Energy management system and method for a hydrogen fuel cell module

Abstract

The present application relates to the technical field of battery energy management, in particular to an energy management system and method for hydrogen fuel cell module, the system comprises: a monitoring and collecting module, based on the voltage parameter, current parameter and temperature parameter of the hydrogen fuel cell module, the sensor readings are collected and compared, the original monitoring data are obtained, the numerical comparison of the power distribution network load demand is carried out, and the supply-demand matching information is generated, in the present application, the differential statistics and difference operation improve the accuracy of energy distribution, so that the matching between energy distribution and actual load demand is more accurate, energy waste is reduced and overall energy efficiency is improved, frequency band extraction and rate recording can capture the subtle changes of battery output, and response speed is improved, in addition, the analysis and adjustment of phase feature group provide adaptive ability, and the stability and reliability of the change of power grid load are enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery energy management, and particularly relates to an energy management system and method of a hydrogen fuel cell module. BACKGROUND

[0002] The energy management system of the hydrogen fuel cell module is a comprehensive energy monitoring and regulation system specially designed for hydrogen fuel cells, aiming to maximize the performance and efficiency of the battery while ensuring its safe and stable operation. The system monitors key parameters of the hydrogen fuel cell such as voltage, current and temperature, and adjusts energy output in real time to adapt to different load demands.

[0003] However, the existing technology has obvious shortcomings in dynamic load change response and energy output adjustment. The output strategy is usually fixed, limiting the flexibility in responding to sudden high load demands. In high load burst events, the output may not be adjusted effectively due to insufficient response, resulting in energy supply shortage or over-discharge, affecting the performance and life of the battery. Therefore, improvements are needed. SUMMARY

[0004] The purpose of the present application is to solve the shortcomings in the prior art and propose an energy management system and method of a hydrogen fuel cell module.

[0005] In order to achieve the above-mentioned purpose, the present application adopts the following technical solution: an energy management system of a hydrogen fuel cell module comprises:

[0006] A monitoring and collecting module based on voltage parameters, current parameters and temperature parameters of the hydrogen fuel cell module collects sensor readings and performs comparison to obtain original monitoring data, compares the load demand of the power distribution network, generates supply-demand matching information, synchronously merges the original monitoring data and the supply-demand matching information, arranges them in time sequence and checks for abnormal values to obtain a preliminary state sequence;

[0007] A dynamic allocation module based on the preliminary state sequence performs interval division, difference statistics and difference operation on the current parameters, temperature parameters and voltage parameters to obtain energy allocation basis, performs correlation comparison, discharge path screening, multi-dimensional coordinate comparison and abnormal value screening on the energy allocation basis and the load demand of the power distribution network to establish discharge allocation values;

[0008] A high-frequency regulation module based on the discharge allocation values performs frequency band extraction and rate recording on the output power of the hydrogen fuel cell to obtain power segmented data, compares the power segmented data with the load change of the power distribution network and measures the phase to obtain a phase feature group, performs frequency screening and amplitude adjustment on the phase feature group to establish a high-frequency adjustment signal;

[0009] The health management module continuously monitors the temperature parameter and output power of the hydrogen fuel cell, judges the amplitude and compares with the safety threshold based on the high-frequency adjustment signal, obtains safety monitoring information, analyzes the state and compares the index of the safety monitoring information and the residual life parameter, and obtains life evaluation data.

[0010] Preferably, the step of obtaining the preliminary state sequence is:

[0011] The voltage parameter, current parameter and temperature parameter of the hydrogen fuel cell are received, and data fusion is performed with the supply-demand matching information, time sequence arrangement is performed on the collected data points through time stamp, and a time-ordered data set is obtained;

[0012] Based on the time-ordered data set, the average value and standard deviation of each parameter are calculated, abnormal values outside the range of twice the standard deviation of the average value are identified and removed, and a data set processed by abnormal values is obtained;

[0013] Based on the data set processed by abnormal values, the maximum value, minimum value, mean value and coefficient of variation of each parameter are calculated, and a preliminary state sequence is formed.

[0014] Preferably, the step of obtaining the energy distribution basis is:

[0015] Based on the preliminary state sequence, interval division is performed on each parameter, threshold values are set according to the historical data range of each parameter and the current operating environment, the parameters are divided into multiple working intervals, each interval represents a different operating state, and a parameter interval data set is generated;

[0016] According to the parameter interval data set, interval difference statistical values are calculated, and the calculation formula is:

[0017]

[0018] Wherein, represents the value of the i-th parameter in the j-th interval, represents the average value of the i-th parameter in the interval, is the interval difference statistical value of the i-th parameter, and N is the number of data points in each interval;

[0019] Based on the interval difference statistical values of each parameter, the sizes of the interval difference values of each parameter are compared, the relative change rates and fluctuation characteristics between the parameters are analyzed, and an energy distribution basis is formed.

[0020] Preferably, the step of obtaining the discharge allocation value is:

[0021] Based on the energy allocation basis, a comparison is performed with the load demand of the distribution network. By analyzing the difference between energy output and demand in each period, the periods when demand is exceeded or not met are identified to form a preliminary difference data set;

[0022] Based on the preliminary difference data set, the overall matching degree between energy output and demand in each period is calculated using the following formula:

[0023]

[0024] in, is the energy output in period i, is the load demand in period i, is the total number of periods evaluated, Indicates the overall matching degree;

[0025] Based on the overall matching degree, discharge path screening and multi-dimensional coordinate comparison are performed, and the discharge strategy is optimized in combination with historical data. After screening out abnormal values, the discharge allocation value is formed.

[0026] Preferably, the steps of obtaining the phase feature group are:

[0027] Based on the discharge allocation value, by setting the time window, measuring the power peak and valley in each window, and obtaining the power segmentation data;

[0028] Comparing the power segment data with the load change data of the power distribution network, performing time synchronization on each power segment, and obtaining an aligned comparison result;

[0029] Based on the alignment comparison result, the phase between power and load is measured, and the periodicity and consistency of the changes are quantified by analyzing the time offset and phase difference between the power change and the load change to form a phase feature group.

[0030] Preferably, the step of obtaining the high-frequency adjustment signal is:

[0031] Based on the phase feature group, performing frequency division screening on the data, selecting the frequency interval with the maximum energy response, and obtaining filtered frequency data;

[0032] According to the filtered frequency data, the adjusted total amplitude is calculated using the following formula:

[0033]

[0034] in, represents the response amplitude of the ith frequency, is the maximum response amplitude among all frequencies, represents the corresponding time delay, is a decay factor, is a total frequency number, is an adjusted total amplitude;

[0035] Based on the adjusted total amplitude, the output power of the hydrogen fuel cell is optimized, the high-frequency signal is adjusted, the real-time load demand of the power grid is matched, and a high-frequency adjustment signal is formed.

