A method, processing method and system for predicting low voltage of a fuel cell single body

Abstract

This invention relates to the field of fuel cell technology, specifically to a method, processing method, and system for predicting low voltage in a fuel cell cell. The prediction method includes: S1, starting up the fuel cell stack; S2, applying current while simultaneously applying AC excitation; S3, acquiring cell voltage data and calculating the current average cell voltage V. avg With the lowest single-chip voltage V min And the standard deviation σ of the current single-chip voltage; S4, determine |V avg -V min If the equation |≤3σ| holds true, the fuel cell stack is considered to be in good health and there is no single-cell low-voltage fault. If the equation does not hold true, the fuel cell stack is considered to have a single-cell low-voltage fault. This invention also provides a method for handling low-voltage single cells in a fuel cell and a fuel cell system. Compared with existing technologies, by using normal distribution theory to identify the number of cells potentially prone to low voltage before the single-cell voltage drops below the shutdown limit and implementing intervention and recovery measures, the failure rate is significantly reduced, thereby improving the stability and reliability of the fuel cell system.

Description

Technical Field

[0001] This invention relates to the field of fuel cell technology, specifically to a method, processing method, and system for predicting low voltage in a fuel cell cell. Background Technology

[0002] Hydrogen fuel cell technology is a core technology in hydrogen energy applications. Due to its advantages such as high specific energy, high specific power, rapid low-temperature start-up, and clean and pollution-free operation, it has attracted the attention of major research institutions and companies worldwide. Currently, fuel cell systems are mainly used in the transportation sector. Depending on the vehicle model and power requirements, the rated power of fuel cell systems ranges from 30kW to 200kW. Typically, a fuel cell stack requires 300 to 500 individual cells connected in series. This series connection means that if one cell fails, the entire fuel cell stack will fail, leading to a system shutdown. Therefore, the consistency of the fuel cell stack has a significant impact on the stability and reliability of the system operation.

[0003] For proton exchange membrane fuel cells (PEMFCs), researchers typically assess the consistency of the fuel cell stack by detecting the voltage of individual cells. If a cell's voltage is too low, fault diagnosis or emergency shutdown is triggered. For example, in an online diagnosis and handling method and system for hydrogen fuel cell cell faults disclosed in publication number CN 117219820 A, a single-cell low voltage fault is identified when the lowest cell voltage is <520mV and the difference between the lowest cell voltage and the average voltage is >150mV. In a fuel cell single-cell low voltage fault diagnosis method and device disclosed in publication number CN118156550A, a single-cell low voltage fault is identified by checking whether the lowest cell voltage is less than 450mV and further determining the deviation from the average voltage (i.e., whether the average cell voltage minus the lowest cell voltage) is greater than 80mV. In a fuel cell system health status detection method and device disclosed in publication number CN115113056 A, the existence of a single-cell low voltage fault is determined by checking whether the voltage variance and deviation from the average voltage in the stack are higher than set values. However, in actual operation, the phenomenon of single low temperature in fuel cells has a certain nonlinearity. In the early stage of the phenomenon, the faulty section will decrease slowly, but in the later stage it will decrease rapidly, triggering an emergency shutdown within a few seconds. Therefore, judging the fault by absolute value has a certain risk. If the limit is too high, false alarms are likely to occur. If the limit is too low, there will be lag, making it difficult to avoid the result of single low temperature emergency shutdown through diagnosis and recovery.

[0004] Therefore, a system and method are needed to predict low voltage in fuel cell cells in advance, identify it before an emergency shutdown occurs, and prevent the failure from escalating through diagnostic and intervention measures, thereby improving the stability and reliability of fuel cell system operation. Summary of the Invention

[0005] To overcome the shortcomings of the prior art, this invention provides a method, processing method, and system for predicting low voltage in fuel cell cells.

[0006] To achieve the above objective, a method for predicting low voltage in a fuel cell cell is provided, comprising:

[0007] S1, fuel cell stack startup;

[0008] S2, AC excitation is performed simultaneously with current loading;

[0009] S3, Obtain individual cell voltage data and calculate the current average individual cell voltage V. avg With the lowest single-cell voltage V min And the standard deviation σ of the current unit voltage;

[0010] S4, Determine |V avg -V min If |≤3σ, the stack is considered to be in good health and there is no individual low voltage fault. If the equation does not hold, the stack is considered to have an individual low voltage fault.

