Intelligent detection system and method for fuel cell test jig

Abstract

The invention relates to the technical field of fuel cell detection, in particular to an intelligent detection system and method for a fuel cell test jig, and the system comprises a temperature gradient monitoring module, a power fluctuation analysis module, a health degree evaluation module, an abnormal trend recognition module and an intelligent early warning module. According to the invention, by accurately measuring the temperature data of the fuel cell and comparing the temperature difference, the multi-point temperature gradient monitoring is realized, the temperature balance is ensured, the local abnormity is reduced, the cooling response is dynamically adjusted, the voltage, current and temperature data are combined, the power fluctuation characteristics are accurately identified, the monitoring precision is improved, the internal resistance and the power reduction rate are analyzed, and the performance degradation is identified. The health state evaluation accuracy is improved, the short-time power change standard deviation is calculated, the load change influence is recognized, the fluctuation distribution rule is established, the abnormity recognition capability is improved, the early warning threshold value is set through the health state data, intelligent life early warning is achieved, and the system stability and safety are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of fuel cell detection, and in particular to an intelligent detection system and method for a fuel cell test stand. Background Art

[0002] The field of fuel cell testing technology includes the research and development, manufacturing, testing and application of fuel cell-related equipment and technologies. The core content of this technical field is to ensure the performance and reliability of fuel cell systems in actual applications, and to conduct comprehensive testing and monitoring of their various parameters. Fuel cell testing includes the detection of key performance indicators such as battery voltage, current, temperature, gas flow, etc., aiming to evaluate the battery's working status, efficiency and lifespan. The technical means involved in this field include sensor technology, data acquisition and analysis technology, automated detection systems and precise testing equipment.

[0003] Among them, the intelligent detection system of the fuel cell test stand refers to a technical system that integrates automated detection, data acquisition and real-time monitoring, which is specially used for fuel cell performance detection and evaluation. The system performs real-time detection on the working status and performance indicators of the fuel cell, and can be flexibly configured for different testing requirements. Its technical measures cover the monitoring of multiple parameters such as temperature, pressure, and gas flow, uses sensors and data acquisition for data processing, and realizes fault diagnosis and performance evaluation through an intelligent algorithm analysis system.

[0004] Existing technologies primarily rely on the measurement of static parameters, failing to fully consider the temperature gradient distribution of fuel cells under dynamic operating conditions. This makes it difficult to effectively identify areas of localized temperature anomalies, leading to delayed cooling system adjustments and increased risk of performance degradation. Traditional power fluctuation analysis relies on data acquisition within a single time period, failing to accurately extract power variation characteristics. Its ability to identify power fluctuation amplitudes and trends is limited, making it difficult to accurately predict power instability. Health assessments primarily rely on static test data, lacking in-depth analysis of power drop rates and internal resistance changes, making it difficult to effectively determine fuel cell performance degradation patterns. This results in inaccurate health status assessments. Existing detection systems lack standardized assessment methods for short-term power fluctuations in terms of identifying abnormal trends, failing to establish the dynamic distribution patterns of abnormal fluctuation areas, impacting the ability to identify abnormal conditions in fuel cell test racks. Health warnings primarily rely on fixed threshold settings and fail to dynamically adjust based on real-time data, resulting in delayed warning responses to fuel cell life degradation trends, impacting system safety and predictability. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an intelligent detection system and method for a fuel cell test stand.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical solution: an intelligent detection system for a fuel cell test stand comprises: The temperature gradient monitoring module, based on fuel cell temperature data, identifies the temperature difference at the measurement point and compares it with the fuel cell temperature gradient distribution. It analyzes the local temperature anomaly area and adjusts the coolant flow rate and air cooling fan speed. By adjusting the temperature change rate before and after, it determines the cooling adjustment effect and obtains the fuel cell temperature gradient offset. The power fluctuation analysis module extracts the fuel cell voltage, current, and temperature data based on the fuel cell temperature gradient offset, sets a power fluctuation monitoring window, extracts power fluctuation characteristics, identifies the amplitude and frequency of the fluctuation, and obtains the power dynamic change rate; The health assessment module calculates the fuel cell internal resistance change rate based on the power dynamic change rate, analyzes the fuel cell performance degradation process in combination with the power drop rate, compares the temperature data of the power drop area, determines the impact of local overheating on the battery health, and obtains the change in the fuel cell health status; The abnormal trend identification module evaluates the standard deviation of short-term power changes based on the change in the health status of the fuel cell, determines the impact of load mutations on power fluctuations, analyzes the frequency of occurrence of abnormal power fluctuation areas, and obtains the abnormal fluctuation trend of the fuel cell test rack.

[0007] As a further solution of the present invention, the fuel cell temperature gradient offset includes temperature difference distribution, local overheating area, and cooling adjustment deviation; the power dynamic change rate includes power fluctuation amplitude, fluctuation frequency, and fluctuation trend; the fuel cell health status change includes internal resistance change rate, power attenuation rate, and thermal imbalance effect; the fuel cell test stand abnormal fluctuation trend includes power short-term fluctuation deviation, voltage fluctuation range, current fluctuation range, and abnormal fluctuation frequency.

[0008] As a further solution of the present invention, the temperature gradient monitoring module includes: The temperature measurement data capture submodule extracts the temperature data of the fuel cell input end, reaction area, cooling channel, and exhaust end based on the fuel cell temperature data, detects the temperature change value of each measurement point, filters the valid measurement point data, eliminates the abnormal measurement point data, and calculates the temperature difference value of the measurement point; The temperature gradient calculation submodule analyzes the temperature gradient changes of the region based on the temperature difference of the measuring points and compares the temperature gradient of each region using the formula: ; Obtain temperature gradient distribution value; in, represents the temperature gradient distribution value, represents the temperature of the i-th measuring point, Indicates the The temperature of each measuring point, Represents the total number of measurement points, represents the weight coefficient of measurement point i, represents the temperature gradient adjustment parameter; The cooling adjustment judgment submodule calls the temperature gradient distribution value, analyzes the local temperature abnormality area, adjusts the coolant flow rate and the air cooling fan speed, monitors the temperature change rate before and after, and obtains the fuel cell temperature gradient offset.