[0036] Preferably, the step of obtaining safety monitoring information is:

[0037] Based on the high-frequency adjustment signal, real-time data points per minute are recorded for the temperature parameters and output power of the hydrogen fuel cell, including peak values, valley values and average values, to obtain real-time monitoring data;

[0038] Using the real-time monitoring data, it is checked whether the temperature and power of each data point exceed the predetermined safety threshold, the out-of-limit data is marked, and the time stamp and parameter value of the out-of-limit event are recorded to form a preliminary abnormal result;

[0039] According to the preliminary abnormal result, the marked out-of-limit event is analyzed to verify the severity and risk level of the out-of-limit event, and safety monitoring information is formed.

[0040] Preferably, the step of obtaining life evaluation data is:

[0041] Based on the safety monitoring information, the current state of the hydrogen fuel cell is analyzed, events and performance trends associated with safety are recorded, including temperature exceeding events and power fluctuations, and a state analysis result is generated;

[0042] According to the state analysis result, the percentage of remaining life is calculated, and the calculation formula is:

[0043]

[0044] wherein, represents the safety score of the i-th safety monitoring event, is a weight factor of the i-th safety monitoring event, is the total number of events, represents the estimated percentage of remaining life;

[0045] Based on the percentage of remaining life, the overall health status and expected life of the hydrogen fuel cell module are evaluated to form life evaluation data.

[0046] The present application provides an energy management method for a hydrogen fuel cell module, comprising the following steps:

[0047] Based on the voltage parameter, current parameter and temperature parameter of the hydrogen fuel cell, raw monitoring data is collected from sensor readings, compared with the load demand of the power distribution network, supply-demand matching information is generated, and the raw monitoring data and the supply-demand matching information are synchronized, merged, time-sequenced and abnormal value checked to obtain a preliminary state sequence;

[0048] Based on the preliminary state sequence, interval division, difference statistics and difference operation are performed on the current parameter, temperature parameter and voltage parameter to obtain energy distribution basis, and the energy distribution basis is associated and compared with the load demand of the power distribution network to establish discharge adjustment values through discharge path screening, multi-dimensional coordinate comparison and abnormal value screening.

[0049] Based on the discharge adjustment values, frequency band extraction and rate recording are performed on the output power of the hydrogen fuel cell to obtain power segmented data, which is compared and phase measured with the load change of the power distribution network to obtain a phase feature group, and the phase feature group is frequency-screened and amplitude-adjusted to establish a high-frequency adjustment signal.

[0050] Based on the high-frequency adjustment signal, the temperature parameter and output power of the hydrogen fuel cell are continuously monitored, and the safety monitoring information is obtained through amplitude limiting and safety threshold comparison, and the safety monitoring information and the remaining life parameter are analyzed and compared to obtain life evaluation data.

[0051] Based on the life evaluation data, the health management of the hydrogen fuel cell module is performed, and the battery state is evaluated and monitored through the battery temperature, output power and safety monitoring information.

[0052] Compared with the prior art, the advantages and positive effects of the present application are:

[0053] In the present application, difference statistics and difference operation improve the accuracy of energy distribution, making the matching between energy distribution and actual load demand more accurate, reducing energy waste and improving overall energy efficiency. Frequency band extraction and rate recording can capture subtle changes in battery output, improving response speed. In addition, the analysis and adjustment of the phase feature group provide adaptive ability, enhancing the stability and reliability of the grid load change. Comprehensive monitoring and safety threshold comparison enhance safety, and the introduction of life evaluation makes maintenance decisions more scientific, prolonging the service life of the equipment. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 The system flowchart of the present application. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0056] Please refer to Figure 1 The present application provides a technical scheme: an energy management system of a hydrogen fuel cell module comprises:

[0057] The monitoring acquisition module acquires sensor readings and performs comparison based on the voltage parameter, current parameter and temperature parameter of the hydrogen fuel cell module to obtain original monitoring data, performs numerical comparison on the load demand of the power distribution network, generates supply-demand matching information, synchronously merges the original monitoring data and the supply-demand matching information, arranges them in time sequence and investigates abnormal values to obtain a preliminary state sequence.

[0058] The dynamic allocation module performs interval division, difference statistics and difference operation on the current parameter, temperature parameter and voltage parameter based on the preliminary state sequence to obtain energy allocation basis, performs associated comparison, discharge path screening, multi-dimensional coordinate comparison and abnormal value screening on the energy allocation basis and the load demand of the power distribution network to establish discharge allocation values.

[0059] The high-frequency regulation module performs frequency band extraction and rate recording on the output power of the hydrogen fuel cell based on the discharge allocation values to obtain power segmented data, compares the power segmented data with the load change of the power distribution network and measures the phase to obtain a phase feature group, performs frequency screening and amplitude adjustment on the phase feature group to establish a high-frequency adjustment signal.

[0060] The health management module performs continuous monitoring, amplitude limiting judgment and safety threshold comparison on the temperature parameter and output power of the hydrogen fuel cell based on the high-frequency adjustment signal to obtain safety monitoring information, performs state analysis and index comparison on the safety monitoring information and the remaining life parameter to obtain life evaluation data.

[0061] The obtaining step of the preliminary state sequence is:

[0062] The voltage parameter, current parameter and temperature parameter of the hydrogen fuel cell are received, and data fusion is performed with the supply-demand matching information, the collected data points are arranged in time sequence through time stamp to obtain a data set arranged in time sequence;

[0063] Based on the data set arranged in time sequence, the average value and standard deviation of each parameter are calculated, abnormal values exceeding twice the standard deviation of the average value are identified and removed to obtain a data set processed by abnormal values;

[0064] Based on the data set processed by abnormal values, the maximum value, minimum value, mean value and coefficient of variation of each parameter are calculated to form a preliminary state sequence.

[0065] Specifically, after setting sensors capable of measuring 0V to 24V voltage, 0A to 10A current, and 0℃ to 90℃ temperature range, the voltage, current, and temperature information generated by the hydrogen fuel cell in different working stages are recorded respectively, and the supply-demand matching information of the power distribution network is processed correspondingly. First, the sensor collection frequency and the synchronization reference time are determined, each record is compared with the previously established measurement range, for example, the temperature is compared with the 0℃ to 90℃ range, the voltage is compared with the 0V to 24V range, and the current is compared with the 0A to 10A range. When the data is within the preset range, it is directly associated with the subsequent calculation. When a certain data item exceeds any threshold, the source of the threshold needs to be determined first. For example, the threshold is set based on the combination of previous experimental results and industry experience, including additional safety margin. If it is found that it is not within the safety range, the parameter needs to be re-verified and the sensor needs to be reset or recalibrated if necessary. To ensure that the supply-demand matching information can be used for fusion, the time scale of the supply-demand matching information needs to be sorted first. For example, the external load information is aligned with the time label in whole seconds or whole minutes, and then compared with the timestamp in the sensor data. During the comparison process, the supply-demand records of different time periods are interpolated or merged to form a consistent time sequence index. Then, the voltage, current, and temperature parameters of the hydrogen fuel cell are associated with the aligned supply-demand matching information and merged into the same sequence structure. Then, all data points are reordered from the earliest to the latest according to the timestamp, and the time order information of each record is marked. Finally, the time-ordered data set is obtained.

