A fuel cell health state evaluation method, system and electronic device

Abstract

The application discloses a fuel cell health state evaluation method, system and electronic equipment, determines a to-be-analyzed working current interval of a fuel cell, determines a working condition screening judgment standard of the fuel cell, obtains real-time sensor data of the fuel cell, determines whether the real-time sensor data meets the to-be-analyzed working current interval and the working condition screening judgment standard, the real-time sensor data at least including performance index data of the fuel cell, determines a fuel cell health factor based on the real-time sensor data meeting the to-be-analyzed working current interval and the working condition screening judgment standard, and determines a health state of the fuel cell based on the fuel cell health factor.

Description

Technical Field

[0001] This application relates to the field of fuel cells, and in particular to a method, system, and electronic device for assessing the health status of a fuel cell. Background Technology

[0002] As fuel cells are used and operated, the materials of their core components will inevitably undergo irreversible degradation. This results in a significant decrease in the power output of a fuel cell system after prolonged operation, under the same working conditions, compared to its initial operating power. Maintaining the factory-set control parameters at this point will accelerate the degradation of the fuel cell system. However, by adjusting the control parameters in a timely manner based on the health status of the fuel cell, the operating efficiency of the fuel cell system can be improved and its lifespan extended. Summary of the Invention

[0003] In view of this, this application provides a method, system, and electronic device for assessing the health status of a fuel cell, the specific solutions of which are as follows:

[0004] A method for assessing the health status of a fuel cell, comprising:

[0005] Determine the operating current range of the fuel cell to be analyzed;

[0006] Determine the screening criteria for the operating conditions of fuel cells;

[0007] Obtain real-time sensor data of the fuel cell, determine whether the real-time sensor data meets the operating current range to be analyzed and the operating condition screening criteria, wherein the real-time sensor data includes at least the performance index data of the fuel cell;

[0008] The fuel cell health factor is determined based on real-time sensor data that meets the specified operating current range and the screening criteria for operating conditions.

[0009] The health status of the fuel cell is determined based on the fuel cell health factors.

[0010] Furthermore, determining the operating current range of the fuel cell to be analyzed includes:

[0011] Determine the operating current variation range of the fuel cell;

[0012] N current values ​​are selected from the operating current variation range to form a candidate set of operating currents;

[0013] Determine the operating current range for each current value in the candidate set of operating currents;

[0014] Determine the performance index data of the fuel cell within the operating current range for each current value;

[0015] Based on the performance index data of the fuel cell within the operating current range for each current value, the first operating current range corresponding to the first current value in the candidate set of operating currents is determined as the operating current range to be analyzed.

[0016] Furthermore, determining the first operating current interval corresponding to the first current value in the candidate set of operating currents as the operating current interval to be analyzed based on the performance index data of the fuel cell within the operating current interval of each current value includes:

[0017] The performance index data of the fuel cell within the operating current range of each current value are respectively determined as the first time series, resulting in N first time series;

[0018] Obtain the variance and life-cycle discrimination of each first time series;

[0019] Based on the variance and full life cycle discrimination of the N first time series, one first time series is determined from the N first time series, and the operating current range corresponding to the determined first time series is determined as the operating current range to be analyzed.

[0020] Furthermore, the criteria for determining the operating conditions of the fuel cell include:

[0021] Determine the range of variation of the screening criteria values ​​for the operating conditions of fuel cells;

[0022] M standard values ​​are selected from the variation range of the working condition screening standard values ​​to form a candidate set of working condition screening standard values;

[0023] Determine the original data time series under the first working condition;

[0024] Based on the original data time series under the first working conditions and each of the M standard values, a first standard value is determined among the M standard values, and the first standard value is determined as the screening and judgment standard for the working conditions of the fuel cell.

[0025] Furthermore, the step of determining a first standard value among the M standard values ​​based on the original data time series under the first operating conditions and each standard value among the M standard values, and determining the first standard value as the screening criterion for the operating conditions of the fuel cell, includes:

[0026] Based on each standard value, the original data time series under the first working condition is filtered according to the set data filtering criteria to obtain the first dimensionless time series corresponding to each standard value;

[0027] Calculate the variance of the first dimensionless time series and the number of remaining samples corresponding to each standard value;

[0028] Based on the variance of the first dimensionless time series corresponding to each of the M standard values ​​and the number of remaining samples, a first standard value is determined among the M standard values, and the first standard value is determined as the screening and judgment standard for the working conditions of the fuel cell.

[0029] Furthermore, the determination of fuel cell health factors based on real-time sensor data that meets the screening criteria for the operating current range to be analyzed and the operating conditions includes:

[0030] Real-time sensor data that meets the operating current range to be analyzed and the operating condition screening criteria are input into the first formula to obtain the current health factor of the fuel cell. The first formula is used to characterize the relationship between the sensor data and the fuel cell health factor.

[0031] Furthermore, the determination of fuel cell health factors based on real-time sensor data that meets the screening criteria for the operating current range to be analyzed and the operating conditions includes:

[0032] The real-time sensor data that meets the specified operating current range and the specified operating condition screening criteria are subjected to moving average filtering to obtain filtered sensor data.

[0033] The filtered sensor data is input into the first formula to obtain the current health factor of the fuel cell.

[0034] Furthermore, determining the health status of the fuel cell based on the fuel cell health factors includes:

[0035] Obtain the health factors of each fuel cell in the fuel cell health factor queue, wherein the fuel cell health factor queue is a first-in-first-out queue.

[0036] The health status of the fuel cell is determined based on the health factors of each fuel cell in the fuel cell health factor queue.

[0037] A fuel cell health status assessment system, comprising:

[0038] The first determining unit is used to determine the operating current range of the fuel cell to be analyzed.

[0039] The second determining unit is used to determine the screening criteria for the working conditions of the fuel cell.

[0040] The judgment unit is used to obtain real-time sensor data of the fuel cell and determine whether the real-time sensor data meets the operating current range to be analyzed and the operating condition screening and judgment criteria. The real-time sensor data includes at least the performance index data of the fuel cell.

[0041] The third determining unit is used to determine the fuel cell health factor based on real-time sensor data that meets the screening and judgment criteria of the working current range to be analyzed and the working conditions.

[0042] The fourth determining unit is used to determine the health status of the fuel cell based on the fuel cell health factors.