[0009] As a further solution of the present invention, the power fluctuation analysis module includes: The fuel cell data extraction submodule extracts the fuel cell voltage, current, and temperature data based on the fuel cell temperature gradient offset, filters the valid data set, removes invalid and abnormal data, extracts the key parameter data of the battery within the key time window, and obtains the fuel cell data sequence; The power fluctuation identification submodule sets a power fluctuation monitoring window based on the fuel cell data sequence, extracts power data within the window, identifies the fluctuation amplitude and frequency, compares the fluctuation characteristics under the differentiated time windows, determines the fluctuation trend change, and obtains the power fluctuation characteristic parameters; The power change rate identification submodule calls the power fluctuation characteristic parameters, extracts the numerical changes of the fluctuation trend, sets a benchmark value for comparison, and judges the dynamic change of power using the formula: ; Calculate and obtain the dynamic rate of change of power; in, Represents the dynamic rate of change of power, Representative time The fuel cell power value at the moment, Representative time The fuel cell power value at the moment, Represents the voltage change, represents the change in current, Represents the change in temperature gradient offset.

[0010] As a further solution of the present invention, the health assessment module includes: The internal resistance change rate calculation submodule monitors the internal impedance change rate of the fuel cell based on the dynamic power change rate, extracts the internal impedance values in differentiated time periods, analyzes the corresponding change amplitudes, and evaluates the impedance change per unit time using the formula: ; Calculate and obtain the fuel cell internal resistance change rate; in, represents the rate of change of the internal resistance of the fuel cell, Representative Moment The internal impedance of the fuel cell at Representative Moment The internal impedance of the fuel cell at Represents the number of time points during the fuel cell operation period, Representative The internal impedance value at a time point, represents the average internal impedance value during the period; The power decay analysis submodule uses the fuel cell power data to call the fuel cell internal resistance change rate, identifies the power decline in differentiated time periods, extracts the power change rate per unit time, and analyzes the power decline trend to obtain the fuel cell power decline rate; The local overheating risk assessment submodule calls the fuel cell power reduction rate, compares the temperature data of the differentiated power reduction area, identifies the regional temperature gradient, determines whether the local temperature change exceeds the set range, identifies the impact of the local overheating area, extracts the local overheating impact coefficient, and obtains the change in the fuel cell health status.

[0011] The local overheating risk assessment submodule calls the fuel cell power reduction rate, compares the temperature data of the differentiated power reduction area, identifies the regional temperature gradient, determines whether the local temperature change exceeds the set range, identifies the impact of the local overheating area, extracts the local overheating impact coefficient, and obtains the change in the fuel cell health status.

[0012] As a further solution of the present invention, the abnormal trend identification module includes: The short-term power distribution identification submodule extracts the fluctuation data of the fuel cell power in a short period of time based on the change in the health status of the fuel cell, calculates the standard deviation of the power change in the time window, screens the abnormal power change values in the key time window, and obtains the short-term power change distribution interval; The load impact judgment submodule calls the short-term power change distribution interval, compares the change amplitude of the fuel cell voltage and current, evaluates the load change rate within the short time window, and combines the standard deviation of the power change to judge the impact of the load mutation on the power fluctuation using the formula: ; Calculate and obtain the load mutation impact factor; in, represents the load mutation impact factor, Representative Moment The voltage, Indicates time The voltage, Representative Moment The current, Representative Moment The current, Represents the standard deviation of power variation within a short time window, is the load adjustment factor, Represents the load mutation rate; The power fluctuation frequency analysis submodule calls the load mutation influencing factor, counts the frequency of abnormal power fluctuation areas in the fuel cell test rack within the key time window, compares the distribution of abnormal power fluctuation areas, and obtains the abnormal fluctuation trend of the fuel cell test rack.

[0013] As a further solution of the present invention, the system also includes an intelligent early warning module: The intelligent early warning module compares the abnormal fluctuation growth rate with the health status change based on the abnormal fluctuation trend of the fuel cell test rack, sets the health warning threshold according to the health status change of the fuel cell test rack, and performs the fuel cell test rack life warning; The fuel cell test stand life warning includes abnormal fluctuation growth rate, health decay rate, and warning threshold.

[0014] As a further solution of the present invention, the intelligent early warning module includes: The fluctuation trend analysis submodule extracts voltage fluctuation, temperature change and pressure amplitude values based on the abnormal fluctuation trend of the fuel cell test rack, analyzes the growth rate of each fluctuation parameter, and screens the time interval with abnormal fluctuation growth rate using the formula: ; Calculate the fluctuation growth rate and obtain the abnormal range of the fluctuation growth rate; in, represents the fluctuation growth rate, Representative The voltage fluctuation value of the fuel cell test stand at each moment, Indicates the The voltage fluctuation value of the fuel cell test stand at each moment, Representative The temperature change value of the fuel cell test stand at each moment, Indicates the The temperature change value of the fuel cell test stand at each moment, represents the total number of time series, is the voltage fluctuation weight coefficient, is the temperature change adjustment coefficient, Adjust parameters for fluctuation normalization; The health status assessment submodule calls the abnormal interval of the fluctuation growth rate, analyzes the health status of the fuel cell test rack in the corresponding period, determines the extent of the health status decline, extracts the key features of the health status change, and obtains the health status evolution features; The life warning setting submodule calls the health status evolution characteristics, calculates the health warning threshold, sets the fuel cell test stand life warning benchmark, compares the health status evolution characteristics with the set threshold, determines the warning level, and executes the fuel cell test stand life warning.