[0066] After obtaining the time-ordered data set, determine the parameters to be processed, including the voltage parameters, current parameters, and temperature parameters obtained based on the previous parameters. Compare the historical range of these parameters with the current measured value range, for example, the temperature corresponds to 0℃ to 90℃, the current corresponds to 0A to 10A, and the voltage corresponds to 0V to 24V. When all data is within the respective range, the average value and standard deviation of each parameter are calculated. Here, the average value is obtained by dividing the sum of samples within the same time window by the total number of samples, and the standard deviation is obtained by squaring the difference between each sample and the average value, and then dividing the sum by the total number of samples minus one. To distinguish data points with obvious abnormal fluctuations, each data point is compared with the average value. If its deviation exceeds twice the standard deviation of the average value, it is considered to be outside the experience range. The "twice standard deviation" threshold is derived from the statistical results of the fluctuation of the hydrogen fuel cell running data in the past, and is conservatively corrected by referring to the safety requirements of the equipment. Therefore, when a data point is detected to exceed the set threshold, it is marked as abnormal. Then, these data points marked as abnormal are directly excluded from the subsequent calculation, only the unmarked data points are retained, and finally the overall data after excluding the abnormal values is re-summarized and recorded to obtain the data set after abnormal value processing.

[0067] After obtaining the data set after the outlier processing, the voltage parameter, the current parameter and the temperature parameter remaining therein are continuously segmented and counted to calculate the maximum value, the minimum value, the average value and the coefficient of variation of each parameter. For the maximum value and the minimum value, each time interval can be screened in the previous range to determine, for example, when processing the temperature, all measured results in the range of 0°C to 90°C are searched one by one, and the highest value and the lowest value appearing therefrom are selected, and the coefficient of variation can be obtained using the ratio of the standard deviation of the parameter to the average value thereof. If the standard deviation is and the average value is the coefficient of variation can be defined as Each symbol in the calculation formula is obtained under the statistical definition, wherein is the square sum of the deviation of all sampling points of the parameter from the average value divided by the square root of the sampling quantity minus one, and is the total sum of the sampling divided by the sampling quantity. Combining the statistical results of the voltage, the current and the temperature three parameters, the four important indicators of the maximum value, the minimum value, the average value and the coefficient of variation of each parameter can be recorded, and finally the data of these indicators are summarized to obtain the preliminary state sequence.

[0068] The acquisition steps of the energy distribution basis are as follows:

[0069] Based on the preliminary state sequence, interval division is performed on each parameter, threshold values are set according to the historical data range of each parameter and the current operating environment, the parameters are divided into multiple working intervals, each interval represents a different operating state, and a parameter interval data set is generated;

[0070] According to the parameter interval data set, interval difference statistical values are calculated, and the calculation formula is as follows:

[0071]

[0072] wherein, represents the value of the i-th parameter in the j-th interval, represents the average value of the i-th parameter in the interval, is the interval difference statistical value of the i-th parameter, and N is the number of data points in each interval;

[0073] Based on the interval difference statistical values of each parameter, the sizes of the interval difference values of the parameters are compared, the relative change rates and the fluctuation characteristics between the parameters are analyzed, and the energy distribution basis is formed.

[0074] Specifically, based on the preliminary state sequence obtained before, the division of multiple working intervals is carried out for each parameter. Firstly, the historical data range of the parameters needs to be determined, for example, the effective range of temperature is set to 0-90℃, the effective range of voltage is set to 0-24V, and the effective range of current is set to 0-10A. These ranges are obtained by statistical analysis of a large amount of data recorded during multiple start-ups and continuous operation of the hydrogen fuel cell, and are formed by adding a safety margin on the basis of the minimum and maximum temperatures, the upper and lower limits of voltage and the current extreme values allowed by the comprehensive equipment. In addition, the upper and lower limits of each parameter are also determined in combination with the current operating environmental factors such as ambient temperature, air pressure and load fluctuation. For example, the working temperature of the hydrogen fuel cell measured in high-altitude areas is often closer to its maximum allowable temperature value. In order to prevent the parameter from exceeding the historical limit, the temperature threshold is moderately adjusted to 85℃ to avoid overheating. The working interval determination method for current and voltage is similar to that for temperature, and the upper limits of current and voltage are set to 9A and 23V respectively as the actual measurement threshold. Then, according to the collection time and value distribution of the parameters, the parameters are divided into multiple working intervals, such as temperature intervals of 0-30℃, 30-60℃ and 60-85℃, current intervals of 0-3A, 3-6A and 6-9A, and voltage intervals of 0-8V, 8-16V and 16-23V. The relevant measurement times and data fluctuation distributions in each interval are recorded synchronously. Finally, the division results are unified and arranged, and the upper and lower boundaries of the intervals and the corresponding interval numbers are identified to generate the parameter interval data set.

[0075] The symbol parameters of the formula are introduced as follows, represents the sampling value of the th parameter in the th interval, which is a value recorded by detecting the temperature or voltage or current in the corresponding interval multiple times, represents the average value of the th parameter in the interval, which is obtained by summing all samples and dividing by the total number of samples after multiple observations in the same interval, is the number of data points in the interval, which is obtained by counting the number of data corresponding to the time stamp in the same interval.

[0076] Calculation process: the observation value of the temperature parameter in a certain interval is , which is recorded as 25℃, 28℃, 30℃, 29℃ and 27℃. Then is equal to:

[0077]

[0078] Then is calculated:

[0079] ​

[0080]

[0081]

[0082] Then put in the denominator , we can get:

[0083]

[0084] Finally, we get the square root Right now:

[0085]

[0086] The results show that the interval difference statistics can measure the fluctuation amplitude of the parameters within the interval. The larger the value, the more obvious the temperature fluctuation within the interval. For other parameters such as current or voltage, the same method can be used to perform interval difference calculation and obtain corresponding results.