[0043] An electronic device, comprising:

[0044] The processor is configured to: determine the operating current range to be analyzed for the fuel cell; determine the screening criteria for the operating conditions of the fuel cell; obtain real-time sensor data of the fuel cell; determine whether the real-time sensor data conforms to the operating current range to be analyzed and the screening criteria for the operating conditions, wherein the real-time sensor data includes at least the performance index data of the fuel cell; determine the fuel cell health factors based on the real-time sensor data conforming to the operating current range to be analyzed and the screening criteria for the operating conditions; and determine the health status of the fuel cell based on the fuel cell health factors.

[0045] The memory is used to store the program used by the processor to execute the above-described processing procedure.

[0046] As can be seen from the above technical solutions, the fuel cell health status assessment method, system, and electronic equipment disclosed in this application determine the operating current range to be analyzed for the fuel cell, determine the screening criteria for the operating conditions of the fuel cell, obtain real-time sensor data of the fuel cell, determine whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions, and the real-time sensor data includes at least the performance index data of the fuel cell. Based on the real-time sensor data that meets the screening criteria for the operating current range and operating conditions, a fuel cell health factor is determined, and based on the fuel cell health factor, the health status of the fuel cell is determined. This solution screens the real-time sensor data of the fuel cell, and after screening, determines the health factor of the fuel cell based on the real-time sensor data, thereby determining the health status of the fuel cell. It realizes the detection of the performance index data of the fuel cell under specific operating conditions by sensors, and determines the health status of the fuel cell based on the performance index data under specific operating conditions. It realizes the quantitative assessment of the health status of the fuel cell through real-time sensor data, so as to adjust the control strategy based on the health status of the fuel cell, thereby ensuring operating efficiency and service life. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of a fuel cell health status assessment method disclosed in an embodiment of this application;

[0049] Figure 2 This is a flowchart of a fuel cell health status assessment method disclosed in an embodiment of this application;

[0050] Figure 3 This is a schematic diagram illustrating the relationship between the variance of a physical quantity and its life-cycle discrimination in a first time series disclosed in an embodiment of this application.

[0051] Figure 4 This is a flowchart of a fuel cell health status assessment method disclosed in an embodiment of this application;

[0052] Figure 5 This is a flowchart of a fuel cell health status assessment method disclosed in an embodiment of this application;

[0053] Figure 6 This is a schematic diagram comparing the variance of voltage sequences under different currents and the voltage difference over the entire life cycle, as disclosed in an embodiment of this application.

[0054] Figure 7 This is a schematic diagram illustrating the relationship between normalized voltage sequence variance, normalized sample size, data screening optimization objective, and working condition screening criteria disclosed in an embodiment of this application.

[0055] Figure 8 This is a schematic diagram illustrating the relationship between the average voltage after screening within the operating current range and the number of voltage endurance cycles, as disclosed in an embodiment of this application.

[0056] Figure 9 This is a schematic diagram illustrating the relationship between the average voltage and the number of voltage endurance cycles after screening by operating current range and operating condition screening criteria, as disclosed in an embodiment of this application.

[0057] Figure 10 This is a schematic diagram illustrating the changing trend of fuel cell health factors over operating time, as disclosed in this embodiment.

[0058] Figure 11 This is a schematic diagram illustrating the real-time changing trend of a fuel cell health factor as disclosed in an embodiment of this application.

[0059] Figure 12 This is a schematic diagram of the structure of a fuel cell health status assessment system disclosed in an embodiment of this application;

[0060] Figure 13 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. Detailed Implementation

[0061] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] This application discloses a method for assessing the health status of a fuel cell, the flowchart of which is shown below. Figure 1 As shown, it includes:

[0063] Step S11: Determine the operating current range of the fuel cell to be analyzed;

[0064] Step S12: Determine the screening criteria for the working conditions of the fuel cell;

[0065] Step S13: Obtain real-time sensor data of the fuel cell and determine whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions to be analyzed. The real-time sensor data shall include at least the performance index data of the fuel cell.

[0066] Step S14: Determine the fuel cell health factors based on real-time sensor data that meets the screening and judgment criteria for the operating current range and operating conditions to be analyzed.

[0067] Step S15: Determine the health status of the fuel cell based on fuel cell health factors.

[0068] Currently, the health status of fuel cells is mainly assessed through polarization curves or AC impedance spectroscopy. However, both methods require high precision of the measurement equipment and have long measurement times, which cannot meet the economic and real-time requirements of fuel cells in vehicle applications.

[0069] The solution disclosed in this embodiment directly measures data in real time through sensors and calculates the health status of the fuel cell based on the measured sensor data. It does not require high-precision measurement equipment or long measurement time, and can meet the requirements of economy and real-time performance.

[0070] Specifically, the health factor of a fuel cell is first defined, which can be: the long-term trend of apparent electrochemical performance or its derived indicators under similar operating conditions, specifically defined as:

[0071]

[0072] Where SOH(t) is the health factor of the fuel cell, physical quantity X is the apparent electrochemical performance of the fuel cell, 2ω is the moving average window size, t is the current time, and dτ is also a variable on the time scale, but it has no physical meaning and is only used as a mathematical integral symbol.

[0073] Since the physical quantity X obtained from actual sampling is discrete data, the fuel cell health factor can be defined by the moving average of the discrete physical quantity X, as follows:

[0074]

[0075] Where i is a positive integer.

[0076] Apparent electrochemical performance refers to the performance indicators of a fuel cell, which are relative to intrinsic electrochemical performance, such as open-circuit voltage, limiting diffusion current, and electrochemical active area. Apparent electrochemical performance refers to the electrical signals that can be directly measured during the operation of the fuel cell system. Depending on the fuel cell control strategy, apparent electrochemical performance can be, for example, the average voltage of the fuel cell stack, the current of the fuel cell stack, or the power of the fuel cell stack. In addition, derived indicators of apparent electrochemical performance can be, for example, the voltage variance of a single fuel cell segment and the electrochemical impedance of the fuel cell.

[0077] The health status of a fuel cell can be assessed based on its health factors. Therefore, this scheme first determines the fuel cell health factors and then determines the health status of the fuel cell based on these factors.

[0078] Since the health factor of a fuel cell is the long-term trend of its apparent electrochemical performance or its derived indicators under similar operating conditions, it is necessary to first obtain the apparent electrochemical performance of the fuel cell under specific operating conditions. Apparent electrochemical performance can be directly measured, that is, by obtaining sensor data under specific operating conditions. The obtained real-time sensor data includes apparent electrochemical performance, which is the performance index data.