[0011] The fuel cell stack consists of two or more individual cells, and the average voltage V of each cell in step S3 is... avg Minimum single-unit voltage V min The calculation methods for the standard deviation σ of a single-unit voltage include:

[0012] S3.1, Obtain the voltage data of a series connection of n sections, with the lowest group voltage being V. i The lowest voltage group is numbered i, and the average voltage V of the group is calculated. avgn V avg =V avgn / n, calculate the standard deviation σ of the grouped voltages. n The standard deviation of the individual voltage σ = σ n / n;

[0013] S3.2, Obtain the voltage signals V of the two nearest groups based on the sequence number of the lowest voltage group. i-1 V i+1 ;

[0014] S3.3, Determine the proximity voltage signal |V i-1 -V avgn |≤3σ n or |V i+1 -V avgn |≤3σ n If the equation holds, then V min =V i -(n-1)×V avgn / n; if the equation does not hold, then Vmin =V i / n.

[0015] The present invention also provides a method for processing low voltage in a fuel cell cell, including the above-mentioned method for predicting low voltage in a fuel cell cell, and further comprising:

[0016] S5, determine the cause of low voltage in a single unit;

[0017] S6, Process according to the judgment result;

[0018] S7, determine V min If the decrease continues, return to step S4 for continuous monitoring.

[0019] S8, if V min The voltage continues to decrease, triggering a low-voltage preload on individual units and limiting power operation.

[0020] S9, determine V min Should it be reduced to V? limit Below, if it does not drop to V limit Now, let's return to step S7;

[0021] S10, if V min Decrease to V limit The following is a shutdown operation.

[0022] Step S5 specifically includes: obtaining the high-frequency and low-frequency impedance information of the faulty cell; if the high-frequency impedance exceeds 55 mΩ cm... 2 If the low-frequency impedance exceeds 450 mΩ / cm, then film dryness is detected. 2 This indicates that flooding has occurred.

[0023] Step S6 specifically includes:

[0024] S6.1 If membrane dryness is detected, perform fuel cell humidification intervention and continuously check whether the equation in S4 is true for 10~30 seconds. If the equation is true, continue running according to the existing procedure.

[0025] S6.2 If flooding is detected, perform fuel cell drainage intervention and continuously check whether the equation in S4 is true for 10~30 seconds. If the equation is true, continue running according to the existing procedure.

[0026] The fuel cell stack humidification intervention method in step S6.1 includes at least one of increasing air humidity and decreasing fuel cell stack temperature.

[0027] The fuel cell stack drainage intervention method in step S6.2 includes at least one of increasing the hydrogen venting frequency, reducing air humidity, and increasing the fuel cell stack temperature.

[0028] The present invention also provides a fuel cell system, characterized in that it comprises:

[0029] A fuel cell stack comprising multiple fuel cell cells connected in series for generating electrical energy;

[0030] A DC / DC converter is used to convert the electrical energy generated by the fuel cell stack into voltage, and at the same time provide AC signal excitation to the fuel cell stack.

[0031] The CVM voltage monitoring module is electrically connected to the fuel cell stack via a low-voltage wiring harness and is used to detect the voltage information of a single fuel cell or multiple fuel cells.

[0032] An air subsystem, which is connected to the fuel cell stack via piping, is used to provide the fuel cell stack with air at the required pressure, temperature, flow rate, and humidity.

[0033] A hydrogen subsystem, which is connected to the fuel cell stack via pipelines, is used to provide the fuel cell stack with hydrogen at the required pressure, temperature, flow rate, and humidity.

[0034] A cooling subsystem, which is connected to the fuel cell stack via piping, is used to provide the coolant required for heat dissipation from the fuel cell stack.

[0035] The control module is connected to the DC / DC converter, CVM voltage monitoring module, air subsystem, hydrogen subsystem, and cooling subsystem via CAN, and is used to control the system to execute the above-mentioned method for predicting low voltage of fuel cell cells or the above-mentioned method for handling low voltage of fuel cell cells.

[0036] Compared with existing technologies, this invention uses a combination of CVM voltage monitoring module and DC / DC converter to obtain cell voltage information. By using proximity information, it more accurately predicts the actual minimum cell voltage. By using normal distribution theory, it identifies the number of cells that may experience a low voltage before the cell voltage drops below the shutdown limit and takes intervention and recovery measures, which greatly reduces the failure rate and improves the stability and reliability of the fuel cell system. Attached Figure Description

[0037] Figure 1 A flowchart of a method for predicting low voltage in a fuel cell cell.

[0038] Figure 2 A flowchart illustrating the calculation method for the minimum single-cell voltage of a fuel cell.