[0087] Based on the interval difference statistics obtained above, it is necessary to further compare the difference values ​​of each parameter in each interval. Here, the temperature, voltage, and current in different intervals are collected. The results are listed one by one, such as the temperature in each interval 、 、 , the voltage is obtained in each interval 、 、 , the current in each interval is obtained 、 、 During the comparison process, the differential statistics of the same type of parameters in all intervals need to be arranged in the same order. For example, the numerical values ​​of temperature are checked one by one in the order from the first interval to the third interval, then to the three intervals of current, and finally to the three intervals of voltage. Through numerical comparison, it can be found that the temperature differential value reaches 2.5 in the second interval, which has exceeded the preset maximum safety fluctuation threshold of 2.4. This threshold is obtained by averaging the temperature fluctuation records in multiple intervals during long-term monitoring of the equipment and adding a safety factor of 0.2. When it is detected that this threshold is exceeded, it is necessary to re-determine whether an overheating problem has occurred and record the abnormality. The same process is applied to other parameters. When the voltage or current differential value exceeds the reference threshold of 1.5 and 2.0, the corresponding interval number and measurement time are recorded. Then, after the differential value comparison of all intervals is completed, based on the fluctuation rate information of temperature, current, and voltage in different intervals and their relative change characteristics, this information is summarized and the fluctuation characteristics of each interval are marked to form the basis for energy allocation.

[0088] The obtaining step of the discharge allocation value is:

[0089] Based on the energy allocation basis, a comparison with the load demand of the power distribution network is performed, the periods of demand exceeding and not being met are identified by analyzing the energy output and demand difference in each period, and a preliminary difference data set is formed;

[0090] According to the preliminary difference data set, the overall matching degree of energy output and demand of each period is calculated, and the calculation formula is:

[0091]

[0092] Among them, is the energy output of the i-th period, is the load demand of the i-th period, is the total number of periods evaluated, indicates the overall matching degree;

[0093] Based on the overall matching degree, the discharge path screening and multi-dimensional coordinate comparison are performed, the discharge strategy is optimized combined with historical data, and after excluding abnormal values, the discharge allocation value is formed.

[0094] Specifically, based on the energy allocation basis obtained in the foregoing, the energy output data of the hydrogen fuel cell in each period and the load demand data of the power distribution network are obtained, the actual load demand value of each period is first searched according to the time sequence, and then the hydrogen fuel cell energy output value recorded in each period is compared. Each record is compared with the predetermined numerical interval, for example, the energy output value is set to 0kWh to 10kWh interval, and the load demand value is set to 0kWh to 12kWh interval. These interval ranges are obtained through multiple observations and empirical statistics. When it is detected that the energy output value is less than the lower limit or exceeds the upper limit, it is necessary to further determine whether the collection time is misaligned or the equipment is malfunctioning. If it indeed deviates from the maximum allowable range, it is marked as abnormal and the data and time label are recorded. Continue to compare the data of the remaining periods. If it is found that the output value is higher than the demand value of the corresponding period, it is identified as a demand exceeding period, and these periods are collectively classified into the "output surplus period" list. Conversely, if the output value is lower than the demand value, it is identified as an unmet period, and these periods are collectively classified into the "output shortage period" list. By summarizing the above two lists, it is determined whether there is a continuous surplus or shortage in the adjacent periods, and the detailed difference degree is listed period by period, and finally a preliminary difference data set is formed.

[0095] The beneficial effect of the formula is that by introducing This difference ratio and combining exponential mapping, the matching of energy output and demand can be nonlinearly converted, thereby avoiding the problem that simple linear comparison cannot reflect the data distribution characteristics.

[0096] The acquisition step is to record the actual output energy of each time period of the hydrogen fuel cell, read the values by the on-site installed electric energy metering device with a sampling period of half an hour or one hour, count and store the collected output values of each time period, and obtain ; The acquisition step is to calculate the average load value by monitoring the real-time load demand of the power distribution network in the same time interval, and correct it to obtain a more accurate load demand value combined with the peak and valley values, to form ; The acquisition step is to count the total number of all valid time periods in the evaluation period of the day or week, for example, starting from 0 o'clock every day, one time period every 1 hour, ending at 24 o'clock, 24 time periods can be obtained, and more time periods can be obtained for a whole week, and finally accumulated into .

[0097] Calculation process: let the total number of time periods evaluated be , the energy output , , , and the load demand , ,

[0098] First, calculate the first time period :

[0099]

[0100] The corresponding index term , so ;

[0101] Then calculate the second time period:

[0102]

[0103] , ;

[0104] Calculate the third time period:

[0105]

[0106] , ;

[0107] Sum the three and divide by 3:

[0108]

[0109] The result shows that the overall matching degree is about 0.5567, if the value is between 0.5 and 0.6, it means that the output and demand still have a certain gap, and a more refined scheduling strategy is needed, when the value is higher than 0.8, it means that the matching degree of output and demand is higher.

[0110] Based on the overall matching degree obtained in the foregoing, the discharge path is screened and compared with the multi-dimensional coordinates, and the load data of each period in the historical record and the output value of the hydrogen fuel cell corresponding to the period are arranged in advance, for example, each day is divided into 24 whole point periods, and the load demand and output value of each period are recorded by mutual comparison, then in the comparison process of multiple periods, some abnormal data exceeding the predetermined fluctuation threshold are identified, for example, the upper and lower limits of voltage are set to 0V to 24V, if the voltage of a period exceeds 25V, it means that there may be collection error in measurement, and the record needs to be marked and excluded, and the key data such as power output are checked in the same way, the threshold is set by statistical analysis of historical measurement data, and the battery type and design parameters are considered, then after the abnormal records are screened out, the current, voltage, temperature and load change and other indicators are arranged uniformly by using the remaining multi-dimensional coordinate data, the data of adjacent periods are represented by horizontal coordinates and vertical coordinates, and the change trend is judged, if there is a serious deviation in any key period, the period needs to be further verified, finally, the discharge rule is compared with the historical data, and the temperature and pressure data collected under different seasons and operating conditions are combined to correct the energy output arrangement, and the discharge deployment value is obtained by system summarization.

[0111] The acquisition step of the phase feature group is:

[0112] Based on the discharge deployment value, the power peak and trough in each window are measured by setting a time window, and power segmentation data is obtained;

[0113] According to the power segmentation data, the load change data of the power distribution network are compared, the time synchronization of each power segmentation is obtained, and the aligned comparison result is obtained;

[0114] Based on the aligned comparison result, the phase measurement between power and load is performed, the time offset and phase difference between power change and load change are analyzed, the periodicity and consistency of change are quantified, and the phase feature group is formed.