[0079] To obtain sensor data of a fuel cell under specific operating conditions, the fuel cell needs to operate within a specific current range and under specific operating conditions. Therefore, the operating current range of the fuel cell to be analyzed and the screening criteria for the operating conditions of the fuel cell can be determined to make it conform to the specific operating conditions.

[0080] Therefore, after obtaining the real-time sensor data of the fuel cell, the obtained real-time sensor data is filtered to obtain real-time sensor data that meets the specific operating conditions, that is, the screening and judgment criteria of the operating current range and operating conditions to be analyzed. Real-time sensor data that does not meet the specific operating conditions will not be further analyzed and judged.

[0081] For real-time sensor data that meets this specific operating condition, it is input into the first formula to obtain the fuel cell health factor, where the first formula is the formula for the fuel cell health factor mentioned above, namely:

[0082]

[0083] After obtaining the fuel cell health factors, the health status of the fuel cell is determined based on these factors.

[0084] The fuel cell health status assessment method disclosed in this solution first defines fuel cell health factors, thereby enabling the extraction of relevant performance index data during the dynamic operation of the fuel cell to predict its health status, thereby achieving real-time adaptation of its control strategy to improve system durability.

[0085] This embodiment discloses a fuel cell health status assessment method that determines the operating current range to be analyzed for the fuel cell, establishes screening criteria for the fuel cell's operating conditions, obtains real-time sensor data of the fuel cell, and determines whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions. The real-time sensor data includes at least the fuel cell's performance index data. Based on the real-time sensor data that meets the screening criteria for the operating current range and operating conditions, a fuel cell health factor is determined, and based on the fuel cell health factor, the fuel cell's health status is determined. This scheme filters the real-time sensor data of the fuel cell. After filtering, it determines the fuel cell's health factor based on the real-time sensor data, thereby determining the fuel cell's health status. It achieves quantitative assessment of the fuel cell's health status through real-time sensor data, enabling adjustments to the control strategy based on the fuel cell's health status to ensure operating efficiency and service life.

[0086] This embodiment discloses a method for assessing the health status of a fuel cell, the flowchart of which is shown below. Figure 2 As shown, it includes:

[0087] Step S21: Determine the operating current variation range of the fuel cell;

[0088] Step S22: Select N current values ​​from the operating current variation range to form a candidate set of operating currents;

[0089] Step S23: Determine the operating current range for each current value in the candidate set of operating currents, and determine the performance index data of the fuel cell within the operating current range for each current value.

[0090] Step S24: Based on the performance index data of the fuel cell within the operating current range of each current value, determine the first operating current range corresponding to the first current value in the candidate set of operating current as the operating current range to be analyzed.

[0091] Step S25: Determine the screening criteria for the working conditions of the fuel cell;

[0092] Step S26: Obtain real-time sensor data of the fuel cell, and determine whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions to be analyzed. The real-time sensor data shall include at least the performance index data of the fuel cell.

[0093] Step S27: Determine the fuel cell health factors based on real-time sensor data that meets the screening and judgment criteria for the operating current range and operating conditions to be analyzed;

[0094] Step S28: Determine the health status of the fuel cell based on fuel cell health factors.

[0095] The operating current range to be analyzed can be set based on the performance index data of fuel cells, namely, the principle that the physical quantity X fluctuates little and has a large degree of differentiation throughout its entire life cycle.

[0096] Specifically, determine the range of variation of the operating current of the fuel cell [I] min ,I max ], that is, the operating current of a fuel cell is typically in [I min ,I max The variation occurs within this range.

[0097] N values ​​are selected from the aforementioned operating current variation range to form a candidate set of operating currents. The N current values ​​can be taken at equal intervals from the operating current variation range, thus forming the candidate set S1 of operating currents, where N can be any positive integer.

[0098] Determine the operating current fluctuation range, and based on the operating current fluctuation range, determine the operating current range for each current value in the candidate set of operating currents.

[0099] If the fluctuation range of the operating current is δ, the fluctuation range of the operating current needs to meet the following requirements. Among them, I min I is greater than or equal to the idle current. maxLess than or equal to the peak current.

[0100] After determining the operating current fluctuation amplitude δ, the operating current range for each current value ε is determined, namely [ε-δ, ε+δ], and the numerical value of the operating current fluctuation amplitude is the same for each current value.

[0101] Then, the performance index data of each current value ε in the working current candidate set S1 is obtained sequentially. The performance index data of each current value in the working current interval is arranged according to time to obtain the first time series. The performance index data of the working current interval of N current values ​​can correspond to N first time series.

[0102] Based on N first time series, one current value can be selected from the N current value operating current ranges as the operating current range to be analyzed. Specifically, the operating current range with small fluctuations in physical quantity X and high full life cycle discrimination in the first time series can be determined as the operating current range to be analyzed.

[0103] Specifically, the variance and life-cycle discrimination of each first time series are obtained. Based on the variance and life-cycle discrimination of N first time series, one first time series is determined from the N first time series. The operating current range corresponding to the determined first time series is determined as the operating current range to be analyzed.

[0104] Sequentially select a current value from the candidate set of operating currents for calculation to determine the first time series corresponding to the selected operating current range. The first time series is defined as {x}. i}, where i is a positive integer. The first time series represents the value of the physical quantity X corresponding to the working current in the above-mentioned working current range, which corresponds one-to-one with time. As shown in Table 1, it is the time series of physical quantity X as the average voltage of the fuel cell stack.

[0105] Table 1

[0106]

[0107]

[0108] Where x1 = 0.74, x2 = 0.78, ..., this forms a time series {x i}. Here, actual time (e.g., 2021 / 9 / 26 16:23:06) and time intervals are not considered; only {x} are considered. i} represents different discrete values ​​on the time axis.

[0109] Let time series {x} iIf the length of} is L, and a range W is defined, then the life-cycle discrimination of physical quantity X can be calculated according to the following formula:

[0110]

[0111] Where i is a positive integer, and the length L is the time series {x} i There are L data points in total, and the range W only reflects the time series {x}. i The number of discrete data selected in the} can be adjusted according to the different characteristics of the data. Usually, the value of W can be between [0, 0.01*L].

