to the rate of rise of the high-frequency resistance. The faster the high-frequency resistance rises (i.e., the faster it rises...), the more predictive power is applied to the high-frequency resistance. The larger the value of the air supply, the greater the amount of air that the controller can adjust to reduce the air supply (i.e., the negative adjustment amount of the adjustment). The more [the system], the better, thus enabling direct and rapid compensation for disturbances in the fuel cell system, effectively improving the timeliness and accuracy of control.

[0061] Feedback items Important parameters used for fine-tuning, among which, The deviation between the actual and target values ​​of the high-frequency resistance is calculated. By calculating this deviation, the high-frequency resistance of the fuel cell stack can be continuously monitored and dynamically adjusted, thereby ensuring that the high-frequency resistance remains stable within an ideal range to improve overall performance and operating efficiency.

[0062] The calculation for increasing the humidification capacity of a humidifier is as follows: Feedforward term: ; Feedback items: .

[0063] in, The difference between the next set value for adjusting the humidity of the air at the humidifier outlet and the actual value detected in real time.

[0064] feedforward term The humidity compensation amount can be provided based on the severity of the dryness trend (i.e., the amount by which the set value of the humidifier outlet air humidity needs to be increased from the current value). The larger the value, the greater the compensation.

[0065] Feedback items Fine-tuning can be performed based on the deviation between the actual value and the target value of the high-frequency resistor to ensure that the high-frequency resistor is always stable within an ideal range.

[0066] In one possible implementation, in other embodiments, the water content state prediction signal indicates that the proton exchange membrane tends to be too wet. Based on the water content state prediction signal, a preset feedforward-feedback control strategy is used to generate control instructions for the gas path subsystem in the fuel cell system. These instructions include: if a preset flooding risk is determined to exist at the cathode of the fuel cell stack, a third instruction is generated based on the water content state prediction signal and the preset feedforward-feedback control strategy to increase the speed of the air compressor in the gas path subsystem, and / or a fourth instruction is generated to reduce the inlet humidity of the cathode of the fuel cell stack; if a preset flooding risk is determined to exist at the anode of the fuel cell stack, a fifth instruction is generated based on the water content state prediction signal and the preset feedforward-feedback control strategy to increase the purging frequency and purging duration of the anode in the fuel cell stack, and / or a sixth instruction is generated to increase the speed of the hydrogen circulation pump in the gas path subsystem.

[0067] Specifically, upon receiving a cathode flooding warning signal (i.e., a moisture content prediction signal indicating that the proton exchange membrane is becoming too wet and there is a risk of cathode flooding), the controller can generate a third command to increase the speed of the air compressor in the air path subsystem via feedforward control. Based on this third command, the operating speed of the air compressor can be increased to increase the airflow supplied to the cathode. Furthermore, if the system is equipped with a humidifier, the controller can also generate a fourth command to reduce the inlet humidity of the cathode in the fuel cell stack. Based on this fourth command, the humidity level at the cathode inlet can be lowered. If the system does not have a humidifier installed, no adjustment operation will be performed on the cathode inlet humidity.

[0068] It should be noted that while performing the above operations, the system will continuously monitor the voltage fluctuations of the fuel cell stack and the changes in the differential pressure within the cathode flow channel to determine whether they are gradually stabilizing. This will occur when the following conditions are simultaneously met: ,and ,and ,and Furthermore, after the above state remains stable for a period of time t, it can be determined that the cathode flooding situation has been effectively alleviated. At this point, the compensation effect introduced in the feedforward control can be gradually reduced and eventually withdrawn. The specific value reflects the maximum residual deviation of the actual cathode pressure difference relative to the target value during actual operation. This value can be determined comprehensively based on the control accuracy of the gas circuit subsystem and the measurement accuracy of the cathode pressure sensor used.

[0069] The calculation for increasing the air compressor speed (i.e., increasing the air flow rate) is as follows: Feedforward term: ; Feedback items: ,in, This represents the deviation between the actual value and the target value of the cathode pressure difference.