[0015] The intelligent detection method of the fuel cell test stand is performed based on the intelligent detection system of the fuel cell test stand, and includes the following steps: S1: Based on the fuel cell temperature data, extract the temperature measurement data of the fuel input end, reaction area, cooling channel, and exhaust end, compare the fuel cell temperature gradient distribution, determine the temperature difference characteristics of the local area, adjust the coolant flow rate and air cooling fan speed, compare the temperature change rate before and after the adjustment, and obtain the fuel cell temperature gradient offset; S2: Based on the fuel cell temperature gradient offset, extract the fuel cell voltage, current, and temperature data, set a power fluctuation monitoring window, extract the power fluctuation characteristics, calculate the fluctuation amplitude and frequency, compare the power change trend, and obtain the power dynamic change rate; S3: Based on the dynamic power change rate, extract the fuel cell internal resistance change rate, compare the load change data of the fuel cell test stand, analyze the fuel cell performance degradation process, extract the temperature data corresponding to the power drop area, analyze the impact of local temperature rise on battery loss, and obtain the change in the fuel cell health status; S4: Based on the change in the fuel cell health status, comparing the battery voltage and current change amplitudes of the fuel cell test rack, analyzing the impact of the sudden load change on the power fluctuation of the fuel cell test rack, counting the frequency of occurrence of abnormal power fluctuation areas, and obtaining the abnormal fluctuation trend of the test rack; S5: Based on the abnormal fluctuation trend of the test rack, calculate the abnormal fluctuation growth rate, compare the health status change, set the health warning threshold, compare the threshold with the current status, and obtain the fuel cell test rack life warning information.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: The present invention uses precise fuel cell temperature data measurement and temperature difference comparison to monitor multi-point temperature gradients across the fuel input, reaction zone, cooling channel, and exhaust port, ensuring balanced temperature distribution across all fuel cell regions and effectively reducing localized temperature anomalies. Dynamic adjustment of the coolant flow rate and air cooling fan speed optimizes cooling responsiveness and reduces performance degradation caused by temperature imbalance. Combining fuel cell voltage, current, and temperature data, a power fluctuation monitoring window is constructed to extract power variation characteristics, enabling precise identification of power fluctuation amplitude, frequency, and trend, improving the accuracy and stability of dynamic power monitoring. The dynamic power change rate is used to analyze the fuel cell internal resistance variation trend. Combined with the power drop rate, performance degradation patterns are identified, quantifying the impact of localized overheating on fuel cell lifespan and improving the accuracy of health status assessment. By calculating the standard deviation of short-term power variation and combining it with the voltage and current fluctuation amplitudes, the impact of sudden load changes on power stability is identified. The distribution pattern of abnormal power fluctuation areas is established, improving the comprehensiveness and reliability of abnormal trend identification. Combined with fuel cell health status change data, the abnormal fluctuation growth rate is compared to set health warning thresholds, enabling intelligent fuel cell lifespan warning and improving the long-term stability and safety of the fuel cell system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a system flow chart of the present invention; Figure 2 This is a flow chart of the temperature gradient monitoring module in the present invention; Figure 3 This is a flow chart of the power fluctuation analysis module in the present invention; Figure 4 This is a flow chart of the health assessment module in the present invention; Figure 5 This is a flow chart of the abnormal trend identification module in the present invention; Figure 6 This is a flow chart of the intelligent early warning module in the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0020] See also Figure 1 The present invention provides a technical solution: an intelligent detection system for a fuel cell test stand includes: Based on the fuel cell temperature data, the temperature gradient monitoring module extracts temperature measurement data from the fuel input end, reaction area, cooling channel, and exhaust end. It identifies the temperature difference at the measurement point and compares it with the fuel cell temperature gradient distribution. It analyzes the local temperature anomaly area and adjusts the coolant flow rate and air cooling fan speed. By adjusting the temperature change rate before and after, it determines the cooling adjustment effect and obtains the fuel cell temperature gradient offset. The power fluctuation analysis module extracts fuel cell voltage, current, and temperature data based on the fuel cell temperature gradient offset, sets a power fluctuation monitoring window, extracts power fluctuation characteristics, identifies the amplitude and frequency of fluctuations, determines the power fluctuation trend, and obtains the power dynamic change rate; The health assessment module calculates the fuel cell internal resistance change rate based on the dynamic power change rate. Combined with the power drop rate, it analyzes the fuel cell performance degradation process. By comparing the temperature data in the power drop area, it determines the impact of local overheating on battery health and obtains the change in the fuel cell health status. The abnormal trend identification module evaluates the standard deviation of short-term power changes based on the change in fuel cell health status, compares the battery voltage and current change amplitudes, determines the impact of load mutations on power fluctuations, analyzes the frequency of abnormal power fluctuation areas, and obtains abnormal fluctuation trends of the fuel cell test rack; The intelligent early warning module is based on the abnormal fluctuation trend of the fuel cell test rack, compares the abnormal fluctuation growth rate with the change in health status, sets the health warning threshold through the change in health status of the fuel cell test rack, and executes the fuel cell test rack life warning.

[0021] The fuel cell temperature gradient offset includes temperature difference distribution, local overheating area, and cooling adjustment deviation. The power dynamic change rate includes power fluctuation amplitude, fluctuation frequency, and fluctuation trend. The fuel cell health status change includes internal resistance change rate, power attenuation rate, and thermal imbalance impact. The fuel cell test stand abnormal fluctuation trend includes power short-term fluctuation deviation, voltage fluctuation range, current fluctuation range, and abnormal fluctuation frequency. The fuel cell test stand life warning includes abnormal fluctuation growth rate, health attenuation rate, and warning threshold.

[0022] See also Figure 2 , the temperature gradient monitoring module includes: The temperature measurement data capture submodule extracts the temperature data of the fuel cell input end, reaction area, cooling channel, and exhaust end based on the fuel cell temperature data, detects the temperature change value of each measurement point, filters the valid measurement point data, eliminates the abnormal measurement point data, and calculates the temperature difference value of the measurement point; First, temperature data from different parts of the fuel cell (such as the fuel input end, reaction area, cooling channel and exhaust end) is collected to ensure the accuracy and representativeness of the data. For example, in actual industrial applications, high-precision temperature sensors can be installed at key locations of the fuel cell to monitor its temperature changes, thereby effectively controlling and preventing overheating or insufficient cooling. In this way, the operator can monitor the temperature readings of each sensor in real time and determine whether there are abnormal temperature points. For example, if a sensor suddenly reads too high, it indicates that there is a blockage or leakage problem in the fuel cell at that location. For the data, the operator needs to filter the data to filter out abnormal data caused by faults or external interference. For example, if the temperature value of a sensor suddenly differs from the temperature of the sensor by more than a set threshold, such as 10°C, this is an abnormal point and should be eliminated from the data analysis. Then, the temperature difference is calculated using the remaining valid data to identify the temperature distribution in the entire fuel cell, thereby obtaining the temperature difference value of the measuring point.

