Hydrogen fuel cell evaluation method based on entropy weighted stack stability index

Abstract

The invention discloses a hydrogen fuel cell evaluation method based on an entropy weighted stack stability index, and the method comprises the following steps: S1, inputting a voltage matrix of each single cell, calculating an average value of each single voltage, and constructing an average value matrix of the single voltages; s2, calculating the mean deviation of the single voltage, and constructing a mean matrix of the single voltage; s3, performing interval division on the average deviation of the single voltage, and constructing an average deviation interval matrix of the single voltage; s4, calculating the frequency of each off-average difference interval region, and constructing a frequency matrix of each interval data; s5, calculating the Shannon entropy of each single battery, and constructing a Shannon entropy value matrix; s6, the deviation value of the mean deviation interval relative to the target interval is calculated, and a weighted deviation degree matrix is constructed; s7, calculating an entropy weighted stack stability index, and constructing an entropy weighted stack stability index matrix; and S8, evaluating the consistency and stability among a plurality of single cells in the hydrogen fuel cell system by adopting the entropy weighted stack stability index.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and particularly relates to a method for evaluating a hydrogen fuel cell based on an entropy-weighted stack stability index. Background Art

[0002] At present, as one of the key components of a hydrogen fuel cell engine system, a hydrogen fuel cell stack is composed of multiple single cells connected in series. The inconsistency of the single cell voltage will directly affect the performance stability and consistency of the entire stack. During actual operation, the stack has a "barrel effect", and its operating life depends on the single cell with the worst performance. Therefore, it is very necessary to carry out an evaluation of the voltage consistency of the fuel cell stack to improve performance and extend the life.

[0003] Traditional stack consistency evaluation methods usually adopt graphical methods, the fluctuation amplitude of the single cell voltage, the root mean square of the single cell voltage, etc. Existing methods mostly rely on intuitive test data, such as changes in parameters such as voltage, current, and temperature. However, these methods do not fully consider the relative differences between cells and ignore the comprehensive consideration of multi-dimensional data. Specifically as follows: The graphical method can intuitively reflect the voltage fluctuation of the single cell. However, whenever an additional factor affecting the performance of the single cell is added, the number of graphs will increase significantly, or a dimension will be added to the coordinate. Therefore, the graphical method is not suitable for the analysis of voltage consistency under multiple factors. At the same time, without the aid of other evaluation indicators, it is difficult to quantitatively analyze voltage consistency by the graphical method. The fluctuation amplitude of the single cell voltage and the root mean square of the single cell voltage are not suitable for directly measuring the situation where the data measurement range is too large and the data dimensions are different; when the voltage fluctuation range is too large, the fluctuation amplitude and the root mean square will overly rely on extreme values, and extreme maximum or minimum values will distort the fluctuation amplitude or the root mean square value, making these indicators no longer able to truly reflect the uniformity of battery performance; when the data dimensions are different, since these two methods do not normalize or standardize the voltage values of different dimensions, simply using the fluctuation amplitude or the root mean square value cannot provide a fair evaluation.

[0004] With the gradual development of fuel cell technology, how to objectively and accurately evaluate the consistency of the stack has become a technical problem to be solved urgently. Traditional evaluation methods are difficult to comprehensively consider various factors of the battery. Especially when considering the dynamic behavior of the battery and the complex working environment, there is a lack of a scientific evaluation method that can comprehensively consider multiple factors, which is not conducive to the management, supervision, and prediction of the performance of fuel cells and difficult to meet the actual use requirements. Summary of the Invention