[0070] feedforward term It is the core predictive component of the control strategy, capable of adjusting operating parameters in real time based on the rate and magnitude of the increase in cathode differential pressure. That is, the faster the cathode differential pressure rises (i.e., the higher the rate and magnitude of the increase), the better. The larger the value of the cathode airflow, the greater the adjustment range of the controller can be, which means a positive increment in the cathode airflow. This is more significant, enabling a rapid response to and effective suppression of clogging trends, and maintaining the stable operation and high efficiency of the fuel cell system.

[0071] Feedback items Fine-tuning can be performed based on the deviation between the actual and target values ​​of the cathode pressure difference to ensure that the cathode pressure difference remains stable within an ideal range.

[0072] The calculation for reducing the humidifier's humidification capacity is as follows: Feedforward term: ; Feedback items: ,in, This represents the deviation between the actual value and the target value of the cathode pressure difference.

[0073] in, The difference between the next set value for adjusting the humidity of the air at the humidifier outlet and the actual value detected in real time.

[0074] feedforward term It can provide humidity compensation based on humidity trends. The larger the value, the smaller the compensation amount generated by the feedforward term will be, and the more significant the trend of change, the greater the reduction in compensation amount. This reverse adjustment strategy helps the system achieve more stable and accurate dynamic balance control of humidity.

[0075] Feedback items Fine-tuning can be performed based on the deviation between the actual and target values ​​of the cathode pressure difference to ensure that the cathode pressure difference remains stable within an ideal range.

[0076] Upon receiving an anode flooding warning signal (i.e., a water content prediction signal indicating that the proton exchange membrane is becoming too wet and the anode is at risk of flooding), the controller can generate a fifth command via feedforward control to increase the purging frequency and duration of the fuel cell stack's anode. Based on this fifth command, the anode purging frequency and duration can be increased to enhance drainage. The core of this measure is to proactively open the anode vent valve before liquid water accumulates to a dangerous level, using the pressure of hydrogen itself to quickly expel any accumulated trace amounts of liquid water or high-humidity gas from the system, thereby effectively reducing the risk of anode flooding. Furthermore, the controller can also generate a sixth command to increase the speed of the hydrogen circulation pump in the gas path subsystem. Based on this sixth command, the operating speed of the hydrogen circulation pump can be increased to enhance the gas flow velocity and circulation flow rate within the anode cavity, further promoting water and gas discharge and improving flow field uniformity.

[0077] It should be noted that while performing the above operations, the system will also continuously monitor the status parameters of the fuel cell system, and will respond when the following conditions are met simultaneously: ,and ,and Furthermore, after the above state remains stable for a period of time t, it can be determined that the anode flooding situation has been effectively alleviated. At this time, the compensation amount in the feedforward control can be gradually reduced according to the actual operating conditions until it is completely withdrawn, so as to achieve the smooth recovery and efficient operation of the fuel cell system.

[0078] Upon outputting an anode flooding warning signal, a purging operation can be triggered immediately. The duration of this purging operation is [duration to be specified]. ,in, The base duration for purging, This represents the increment of the purging time relative to the base time. After the initial purging process, wait 2 seconds for the fuel cell system to stabilize, and then re-measure the target single-cell voltage (i.e., the minimum single-cell voltage) in the fuel cell system. and the anode pressure difference of the fuel cell stack If detected ,and If the anode water level is not completely cleared, the anode vent valve can be restarted to trigger a new round of purging. At this point, the anode can be... Appropriately increase, that is, This makes the new round of purging more powerful, forming an "adaptive strong purging" strategy.

[0079] The calculation for increasing the anode purging time is as follows: Feedforward term: ; Feedback items: ,in, This represents the deviation between the actual and target values ​​of the anode pressure difference.