[0023] The temperature gradient calculation submodule analyzes the temperature gradient changes in the region based on the temperature difference of the measuring points and compares the temperature gradient of each region using the formula: ; Obtain temperature gradient distribution value; in, represents the temperature gradient distribution value, represents the temperature of the i-th measuring point, Indicates the The temperature of each measuring point, Represents the total number of measurement points, represents the weight coefficient of measurement point i, represents the temperature gradient adjustment parameter; Calculate the temperature gradient distribution of the entire fuel cell. In practical applications, the temperature gradient is evaluated by analyzing the temperature difference between different measuring points. This helps to determine which areas need adjustment due to insufficient cooling. By introducing weight coefficients , adjust the contribution of each measuring point in the overall temperature gradient calculation according to its location and impact on performance. For example, the temperature monitoring point in the reaction area is more important than that in the cooling channel, so it can be given a higher weight. Assume there are 5 measuring points and the temperature readings are 、 、 、 、 , the weights are set to 、 、 、 、 , adjust the parameters It is mainly used to correct the denominator to avoid the distribution evaluation error caused by a small number of measurement points. , use the following formula to calculate the temperature gradient distribution: ; The calculation process is as follows: ; This approach not only enhances the applicability of the model, but also makes the calculation of the temperature gradient more in line with the needs of the actual operating state, thereby accurately obtaining the temperature gradient distribution value. ,This result provides a quantitative basis for subsequent cooling strategies, ,and proves the practicality of the model and the necessity of ,adjustment.

[0024] The cooling adjustment judgment submodule calls the temperature gradient distribution value, analyzes the local temperature abnormality area, adjusts the coolant flow rate and air cooling fan speed, monitors the temperature change rate before and after, and obtains the fuel cell temperature gradient offset; Using data to judge and adjust cooling settings is the key to ensuring that the fuel cell system operates in the best working state. For example, in specific operations, by comparing the calculated temperature gradient distribution with the set safe or optimal operating range, the operator can determine whether it is necessary to adjust the coolant flow rate or fan speed. This adjustment is based on the monitoring results of the temperature change rate before and after. For example, if the temperature gradient of a certain area is detected to exceed the preset maximum safety gradient of 5°C / m, the coolant flow rate of the area needs to be increased, while the flow rate of the adjacent area needs to be reduced to achieve a more uniform temperature distribution. Through this dynamic adjustment, the fuel cell temperature gradient offset is finally obtained. The offset is a direct indicator for evaluating the adjustment effect. If the offset is reduced, it proves that the cooling system adjustment is effective, thereby ensuring the stability and efficiency of the fuel cell system.

[0025] See also Figure 3 , the power fluctuation analysis module includes: The fuel cell data extraction submodule extracts fuel cell voltage, current, and temperature data based on the fuel cell temperature gradient offset, filters valid data sets, removes invalid and abnormal data, extracts key battery parameter data within the critical time window, and obtains the fuel cell data sequence; In the fuel cell monitoring used in electric vehicles, the voltage, current and temperature data of the battery are monitored in real time during driving, and a time window is set to collect data every 5 minutes to analyze the working status of the battery at different speeds and ambient temperatures. By comparing the temperature gradient offset in historical data, signs of battery performance degradation are discovered in a timely manner. In this process, the baseline threshold of the data set must be determined first. For example, the threshold for abnormal current fluctuations is set to ±10%, the voltage fluctuation threshold is set to ±5%, and the temperature fluctuation threshold is set to ±2°C. Valid monitoring data is filtered out according to the threshold, and abnormal data is marked and excluded, thereby ensuring the accuracy and efficiency of subsequent analysis and obtaining the fuel cell data sequence.

[0026] The power fluctuation identification submodule sets a power fluctuation monitoring window based on the fuel cell data sequence, extracts the power data within the window, identifies the fluctuation amplitude and frequency, compares the fluctuation characteristics under differentiated time windows, determines the fluctuation trend change, and obtains the power fluctuation characteristic parameters; When using fuel cells for load regulation in grid energy storage, it is necessary to accurately monitor the battery's power output fluctuations and determine whether the power output is stable based on real-time data. The monitoring window is set to 10 minutes, and power data is recorded once a minute. By analyzing data changes within 10 minutes, the average amplitude and frequency of power fluctuations are calculated. If the power fluctuation is found to exceed the set threshold of ±5%, further trend analysis is performed to predict future performance degradation or failure. This analysis can help operators adjust control strategies in a timely manner, ensure efficient and stable operation of the system, and obtain power fluctuation characteristic parameters.