[0115] Specifically, based on the discharge allocation value obtained in the foregoing, firstly, the entire process needs to be divided into several time windows according to the actual running time, and then the power output data of the hydrogen fuel cell in each time window is recorded. In order to ensure the accuracy of the data, a fixed time interval, for example, 1 minute or 5 minutes, is usually set as the length of each window, and the output power generated by the hydrogen fuel cell at different time points in the window is collected. Then, the maximum value and the minimum value are retrieved from the collected output power data one by one, and are marked as power peak and power valley respectively. When the maximum value or the minimum value in some windows fluctuates extremely, it is also necessary to verify whether the threshold setting is reasonable. The threshold is generally determined by collecting the running data of previous tests and referring to the actual performance of the hydrogen fuel cell under different load conditions. In order to maintain the accuracy of power measurement, the signal sampling frequency also needs to be set. For example, when monitoring the power output in the range of 0kW to 10kW, the high power is included in the abnormal range and recorded. According to the time position of the peak and valley recorded in each window, these segmented information is summarized. If it is found that the extreme value data in some windows is out of the safe range, for example, exceeds the predetermined upper limit of 10kW or is lower than 0kW, it is necessary to judge whether there is instrument error or failure in combination with the previously statistical power monitoring upper limit data. Finally, the peak and valley values of each time window are arranged in order to obtain the power segmented data.

[0116] According to the power segmented data obtained in the foregoing, it needs to be corresponded one by one with the load change data of the power distribution network and time aligned. Firstly, the load change data needs to be collected and ensured to have the same or convertible time mark. For example, when 1 minute is recorded as the period, the average value or instantaneous value of the load change in the period should be obtained, and the time stamp consistent with the power segmented data is added. Then, the time fields in the two data sources are sorted. If there is a time period missing or record delay, it is necessary to determine whether to interpolate or replace with the last time record, otherwise it may cause deviation in the comparison process. For the reference range of load change used in these steps, such as 0kW to 12kW, the appropriate upper and lower limits can be determined based on historical running statistics when setting the threshold. The power peak and valley values of each window in the alignment process are respectively compared with the load change in the corresponding period. If there is a power difference or load difference exceeding the upper limit, this period is marked as an abnormal period and the detailed time is recorded. If the comparison result is in the reasonable interval, the original record is continued. Finally, the alignment is completed and the comparison result is output. When the time sequence and the value are fully matched, the aligned data can be further summarized for subsequent measurement analysis.

[0117] Based on the aligned comparison results, the phase measurement between power and load is carried out, which needs to extract the fluctuation period of power in each time window and the variation period of load in the corresponding period, and to offset and compare them on the same time axis. In order to obtain the time offset between power and load, each record needs to be compared with the effective range set in advance, for example, the power is compared with the range of 0kW to 10kW, and the load is compared with the range of 0kW to 12kW. The positions of the peak and valley points of each are found in the graphical mode or numerical sequence, and these peak and valley points are marked and aligned with each other, and then the time difference between them is calculated. Then the phase difference is calculated to quantify the difference, for example, the peak value appears 10 seconds apart in a certain period, or 0.1 cycles apart when the frequency unit is millihertz. These difference values can be converted into phase angles for recording. After completing the statistics of all time windows, the power and load phase differences of each window are grouped and sorted, and the periodicity and consistency indexes are calculated according to the fluctuation law. If the phase difference of some windows is too large, it is necessary to check whether there is a delay configuration problem in the power output or load acquisition link. Finally, the phase feature group is formed according to the analysis results of periodicity and consistency.

[0118] The acquisition step of the high-frequency adjustment signal is:

[0119] Based on the phase feature group, frequency data after screening is obtained by selecting the frequency interval with the maximum energy response.

[0120] According to the screened frequency data, the adjusted total amplitude is calculated, and the calculation formula is:

[0121]

[0122] Among them, represents the response amplitude of the i-th frequency, is the maximum response amplitude in all frequencies, represents the corresponding time delay, is a decay factor, is the total frequency number, is the adjusted total amplitude.

[0123] Based on the adjusted total amplitude, the output power of the hydrogen fuel cell is optimized, the high-frequency signal is adjusted, the real-time load demand of the power grid is matched, and the high-frequency adjustment signal is formed.

[0124] Specifically, based on the phase feature group obtained in the foregoing, the data is frequency-division filtered, and it is necessary to first check the power performance data of different frequency components in the existing records, and preliminarily distinguish all frequency components in the range of 0 Hz to 50 Hz, for example, 0 Hz to 5 Hz is regarded as a low frequency band, 5 Hz to 20 Hz is regarded as a medium frequency band, and 20 Hz to 50 Hz is regarded as a high frequency band. The frequency interval here is set based on the power frequency range that the hydrogen fuel cell can generate under different working conditions, and the variation of the hydrogen fuel cell output waveform is collected in multiple tests to determine the appropriate upper and lower limits. Then, the power amplitude value in each frequency band is checked. If it is found that some amplitude value deviates too much, it is necessary to compare the threshold value set in advance to determine whether it is a measurement error. For example, the amplitude threshold value can be defined in the range of 0 to 100 units. When the amplitude value exceeds 120, it is marked as abnormal and the time tag of the measurement is recorded. After the statistics of all frequency bands are completed, the frequency band with the highest energy response value is selected and marked as the maximum energy response frequency interval. The remaining frequency band data is kept in the database as a reference. If the maximum response frequency interval occurs at 15 Hz to 22 Hz, the amplitude value, phase feature and other information of the interval are summarized and compared with the load response record of the same time period. When the time delay exceeds 0.5 seconds or the phase offset exceeds 10°, it is necessary to reconfirm whether the signal acquisition frequency setting meets the current monitoring requirements. Finally, the frequency interval with the maximum energy response is selected after comprehensive comparison, and the filtered frequency data is obtained.

[0125] The advantage of the formula is that by introducing the attenuation factor in the amplitude ratio and considering the time delay, the energy characteristics and relative response time of different frequency components are considered, and the problem that only amplitude comparison cannot reflect the time difference is avoided.

[0126] The acquisition step is to collect the amplitude values of the output frequency point by point in the maximum energy response frequency interval obtained in the foregoing, to perform discrete sampling on the actual power waveform by a high-precision measuring device, and to accumulate each recorded amplitude value into an array to obtain ; The acquisition step is to determine the highest amplitude value in all the collected frequency components, and record the highest amplitude value as ; The acquisition step is to collect the amplitude values of the output frequency point by point in the maximum energy response frequency interval obtained in the foregoing, to perform discrete sampling on the actual power waveform by a high-precision measuring device, and to accumulate each recorded amplitude value into an array to obtain ; ; The acquisition step is to observe the attenuation phenomenon in different time periods, count the energy change trend of the same frequency in different time windows, and obtain an attenuation factor interval after multiple comparisons. Then, a typical value is taken according to the equipment operating state to cover the common frequency attenuation characteristics. For the total number of frequencies identified in the selected frequency interval, the counting method is used after scanning the waveform characteristics.