[0080] feedforward term As a core component of system prediction and regulation, its key parameter is the minimum single-cell voltage. And the set critical lower limit of single-cell voltage .when Gradually approaching Time (i.e.) The smaller, The value is a positive number less than 1 (the specific value can be set according to the actual operating state and characteristics of the fuel cell stack), and the system can correspondingly increase the intensity of the exhaust operation. At this time, the adjustment range of the exhaust duration will increase significantly (manifested as a positive increment in the exhaust duration). (More obvious), thus more quickly and effectively suppressing the potential blockage trend in the fuel cell stack, ensuring the stability and efficiency of the fuel cell system operation.

[0081] Feedback items Fine-tuning can be performed based on the deviation between the actual and target values ​​of the anode pressure difference to ensure that the anode pressure difference remains stable within an ideal range.

[0082] The calculation for increasing the speed of the hydrogen circulation pump is as follows: Feedforward term: ; Feedback items: ,in, This represents the deviation between the actual and target values ​​of the anode pressure difference.

[0083] feedforward term As a core component of system prediction and regulation, its key mechanism lies in the fact that the faster the control pressure differential rises (i.e., the more rapid the rise), the more critical the mechanism. When the value is larger, the magnitude of the increase in the hydrogen circulation pump speed is also larger (i.e., the increment in the hydrogen circulation pump speed). (This is presented as a positive trend with a more significant increase in value), thereby enhancing the efficiency of gas circulation, enabling rapid detection and effective suppression of blockage trends, and ensuring the stability and response speed of the fuel cell system. Feedback items It can make fine adjustments based on the deviation between the actual value and the target value of the anode pressure difference, ensuring that the anode pressure difference is always stable within an ideal range.

[0084] Therefore, by adopting a composite control strategy combining feedforward and feedback, the feedforward control section can generate rapid and reasonably sized preliminary compensation based on changes in system input, effectively addressing major disturbances. The feedback control, on the other hand, detects the deviation between the output and the target in real time, performing more precise and smooth fine-tuning, thus significantly improving the accuracy and stability of the control. Under this control structure, the behavior of various system components (such as air compressors and valves) is optimized, no longer exhibiting the abrupt and drastic "switching" transitions common in traditional control, but instead transforming into a continuous, smooth, and highly coordinated adjustment mode. This improvement not only effectively reduces the fluctuation amplitude of key system parameters but also significantly reduces mechanical wear on actuators, thereby extending the overall service life of the equipment. Simultaneously, it completely eliminates potential oscillations in the control loop, making the system operation more stable and reliable.

[0085] According to the fuel cell system water balance control method proposed in this application, a high-frequency resistance data sequence within a preset time window of the fuel cell stack is obtained, and the rate of change and acceleration of change of the high-frequency resistance are calculated accordingly. Based on the rate of change and acceleration of change, a predicted signal of the proton exchange membrane water content is generated using a preset water content state prediction model. According to the predicted signal, a preset feedforward-feedback control strategy is used to generate a gas path subsystem control command, and the preset gas path subsystem is adjusted based on the command to maintain the proton exchange membrane water content within the target range. Thus, by monitoring the rate of change of high-frequency resistance with load / time in real time, the future trend of water content in the membrane electrode is predicted, and the gas path parameters are adjusted before the water state in the stack deteriorates to an irreversible state. This solves the problems of lag in water management response and poor adaptability in the prior art, thereby achieving precise and stable control of water content.

[0086] Next, the water balance control device for a fuel cell system proposed according to an embodiment of this application is described with reference to the accompanying drawings.

[0087] Figure 2 This is a block diagram of a fuel cell system water balance control device according to an embodiment of this application.

[0088] like Figure 2 As shown, the fuel cell system water balance control device 10 includes: a calculation module 100, a generation module 200, and an adjustment module 300.