[0039] Figure 3 A flowchart of a method for handling low voltage in a fuel cell cell.

[0040] Figure 4 This is a schematic diagram of a fuel cell system. Detailed Implementation

[0041] The present invention will now be further described with reference to the accompanying drawings.

[0042] like Figure 1 As shown, the present invention provides a method for predicting low voltage in a fuel cell cell, comprising:

[0043] S1, fuel cell stack startup;

[0044] S2, AC excitation is performed simultaneously with current loading;

[0045] S3, Obtain individual cell voltage data and calculate the current average individual cell voltage V. avg With the lowest single-cell voltage V min And the standard deviation σ of the current unit voltage;

[0046] S4, Determine |V avg -V min If |≤3σ, the stack is considered to be in good health and there is no individual low voltage fault. If the equation does not hold, the stack is considered to have an individual low voltage fault.

[0047] like Figure 2 As shown, for voltage data consisting of two or more individual cells tested individually, the average individual cell voltage V in step S3 is... avg Minimum single-unit voltage V min The calculation methods for the standard deviation σ of a single-unit voltage include:

[0048] S3.1, Obtain the voltage data of a series connection of n sections, with the lowest group voltage being V. i The lowest voltage group is numbered i, and the average voltage V of the group is calculated. avgn V avg =V avgn / n, calculate the standard deviation σ of the grouped voltages. n The standard deviation of the individual voltage σ = σ n / n;

[0049] S3.2, Obtain the voltage signals V of the two nearest groups based on the sequence number of the lowest voltage group. i-1 V i+1 ;

[0050] S3.3, Determine the proximity voltage signal |V i-1 -V avgn |≤3σ n or |V i+1 -V avgn |≤3σ n If the equation holds, then V min =V i-(n-1)×V avgn / n; if the equation does not hold, then V min =V i / n.

[0051] Step S3.3 on |V i-1 -V avgn |≤3σ n and |V i+1 -V avgn |≤3σ n The results of two equations are ORed together. If either equation's result is true, then the equation is true, and V is the result of the OR operation. min =V i -(n-1)×V avgn / n; if neither equation is true, then V min =V i / n.

[0052] Specifically, if i is the last group in the stack, then the two nearest voltage signals to it are V. i-1 V i-2 The conditional statement executed in S3.3 is |V i-1 -V avgn |≤3σ n or |V i+1 -V avgn |≤3σ n If i is the first group in the stack, then the two nearest voltage signals to it are V. i+1 V i+2 The conditional statement executed in S3.3 is |V i+1 -V avgn |≤3σ n or |V i+2 -V avgn |≤3σ n .

[0053] like Figure 3 As shown, the present invention provides a method for processing low voltage in a fuel cell cell: including the low voltage prediction method for a fuel cell cell as described above, and further including:

[0054] S5, determine the cause of low voltage in a single unit;

[0055] S6, Process according to the judgment result;

[0056] S7, determine V min If the decrease continues, return to step S4 for continuous monitoring.

[0057] S8, if V min The voltage continues to decrease, triggering a low-voltage preload on individual units and limiting power operation.

[0058] S9, determine V min Should it be reduced to V? limit Below, if it does not drop to V limit Now, let's return to step S7;

[0059] S10, if V min Decrease to V limit The following is a shutdown operation.

[0060] Step S5 specifically includes: obtaining the high-frequency and low-frequency impedance information of the faulty cell; if the high-frequency impedance exceeds 55mΩ / cm... 2 If the low-frequency impedance exceeds 450 mΩ / cm, then film dryness is detected. 2 This indicates that flooding has occurred.

[0061] Step S6 specifically includes:

[0062] S6.1 If membrane dryness is detected, perform fuel cell humidification intervention and continuously check whether the equation in S4 is true for 10~30 seconds. If the equation is true, continue running according to the existing procedure.

[0063] S6.2 If flooding is detected, perform fuel cell drainage intervention and continuously check whether the equation in S4 is true for 10~30 seconds. If the equation is true, continue running according to the existing procedure.

[0064] The fuel cell stack humidification intervention method in step S6.1 includes at least one of increasing air humidity and decreasing fuel cell stack temperature.

[0065] The fuel cell stack drainage intervention method in step S6.2 includes at least one of increasing the hydrogen venting frequency, reducing air humidity, and increasing the fuel cell stack temperature.