[0027] The power change rate identification submodule calls the power fluctuation characteristic parameters, extracts the numerical changes of the fluctuation trend, sets the benchmark value for comparison, and judges the dynamic change of power using the formula: ; Calculate and obtain the dynamic rate of change of power; in, Represents the dynamic rate of change of power, Representative time The fuel cell power value at the moment, Representative time The fuel cell power value at the moment, Represents the voltage change, represents the change in current, Represents the change value of the temperature gradient offset; Temperature gradient offset change: This value is obtained by continuously measuring and comparing temperature data at different measurement points in the fuel cell (such as the fuel input, reaction area, cooling channel, and exhaust), calculating the temperature difference and temperature gradient distribution between each measurement point, and dynamically adjusting the coolant flow rate and fan speed, monitoring the temperature change rate before and after the adjustment. This process is based on comparing temperature data at different times and is therefore time-dependent. Voltage and current changes: These two values are extracted from the fuel cell's voltage and current data through the power fluctuation analysis module. A time window is set to monitor power fluctuation characteristics, identify the amplitude and frequency of the fluctuation, and calculate the dynamic rate of power change. The calculation of these rates of change depends on the power, temperature, and current data before and after the moment, and is therefore also time-dependent. How to obtain the change value: The value is calculated by setting a specific time window, collecting data in real time over a continuous period of time, and using a mathematical formula. For example, the power change rate calculation formula explicitly includes the change values of power value, voltage change, current change, and temperature gradient offset at different times, emphasizing the impact of time on data changes; In fuel cell power management, real-time monitoring of power fluctuations and calculation of their dynamic change rate are performed. For a fuel cell with a rated power of 2000W, at a certain moment , the power measurement value is , at the next moment The measured power value is , the measured voltage changed from 48V to 49V, the current changed from 32A to 33A, and the temperature changed from 25℃ to 26℃; Enter specific data to calculate: ; Calculate the power dynamic change rate This value is used to evaluate the fluctuation of fuel cell power and compare it with the preset reference value. If the fuel consumption is higher than the set safety threshold (such as 30), regulatory measures need to be taken, such as adjusting the fuel supply rate or optimizing the power output strategy. If the value is lower than the set lower limit (such as 10), it indicates that the fuel cell is operating stably and there is no need to adjust the output parameters. This calculation process is applicable to scenarios such as smart microgrids and fuel cell vehicles to ensure that power fluctuations are within a reasonable range and to calculate the dynamic power change rate.

[0028] See also Figure 4 , the health assessment module includes: The internal resistance change rate calculation submodule monitors the internal impedance change rate of the fuel cell based on the dynamic power change rate, extracts the internal impedance value in different time periods, analyzes the corresponding change amplitude, and evaluates the impedance change per unit time using the formula: ; Calculate and obtain the fuel cell internal resistance change rate; in, represents the rate of change of the internal resistance of the fuel cell, Representative Moment The internal impedance of the fuel cell at Representative Moment The internal impedance of the fuel cell at Represents the number of time points during the fuel cell operation period, Representative The internal impedance value at a time point, represents the average internal impedance value during the period; The process of calling the fuel cell operating time series data and calculating the impedance change per unit time involves collecting and analyzing the time series data to obtain the internal impedance of the battery at different time points. The data is recorded by online monitoring and stored in the database. Assume that in a certain fuel cell, the internal impedance values monitored for 10 days are: ; According to the calculation formula of impedance change rate: ; First calculate the mean impedance: ; Then calculate the impedance variance: ; Its square root is: ; Assume that the impedance change rate on day 3 and day 2 is calculated as follows: ; Similarly, the impedance change rate can be calculated for another day, ultimately yielding the fuel cell internal resistance change rate. This calculated value is used for subsequent analysis. For example, in battery management, this change rate can be used to determine whether the battery is experiencing abnormal attenuation. If the impedance change rate continues to rise, it indicates battery aging. Engineers can then take maintenance measures in advance based on the data to ensure stable fuel cell operation.

[0029] The power decay analysis submodule uses the fuel cell power data and the fuel cell internal resistance change rate to identify power declines in differentiated time periods, extract the power change rate per unit time, and analyze the power decline trend to obtain the fuel cell power decline rate. Power data is obtained from battery management. The data includes power readings at each time point. Assume that the battery power drops from 100 watts to 90 watts during continuous operation. The change is related to battery aging or changes in external conditions. The power drop rate is calculated. By comparing the power values at adjacent time points, a clear downward trend can be obtained, which helps to detect performance problems in a timely manner and take measures to adjust them. To illustrate with a specific numerical example, assume that the power drops from 100 watts to 95 watts and then to 90 watts within two hours, and the fuel cell power drop rate is obtained. The drop rate is critical for evaluating the health of the battery and predicting future performance degradation. The analysis of the drop rate not only helps technicians monitor the immediate status of the battery, but also serves as an important indicator for evaluating the long-term performance of the battery.

[0030] The local overheating risk assessment submodule uses the fuel cell power drop rate to compare the temperature data of the differentiated power drop areas, identify the regional temperature gradient, determine whether the local temperature change exceeds the set range, identify the impact of the local overheating area, extract the local overheating impact coefficient, and obtain the change in the fuel cell health status; Relevant temperature data is extracted from battery management. The data is displayed in different parts of the battery module and obtained through real-time monitoring or periodic data recording. For example, during a day's operation, it is observed that the temperature of some modules in the battery pack continues to rise. This is due to the failure of internal components or the influence of ambient temperature. The temperature data is compared with the power reduction data to analyze the correlation between the two. A specific numerical example is used to illustrate. Assume that when the power of a module drops by 10%, its temperature rises from the normal 30 degrees to 40 degrees. The obvious temperature change indicates that the module has an overheating problem. By calculating the temperature gradient and setting the threshold, the severity of the overheating and the affected area can be more accurately judged. The analysis of the judgment of local overheating impact helps the maintenance team to perform fault diagnosis and necessary maintenance work in a timely manner, thereby ensuring the safe operation and long-term performance of the battery pack and obtaining the change in the health status of the fuel cell.

[0031] See also Figure 5 ,The abnormal trend identification module includes: The short-term power distribution identification submodule extracts the fluctuation data of fuel cell power in a short period of time based on the change in fuel cell health status, calculates the standard deviation of power change within the time window, filters out abnormal power change values within the key time window, and obtains the short-term power change distribution interval; By continuously recording the output power of the fuel cell through real-time monitoring equipment, extracting the maximum and minimum power values within each minute, and calculating the power fluctuation rate within this interval, an intuitive understanding of the fuel cell's response speed and stability in actual operation can be obtained. For example, in a typical bus fuel cell test, the power fluctuation data is monitored every minute, and the standard deviation of the data is calculated. By comparing the standard deviations of different time windows (such as morning and evening peaks), the performance of the fuel cell under different load conditions can be evaluated. To further analyze the data, the standard deviation is used to filter out abnormal power change values within a specific time window. The value exceeds a set threshold, such as setting 2 times the standard deviation as the abnormal threshold. This method can quickly identify periods of equipment failure or performance degradation, and ultimately obtain the short-term power change distribution range, and archive the data for subsequent maintenance decision support.