[0112] It should be noted that the range W is subject to two constraints: First, W cannot be too large, otherwise the physical quantity sequence data at this stage will not meet the stationarity requirement, i.e., there will be a trend change. At this time, the multiple observations of physical quantity X are not in the same state, so the true value cannot be estimated from the observations at this time. Second, W cannot be too small; it should make the sample distribution not too different from the population distribution to ensure that the sample mean is closer to the population mean. Specifically, the range W is unrelated to the degree of dispersion and discrimination, but only to the characteristics of the time series of physical quantity X. The selection principle of W can be defined as: under the premise of satisfying the stationarity of the observation sequence of physical quantity X, select the largest W. If such W does not exist, then let W = 0.01*L. Here, the observation sequence of physical quantity X refers to the physical quantity within the range W.

[0113] Wherein, the variance of the first time series refers to the variance of the first time series {x}. i The degree of dispersion of each physical quantity X in}; the life cycle discrimination is the difference between physical quantities X, which is usually the physical quantity X(t) at the current time minus the physical quantity X(0) at the beginning of life. Since the true value of the physical quantity X(t) at the current time cannot be directly measured, it can be estimated by measuring the value and the true value can be obtained by averaging multiple measurements.

[0114] like Figure 3 The diagram shows the relationship between the variance of physical quantity X and the life-cycle discrimination in the first time series. There are L discrete data in the first time series, that is, the number of physical quantities X is L. The degree of dispersion of the L discrete data is determined to obtain the variance of the first time series. W discrete data from the early stage of life and W discrete data at the current time are selected to obtain the difference between the two, which is to determine the life-cycle discrimination of the first time series.

[0115] Obtain the variance and full life cycle discrimination of each of the N first time series. Based on the magnitude of the variance and full life cycle discrimination of each first time series, select a first time series and determine the working current range of its corresponding current value as the working current range to be analyzed.

[0116] Based on the principle that the physical quantity X has small fluctuations and large life cycle discrimination, the working current range to be analyzed is selected. Specifically, the variance and life cycle discrimination of N first time series can be ranked. According to the principle that the physical quantity X has small fluctuations and large life cycle discrimination, the discrimination ranking of each first time series is subtracted from the fluctuation ranking to obtain the difference. The size of the N differences is compared, and the working current range corresponding to the smallest difference is selected as the working current range to be analyzed.

[0117] Then, within the determined operating current range to be analyzed, health factors are identified in order to assess the health status of the fuel cell.

[0118] This embodiment discloses a fuel cell health status assessment method that determines the operating current range to be analyzed for the fuel cell, establishes screening criteria for the fuel cell's operating conditions, obtains real-time sensor data of the fuel cell, and determines whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions. The real-time sensor data includes at least the fuel cell's performance index data. Based on the real-time sensor data that meets the screening criteria for the operating current range and operating conditions, a fuel cell health factor is determined, and based on the fuel cell health factor, the fuel cell's health status is determined. This scheme filters the real-time sensor data of the fuel cell. After filtering, it determines the fuel cell's health factor based on the real-time sensor data, thereby determining the fuel cell's health status. It achieves quantitative assessment of the fuel cell's health status through real-time sensor data, enabling adjustments to the control strategy based on the fuel cell's health status to ensure operating efficiency and service life.

[0119] This embodiment discloses a method for assessing the health status of a fuel cell, the flowchart of which is shown below. Figure 4 As shown, it includes:

[0120] Step S41: Determine the operating current range of the fuel cell to be analyzed;

[0121] Step S42: Determine the range of variation of the screening criteria values ​​for the operating conditions of the fuel cell;

[0122] Step S43: Select M standard values ​​from the variation range of the working condition screening standard values ​​to form a candidate set of working condition screening standard values;

[0123] Step S44: Determine the original data time series under the first working condition;

[0124] Step S45: Based on the original data time series under the first working condition and each of the M standard values, determine the first standard value among the M standard values, and determine the first standard value as the screening and judgment standard for the working conditions of the fuel cell;

[0125] Step S46: Obtain real-time sensor data of the fuel cell, and determine whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions to be analyzed. The real-time sensor data shall include at least the performance index data of the fuel cell.

[0126] Step S47: Determine the fuel cell health factors based on real-time sensor data that meets the screening and judgment criteria for the operating current range and operating conditions to be analyzed;

[0127] Step S48: Determine the health status of the fuel cell based on fuel cell health factors.

[0128] The operating conditions of a fuel cell also include: hydrogen flow rate, hydrogen inlet pressure, hydrogen inlet temperature, hydrogen flow rate, air flow rate, air inlet temperature, air outlet temperature, coolant inlet temperature, and coolant outlet temperature.

[0129] Determine the range of variation for a screening criterion value for a working condition [C] min C max From this, M values ​​are selected, which can be chosen at equal intervals within the range of variation, thus forming a candidate set S2 of working condition screening standard values, where C min ≥0, C max ≤1, where M is a positive integer provided that computational performance is satisfied.

[0130] Select a working condition and extract the raw data time series {y} under that working condition. i Then, based on the original data time series and M standard values, the first standard value is determined from the M standard values, and the first standard value is determined as the screening and judgment standard for the working conditions of the fuel cell.

[0131] Specifically, based on each standard value, the original data time series under the first working condition is filtered according to the set data filtering criteria to obtain the first dimensionless time series corresponding to each standard value. The variance and the number of remaining samples of the first dimensionless time series corresponding to each standard value are calculated. Based on the variance and the number of remaining samples of the first dimensionless time series corresponding to each of the M standard values, the first standard value among the M standard values ​​is determined, and the first standard value is determined as the screening and judgment criterion for the working conditions of the fuel cell.

[0132] The original time series data is data-centric to obtain a dimensionless second time series with values ​​in the range [0,1]. Then, select a standard value ξ from the candidate set S2 of working condition screening standard values, satisfying ξ∈[0,1]. Set the screening standard according to the selected standard value, and filter out the dimensionless data within the screening standard, that is, filter out the data that does not meet the screening standard, so as to obtain the filtered dimensionless time series and the first dimensionless time series.

[0133] The screening criterion is determined based on a standard value. If the standard value is ξ, then the screening criterion can be... The first dimensionless time series that meets the screening criteria is:

[0134] Calculate the first dimensionless time series variance And the number of remaining samples N, where the number of remaining samples N is the number of dimensionless data remaining after screening according to the screening criteria.

[0135] Repeat the above steps to obtain the variance of the first dimensionless time series corresponding to each of the M standard values ​​and the number of remaining samples, thereby determining a standard value from the M standard values ​​as the screening criterion for the working conditions of the fuel cell.