[0089] The calculation module 100 is used to acquire the high-frequency resistance data sequence of the fuel cell stack within a preset time window, and to calculate the rate of change of high-frequency resistance and the acceleration of change of high-frequency resistance based on the high-frequency resistance data sequence. The generation module 200 is used to generate a water content state prediction signal of the proton exchange membrane in the fuel cell stack based on the rate of change and acceleration of the change of the high-frequency resistance and using a preset water content state prediction model. The adjustment module 300 is used to generate control commands for the gas path subsystem in the fuel cell system based on the water content state prediction signal and a preset feedforward-feedback control strategy, and adjust the preset gas path subsystem based on the control commands so that the water content of the proton exchange membrane is maintained within the target range.

[0090] Optionally, in some embodiments, the generation module 200 is specifically used for: Based on the rate of change and acceleration of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is greater than a first preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between the second preset threshold and the first preset safety margin. If the rate of change of the high-frequency resistance is greater than the first preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between the second preset threshold and the first preset safety margin, then a water content state prediction signal indicating that the proton exchange membrane is tending to dry out is generated.

[0091] Optionally, in some embodiments, the generation module 200 is specifically used for: Based on the rate of change and acceleration of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin. If the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum between the fourth preset threshold and the second preset safety margin, then it is determined whether the single cell voltage fluctuation variance of the fuel cell stack is greater than the preset voltage fluctuation variance threshold, whether the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference threshold, and whether the rate of change of the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference rate of change threshold. If the single-cell voltage fluctuation variance of the fuel cell stack is greater than a preset voltage fluctuation variance threshold, and / or the cathode voltage difference of the fuel cell stack is greater than a preset cathode voltage difference threshold, and / or the cathode voltage difference change rate of the fuel cell stack is greater than a preset cathode voltage difference change rate threshold, then it is determined that the cathode of the fuel cell stack has a preset flooding risk, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

[0092] Optionally, in some embodiments, the generation module 200 is specifically used for: Based on the rate of change and acceleration of the high-frequency resistance, it is determined whether the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin. If the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin, then it is determined whether the target single cell voltage of the fuel cell stack is less than the preset voltage threshold, whether the anode differential pressure of the fuel cell stack is greater than the preset anode differential pressure threshold, and whether the rate of change of the anode differential pressure of the fuel cell stack is greater than the preset anode differential pressure change rate threshold. If the target single cell voltage of the fuel cell stack is less than a preset voltage threshold, and / or the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure threshold, and / or the anode differential pressure change rate of the fuel cell stack is greater than a preset anode differential pressure change rate threshold, then it is determined that the anode of the fuel cell stack has a preset risk of flooding, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

[0093] Optionally, in some embodiments, the water content state prediction signal indicates that the proton exchange membrane is tending to dry out, and the adjustment module 300 is specifically used for: Based on the moisture content state prediction signal, a first instruction is generated to increase the outlet humidity of the humidifier in the gas path subsystem using a preset feedforward-feedback control strategy, and / or a second instruction is generated to reduce the air metering ratio of the cathode of the fuel cell stack.

[0094] Optionally, in some embodiments, the moisture content state prediction signal indicates that the proton exchange membrane is becoming too wet, and the adjustment module 300 is specifically used for: If it is determined that there is a preset risk of flooding at the cathode of the fuel cell stack, a third command is generated to increase the speed of the air compressor in the gas circuit subsystem and / or a fourth command is generated to reduce the inlet humidity of the cathode of the fuel cell stack, based on the water content state prediction signal and using a preset feedforward-feedback control strategy. If a pre-defined risk of flooding is determined to exist at the anode of the fuel cell stack, a fifth instruction is generated based on the water content state prediction signal and a pre-defined feedforward-feedback control strategy to increase the purging frequency and purging duration of the anode of the fuel cell stack, and / or a sixth instruction is generated to increase the speed of the hydrogen circulation pump of the gas circuit subsystem.

[0095] It should be noted that the foregoing explanation of the embodiment of the fuel cell system water balance control method also applies to the fuel cell system water balance control device of this embodiment, and will not be repeated here.