[0127] Calculation process: let , the amplitude data , , , where , the time delay is measured as , , , the attenuation factor is

[0128] First, calculate :

[0129]

[0130]

[0131] Then calculate :

[0132]

[0133]

[0134] Then calculate :

[0135]

[0136]

[0137] Add them together:

[0138]

[0139] Finally, take the square root:

[0140]

[0141] The result shows that the adjusted total amplitude of the current frequency component is about 1.10 after considering the attenuation factor and time delay. If the value is greater than 1.0, it means that the comprehensive intensity is relatively higher. If it is lower than 0.5, it means that the amplitude response in the same range may be significantly weakened.

[0142] Based on the adjusted total amplitude, when optimizing the output power of the hydrogen fuel cell, the power situation of each main period needs to be checked. First, the output load situation of the current period is sorted out from the statistical data, for example, the power value is compared with the effective range set in advance, such as the output power is compared with the interval of 0kW to 10kW. If the power value is lower than 0kW or exceeds 10kW, record this point as abnormal collection and mark the reason. Then, for the power value in the normal range, combined with the adjusted total amplitude calculated, the amplitude setting of the high-frequency signal part is gradually corrected. For example, when the frequency component of a period reaches 40Hz and the measured delay time is 1.2 seconds, the amplitude of this component is reduced or increased according to the previously obtained attenuation factor information, and different frequency components are sequentially corrected according to the time sequence. If the power in the high frequency segment exceeds the empirical threshold, such as breaking through 12kW, it is necessary to compare again with the actual equipment carrying capacity and make convergence adjustment to the output. Finally, all frequency adjustment operations are collected to establish a new high-frequency signal output configuration for the hydrogen fuel cell under the current load demand conditions, and then obtain the high-frequency adjustment signal.

[0143] The acquisition step of safety monitoring information is:

[0144] Based on the high-frequency adjustment signal, the real-time data points of each minute are recorded for the temperature parameter and output power of the hydrogen fuel cell, including the peak value, the valley value and the average value, to obtain real-time monitoring data.

[0145] Using the real-time monitoring data, whether the temperature and power of each data point exceed the predetermined safety threshold is checked. The out-of-limit data is marked, and the time stamp and parameter value of the out-of-limit event are recorded to form a preliminary abnormal result.

[0146] According to the preliminary abnormal result, the marked out-of-limit event is analyzed to verify the severity and risk level of the out-of-limit event, and safety monitoring information is formed.

[0147] Specifically, based on the high-frequency adjustment signal, the temperature parameter and the output power of the hydrogen fuel cell need to be selected as the recording period per minute, and the temperature and power are sampled multiple times within the period. In order to make the data comparable, the temperature range of 0°C to 90°C and the output power range of 0kW to 10kW are set in advance, which are derived from the actual range statistics of the device in the past multiple runs. Then, all the sampling values within the period are searched at the end of each minute, and the peak value, the valley value and the average value are identified respectively, and these data are recorded in time sequence. If a peak value exceeds the preset upper limit, such as temperature exceeding 90°C or output power exceeding 10kW, the sampling timestamp and the corresponding sensor need to be verified to confirm whether there is a sensor deviation or abnormal reading. If the reading is accurate after verification, the value is marked as exceeding the acceptable range. Otherwise, it is marked as data anomaly and separately summarized. In order to prevent the interference of frequent temperature or power fluctuations, the difference statistics of the collection quantity of different measuring points are also needed in the final summary, for example, in a daily monitoring period, if the peak value of the hydrogen fuel cell temperature reaches 85°C to 90°C multiple times, it means that it is near the upper limit of the temperature, so the adjacent period data should be recorded and other environmental conditions should be checked for changes. Finally, the minute-level real-time monitoring data is classified and saved in the database by day, hour or minute, and the real-time monitoring data is obtained by combining the temperature, power and high-frequency signal correlation information.

[0148] Using the above real-time monitoring data, the two key items of temperature and output power are searched for each data point. First, each record is compared with the safety threshold set in advance, for example, the temperature threshold is set within 90°C, which is obtained by combining the heat resistance of the device material and the maximum safe temperature tested on site. The output power threshold is set within 10kW, which is based on the optimal output range of the hydrogen fuel cell under different test loads and is modified by referring to the operating conditions. Then, the peak value, the valley value and the average value of the temperature and the power in the record are scanned. If the peak value or the average value exceeds the above set threshold, the record is marked as "over-limit data", and the timestamp and the corresponding temperature or power value are noted in the database for subsequent reference. When the valley value also appears abnormal, such as temperature valley below 0°C or power valley below 0kW, the same over-limit logic is used for judgment and marking to ensure that any data deviating from the acceptable range is identified separately. If over-limit phenomenon occurs in continuous multiple periods, further analysis is needed. Finally, all the marked over-limit events are summarized and distinguished from the normal records to form the preliminary abnormal results.

[0149] According to the preliminary abnormal results obtained above, subsequent analysis needs to be performed according to the time stamp and parameter value of each over-limit event. In order to determine the severity and risk level of each abnormal event, first of all, it is necessary to check how much the temperature or power value involved in the event deviates from the range, for example, if the temperature peak in a certain record exceeds the threshold of 90℃ by more than 2℃, and appears for three times in succession, it can be rated as a high-risk event, while if a certain record is only slightly higher than the upper limit threshold by 0.5℃ or the power exceeds 10kW by no more than 0.3kW, it is classified into a lower risk group. In addition, it is necessary to refer to factors such as the continuous running time of the equipment, the environmental temperature and the load condition to confirm whether these events can be considered as occasional in a short period of time or whether there are further signs of escalation. If there are multiple high temperature or high power abnormalities occurring in adjacent periods, they are combined for event correlation analysis to find out whether they are caused by the same fault reason. Finally, the events are divided into high-risk or low-risk according to the risk level, and the relevant information is recorded. The result obtained after this procedure is the safety monitoring information.

[0150] The life assessment data acquisition step is:

[0151] Based on the safety monitoring information, the current state of the hydrogen fuel cell is analyzed, the events and performance trends related to safety are recorded, including temperature exceeding events and power fluctuations, and the state analysis result is generated.

[0152] According to the state analysis result, the remaining life percentage is calculated, and the calculation formula is:

[0153]

[0154] Among them, represents the safety score of the i-th safety monitoring event, is the weight factor of the i-th safety monitoring event, is the total number of events, represents the estimated remaining life percentage;

[0155] Based on the remaining life percentage, the overall health status and expected life of the hydrogen fuel cell module are evaluated, and the life assessment data is formed.