[0136] The first standard value can be selected based on the fundamental principles of small variance and large number of remaining samples, using the following formula as the optimization objective:

[0137]

[0138] Based on this first standard value, the range of fluctuations in operating conditions can be reduced, thereby reducing the range of fluctuations in fuel cell performance.

[0139] This embodiment discloses a fuel cell health status assessment method that determines the operating current range to be analyzed for the fuel cell, establishes screening criteria for the fuel cell's operating conditions, obtains real-time sensor data of the fuel cell, and determines whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions. The real-time sensor data includes at least the fuel cell's performance index data. Based on the real-time sensor data that meets the screening criteria for the operating current range and operating conditions, a fuel cell health factor is determined, and based on the fuel cell health factor, the fuel cell's health status is determined. This scheme filters the real-time sensor data of the fuel cell. After filtering, it determines the fuel cell's health factor based on the real-time sensor data, thereby determining the fuel cell's health status. It achieves quantitative assessment of the fuel cell's health status through real-time sensor data, enabling adjustments to the control strategy based on the fuel cell's health status to ensure operating efficiency and service life.

[0140] This embodiment discloses a method for assessing the health status of a fuel cell, the flowchart of which is shown below. Figure 5 As shown, it includes:

[0141] Step S51: Determine the operating current range of the fuel cell to be analyzed;

[0142] Step S52: Determine the screening criteria for the working conditions of the fuel cell;

[0143] Step S53: Obtain real-time sensor data of the fuel cell, and determine whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions to be analyzed. The real-time sensor data shall include at least the performance index data of the fuel cell.

[0144] Step S54: Determine the fuel cell health factors based on real-time sensor data that meets the screening and judgment criteria for the operating current range and operating conditions to be analyzed;

[0145] Step S55: Obtain each fuel cell health factor in the fuel cell health factor queue. The fuel cell health factor queue is a first-in-first-out queue.

[0146] Step S56: Determine the health status of the fuel cell based on the health factors of each fuel cell in the fuel cell health factor queue.

[0147] The health status of a fuel cell can be determined specifically through a set of continuous health factors. Therefore, storage space can be pre-set in the system memory to store a queue of fuel cell health factors. Additionally, storage space can be set up to store an array of physical quantities collected in real-time by the fuel cell. The queue of fuel cell health factors can be array I, and the array of physical quantities collected in real-time by the fuel cell can be array II.

[0148] After obtaining sensor data from the fuel cell in real time, and if the obtained sensor data includes at least the physical quantity X, then the sensor data is stored in array II.

[0149] The real-time physical quantity X is determined by screening and judging criteria based on the operating current range and operating conditions. If it does not meet the criteria, the data collection and judgment process described above is repeated. If the physical quantity meets the criteria, the fuel cell health factor is determined based on the physical quantity. Alternatively, the judgment parameter of the loading rate range can be added to determine whether the physical quantity X meets the criteria; that is, whether the physical quantity X conforms to the loading rate range.

[0150] Before obtaining the health factor, a dequeue operation is first performed on array I, that is, the health factor that entered the queue earliest is removed from the queue, so that the health factor calculated this time can be stored in the queue, that is, the queue is a first-in-first-out queue.

[0151] The remaining data in array I and the physical quantity X that meets the standard are filtered and calculated based on the sliding data window to obtain the real-time fuel cell health factor. Then, the obtained real-time fuel cell health factor is enqueued so that the changing trend of the fuel cell health factor can be determined based on array I.

[0152] Among them, filtering calculation is to eliminate random fluctuations in the collected data and extract the long-term trend of the time series changes of the collected data. Filtering can be done using moving average filters, low-pass filters, band-pass filters, etc.

[0153] The size of the sliding window can be the same as the size of array I.

[0154] When a fuel cell is dynamically operated in current control mode between 30-330A, its health status can be assessed through the following steps:

[0155] The fuel cell test bench acquires and stores sensor setpoints and feedback values ​​for operating current, operating voltage, air flow rate, hydrogen flow rate, air inlet / outlet pressure, hydrogen inlet / outlet pressure, coolant inlet / outlet pressure, air inlet / outlet temperature, hydrogen inlet / outlet temperature, coolant inlet / outlet temperature, and single-cell voltage, among others, within the test system. The test bench is the hardware framework upon which the system for testing the health status of fuel cells is based.

[0156] Select the operating current range to be analyzed: Quantifying the current fluctuation δ at 30A using standard deviation, it is found to be 2.733A. Therefore, the operating current screening range [ε-δ, ε+δ] is set to ±3A. For example... Figure 6 As shown, voltage sequences x under different currents are selected. i A comparison is made between variance and lifetime voltage difference ΔV. If, at a current of 190A, the lifetime voltage difference (i.e., the lifetime discrimination ΔV) ranks first, and the voltage series variance ranks seventh among the selected currents, then the difference between lifetime discrimination and variance is -6, which is the smallest among the considered candidate operating current sets. Therefore, it can be considered that the variance at 190A is small, the lifetime discrimination is large, and there is a significant difference between the two. Thus, 190±3A is used as the operating current range to be analyzed, and the operating voltage at 190±3A is used as the standard for evaluating the health status of the fuel cell, i.e., the health factor.

[0157] Select working condition screening criteria, optimize the screening based on data, compare the effects of different screening thresholds on the variance and sample size of the selected voltage sequence, and select the required working condition screening criteria.

[0158] Taking the determination of hydrogen flow rate screening criteria as an example, Figure 7 This study explores the relationship between normalized voltage sequence variance, normalized sample size, data screening optimization objectives, and working condition screening criteria.

[0159] Figure 7 The horizontal axis represents the working condition screening criteria, and the vertical axis represents the normalized voltage series variance and the normalized voltage series sample size. Specifically, for hydrogen flow rate, when the working condition screening criterion is 0.87, the normalized data sample size N / max[N] and the normalized voltage series variance VAR(V i ) / max[VAR(V i The ratio of [] is 1.045, which is the largest within the range (0,1) of the working condition screening criteria. Therefore, the working condition screening criteria for hydrogen flow rate can be set to 0.87. Based on this, the working condition screening criteria are as follows: hydrogen flow rate 0.69, air flow rate 0.98, air inlet temperature 0.9, hydrogen inlet temperature 0.98, air inlet pressure 0.74, hydrogen inlet pressure 0.7, coolant inlet temperature 0.28, coolant outlet temperature 0.98, and coolant inlet pressure 0.98.