[0005] To solve the above problems, the present invention proposes a method for evaluating a hydrogen fuel cell based on an entropy-weighted stack stability index.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A hydrogen fuel cell evaluation method based on an entropy-weighted stack stability index, comprising the following steps: S1. Input the voltage matrix of each single cell, calculate the average value of each single cell voltage, and construct an average value matrix of the single cell voltages; wherein, the voltage matrix of each single cell is: , where is the voltage matrix of each single cell; is an element in the voltage matrix, i = 1, 2,..., m, i represents different moments or different steady states of the cell, j = 1, 2,..., n, j represents the number of the single cell; S2. Calculate the deviation of the single cell voltage from the mean value, and construct a deviation from the mean value matrix of the single cell voltages; wherein, the deviation from the mean value matrix of the single cell voltages is: , where is the deviation from the mean value matrix of the single cell voltages; is an element in the deviation from the mean value matrix of the single cell voltages, and the calculation formula is: , where is an element in the average value matrix of the single cell voltages; S3. Divide the deviation of the single cell voltage from the mean value into intervals, and construct an interval matrix of the deviation of the single cell voltage from the mean value; wherein, the interval matrix of the deviation of the single cell voltage from the mean value is: , where is the interval matrix of the deviation of the single cell voltage from the mean value; is an element in the interval matrix of the deviation of the single cell voltage from the mean value, and the calculation formula is: ; S4. Calculate the frequency of each deviation interval region, and construct a frequency matrix of each interval data; wherein, the frequency matrix of the interval data is: , where is the frequency matrix of the interval data; is an element in the frequency matrix of the interval data, and the calculation process is: , where is an indicator function, defined as follows: , where S is a set; x is an element; S5. Calculate the Shannon entropy of each single cell, and construct a Shannon entropy value matrix; wherein, the Shannon entropy value matrix is: , where is the Shannon entropy value matrix; is the Shannon entropy value matrix when calculating the inconsistency between each single cell; is the Shannon entropy value matrix when calculating the stability of the j-th single cell; is an element in the Shannon entropy value matrix, and the calculation process is: ; S6. Calculate the deviation value of the deviation-from-mean interval relative to the target interval, and construct a weighted deviation matrix; where the weighted deviation matrix is: , where in the formula, is the weighted deviation matrix; is the weighted deviation matrix when calculating the inconsistency between individual battery cells; is the weighted deviation matrix when calculating the stability of the j-th individual battery cell; is an element in the weighted deviation matrix, and the calculation process is: ; S7. Calculate the entropy-weighted stack stability index, and construct an entropy-weighted stack stability index matrix; where the entropy-weighted stack stability index matrix is: , where in the formula, is the entropy-weighted stack stability index matrix; is the entropy-weighted stability index matrix when calculating the inconsistency between individual battery cells; is the entropy-weighted stability index matrix when calculating the stability of each battery during operation; is an element in the entropy-weighted stack stability index matrix, and the calculation process is: , where the exponential term is the weighting term, which is used to amplify the influence of data deviating from the target interval, so that data points far from the target interval have a greater impact on the result; β is an adjustment parameter, which is used to regulate the influence of the weighted deviation on the entropy-weighted stack stability index. By adjusting β balance the dispersion and deviation degree; S8. Use the entropy-weighted stack stability index to evaluate the consistency and stability among multiple individual battery cells in the hydrogen fuel cell system.

[0007] Preferably, the average matrix of the individual voltages in step S1 is: , where in the formula, The calculation formula of is:

[0008] Preferably, the specific process of dividing the deviation-from-mean of the individual voltages into intervals in step S3 is: divide the deviation-from-mean voltage values into multiple intervals at a length of 5 mV, and the starting value of the interval is 0 mV.