[0032] The load impact judgment submodule calls the short-term power change distribution interval, compares the change amplitude of the fuel cell voltage and current, evaluates the load change rate within the short time window, and combines the standard deviation of the power change to judge the impact of the load mutation on the power fluctuation. The formula is: ; Calculate and obtain the load mutation impact factor; in, represents the load mutation impact factor, Representative Moment The voltage, Indicates time The voltage, Representative Moment The current, Representative Moment The current, Represents the standard deviation of power variation within a short time window, is the load adjustment factor, Represents the load mutation rate; The sampling period of the fuel cell operation is set to 1 second. The voltage, current, and power values are recorded once per second to form a time series data set. Assuming that within a certain time window (10 seconds), the voltage of the fuel cell drops from 72V to 68V and the current increases from 120A to 135A, the voltage change is: ; The current change is: ; At the same time, the power change standard deviation of the time window is obtained from the short-term power change data. Assuming that the power standard deviation of the window is 3.2kW, the load mutation rate The calculation method is the load change within a unit time (10 seconds): ; Among them, set the load adjustment coefficient , the load mutation impact factor is calculated as follows: ; The calculated load mutation impact factor F=5.88 can be used to further determine the degree of load mutation impact on the fuel cell within this time window. If this value exceeds a certain set threshold, for example, the set threshold is 5.0, it can be determined that the current fuel cell load mutation impact is relatively severe and additional adjustment or optimization is required to ensure the stability of the fuel cell under sudden load conditions. The calculated result is the load mutation impact factor, which will be used for subsequent abnormal fluctuation trend analysis.

[0033] The power fluctuation frequency analysis submodule calls the load mutation influencing factor to count the frequency of abnormal power fluctuation areas in the fuel cell test rack within the critical time window, compares the distribution of abnormal power fluctuation areas, and obtains the abnormal fluctuation trend of the fuel cell test rack; The frequency of abnormal power fluctuation areas in the fuel cell test rack within a specific time window is counted, and the distribution of abnormal power fluctuation areas is compared. For example, after one month of continuous monitoring, the analyzed data shows that power fluctuations frequently occur during specific time periods every day (such as 2 pm to 4 pm). The data is analyzed using statistical methods to determine the frequency of abnormal fluctuations and their corresponding time periods. This analysis helps the technical team determine the maintenance time window or conduct further system optimization, and ultimately obtains the abnormal fluctuation trend of the fuel cell test rack, providing data support for future operations.

[0034] See also Figure 6 , the intelligent early warning module includes: The fluctuation trend analysis submodule extracts voltage fluctuations, temperature changes, and pressure amplitude values based on the abnormal fluctuation trend of the fuel cell test rack, analyzes the growth rate of each fluctuation parameter, and screens the time interval with abnormal fluctuation growth rate using the formula: ; Calculate the fluctuation growth rate and obtain the abnormal range of the fluctuation growth rate; in, represents the fluctuation growth rate, Representative The voltage fluctuation value of the fuel cell test stand at each moment, Indicates the The voltage fluctuation value of the fuel cell test stand at each moment, Representative The temperature change value of the fuel cell test stand at each moment, Indicates the The temperature change value of the fuel cell test stand at each moment, represents the total number of time series, is the voltage fluctuation weight coefficient, is the temperature change adjustment coefficient, Adjust parameters for fluctuation normalization; In the fuel cells of the automotive industry, data can be monitored in real time through sensors. For example, the operating voltage of a certain type of fuel cell fluctuates between 72V and 80V, the temperature range is between 60℃ and 80℃, and the pressure amplitude is between 0.5MPa and 0.9MPa. The parameters can be collected once per second in the fuel cell through the built-in sensor and recorded in the monitoring. After the sensor data is stored, it can be used to analyze the operating status and abnormal fluctuation trends of the fuel cell. During the test, the voltage fluctuation value is calculated by calculating the voltage change between any two consecutive time points. For example, When the voltage is 75V, When the voltage becomes 76V, the voltage change at this moment is , the temperature change and pressure amplitude change are calculated in the same way. Through the change value, the growth rate of each fluctuation parameter can be calculated. The growth rate is obtained by comparing the real-time data with the measurement values before and after. Assume that the experimental data are as follows: ; ; set up , , , the total number of time series ; Calculate in sequence: The first set of data ( ): Voltage fluctuations ; Temperature fluctuations ; Weighted calculation: ; The second set of data ( ): Voltage fluctuations ; Temperature fluctuations ; Weighted calculation: ; The third set of data ( ): Voltage fluctuations ; Temperature fluctuations ; Weighted calculation: ; Calculate the composite volatility growth rate: ; Finally, the fluctuation growth rate is calculated is 1.58. Based on this calculated value, we can filter out the time interval with abnormal fluctuation growth rate. For example, we can set the threshold ,when When it is greater than this threshold, it is marked as an abnormal interval, and the data point is judged to be abnormal, and the abnormal interval of the fluctuation growth rate is obtained.

[0035] The health status assessment submodule calls the abnormal interval of the fluctuation growth rate, analyzes the health status of the fuel cell test rack in the corresponding period, determines the extent of the health status decline, extracts the key features of the health status change, and obtains the health status evolution characteristics; The health status parameters of the fuel cell test stand for the corresponding time period are calculated. The health status parameters are calculated using multiple sensor data sets. For example, in the fuel cell module, the health status parameters include the number of charge and discharge cycles of the battery, the voltage difference between battery cells, etc. The data is automatically aggregated through data collection to calculate the health status change. The decline in health status is calculated by comparing historical data with current data. The extraction of key features, such as the maximum voltage drop rate of the battery cell, provides engineers with a basis for judging the health status of the battery, and ultimately obtains the health status change characteristic.