[0096] According to the fuel cell system water balance control device proposed in this application, a high-frequency resistance data sequence within a preset time window of the fuel cell stack is acquired, and the rate of change and acceleration of change of the high-frequency resistance are calculated accordingly. Based on the rate of change and acceleration of change, a proton exchange membrane water content prediction signal is generated using a preset water content state prediction model. According to the prediction signal, a gas path subsystem control command is generated using a preset feedforward-feedback control strategy. Based on the command, the preset gas path subsystem is adjusted to maintain the proton exchange membrane water content within the target range. Thus, by real-time monitoring of the rate of change of high-frequency resistance with load / time, the future trend of water content in the membrane electrode is predicted, and the gas path parameters are adjusted before the water state in the stack deteriorates to an irreversible state. This solves the problems of delayed response and poor adaptability in water management in the prior art, thereby achieving precise and stable control of water content.

[0097] Figure 3 A schematic diagram of the structure of a vehicle provided in an embodiment of this application. The vehicle may include: The memory 301, the processor 302, and the computer program stored on the memory 301 and capable of running on the processor 302.

[0098] When the processor 302 executes the program, it implements the fuel cell system water balance control method provided in the above embodiments.

[0099] Furthermore, the vehicle also includes: Communication interface 303 is used for communication between memory 301 and processor 302.

[0100] The memory 301 is used to store computer programs that can run on the processor 302.

[0101] The memory 301 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0102] If the memory 301, processor 302, and communication interface 303 are implemented independently, then the communication interface 303, memory 301, and processor 302 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0103] Optionally, in a specific implementation, if the memory 301, processor 302, and communication interface 303 are integrated on a single chip, then the memory 301, processor 302, and communication interface 303 can communicate with each other through an internal interface.

[0104] Processor 302 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of this application.

[0105] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described fuel cell system water balance control method.

[0106] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0107] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0108] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A water balance control method for a fuel cell system, characterized in that, Includes the following steps: The high-frequency resistance data sequence of the fuel cell stack within a preset time window is obtained, and the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance are calculated based on the high-frequency resistance data sequence. Based on the rate of change and acceleration of the high-frequency resistance, a water content state prediction signal for the proton exchange membrane in the fuel cell stack is generated using a preset water content state prediction model. Based on the water content state prediction signal, a preset feedforward-feedback control strategy is used to generate a control command for the gas path subsystem in the fuel cell system, and based on the control command, the gas path subsystem is adjusted so that the water content of the proton exchange membrane is maintained within the target range. The step of generating a water content state prediction signal for the proton exchange membrane in the fuel cell stack using a preset water content state prediction model includes: determining whether the rate of change of the high-frequency resistance is greater than a first preset threshold and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between a second preset threshold and a first preset safety margin, based on the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance; if the rate of change of the high-frequency resistance is greater than the first preset threshold and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between the second preset threshold and the first preset safety margin, then generating a water content state prediction signal indicating that the proton exchange membrane is tending to dry out.

2. The method according to claim 1, characterized in that, The step of generating a water content state prediction signal for the proton exchange membrane in the fuel cell stack using a preset water content state prediction model includes: Determine whether the rate of change of the high-frequency resistance is less than or equal to a third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin. If the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin, then it is determined whether the single cell voltage fluctuation variance of the fuel cell stack is greater than the preset voltage fluctuation variance threshold, whether the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference threshold, and whether the rate of change of the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference rate of change threshold. If the single-cell voltage fluctuation variance of the fuel cell stack is greater than the preset voltage fluctuation variance threshold, and / or the cathode voltage difference of the fuel cell stack is greater than the preset cathode voltage difference threshold, and / or the cathode voltage difference change rate of the fuel cell stack is greater than the preset cathode voltage difference change rate threshold, then it is determined that the cathode of the fuel cell stack has a preset flooding risk, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

