Inspection method and system for hydrogen fuel cell stack

Abstract

The invention provides a hydrogen fuel cell stack inspection method and system, and the method comprises the steps: dividing a hydrogen fuel cell stack into a plurality of groups, each group is formed by connecting a plurality of single cells in series, each group corresponds to a differential amplifier, and the plurality of groups of differential amplifiers work at the same time for inspection; during inspection, the positive electrode and the negative electrode of each single battery in each group are respectively connected with the corresponding differential amplifier by utilizing an optical coupling switch, and the differential amplifier inspects the voltage of all the single batteries in the group by controlling the on-off of the optical coupling switch end; performing analog-to-digital conversion on the voltage of all the inspected single batteries; performing SG filtering on the voltages of all the single batteries after analog-to-digital conversion to obtain the voltages of all the single batteries after filtering; wherein the window size of the SG filtering and the fitting order of the polynomial are determined through the improved quantum genetic algorithm. According to the invention, data accuracy can be improved and cost can be reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of voltage signal acquisition of hydrogen fuel cells, and particularly relates to a method and system for inspecting a hydrogen fuel cell stack. Background Art

[0002] A hydrogen fuel cell is a device that directly converts chemical energy into electrical energy, and has the advantages of high efficiency, cleanliness, and renewability, and is considered to be an ideal energy source in the future. A hydrogen fuel cell system usually consists of a hydrogen fuel cell stack, an air path, a fuel path, a cooling path, a control system, etc. Among them, the hydrogen fuel cell stack usually needs dozens or even hundreds of PEMFC (hydrogen fuel cell) monomers connected in series. If the voltage of a single PEMFC is unstable, it will directly affect the performance and lifespan of the fuel cell stack. At the same time, the voltage change trend of each single cell is one of the important parameters of the fuel cell stack control strategy. Therefore, real-time inspection of the voltage of a single PEMFC is crucial for the entire fuel cell stack.

[0003] Currently, the commonly used methods for detecting the voltage of a single PEMFC include the resistor voltage division method, the differential amplifier method, and the floating ground technology method. Among them, the principle of the resistor voltage division method is to divide the voltage of the fuel cell monomer through resistors to reach the measurement range of the A / D chip and then collect it. This method has the advantages of low cost, simplicity, and speed, but has the disadvantages of large cumulative error and low accuracy. The working principle of the differential amplifier method is to input through a voltage differential amplifier and then perform A / D conversion. This method has the advantage of being able to eliminate the common-mode voltage at the battery terminal, but the disadvantages are high cost and poor system stability. The working principle of the floating ground technology method is to perform floating control on the potential, and has the advantages of being able to eliminate the problem of cumulative potential and high accuracy, but the disadvantages are complex structure and high cost. Summary of the Invention

[0004] The technical problem to be solved by the present invention is: to provide a method and system for inspecting a hydrogen fuel cell stack, which can improve the accuracy of data and reduce costs.

[0005] The technical solution adopted by the present invention to solve the above technical problems is as follows: As the first aspect of the present invention, a method for inspecting a hydrogen fuel cell stack, the method includes: S1. Divide the hydrogen fuel cell stack into several groups, each group is composed of several single cells connected in series, each group corresponds to a differential amplifier, and the differential amplifiers of several groups work simultaneously for inspection; During inspection, each group uses an optocoupler switch to connect the positive and negative electrodes of each single cell in the group to the corresponding differential amplifier respectively, and controls the on / off of the optocoupler switch terminal to make the differential amplifier inspect the voltages of all single cells in the group; Perform analog-to-digital conversion on the voltages of all single cells inspected; S2. Perform SG filtering on the voltages of all individual cells after analog-to-digital conversion to obtain the voltages of all individual cells after filtering; wherein the window size and the fitting order of the polynomial of the SG filtering are determined by an improved quantum genetic algorithm.

[0006] According to the above method, the said S2 specifically includes: S21. Determine the window size M and the fitting order P of the polynomial according to the improved quantum genetic algorithm; S22. Take the voltage value of the corresponding battery collected as a data point, and several data points form a data sequence; for any data point in the data sequence, its window contains (M - 1) / 2 data points before and after plus the data point itself; S23. Within the selected window, fit a polynomial to the data points according to the fitting order P of the polynomial and solve; for each data point in the window, calculate the fitting value of the polynomial at the center of the window, and take the fitting value as the smoothed result; S24. Move the window to the next data point and repeat S23 until the entire data sequence is traversed; take all the smoothed results as the voltages of all individual cells after filtering.