[0156] Specifically, based on the safety monitoring information obtained, the current state of the hydrogen fuel cell is analyzed. First, events and performance trends related to safety are selected from the records. Temperature and power are taken as the main concerns, and they are compared with the previously accumulated temperature range of 0-90°C and power range of 0-10 kW. If the temperature peak exceeds 90°C multiple times or the power fluctuation has a large surge in a short time, it is marked in the record with a time label. Then, it is checked whether these over-limit or high fluctuation events show an increasing trend in the recent period, for example, the number and severity of high temperature events in the last 24 hours are counted and compared with the same period before. If the high temperature over-limit events grow too fast, attention should be paid to whether the cooling system of the device or the ambient temperature is abnormal. In addition, it is checked from the perspective of power fluctuation whether there are frequent large changes in load. The power and load change data are checked against each other, for example, when the power value is close to the 10 kW upper limit for several consecutive times, it is checked whether the power peak or abnormal difference in the same period occurs with the power distribution network demand. If the power fluctuation crosses the peak and trough multiple times in 1 minute or less, it is checked whether there are problems such as unstable hydrogen supply pressure or too dense sensor collection interval. The occurrence time and detailed parameters of all the above abnormal or high-frequency events are recorded, and they are combined with the two main phenomena of temperature over-limit and power severe fluctuation to form a comprehensive analysis of the device in the safety aspect in this period, and finally the state analysis result is generated.

[0157] The advantage of the formula is that by introducing , abnormal events with lower safety scores are highlighted, and weighted accumulation is performed combined with the weight factor, thereby avoiding the problem that simple average cannot reflect the influence of events at different risk levels.

[0158] The obtaining step of the safety score is that, for each over-limit event or safety monitoring event, a score is given according to its severity and duration and recorded as a safety score of 0-100. For example, during the continuous operation of the device for 300 hours, each event in which the temperature exceeds 90°C is considered in combination with the corresponding power fluctuation. If the temperature exceeds the range for a long time, a lower safety score of 50-60 is given. If it is only a slight over-limit, the score can be as high as 80 or more. The obtaining step of the weight value is that, by statistically analyzing the frequency and potential impact of similar events in the historical records, the weight value is compiled. The weight is related to the predictable damage degree of the event. The known failure rate of each type of event is analyzed first, and then the weight value is formed based on industry device safety standards or field evaluation correction. ​The acquisition step is to collect the total number of safety monitoring events registered in this period, including temperature exceeding, power overload and other related abnormalities, and add one for each new event.

[0159] Calculation process: let the total number of registered safety monitoring events be , its safety score , , After multiple statistics, assign weight factors to events 1, 2 and 3 respectively , ,

[0160] First, calculate each :

[0161]

[0162] Then square it and multiply it by the corresponding weight:

[0163]

[0164]

[0165]

[0166] Add the three together:

[0167]

[0168] Then take the square root of the result:

[0169]

[0170] Finally, put it into the formula to calculate :

[0171]

[0172] The result shows that the estimated remaining life percentage is about 76.51. If the value is close to or lower than 50, the device condition needs to be closely monitored and preventive measures need to be taken in time. If the value exceeds 80, it indicates that the overall state is relatively good.

[0173] Based on the percentage of remaining life obtained in the foregoing, the overall health of the hydrogen fuel cell module needs to be evaluated subsequently, and the failure conditions and maintenance records in the past multiple operation cycles are comprehensively considered in the evaluation process. The temperature fluctuations and power changes in the recent period of time are compared first to confirm whether there are more high temperatures or abnormal events in the high load state, and the time period and duration of these events are listed one by one. If it is found that most of the events are concentrated in certain specific time periods, it indicates that there may be external environmental interference or high device load configuration, which needs to be paid special attention in the subsequent operation. On the other hand, if the records show that the hydrogen fuel cell temperature and power have not exceeded the preset safety range for several consecutive days, it indicates that the current state is stable, but the percentage of remaining life still needs to be periodically reviewed. By comparing the newly generated monitoring data with the previous cumulative records to identify trend changes, if the percentage of remaining life shows a significant downward trend, the subsequent operation scheme needs to be adjusted. Finally, the evaluation results of this time are compared with the health status of the previous cycles, and the time points of maintenance and repair are summarized to form the life evaluation data.

[0174] The present application provides an energy management method of a hydrogen fuel cell module, comprising the following steps:

[0175] Based on the voltage parameters, current parameters and temperature parameters of the hydrogen fuel cell, raw monitoring data is collected from sensor readings, and numerical comparison is performed with the load demand of the power distribution network to generate supply-demand matching information. The raw monitoring data and the supply-demand matching information are synchronously merged, time-sequentially arranged and abnormal value checked to obtain a preliminary state sequence.

[0176] Based on the preliminary state sequence, interval division, difference statistics and difference operation are performed on the current parameters, temperature parameters and voltage parameters to obtain energy distribution basis, and the energy distribution basis is associated and compared with the load demand of the power distribution network to establish discharge adjustment values through discharge path screening, multi-dimensional coordinate comparison and abnormal value exclusion.

[0177] Based on the discharge adjustment values, frequency band extraction and rate recording are performed on the output power of the hydrogen fuel cell to obtain power segmented data, which is compared and phase measured with the load change of the power distribution network to obtain a phase feature group. The phase feature group is frequency-screened and amplitude-adjusted to establish a high-frequency adjustment signal.

[0178] Based on the high-frequency adjustment signal, the temperature parameters and output power of the hydrogen fuel cell are continuously monitored, and amplitude limiting and safety threshold comparison are performed to obtain safety monitoring information. The safety monitoring information and the remaining life parameters are analyzed and compared to obtain life evaluation data.

[0179] Based on the life assessment data, the health management of the hydrogen fuel cell module is performed, and the battery state is evaluated and monitored through battery temperature, output power and safety monitoring information.

Claims

1. An energy management system for a hydrogen fuel cell module, characterized in that: The system comprises: A monitoring and acquisition module collects and compares sensor readings based on the voltage, current, and temperature parameters of the hydrogen fuel cell module to obtain raw monitoring data, performs numerical comparison on the load demand of the distribution network, generates supply and demand matching information, synchronizes and merges the raw monitoring data with the supply and demand matching information, arranges them in time series, and checks for abnormal values ​​to obtain a preliminary state sequence; a dynamic allocation module, which, based on the preliminary state sequence, performs interval division, differential statistics, and difference operations on the current parameters, temperature parameters, and voltage parameters to obtain an energy allocation basis, performs correlation comparison, discharge path screening, multi-dimensional coordinate comparison, and abnormal value screening on the energy allocation basis and the load demand of the distribution network to establish a discharge allocation value; A high-frequency control module extracts frequency bands and records the rate of the output power of the hydrogen fuel cell based on the discharge adjustment value, obtains power segmentation data, compares the power segmentation data with the load change of the power distribution network and performs phase measurement to obtain a phase feature group, performs frequency division screening and amplitude adjustment on the phase feature group, and establishes a high-frequency adjustment signal; The health management module continuously monitors the temperature parameters and output power of the hydrogen fuel cell, makes limit judgments, and compares safety thresholds based on the high-frequency adjustment signal, obtains safety monitoring information, performs status analysis and indicator comparison on the safety monitoring information and the remaining life parameters, and obtains life assessment data.