[0066] like Figure 4As shown, the present invention also provides a fuel cell system, wherein the stack comprises multiple fuel cell cells connected in series for generating electrical energy; a DC / DC converter is used to convert the electrical energy generated by the stack into voltage and simultaneously provide AC signal excitation to the stack; a CVM voltage monitoring module is electrically connected to the stack via a low-voltage wiring harness for detecting the voltage information and AC impedance of the fuel cell cells or multiple cells; an air subsystem is connected to the stack via pipelines for providing the stack with air at the required pressure, temperature, flow rate, and humidity; a hydrogen subsystem is connected to the stack via pipelines for providing the stack with hydrogen at the required pressure, temperature, flow rate, and humidity; a cooling subsystem is connected to the stack via pipelines for providing the stack with the coolant required for heat dissipation; and a control module is connected to the DC / DC converter, CVM voltage monitoring module, air subsystem, hydrogen subsystem, and cooling subsystem via CAN bus to control the system to execute the fuel cell cell low voltage prediction method or the fuel cell cell low voltage processing method as described above. Example 1:

[0067] In one embodiment of the present invention, the system includes an electric stack, a DC / DC converter, a CVM voltage monitoring module, an air subsystem, a hydrogen subsystem, a cooling subsystem, and a control system.

[0068] The fuel cell stack, consisting of 330 sections, is the component under test. A DC / DC converter converts the electrical energy generated by the stack into voltage and provides AC signal excitation; it is electrically connected to the stack via a high-voltage wiring harness. A CVM (Continuous Voltage Monitor) detects the voltage information of each individual section within the stack; in this embodiment, the CVM detects the voltage of each individual section and is electrically connected to the stack via a low-voltage wiring harness. An air subsystem provides air to the fuel cell stack at specific pressure, temperature, flow rate, and humidity, and is connected to the stack via piping. A hydrogen subsystem provides hydrogen to the fuel cell stack at specific pressure, temperature, flow rate, and humidity, and is connected to the stack via piping. A cooling subsystem provides the coolant needed for heat dissipation and is connected to the stack via piping. The control module communicates with the DC / DC converter, CVM, air subsystem, hydrogen subsystem, and cooling subsystem via CAN communication.

[0069] The system issued a start-up command, and the fuel cell stack started up and operated until it reached the steady-state operating point of 1.5 A / cm. 2 After running for 20 minutes, the voltage signals of the fuel cell stack were monitored. The average individual cell voltage was 653mV, the standard deviation of the voltage was 5mV, and the lowest individual cell voltage was 632mV. The single-cell low shutdown limit of the fuel cell stack is 300mV. It can be seen that the lowest individual cell voltage has not reached the shutdown limit voltage, but using the 3σ principle, this section shows a trend towards a single low voltage. Therefore, according to S4, the condition is not met, and impedance information is determined. The high-frequency impedance is 85mΩ cm. 2Its target range is 45~55mΩ cm. 2 Therefore, it was determined that the high-frequency impedance inside the fuel cell stack was too high, resulting in membrane dryness. The opening of the humidifier bypass valve in the system was adjusted to increase the intake humidity, and the system was observed for 30 seconds. After humidity adjustment, the minimum single-cell voltage increased from 632mV to 643mV. The S4 condition was met again, indicating that the system was in good health and continued to operate. Example 2:

[0070] In another embodiment of the present invention, the system architecture is the same as that in embodiment 1, except that the voltage information detected by the CVM voltage monitoring module is checked every two sections.

[0071] The system issues a start-up command, and the fuel cell stack starts up and operates until it reaches the steady-state operating point of 1.0 A / cm. 2 After running for 20 minutes, the voltage signals of the fuel cell stack were detected. The average group voltage was 1444mV (two sections), the standard deviation of the individual voltage was 3mV, and the lowest group voltage was 1404mV (two sections), numbered 86. The single-low shutdown limit for the fuel cell stack was 300mV. Since voltage detection was performed in two sections at a time, there was a possibility that both sections might be evenly low or have uneven voltage distributions. Considering the mutual influence of sections approaching each other, the voltage information of sections 85 and 87 near section 86 was obtained, where V... 85 =1442mV, V 87 =1444mV. The equation holds true according to S3.3, indicating a high probability of voltage unevenness in this section. Considering a conservative strategy, the actual lowest single-unit voltage is calculated to be V. min =1404-1444 / 2=682mV. Based on the calculated lowest single-unit voltage, the 3σ principle is applied. This section shows a trend towards a low single-unit voltage; therefore, the S4 condition is not met. Impedance information is then determined, with the high-frequency impedance being 46mΩ cm. 2 Its target range is 45~55mΩ cm. 2 Therefore, it was determined that the high-frequency impedance inside the fuel cell stack was normal, and the low-frequency impedance was then obtained as 500 mΩ cm. 2 Its target range is 200~450mΩ cm 2 Based on the impedance data, it was determined that the fuel cell stack had experienced flooding. The air flow rate and hydrogen purging frequency were increased, and the operation was observed for 30 seconds. After purging, the lowest single-cell voltage increased from 682mV to 720mV. The S4 test was performed again, and the system met the requirements, indicating good system health and continued operation.