[0036] The life warning setting submodule calls the health status evolution characteristics, calculates the health warning threshold, sets the fuel cell test stand life warning benchmark, compares the health status evolution characteristics with the set threshold, determines the warning level, and executes the fuel cell test stand life warning; Calculate the health warning threshold. The setting of the warning threshold is based on the comparative analysis of the characteristic quantity of health status change and historical performance data. By setting the threshold, you can pre-define what level of change will trigger the warning, which is very critical for early diagnosis and maintenance. For example, in a fuel cell test stand, if the battery performance drops below a certain threshold, an automatic warning will be issued, prompting the maintenance department to inspect or replace it. Comparing the health status change with the set threshold can help the maintenance team respond in time to avoid greater losses. The warning level is determined by real-time comparison of the set standard and the measured data to obtain the life warning level and execute the fuel cell test stand life warning.

[0037] The intelligent detection method of the fuel cell test rack is performed based on the intelligent detection system of the fuel cell test rack, and includes the following steps: S1: Based on the fuel cell temperature data, extract the temperature measurement data of the fuel input end, reaction area, cooling channel, and exhaust end, compare the fuel cell temperature gradient distribution, determine the temperature difference characteristics of the local area, adjust the coolant flow rate and air cooling fan speed, compare the temperature change rate before and after the adjustment, and obtain the fuel cell temperature gradient offset; S2: Based on the fuel cell temperature gradient offset, extract the fuel cell voltage, current, and temperature data, set the power fluctuation monitoring window, extract the power fluctuation characteristics, calculate the fluctuation amplitude and frequency, compare the power change trend, and obtain the power dynamic change rate; S3: Based on the dynamic power change rate, extract the fuel cell internal resistance change rate, compare it with the load change data of the fuel cell test rack, analyze the fuel cell performance degradation process, extract the temperature data corresponding to the power drop area, analyze the impact of local temperature rise on battery loss, and obtain the change in the fuel cell health status; S4: Based on the change in fuel cell health status, compare the battery voltage and current change amplitudes of the fuel cell test rack, analyze the impact of sudden load changes on the power fluctuation of the fuel cell test rack, count the frequency of abnormal power fluctuation areas, and obtain the abnormal fluctuation trend of the test rack; S5: Based on the abnormal fluctuation trend of the test rack, calculate the abnormal fluctuation growth rate, compare the health status change, set the health warning threshold, compare the threshold with the current status, and obtain the fuel cell test rack life warning information.

[0038] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. An intelligent detection system for a fuel cell test stand, characterized in that: The system comprises: The temperature gradient monitoring module, based on fuel cell temperature data, identifies the temperature difference at the measurement point and compares it with the fuel cell temperature gradient distribution. It analyzes the local temperature anomaly area and adjusts the coolant flow rate and air cooling fan speed. By adjusting the temperature change rate before and after, it determines the cooling adjustment effect and obtains the fuel cell temperature gradient offset. The power fluctuation analysis module extracts the fuel cell voltage, current, and temperature data based on the fuel cell temperature gradient offset, sets a power fluctuation monitoring window, extracts power fluctuation characteristics, identifies the amplitude and frequency of the fluctuation, and obtains the power dynamic change rate; The health assessment module calculates the fuel cell internal resistance change rate based on the power dynamic change rate, analyzes the fuel cell performance degradation process in combination with the power drop rate, compares the temperature data of the power drop area, determines the impact of local overheating on the battery health, and obtains the change in the fuel cell health status; The abnormal trend identification module evaluates the standard deviation of short-term power changes based on the change in the health status of the fuel cell, determines the impact of load mutations on power fluctuations, analyzes the frequency of occurrence of abnormal power fluctuation areas, and obtains the abnormal fluctuation trend of the fuel cell test rack.

2. The intelligent detection system for a fuel cell test stand according to claim 1, characterized in that: The fuel cell temperature gradient offset includes temperature difference distribution, local overheating area, and cooling adjustment deviation; the power dynamic change rate includes power fluctuation amplitude, fluctuation frequency, and fluctuation trend; the fuel cell health status change includes internal resistance change rate, power attenuation rate, and thermal imbalance impact; the fuel cell test stand abnormal fluctuation trend includes power short-term fluctuation deviation, voltage fluctuation range, current fluctuation range, and abnormal fluctuation frequency.

3. The intelligent detection system for a fuel cell test stand according to claim 1, characterized in that: The temperature gradient monitoring module includes: The temperature measurement data capture submodule extracts the temperature data of the fuel cell input end, reaction area, cooling channel, and exhaust end based on the fuel cell temperature data, detects the temperature change value of each measurement point, filters the valid measurement point data, eliminates the abnormal measurement point data, and calculates the temperature difference value of the measurement point; The temperature gradient calculation submodule analyzes the temperature gradient changes of the region based on the temperature difference of the measuring points and compares the temperature gradient of each region using the formula: ； Obtain temperature gradient distribution value; in, represents the temperature gradient distribution value, represents the temperature of the i-th measuring point, Indicates the The temperature of each measuring point, Represents the total number of measurement points, represents the weight coefficient of measurement point i, represents the temperature gradient adjustment parameter; The cooling adjustment judgment submodule calls the temperature gradient distribution value, analyzes the local temperature abnormality area, adjusts the coolant flow rate and the air cooling fan speed, monitors the temperature change rate before and after, and obtains the fuel cell temperature gradient offset.

4. The intelligent detection system for a fuel cell test stand according to claim 3, characterized in that: The power fluctuation analysis module includes: The fuel cell data extraction submodule extracts the fuel cell voltage, current, and temperature data based on the fuel cell temperature gradient offset, filters the valid data set, removes invalid and abnormal data, extracts the key parameter data of the battery within the key time window, and obtains the fuel cell data sequence; The power fluctuation identification submodule sets a power fluctuation monitoring window based on the fuel cell data sequence, extracts power data within the window, identifies the fluctuation amplitude and frequency, compares the fluctuation characteristics under the differentiated time windows, determines the fluctuation trend change, and obtains the power fluctuation characteristic parameters; The power change rate identification submodule calls the power fluctuation characteristic parameters, extracts the numerical changes of the fluctuation trend, sets a benchmark value for comparison, and judges the dynamic change of power using the formula: ； Calculate and obtain the dynamic rate of change of power; in, Represents the dynamic rate of change of power, Representative time The fuel cell power value at the moment, Representative time The fuel cell power value at the moment, Represents the voltage change, represents the change in current, Represents the change in temperature gradient offset.