2. The energy management system for the hydrogen fuel cell module according to claim 1, characterized in that: The steps for obtaining the preliminary state sequence are: Receive the voltage parameters, current parameters, and temperature parameters of the hydrogen fuel cell, fuse them with the supply and demand matching information, and arrange the collected data points in time sequence using timestamps to obtain a time-sorted data set; Based on the time-sorted data set, by calculating the mean and standard deviation of each parameter, identifying and eliminating outliers that exceed the range of two standard deviations of the mean, to obtain an outlier-processed data set; Based on the data set after outlier processing, the maximum value, minimum value, mean value and coefficient of variation of each parameter are calculated to form a preliminary state sequence.

3. The energy management system for the hydrogen fuel cell module according to claim 1, characterized in that: The steps for obtaining the energy distribution basis are: Based on the preliminary state sequence, each parameter is divided into intervals, and a threshold is set according to the historical data range of each parameter and the current operating environment. The parameter is divided into multiple working intervals, each interval representing a different operating state, and a parameter interval data set is generated; According to the parameter interval data set, the interval difference statistics are calculated, and the calculation formula is: in, represents the value of the i-th parameter in the j-th interval, represents the average value of the i-th parameter in the interval, is the interval difference statistic of the i-th parameter, and N is the number of data points in each interval; Based on the interval difference statistics of each parameter, the sizes of the interval difference values ​​of each parameter are compared, the relative change rates and fluctuation characteristics between the parameters are analyzed, and the basis for energy distribution is formed.

4. The energy management system for a hydrogen fuel cell module according to claim 1, characterized in that: The steps for obtaining the discharge allocation value are as follows: Based on the energy allocation basis, a comparison is performed with the load demand of the distribution network. By analyzing the difference between energy output and demand in each period, the periods when demand is exceeded or not met are identified to form a preliminary difference data set; Based on the preliminary difference data set, the overall matching degree between energy output and demand in each period is calculated using the following formula: in, is the energy output in period i, is the load demand in period i, is the total number of periods evaluated, Indicates the overall matching degree; Based on the overall matching degree, discharge path screening and multi-dimensional coordinate comparison are performed, and the discharge strategy is optimized in combination with historical data. After screening out abnormal values, the discharge allocation value is formed.

5. The energy management system for the hydrogen fuel cell module according to claim 1, characterized in that: The steps for obtaining the phase feature group are: Based on the discharge allocation value, by setting the time window, measuring the power peak and valley in each window, and obtaining the power segmentation data; Comparing the power segment data with the load change data of the power distribution network, performing time synchronization on each power segment, and obtaining an aligned comparison result; Based on the alignment comparison result, the phase between power and load is measured, and the periodicity and consistency of the changes are quantified by analyzing the time offset and phase difference between the power change and the load change to form a phase feature group.

6. The energy management system for a hydrogen fuel cell module according to claim 1, characterized in that: The steps of obtaining the high-frequency adjustment signal are: Based on the phase feature group, performing frequency division screening on the data, selecting the frequency interval with the maximum energy response, and obtaining filtered frequency data; According to the filtered frequency data, the adjusted total amplitude is calculated using the following formula: in, represents the response amplitude of the ith frequency, is the maximum response amplitude among all frequencies, represents the corresponding time delay, is a decay factor, is the total frequency, is the total amplitude after adjustment; Based on the adjusted total amplitude, the output power of the hydrogen fuel cell is optimized, and the high-frequency signal is adjusted to match the real-time load demand of the power grid to form a high-frequency regulation signal.

7. The energy management system for a hydrogen fuel cell module according to claim 1, characterized in that: The steps for obtaining the security monitoring information are as follows: Based on the high-frequency adjustment signal, the data points per minute are recorded in real time for the temperature parameters and output power of the hydrogen fuel cell, including peak values, valley values, and average values, to obtain real-time monitoring data; Using the real-time monitoring data, check whether the temperature and power exceed the predetermined safety thresholds for each data point, mark the exceeded data, and record the timestamp and parameter value of the exceeded event to form a preliminary abnormality result; Based on the preliminary abnormal results, the marked out-of-limit events are analyzed to verify the severity and risk level of the out-of-limit events and form safety monitoring information.

8. The energy management system for a hydrogen fuel cell module according to claim 1, characterized in that: The steps for obtaining the life assessment data are as follows: Based on the safety monitoring information, analyze the current state of the hydrogen fuel cell, record safety-related events and performance trends, including temperature exceeding limit events and power fluctuations, and generate state analysis results; According to the state analysis results, the remaining life percentage is calculated using the following formula: in, represents the safety score of the i-th safety monitoring event, is the weight factor of the i-th safety monitoring event, is the total number of events, Indicates the estimated remaining life percentage; Based on the remaining life percentage, the overall health status and expected life of the hydrogen fuel cell module are evaluated to form life evaluation data.

9. A method for energy management of a hydrogen fuel cell module, characterized in that: The energy management system of the hydrogen fuel cell module according to any one of claims 1 to 8 comprises the following steps: Based on the voltage, current, and temperature parameters of the hydrogen fuel cell, raw monitoring data is collected from sensor readings and numerically compared with the load demand of the distribution network to generate supply and demand matching information. The raw monitoring data and the supply and demand matching information are then synchronized, time-sequenced, and checked for outliers to obtain a preliminary state sequence. Based on the preliminary state sequence, the current, temperature, and voltage parameters are divided into intervals, and differential statistics and difference operations are performed to obtain the basis for energy allocation. This is then correlated with the load demand of the distribution network. Through discharge path screening, multi-dimensional coordinate comparison, and abnormal value screening, the discharge allocation value is established. Based on the discharge adjustment value, the frequency band of the hydrogen fuel cell's output power is extracted and the rate is recorded to obtain power segmentation data. This data is then compared with the load changes of the distribution network and the phase is measured to obtain a phase feature group. The phase feature group is then frequency-filtered and amplitude-adjusted to establish a high-frequency regulation signal. Based on high-frequency regulation signals, the temperature parameters and output power of hydrogen fuel cells are continuously monitored, limit judgments are made, and safety thresholds are compared to obtain safety monitoring information. The safety monitoring information is then analyzed and compared with the remaining life parameters to obtain life assessment data. Based on the life assessment data, the health management of the hydrogen fuel cell module is performed, and the battery status is evaluated and monitored through battery temperature, output power and safety monitoring information.

Images