[0160] Based on the operating current range and operating condition screening criteria set in the above process, data is filtered out. Data with an operating current outside the range of 190±3A and data with a loading rate below 5A / s are excluded. This allows us to obtain the change in voltage endurance cycle count. Figure 8As shown; after further data normalization, data that does not meet the working condition screening criteria are excluded, resulting in the filtered data, as shown. Figure 9 As shown.

[0161] Among these, the durability cycle count is a time-dimensional variable. Based on physical principles, the fuel cell is the site of electrochemical reactions. Under suitable storage conditions, it does not experience lifespan reduction when not in operation. Therefore, non-fuel cell operating time should be excluded from the timeline. Thus, the durability cycle count is used as a substitute for general time to show the true evolution trend of the fuel cell's operating voltage over operating time. As the fuel cell ages, the operating voltage under the same operating current and conditions will gradually decrease. Through qualitative and quantitative analysis of the evolution trend of the fuel cell's operating voltage over time, the health status of the fuel cell can be perceived.

[0162] By setting the sliding window size of the moving average filter to 10 and applying the moving average filter to the selected voltage data values, the changing trend of the fuel cell health factor over operating time can be obtained, such as... Figure 10 As shown.

[0163] In another embodiment, the system memory includes storage spaces for storing queues of fuel cell health factors (array I) and arrays of physical quantities collected in real time by the fuel cell (array II). If the window size is set to 10, then array I has a size of 10. Using a fuel cell stack test bench, real-time sensor settings and feedback values, such as operating voltage, operating current, air flow rate, hydrogen flow rate, air inlet / outlet temperature, hydrogen inlet / outlet temperature, air inlet / outlet pressure, hydrogen inlet / outlet pressure, coolant inlet / outlet temperature, coolant inlet / outlet pressure, and single-cell voltage, are acquired and stored in array II.

[0164] set up x Let x' be the original data before centering, μ be the mean of the original data before centering, and σ be the standard deviation of the original data before centering. All statistical indicators above were calibrated using data from the past 100 hours. The data variables were then processed using the following formula for data centralization:

[0165]

[0166] By comparing the data within the operating current range and loading rate range with preset judgment criteria, the system determines in real time whether the fuel cell's operating status meets the criteria for entering the queue after calculating health factors. If it does not meet the criteria, the data collection and judgment process is repeated; if it does meet the criteria, the process continues to the next step. The preset judgment criteria can be: air flow rate 0.96, hydrogen flow rate 0.87, air inlet temperature 0.98, hydrogen inlet temperature 0.97, air inlet pressure 0.92, hydrogen inlet pressure 0.67, coolant inlet temperature 0.98, coolant outlet temperature 0.98, and coolant inlet pressure 0.98.

[0167] A dequeue operation is performed on array I, where the earliest health factor is dequeued. The remaining data in array I is then averaged with the real-time fuel cell operating voltage to obtain the real-time fuel cell health factor. This obtained real-time fuel cell health factor is then enqueued. During fuel cell operation, the above steps are repeated cyclically to obtain the real-time trend of the fuel cell health factor. Figure 11 As shown.

[0168] From a long-term trend analysis, the voltage of a fuel cell under the same current density and operating conditions gradually decreases as the fuel cell operating time increases. From the perspective of time series decomposition, this downward trend is a long-term trend, which can be extracted using a low-pass filter. The moving average filter is one of the common low-pass filters, and it can effectively reduce the time complexity of the algorithm. In addition, in terms of application, the sliding window method can effectively reduce the space complexity of the algorithm. Combined with the moving average filter algorithm, the hardware dependence of the health factor extraction algorithm is reduced.

[0169] This embodiment discloses a fuel cell health status assessment method that determines the operating current range to be analyzed for the fuel cell, establishes screening criteria for the fuel cell's operating conditions, obtains real-time sensor data of the fuel cell, and determines whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions. The real-time sensor data includes at least the fuel cell's performance index data. Based on the real-time sensor data that meets the screening criteria for the operating current range and operating conditions, a fuel cell health factor is determined, and based on the fuel cell health factor, the fuel cell's health status is determined. This scheme filters the real-time sensor data of the fuel cell. After filtering, it determines the fuel cell's health factor based on the real-time sensor data, thereby determining the fuel cell's health status. It achieves quantitative assessment of the fuel cell's health status through real-time sensor data, enabling adjustments to the control strategy based on the fuel cell's health status to ensure operating efficiency and service life.

[0170] This embodiment discloses a fuel cell health status assessment system, the structural schematic diagram of which is shown below. Figure 12 As shown, it includes:

[0171] The system comprises a first determining unit 121, a second determining unit 122, a judging unit 123, a third determining unit 124, and a fourth determining unit 125.

[0172] The first determining unit 121 is used to determine the operating current range to be analyzed in the fuel cell;

[0173] The second determining unit 122 is used to determine the screening criteria for the working conditions of the fuel cell;

[0174] The judgment unit 123 is used to obtain real-time sensor data of the fuel cell and determine whether the real-time sensor data meets the screening and judgment criteria of the working current range and working conditions to be analyzed. The real-time sensor data includes at least the performance index data of the fuel cell.

[0175] The third determining unit 124 is used to determine the fuel cell health factor based on real-time sensor data that meets the screening and judgment criteria of the working current range and working conditions to be analyzed.

[0176] The fourth determining unit 125 is used to determine the health status of the fuel cell based on fuel cell health factors.

[0177] Furthermore, the first determining unit is used for:

[0178] Determine the operating current variation range of the fuel cell; select N current values ​​from the operating current variation range to form a candidate set of operating currents; determine the operating current range for each current value in the candidate set of operating currents; determine the performance index data of the fuel cell within the operating current range of each current value; based on the performance index data of the fuel cell within the operating current range of each current value in the candidate set of operating currents, determine the first operating current range corresponding to the first current value in the candidate set of operating currents as the operating current range to be analyzed.

[0179] Furthermore, the first determining unit is used for:

[0180] The performance index data of the fuel cell within the operating current range of each current value are determined as the first time series, resulting in N first time series. The variance and life cycle discrimination of each first time series are obtained. Based on the variance and life cycle discrimination of the N first time series, one first time series is determined from the N first time series, and the operating current range corresponding to the determined first time series is determined as the operating current range to be analyzed.