[0007] According to the above method, the said S21 specifically includes: S211. Quantum chromosome initialization: Form the individuals in the population by encoding the combination of the window size and the fitting order in binary form, preset the population size, and determine the number of bits of the binary encoding and the probability amplitude; S212. Binary encoding conversion: Determine the value of each bit of the binary encoding according to the comparison of the random number and the probability amplitude; according to the number of bits corresponding to the window size and the fitting order respectively, divide the binary encoding into 2, and convert them into decimal respectively, so as to obtain the window size and the fitting order of an individual; Obtain the window sizes and fitting orders of several individuals according to different random numbers; S213. Quality determination: According to the preset fitness function, calculate the quality of the window size and the fitting order of each individual; the higher the fitness value, the higher the quality of the individual; S214. Iteration: Update the population by adjusting the quantum rotation gate, and then repeat S212 and S213 for iteration until the preset number of iterations is reached; S215. Output the window size and the fitting order of the polynomial corresponding to the highest final fitness value as the window size and the fitting order of the SG filtering.

[0008] According to the above method, in S214, the rotation angle of the quantum rotation gate is adaptively determined according to the velocity update algorithm in the particle swarm algorithm.

[0009] According to the above method, in the velocity update algorithm, there is a weight parameter, which is dynamically adjusted according to the average fitness of the current population.

[0010] According to the above method, in S214, the population is also updated by using the Hadamard gate to mutate random numbers.

[0011] According to the above method, in S214, when the highest fitness values of three consecutive iterations are the same, then after the third iteration, a part of the individuals with the lowest fitness values are initialized; the part is a preset percentage.

[0012] According to the above method, S23 specifically includes: S231. In the selected window, use the least squares method to perform polynomial fitting on the data points; S232. Construct a Vandermonde matrix, the number of rows of which is the window size, the number of columns of which is the fitting order of the polynomial, each row of the Vandermonde matrix corresponds to a data point in the window, and each column corresponds to a power of the polynomial; S233. Use the least squares method to solve the coefficients of the fitted polynomial; S234. For each data point in the window, use the polynomial with the solved coefficients to calculate the fitting value at the center of the window, and take the fitting value as the smoothed result.

[0013] According to the above method, S1 specifically includes: Divide the hydrogen fuel cell stack into several groups, and each group is composed of several single cells connected in series; In each group, several optocoupler switches are sequentially arranged at the connection nodes and endpoints of the single cells; Control the adjacent optocoupler switches to close in sequence through the encoder, and open the other optocoupler switches, so that only one single cell is connected to the input port of the differential amplifier at a time; The differential amplifier transmits the voltage of the single cell obtained by the inspection to the analog-to-digital conversion unit for analog-to-digital conversion.

[0014] As the second aspect of the present invention, the present invention provides a hydrogen fuel cell stack inspection system, including: A differential amplifier, wherein the hydrogen fuel cell stack is divided into several groups, each group is composed of several single cells connected in series, and the number of differential amplifiers is the same as the number of groups of the hydrogen fuel cell stack; A number of optocoupler switches, one end of each optocoupler switch is successively connected to the connection nodes and endpoints of the single cells in each group of hydrogen fuel cell stacks, and the other end of the optocoupler switch is connected to the input port of the differential amplifier corresponding to this group of hydrogen fuel cell stacks; the on / off of the optocoupler switch enables only one single cell to be connected to the input port of the differential amplifier at a time; An analog-to-digital conversion unit, connected to each differential amplifier, for performing analog-to-digital conversion on the voltages of all single cells collected by the differential amplifier; A data processing unit, for performing SG filtering on the voltages of all single cells after analog-to-digital conversion to obtain the voltages of all single cells after filtering; wherein the window size of the SG filtering and the fitting order of the polynomial are determined by an improved quantum genetic algorithm.

[0015] The beneficial effects of the present invention are as follows: 1. By grouping the hydrogen fuel cell stacks through the inspection circuit and performing simultaneous inspections on multiple groups, high-speed acquisition of the voltages of single cells is achieved. The cooperation of the optocoupler and the differential amplifier is used for inspection, reducing the cost; then, by combining the improved quantum genetic algorithm with SG filtering, the change situation of the voltages of single cells can be accurately obtained.