Claims

1. A method for predicting low voltage in a fuel cell cell, characterized in that, include: S1, fuel cell stack startup; S2, AC excitation is performed simultaneously with current loading; S3, Obtain individual cell voltage data and calculate the current average individual cell voltage V. avg With the lowest single-cell voltage V min And the standard deviation σ of the current unit voltage; S4, Determine If the equation holds true, the fuel cell stack is considered to be in good health and there is no individual low voltage fault. If the equation does not hold true, the fuel cell stack is considered to have an individual low voltage fault. S5, determine the cause of low voltage in a single unit; S6, Process according to the judgment result; S7, determine V min If the decrease continues, return to step S4 for continuous monitoring. S8, if V min The voltage continues to decrease, triggering a low-voltage preload on individual units and limiting power operation. S9, determine V min Should it be reduced to V? limit Below, if it does not drop to V limit Now, let's return to step S7; S10, if V min Decrease to V limit The following is a shutdown operation; The fuel cell stack consists of two or more individual cells, and the average voltage V of each cell in step S3 is... avg Minimum single-unit voltage V min The calculation methods for the standard deviation σ of a single-unit voltage include: S3.1, Obtain the voltage data of a series connection of n sections, with the lowest group voltage being V. i The lowest voltage group is numbered i, and the average voltage V of the group is calculated. avgn , Calculate the standard deviation σ of the grouped voltages. n The standard deviation of the individual voltage σ = σ n / n; S3.2, Obtain the voltage signals V of the two nearest groups based on the sequence number of the lowest voltage group. i-1 V i+1 ; S3.3, Determine the proximity voltage signal. If the equation holds true, then If the equation does not hold, then ; Step S6 specifically includes: S6.1 If membrane dryness is detected, perform fuel cell humidification intervention and continuously check whether the equation in S4 is true for 10~30 seconds. If the equation is true, continue running according to the existing procedure. S6.2 If flooding is detected, perform fuel cell drainage intervention and continuously check whether the equation in S4 is true for 10~30 seconds. If the equation is true, continue running according to the existing procedure.

2. The method for predicting low voltage in a fuel cell cell according to claim 1, characterized in that: Step S5 specifically includes: obtaining the high-frequency and low-frequency impedance information of the faulty cell; if the high-frequency impedance exceeds 55 mΩ cm... 2 If the low-frequency impedance exceeds 450 mΩ / cm, then film dryness is detected. 2 This indicates that flooding has occurred.

3. The method for predicting low voltage in a fuel cell cell according to claim 1, characterized in that: The fuel cell stack humidification intervention method in step S6.1 includes at least one of increasing air humidity and decreasing fuel cell stack temperature.

4. The method for predicting low voltage in a fuel cell cell according to claim 1, characterized in that: The fuel cell stack drainage intervention method in step S6.2 includes at least one of increasing the hydrogen venting frequency, reducing air humidity, and increasing the fuel cell stack temperature.

5. A fuel cell system, characterized in that, include: A fuel cell stack comprising multiple fuel cell cells connected in series for generating electrical energy; A DC / DC converter is used to convert the electrical energy generated by the fuel cell stack into voltage, and at the same time provide AC signal excitation to the fuel cell stack. The CVM voltage monitoring module is electrically connected to the fuel cell stack via a low-voltage wiring harness and is used to detect the voltage information and AC impedance of a single fuel cell or multiple fuel cells. An air subsystem, which is connected to the fuel cell stack via piping, is used to provide the fuel cell stack with air at the required pressure, temperature, flow rate, and humidity. A hydrogen subsystem, which is connected to the fuel cell stack via pipelines, is used to provide the fuel cell stack with hydrogen at the required pressure, temperature, flow rate, and humidity. A cooling subsystem, which is connected to the fuel cell stack via piping, is used to provide the coolant required for heat dissipation from the fuel cell stack. The control module is connected to the DC / DC converter, CVM voltage monitoring module, air subsystem, hydrogen subsystem, and cooling subsystem via CAN, and is used to control the system to execute the low voltage prediction method for a fuel cell cell as described in claims 1-4.

Images