[0181] Furthermore, the second determining unit is used for:

[0182] Determine the variation range of the screening standard values ​​for the working conditions of the fuel cell; select M standard values ​​from the variation range of the screening standard values ​​for the working conditions to form a candidate set of screening standard values ​​for the working conditions; determine the original data time series under the first working conditions; based on the original data time series under the first working conditions and each of the M standard values, determine the first standard value among the M standard values, and determine the first standard value as the screening criterion for the working conditions of the fuel cell.

[0183] Furthermore, the second determining unit is used for:

[0184] Based on each standard value, the original data time series under the first working condition is filtered according to the set data filtering criteria to obtain the first dimensionless time series corresponding to each standard value; the variance and the number of remaining samples of the first dimensionless time series corresponding to each standard value are calculated; based on the variance and the number of remaining samples of the first dimensionless time series corresponding to each of the M standard values, the first standard value among the M standard values ​​is determined, and the first standard value is determined as the screening and judgment criterion for the working conditions of the fuel cell.

[0185] Furthermore, the third determining unit is used for:

[0186] Real-time sensor data that meets the screening criteria for the operating current range and operating conditions to be analyzed is input into the first formula to obtain the current health factor of the fuel cell. The first formula is used to characterize the relationship between the sensor data and the fuel cell health factor.

[0187] Furthermore, the third determining unit is used for:

[0188] The real-time sensor data that meets the screening criteria of the operating current range and operating conditions to be analyzed is processed by moving average filtering to obtain filtered sensor data; the filtered sensor data is then input into the first formula to obtain the current health factor of the fuel cell.

[0189] Furthermore, the fourth determining unit is used for:

[0190] Obtain the health factors of each fuel cell in the fuel cell health factor queue, which is a first-in-first-out queue; determine the health status of the fuel cells based on the health factors of each fuel cell in the fuel cell health factor queue.

[0191] The fuel cell health status assessment system disclosed in this embodiment is based on the fuel cell health status assessment method disclosed in the above embodiment, and will not be described again here.

[0192] This embodiment discloses a fuel cell health status assessment system that determines the operating current range to be analyzed for the fuel cell, establishes screening criteria for the fuel cell's operating conditions, obtains real-time sensor data of the fuel cell, and determines whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions. The real-time sensor data includes at least the fuel cell's performance index data. Based on the real-time sensor data that meets the screening criteria for the operating current range and operating conditions, a fuel cell health factor is determined, and based on the fuel cell health factor, the fuel cell's health status is determined. This scheme filters the real-time sensor data of the fuel cell. After filtering, it determines the fuel cell's health factor based on the real-time sensor data, thereby determining the fuel cell's health status. It achieves quantitative assessment of the fuel cell's health status through real-time sensor data, enabling adjustments to the control strategy based on the fuel cell's health status to ensure operating efficiency and service life.

[0193] This embodiment discloses an electronic device, the structural schematic diagram of which is shown below. Figure 13 As shown, it includes:

[0194] Processor 131 and memory 132.

[0195] The processor 131 is used to determine the operating current range to be analyzed for the fuel cell; determine the screening criteria for the operating conditions of the fuel cell; obtain real-time sensor data of the fuel cell and determine whether the real-time sensor data meets the operating current range to be analyzed and the screening criteria for the operating conditions, wherein the real-time sensor data includes at least the performance index data of the fuel cell; determine the health factors of the fuel cell based on the real-time sensor data that meets the screening criteria for the operating current range to be analyzed and the operating conditions; and determine the health status of the fuel cell based on the health factors.

[0196] The memory 132 is used to store the program for the processor to execute the above-mentioned processing.

[0197] The electronic device disclosed in this embodiment is based on the fuel cell health status assessment method disclosed in the above embodiments, and will not be described again here.

[0198] The electronic device disclosed in this embodiment determines the operating current range of the fuel cell to be analyzed, establishes screening criteria for the fuel cell's operating conditions, obtains real-time sensor data of the fuel cell, and determines whether the real-time sensor data meets the screening criteria for the operating current range and operating conditions. The real-time sensor data includes at least the fuel cell's performance index data. Based on the real-time sensor data that meets the screening criteria for the operating current range and operating conditions, a fuel cell health factor is determined, and based on the fuel cell health factor, the fuel cell's health status is determined. This scheme filters the real-time sensor data of the fuel cell. After filtering, it determines the fuel cell's health factor based on the real-time sensor data, thereby determining the fuel cell's health status. It realizes the ability to detect the fuel cell's performance index data under specific operating conditions through sensors, and to determine the fuel cell's health status based on the performance index data under specific operating conditions. This achieves a quantitative evaluation of the fuel cell's health status through real-time sensor data, so that the control strategy can be adjusted based on the fuel cell's health status, thereby ensuring operating efficiency and service life.

[0199] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0200] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0201] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0202] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for assessing the health status of a fuel cell, characterized in that, include: Determine the operating current variation range of the fuel cell; N current values ​​are selected from the operating current variation range to form a candidate set of operating currents; Determine the operating current range for each current value in the candidate set of operating currents; Determine the performance index data of the fuel cell within the operating current range for each current value; The performance index data of the fuel cell within the operating current range of each current value are respectively determined as the first time series, resulting in N first time series; Obtain the variance and life-cycle discrimination of each first time series; Based on the variance and full lifecycle discrimination of the N first time series, one first time series is determined from the N first time series. The operating current range corresponding to the determined first time series is determined as the operating current range to be analyzed, including: Let the first time series { The length of} is Define a range Calculate the physical quantities in the time series according to the following formula. Lifecycle differentiation: ； in, It is a positive integer, and its length is... For the first time series { There are a total of} Data, range Reflecting from the first time series { The number of discrete data selected; based on the fuel cell performance data, physical quantities The operating current range to be analyzed is set based on the principles of small fluctuations and high differentiation throughout its entire life cycle. Determine the range of variation of the screening criteria values ​​for the operating conditions of fuel cells; M standard values ​​are selected from the variation range of the working condition screening standard values ​​to form a candidate set of working condition screening standard values; Determine the original data time series under the first working condition; Based on each standard value, the original data time series under the first working condition is filtered according to the set data filtering criteria to obtain the first dimensionless time series corresponding to each standard value; Calculate the variance of the first dimensionless time series and the number of remaining samples corresponding to each standard value; Based on the variance of the first dimensionless time series corresponding to each of the M standard values ​​and the number of remaining samples, a first standard value is determined among the M standard values, and the first standard value is determined as the screening and judgment standard for the working conditions of the fuel cell; the first standard value is selected according to the basic principle of small variance and large number of remaining samples. Obtain real-time sensor data of the fuel cell, determine whether the real-time sensor data meets the operating current range to be analyzed and the operating condition screening criteria, wherein the real-time sensor data includes at least the performance index data of the fuel cell; The fuel cell health factor is determined based on real-time sensor data that meets the specified operating current range and the screening criteria for operating conditions. The health status of the fuel cell is determined based on the fuel cell health factors.