[0016] 2. In order to reduce the possibility of the algorithm falling into local optimum, a mutation and disaster mechanism is added: setting a mutation algorithm to optimize random numbers; when the highest fitness values of three consecutive iterations of the algorithm are the same, a disaster mechanism is adopted. According to the principle of survival of the fittest, after this iteration, a part of the individuals with the lowest fitness values are initialized, thereby protecting excellent individuals and increasing the diversity of the population, and further improving the accuracy of the voltages of single cells. Description of the Drawings

[0017] Figure 1 is the inspection circuit of the hydrogen fuel cell stack in an embodiment of the present invention.

[0018] Figure 2 is the encoder control definition diagram in an embodiment of the present invention.

[0019] Figure 3 is the hardware block diagram in an embodiment of the present invention.

[0020] Figure 4 is the filtering algorithm flow chart in an embodiment of the present invention. Detailed Embodiments

[0021] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and 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.

[0022] As a first aspect of the present invention, the present invention provides a method for inspecting a hydrogen fuel cell stack, wherein the hydrogen fuel cell stack is formed by connecting multiple single cells in series; the method includes: S1. Divide the hydrogen fuel cell stack into several groups, each group is formed by connecting several single cells in series, and each group corresponds to a differential amplifier. The differential amplifiers of several groups work simultaneously for inspection.

[0023] During inspection, each group uses an optocoupler switch to connect the positive and negative electrodes of each single cell in the group to the corresponding differential amplifier respectively, and by controlling the on / off of the optocoupler switch terminal, the differential amplifier inspects the voltages of all single cells in the group. Specifically, in each group, several optocoupler switches are sequentially arranged at the connection nodes and endpoints of the single cells, and the adjacent optocoupler switches are sequentially controlled to close by an encoder, and other optocoupler switches are opened, so that only one single cell is connected to the input port of the differential amplifier at a time.

[0024] Perform analog-to-digital conversion on the voltages of all inspected single cells. In this embodiment, the differential amplifier transmits the voltage of the single cell obtained by inspection to the analog-to-digital conversion unit for analog-to-digital conversion.

[0025] Figure 1 Shown in Figure 1 is the inspection circuit for a group of hydrogen fuel cell stacks, where a single area contains 31 single fuel cells, 1 analog-to-digital conversion unit, namely the ADS1115 chip, 1 differential amplifier, namely the INA149AIDR chip, 2 74HC154PW encoders, and 32 optocoupler switches. High-precision voltage measurement is achieved through the differential channel of the INA149AIDR chip. INA149AIDR is a high-precision unity-gain differential amplifier, and this amplifier has a very high input common-mode voltage range. It is a single monolithic device containing a high-precision operational amplifier and an integrated thin-film resistor network. When the common-mode signal voltage is as high as ±275 V, INA149AIDR can accurately measure a small differential voltage. At the same time, its low-power consumption characteristic further optimizes the energy consumption management of the fuel cell system.

[0026] Output digital signals through the 74HC154PW to control the on / off of the optocoupler switch. The optocoupler switch is a device that uses optical signals to achieve electrical isolation and can be used for signal transmission and isolation in fuel cell inspection to ensure the safety of the high-voltage battery system. Isolate the inspection circuit from the main circuit through the optocoupler, which can not only avoid the electrical interference of the main circuit from affecting the signal quality, but also protect the inspection circuit from high-voltage impact. In addition, the optocoupler switch can also play a role in switch control during the inspection process, improving the reliability and anti-interference ability of the system, thereby ensuring the stable operation of the fuel cell.

[0027] The main inspection process for a single area is: When inspecting the first battery section, the encoder is used to control the closure of optocoupler 0 and optocoupler 1, so that the positive and negative poles of the first section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0028] When inspecting the second battery section, the encoder is used to control the closure of optocoupler 1 and optocoupler 2, so that the positive and negative poles of the second section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0029] When inspecting the third battery section, the encoder is used to control the closure of optocoupler 2 and optocoupler 3, so that the positive and negative poles of the third section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0030] When inspecting the fourth battery section, the encoder is used to control the closure of optocoupler 3 and optocoupler 4, so that the positive and negative poles of the fourth section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0031] When inspecting the fifth battery section, the encoder is used to control the closure of optocoupler 4 and optocoupler 5, so that the positive and negative poles of the fifth section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0032] When inspecting the sixth battery section, the encoder is used to control the closure of optocoupler 5 and optocoupler 6, so that the positive and negative poles of the sixth section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0033] When inspecting the seventh battery section, the encoder is used to control the closure of optocoupler 6 and optocoupler 7, so that the positive and negative poles of the seventh section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0034] When inspecting the eighth battery section, the encoder is used to control the closure of optocoupler 7 and optocoupler 8, so that the positive and negative poles of the eighth section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0035] When inspecting the ninth battery section, the encoder is used to control the closure of optocoupler 8 and optocoupler 9, so that the positive and negative poles of the ninth section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0036] When inspecting the tenth battery section, the encoder is used to control the closure of optocoupler 9 and optocoupler 10, so that the positive and negative poles of the tenth section are connected to the input ports of the differential amplifier. The data is read by INA149AIDR and sent to ADS1115.