3. The method according to claim 1, characterized in that, The step of generating a water content state prediction signal for the proton exchange membrane in the fuel cell stack using a preset water content state prediction model includes: Determine whether the rate of change of the high-frequency resistance is less than or equal to a third preset threshold, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of a fourth preset threshold and a second preset safety margin. If the rate of change of the high-frequency resistance is less than or equal to the third preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is less than the sum of the fourth preset threshold and the second preset safety margin, then it is determined whether the target single cell voltage of the fuel cell stack is less than a preset voltage threshold, whether the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure threshold, and whether the rate of change of the anode differential pressure of the fuel cell stack is greater than a preset anode differential pressure change rate threshold. If the target single cell voltage of the fuel cell stack is less than the preset voltage threshold, and / or the anode differential pressure of the fuel cell stack is greater than the preset anode differential pressure threshold, and / or the anode differential pressure change rate of the fuel cell stack is greater than the preset anode differential pressure change rate threshold, then it is determined that the anode of the fuel cell stack has a preset flooding risk, and a water content state prediction signal indicating that the proton exchange membrane tends to be too wet is generated.

4. The method according to claim 1, characterized in that, The water content state prediction signal indicates that the proton exchange membrane is tending to dry out. The step of generating control commands for the gas path subsystem of the fuel cell system based on the water content state prediction signal using a preset feedforward-feedback control strategy includes: Based on the moisture content state prediction signal, a first instruction is generated to increase the outlet humidity of the humidifier in the gas path subsystem using a preset feedforward-feedback control strategy, and / or a second instruction is generated to reduce the air metering ratio of the cathode of the fuel cell stack.

5. The method according to claim 1, characterized in that, The water content state prediction signal indicates that the proton exchange membrane tends to be over-wet. The step of generating control commands for the gas path subsystem of the fuel cell system based on the water content state prediction signal using a preset feedforward-feedback control strategy includes: If it is determined that there is a preset risk of water flooding at the cathode of the fuel cell stack, a third command is generated to increase the speed of the air compressor of the gas circuit subsystem, and / or a fourth command is generated to reduce the inlet humidity of the cathode of the fuel cell stack, based on the water content state prediction signal and using a preset feedforward-feedback control strategy. If it is determined that there is a preset risk of flooding at the anode of the fuel cell stack, a fifth instruction is generated based on the water content state prediction signal and using a preset feedforward-feedback control strategy to increase the purging frequency and purging duration of the anode of the fuel cell stack, and / or a sixth instruction is generated to increase the speed of the hydrogen circulation pump of the gas circuit subsystem.

6. A water balance control device for a fuel cell system, characterized in that, include: The calculation module is used to acquire the high-frequency resistance data sequence of the fuel cell stack within a preset time window, and to calculate the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance based on the high-frequency resistance data sequence. The generation module is used to generate a water content state prediction signal of the proton exchange membrane in the fuel cell stack based on the rate of change and acceleration of the change of the high-frequency resistance and using a preset water content state prediction model. The adjustment module is used to generate a control command for the gas path subsystem in the fuel cell system based on the water content state prediction signal and a preset feedforward-feedback control strategy, and adjust the gas path subsystem based on the control command so that the water content of the proton exchange membrane is maintained within the target range. Specifically, the generation module is used to: determine whether the rate of change of the high-frequency resistance is greater than a first preset threshold based on the rate of change of the high-frequency resistance and the acceleration of the change of the high-frequency resistance, and whether the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between a second preset threshold and a first preset safety margin. If the rate of change of the high-frequency resistance is greater than the first preset threshold, and the high-frequency resistance value at the current moment in the high-frequency resistance data sequence is greater than the difference between the second preset threshold and the first preset safety margin, then a water content state prediction signal indicating that the proton exchange membrane is tending to dry out is generated.

7. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the fuel cell system water balance control method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the fuel cell system water balance control method as described in any one of claims 1-5.

Images