[0009] After adopting the above technical solution, the present invention has the following beneficial effects: The entropy-weighted stack stability index calculated by the present invention can objectively and comprehensively evaluate the consistency, difference among multiple single cells in a hydrogen fuel cell system, as well as the stability of each single cell over time and operating conditions. Through multi-dimensional consistency and stability analysis, the present invention can not only see the differences in certain dimensions (such as time, current, voltage, temperature, humidity) of the hydrogen fuel cell, but also quantify their similarity and consistency in the multi-dimensional data space, thereby providing valuable basis for optimizing the performance of the hydrogen fuel cell system. The present invention is applicable to scenarios that need to comprehensively consider multi-dimensional factors and quantify uncertainties, and can be widely applied to the consistency and stability evaluation of hydrogen fuel cell stacks, which is beneficial to the management, supervision and prediction of the performance of fuel cells. It has important application value especially in the following fields: (1) Battery manufacturing industry: used for the monitoring and control of stack consistency during the production process; (2) Electric vehicle industry: used for the consistency test and monitoring of electric vehicle battery systems to ensure the overall performance and stability of the battery pack; (3) Energy storage system: used for the consistency management of batteries in large-scale energy storage systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a schematic flowchart of the present invention; Figure 2 is a graph showing the change of the entropy-weighted stack stability index of the present invention over time. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0011] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0012] As Figures 1 to 2 shown, a hydrogen fuel cell evaluation method based on the entropy-weighted stack stability index includes the following steps: S1. Input the voltage matrix of each single cell, calculate the average value of each single cell voltage, and construct an average value matrix of the single cell voltage; wherein, the voltage matrix of each single cell is: , where is the voltage matrix of each single cell; is an element in the voltage matrix, i = 1, 2,..., m, i represents different moments or different steady states of the battery, j = 1, 2,..., n, j represents the number of the single cell; The average value matrix of the single cell voltage described in step S1 is: , where The calculation formula of ; S2. Calculate the deviation of the monomer voltage from the mean value and construct a matrix of the deviation of the monomer voltage from the mean value. Among them, the matrix of the deviation of the monomer voltage from the mean value is: , where is the matrix of the deviation of the monomer voltage from the mean value; is an element in the matrix of the deviation of the monomer voltage from the mean value, and the calculation formula is: , where is an element in the matrix of the average value of the monomer voltage; S3. Divide the deviation of the monomer voltage from the mean value into intervals and construct a matrix of the deviation intervals of the monomer voltage. Among them, the matrix of the deviation intervals of the monomer voltage is: , where is the matrix of the deviation intervals of the monomer voltage; is an element in the matrix of the deviation intervals of the monomer voltage, and the calculation formula is: ; The specific process of dividing the deviation of the monomer voltage from the mean value into intervals in step S3 is as follows: Divide the deviation voltage value into multiple intervals at a length of 5 mV, and the starting value of the interval is 0 mV; S4. Calculate the frequency of each deviation interval region and construct a frequency matrix of each interval data. Among them, the frequency matrix of the interval data is: , where is the frequency matrix of the interval data; is an element in the frequency matrix of the interval data, and the calculation process is: , , where is an indicator function, defined as follows: , where S is a set; x is an element; S5. Calculate the Shannon entropy of each monomer cell and construct a Shannon entropy value matrix. Among them, the Shannon entropy value matrix is: , where is the Shannon entropy value matrix; is the Shannon entropy value matrix for calculating the inconsistency between each monomer cell; is the Shannon entropy value matrix for calculating the stability of the jth monomer cell; is an element in the Shannon entropy value matrix, and the calculation process is: ; S6. Calculate the deviation value of the deviation interval relative to the target interval and construct a weighted deviation degree matrix. Among them, the weighted deviation degree matrix is: , where is the weighted deviation degree matrix; is the weighted deviation degree matrix for calculating the inconsistency between each monomer cell; is the weighted deviation degree matrix for calculating the stability of the jth monomer cell; is an element in the weighted deviation matrix, and the calculation process is as follows: ; S7. Calculate the entropy-weighted stack stability index and construct the entropy-weighted stack stability index matrix; where the entropy-weighted stack stability index matrix is: , in the formula, is the entropy-weighted stack stability index matrix; is the entropy-weighted stability index matrix when calculating the inconsistency between individual cells; is the entropy-weighted stability index matrix when calculating the stability of each cell during operation; is an element in the entropy-weighted stack stability index matrix, and the calculation process is as follows: , in the formula, the exponential term is the weighted term, which is used to amplify the influence of data deviating from the target interval, so that data points far from the target interval have a greater impact on the result; β is an adjustment parameter, which is used to regulate the influence of the weighted deviation on the entropy-weighted stack stability index. By adjusting β balance the dispersibility and deviation degree; S8. Use the entropy-weighted stack stability index to evaluate the consistency and stability among multiple individual cells in the hydrogen fuel cell system.