5. The intelligent detection system for a fuel cell test stand according to claim 4, characterized in that: The health assessment module includes: The internal resistance change rate calculation submodule monitors the internal impedance change rate of the fuel cell based on the dynamic power change rate, extracts the internal impedance values in differentiated time periods, analyzes the corresponding change amplitudes, and evaluates the impedance change per unit time using the formula: ； Calculate and obtain the fuel cell internal resistance change rate; in, represents the rate of change of the internal resistance of the fuel cell, Representative Moment The internal impedance of the fuel cell at Representative Moment The internal impedance of the fuel cell at Represents the number of time points during the fuel cell operation period, Representative The internal impedance value at a time point, represents the average internal impedance value during the period; The power decay analysis submodule uses the fuel cell power data to call the fuel cell internal resistance change rate, identifies the power decline in differentiated time periods, extracts the power change rate per unit time, and analyzes the power decline trend to obtain the fuel cell power decline rate; The local overheating risk assessment submodule calls the fuel cell power reduction rate, compares the temperature data of the differentiated power reduction area, identifies the regional temperature gradient, determines whether the local temperature change exceeds the set range, identifies the impact of the local overheating area, extracts the local overheating impact coefficient, and obtains the change in the fuel cell health status.

6. The intelligent detection system for a fuel cell test stand according to claim 5, characterized in that: The abnormal trend identification module includes: The short-term power distribution identification submodule extracts the fluctuation data of the fuel cell power in a short period of time based on the change in the health status of the fuel cell, calculates the standard deviation of the power change in the time window, screens the abnormal power change values in the key time window, and obtains the short-term power change distribution interval; The load impact judgment submodule calls the short-term power change distribution interval, compares the change amplitude of the fuel cell voltage and current, evaluates the load change rate within the short time window, and combines the standard deviation of the power change to judge the impact of the load mutation on the power fluctuation using the formula: ； Calculate and obtain the load mutation impact factor; in, represents the load mutation impact factor, Representative Moment The voltage, Indicates time The voltage, Representative Moment The current, Representative Moment The current, Represents the standard deviation of power variation within a short time window, is the load adjustment factor, Represents the load mutation rate; The power fluctuation frequency analysis submodule calls the load mutation influencing factor, counts the frequency of abnormal power fluctuation areas in the fuel cell test rack within the key time window, compares the distribution of abnormal power fluctuation areas, and obtains the abnormal fluctuation trend of the fuel cell test rack.

7. The intelligent detection system for a fuel cell test stand according to claim 1, characterized in that: The system also includes an intelligent early warning module: The intelligent early warning module compares the abnormal fluctuation growth rate with the health status change based on the abnormal fluctuation trend of the fuel cell test rack, sets the health warning threshold according to the health status change of the fuel cell test rack, and performs the fuel cell test rack life warning; The fuel cell test stand life warning includes abnormal fluctuation growth rate, health decay rate, and warning threshold.

8. The intelligent detection system for a fuel cell test rack according to claim 7, characterized in that: The intelligent early warning module includes: The fluctuation trend analysis submodule extracts voltage fluctuation, temperature change and pressure amplitude values based on the abnormal fluctuation trend of the fuel cell test rack, analyzes the growth rate of each fluctuation parameter, and screens the time interval with abnormal fluctuation growth rate using the formula: ； Calculate the fluctuation growth rate and obtain the abnormal range of the fluctuation growth rate; in, represents the fluctuation growth rate, Representative The voltage fluctuation value of the fuel cell test stand at each moment, Indicates the The voltage fluctuation value of the fuel cell test stand at each moment, Representative The temperature change value of the fuel cell test stand at each moment, Indicates the The temperature change value of the fuel cell test stand at each moment, represents the total number of time series, is the voltage fluctuation weight coefficient, is the temperature change adjustment coefficient, Adjust parameters for fluctuation normalization; The health status assessment submodule calls the abnormal interval of the fluctuation growth rate, analyzes the health status of the fuel cell test rack in the corresponding period, determines the extent of the health status decline, extracts the key features of the health status change, and obtains the health status evolution features; The life warning setting submodule calls the health status evolution characteristics, calculates the health warning threshold, sets the fuel cell test stand life warning benchmark, compares the health status evolution characteristics with the set threshold, determines the warning level, and executes the fuel cell test stand life warning.

9. An intelligent detection method for a fuel cell test stand, characterized in that: The intelligent detection system of the fuel cell test stand according to any one of claims 1 to 8 comprises the following steps: S1: Based on the fuel cell temperature data, extract the temperature measurement data of the fuel input end, reaction area, cooling channel, and exhaust end, compare the fuel cell temperature gradient distribution, determine the temperature difference characteristics of the local area, adjust the coolant flow rate and air cooling fan speed, compare the temperature change rate before and after the adjustment, and obtain the fuel cell temperature gradient offset; S2: Based on the fuel cell temperature gradient offset, extract the fuel cell voltage, current, and temperature data, set a power fluctuation monitoring window, extract the power fluctuation characteristics, calculate the fluctuation amplitude and frequency, compare the power change trend, and obtain the power dynamic change rate; S3: Based on the dynamic power change rate, extract the fuel cell internal resistance change rate, compare the load change data of the fuel cell test stand, analyze the fuel cell performance degradation process, extract the temperature data corresponding to the power drop area, analyze the impact of local temperature rise on battery loss, and obtain the change in the fuel cell health status; S4: Based on the change in the fuel cell health status, comparing the battery voltage and current change amplitudes of the fuel cell test rack, analyzing the impact of the sudden load change on the power fluctuation of the fuel cell test rack, counting the frequency of occurrence of abnormal power fluctuation areas, and obtaining the abnormal fluctuation trend of the test rack; S5: Based on the abnormal fluctuation trend of the test rack, calculate the abnormal fluctuation growth rate, compare the health status change, set the health warning threshold, compare the threshold with the current status, and obtain the fuel cell test rack life warning information.

Images