2. The method according to claim 1, characterized in that, The determination of fuel cell health factors based on real-time sensor data that meets the screening criteria for the operating current range to be analyzed and the operating conditions includes: Real-time sensor data that meets the operating current range to be analyzed and the operating condition screening criteria are input into the first formula to obtain the current health factor of the fuel cell. The first formula is used to characterize the relationship between the sensor data and the fuel cell health factor.

3. The method according to claim 2, characterized in that, The determination of fuel cell health factors based on real-time sensor data that meets the screening criteria for the operating current range to be analyzed and the operating conditions includes: The real-time sensor data that meets the specified operating current range and the specified operating condition screening criteria are subjected to moving average filtering to obtain filtered sensor data. The filtered sensor data is input into the first formula to obtain the current health factor of the fuel cell.

4. The method according to claim 1, characterized in that, Determining the health status of the fuel cell based on the fuel cell health factors includes: Obtain the health factors of each fuel cell in the fuel cell health factor queue, wherein the fuel cell health factor queue is a first-in-first-out queue. The health status of the fuel cell is determined based on the health factors of each fuel cell in the fuel cell health factor queue.

5. A fuel cell health status assessment system, characterized in that, include: The first determining unit is used to determine the operating current range of the fuel cell to be analyzed. The second determining unit is used to determine the screening criteria for the working conditions of the fuel cell. The judgment unit is used to obtain real-time sensor data of the fuel cell and determine whether the real-time sensor data meets the operating current range to be analyzed and the operating condition screening and judgment criteria. The real-time sensor data includes at least the performance index data of the fuel cell. The third determining unit is used to determine the fuel cell health factor based on real-time sensor data that meets the screening and judgment criteria of the working current range to be analyzed and the working conditions. The fourth determining unit is used to determine the health status of the fuel cell based on the fuel cell health factors; The first determining unit is specifically used for: Determine the operating current variation range of the fuel cell; N current values ​​are selected from the operating current variation range to form a candidate set of operating currents; Determine the operating current range for each current value in the candidate set of operating currents; Determine the performance index data of the fuel cell within the operating current range for each current value; The performance index data of the fuel cell within the operating current range of each current value are respectively determined as the first time series, resulting in N first time series; Obtain the variance and life-cycle discrimination of each first time series; Based on the variance and full lifecycle discrimination of the N first time series, one first time series is determined from the N first time series. The operating current range corresponding to the determined first time series is determined as the operating current range to be analyzed, including: Let the first time series { The length of} is Define a range Calculate the physical quantities in the time series according to the following formula. Lifecycle differentiation: ； in, It is a positive integer, and its length is... For the first time series { There are a total of} Data, range Reflecting from the first time series { The number of discrete data selected; based on the fuel cell performance data, physical quantities The operating current range to be analyzed is set based on the principles of small fluctuations and high differentiation throughout its entire life cycle. The second determining unit is specifically used for: Determine the range of variation of the screening criteria values ​​for the operating conditions of fuel cells; M standard values ​​are selected from the variation range of the working condition screening standard values ​​to form a candidate set of working condition screening standard values; Determine the original data time series under the first working condition; Based on each standard value, the original data time series under the first working condition is filtered according to the set data filtering criteria to obtain the first dimensionless time series corresponding to each standard value; Calculate the variance of the first dimensionless time series and the number of remaining samples corresponding to each standard value; Based on the variance of the first dimensionless time series corresponding to each of the M standard values ​​and the number of remaining samples, the first standard value among the M standard values ​​is determined, and the first standard value is determined as the screening and judgment standard for the working conditions of the fuel cell; the first standard value is selected according to the basic principle of small variance and large number of remaining samples.

6. An electronic device, characterized in that, include: The processor is used to determine the operating current variation range of the fuel cell. N current values ​​are selected from the operating current variation range to form a candidate set of operating currents; Determine the operating current range for each current value in the candidate set of operating currents; Determine the performance index data of the fuel cell within the operating current range for each current value; The performance index data of the fuel cell within the operating current range of each current value are respectively determined as the first time series, resulting in N first time series; Obtain the variance and life-cycle discrimination of each first time series; Based on the variance and full lifecycle discrimination of the N first time series, one first time series is determined from the N first time series. The operating current range corresponding to the determined first time series is determined as the operating current range to be analyzed, including: Let the first time series { The length of} is Define a range Calculate the physical quantities in the time series according to the following formula. Lifecycle differentiation: ； in, It is a positive integer, and its length is... For the first time series { There are a total of} Data, range Reflecting from the first time series { The number of discrete data selected; based on the fuel cell performance data, physical quantities The operating current range to be analyzed is set based on the principles of small fluctuations and high differentiation throughout its entire life cycle. Determine the range of variation of the screening criteria values ​​for the operating conditions of fuel cells; M standard values ​​are selected from the variation range of the working condition screening standard values ​​to form a candidate set of working condition screening standard values; Determine the original data time series under the first working condition; Based on each standard value, the original data time series under the first working condition is filtered according to the set data filtering criteria to obtain the first dimensionless time series corresponding to each standard value; Calculate the variance of the first dimensionless time series and the number of remaining samples corresponding to each standard value; Based on the variance of the first dimensionless time series corresponding to each of the M standard values ​​and the number of remaining samples, a first standard value is determined from the M standard values, and this first standard value is used as the screening criterion for the operating conditions of the fuel cell. The first standard value is selected based on the principle of small variance and large number of remaining samples. Real-time sensor data of the fuel cell is obtained, and it is determined whether the real-time sensor data meets the operating current range to be analyzed and the operating condition screening criterion. The real-time sensor data includes at least the performance index data of the fuel cell. A fuel cell health factor is determined based on the real-time sensor data that meets the operating current range to be analyzed and the operating condition screening criterion. The health status of the fuel cell is determined based on the fuel cell health factor. The memory is used to store the program that the processor executes during the processing.

Images