[0013] Specific application example: When the hydrogen fuel cell stack is under the operating conditions of a temperature of 75°C, a humidity of 60%, a back pressure of 150 kPa, and a cathode stoichiometry of 1.4, and the current is as Figure 2 loaded, the change of the entropy-weighted stack stability index (abbreviated as the entropy-weighted index) with time is as Figure 2 shown.

[0014] As described above, it is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A hydrogen fuel cell evaluation method based on an entropy-weighted stack stability index, characterized in that, Including the following steps: S1. Input the voltage matrix of each single cell, calculate the average value of each single cell voltage, and construct the average value matrix of the single cell voltage. Among them, the voltage matrix of each single cell is: , where is the voltage matrix of each single cell; is the element in the voltage matrix, i = 1, 2, …, m, i represents different moments or different steady states of the battery, j = 1, 2, …, n, j represents the number of the single cell; S2. Calculate the deviation of the single-cell voltage from the mean value and construct a matrix of the deviations of the single-cell voltage from the mean value. Among them, the matrix of the deviations of the single-cell voltage from the mean value is: , where is the matrix of the deviations of the single-cell voltage from the mean value; is an element in the matrix of the deviations of the single-cell voltage from the mean value, and the calculation formula is: , where is an element in the matrix of the average values of the single-cell voltage; S3. Divide the interval of the deviation of the single - cell voltage from the mean value, and construct the deviation - from - mean - value interval matrix of the single - cell voltage; among them, the deviation - from - mean - value interval matrix of the single - cell voltage is: , where is the deviation - from - mean - value interval matrix of the single - cell voltage; is an element in the deviation - from - mean - value interval matrix of the single - cell voltage, and the calculation formula is: ; S4. Calculate the frequency of each deviation-from-mean interval region and construct the frequency matrix of the data for each interval; wherein, the frequency matrix of the interval data is as follows: , where is the frequency matrix of the interval data; is an element in the frequency matrix of the interval data, and the calculation process is as follows: , wherein, is an indicator function, defined as follows: , wherein S is a set; x is an element; S5. Calculate the Shannon entropy of each single cell and construct a Shannon entropy value matrix; where the Shannon entropy value matrix is: , where is the Shannon entropy value matrix; is the Shannon entropy value matrix for calculating the inconsistency between single cells; is the Shannon entropy value matrix for calculating the stability of the j-th single cell; is an element in the Shannon entropy value matrix, and the calculation process is: ; S6. Calculate the deviation value of the deviation-from-mean interval relative to the target interval, and construct a weighted deviation matrix; where the weighted deviation matrix is: , in the formula, is the weighted deviation matrix; is the weighted deviation matrix when calculating the inconsistency between individual battery cells; is the weighted deviation matrix when calculating the stability of the j-th individual battery cell; is an element in the weighted deviation matrix, and the calculation process is: ; S7. Calculate the entropy-weighted stack stability index and construct the entropy-weighted stack stability index matrix. The entropy-weighted stack stability index matrix is as follows: , where is the entropy-weighted stack stability index matrix; is the entropy-weighted stability index matrix for calculating the inconsistency between individual cells; is the entropy-weighted stability index matrix for calculating the stability of each cell during operation; is an element in the entropy-weighted stack stability index matrix, and the calculation process is as follows: , where the exponential term is the weighted term, which is used to amplify the influence of data deviating from the target interval, so that data points far from the target interval have a greater impact on the result; β is an adjustment parameter used to control the influence of the weighted deviation degree on the entropy-weighted stack stability index. By adjusting β the balance between dispersion and deviation degree is achieved; S8. Use the entropy-weighted stack stability index to evaluate the consistency and stability among multiple single cells in the hydrogen fuel cell system.

2. The hydrogen fuel cell evaluation method based on the entropy-weighted stack stability index according to claim 1, wherein The average value matrix of the monomer voltages described in step S1 is: , where The calculation formula of .

3. The hydrogen fuel cell evaluation method based on the entropy-weighted stack stability index according to claim 1, wherein The specific process of dividing the deviation from the mean of the single cell voltage in step S3 is as follows: Divide the deviation voltage value from the mean into multiple intervals at a length of 5 mV, and the starting value of the interval is 0 mV.

Images