[0037] When inspecting the 11th battery cell, the optocouplers 10 and 11 are controlled by the encoder to be closed, so that the positive and negative electrodes of the 11th cell are connected to the input ports of the differential amplifier, and data is read through INA149AIDR and sent to ADS1115.

[0038] And so on, closing adjacent two optocouplers in turn. When inspecting the 31st battery cell, the optocouplers 30 and 31 are controlled by the encoder to be closed, so that the positive and negative electrodes of the 31st cell are connected to the input ports of the differential amplifier, and data is read through INA149AIDR and sent to ADS1115.

[0039] The on-off of the optocoupler switch is controlled by the encoder 74HC154PW. The main features of 74HC154PW include high-speed logic processing ability and low-power consumption characteristics, which are suitable for precise encoding and decoding of signals. It supports a wide power supply voltage range (usually 2V to 6V), ensuring adaptation to different circuit conditions. In addition, this chip has strong anti-interference ability and stability, and can work reliably in a complex electromagnetic environment, which is particularly important in the inspection of fuel cells and can achieve precise switching and efficient control of optocoupler signals. Figure 2 It is the control definition diagram of 74HC154PW. The digital signals output are controlled by controlling the input ports A0, A1, A2, and A3, and thus the optocoupler switch is controlled accordingly.

[0040] After building the inspection circuit for a single partition, multiple partitions can be built for simultaneous inspection, and the multi-partition circuit is a repetition of the single-partition circuit. By the above method, the inspection of 31*n single fuel cells can be realized, where n is the number of partitions. When conducting the inspection, the single cells in each area can be inspected simultaneously, greatly improving the voltage acquisition speed.

[0041] The present invention manages multiple fuel cells in groups. The differential amplifiers of different groups are responsible for inspecting the single fuel cells within the group, and multiple groups can conduct inspections synchronously, thus significantly shortening the inspection cycle. Through the control of the optocoupler switch, the positive and negative electrodes of different single cells within the same group are connected to the input end of the amplifier, and finally the output signal of the differential amplifier is collected through the ADS1115 chip. The rapid and accurate acquisition of the voltage values of multiple single cells is realized.

[0042] S2. Perform SG filtering on the voltages of all single cells after analog-to-digital conversion to obtain the voltages of all single cells after filtering; wherein the window size and polynomial fitting order of the SG filtering are determined by the improved quantum genetic algorithm, as Figure 4 shown, specifically including: S21. Determine the window size M and polynomial fitting order P according to the improved quantum genetic algorithm. The specific content of the S21 includes: S211. Quantum chromosome initialization: Combine the combination of window size and fitting order in the form of binary encoding to form individuals in the population, preset the population size, and determine the number of bits of the binary encoding and the probability amplitude.

[0043] Specifically, first determine the size of the population, that is, the number of individuals. Then randomly generate a group of individuals within the preset range to form an initial population. Each individual is a candidate solution, composed of a string of binary encoding. For example, set the window size to the first 5 bits and the fitting order to the last 5 bits, jointly forming a 10-bit binary encoding.

[0044] The present invention uses a quantum state vector (qubit encoding) to replace the original binary encoding. A qubit can be represented by the following formula:

[0045] In the formula, both α and β are complex numbers, representing the spin-up state and the spin-down state respectively, and are probability amplitudes and satisfy .

[0046] The present invention improves the initialization of quantum encoding, divides the qubits into N probability spaces, and divides them through the following formula:

[0047] In the formula, i represents the i-th individual, M is the number of individuals in the population. Divide the probability space through α and β, which can make each sub-population have the same probability during initialization, and can further improve the search ability of the population.

[0048] The probability amplitude encoding corresponding to the individuals in the population is:

[0049] In the formula, i represents the i-th individual, θ ij =2π*rand represents the rotation angle of the j-th bit of the i-th individual, rand is a random number between 0 and 1, and n is the total number of bits of the binary encoding.

[0050] From this, we can get: α = cos(θ), β = sin(θ).

[0051] S212. Binary encoding conversion: Determine the value of each bit of the binary encoding according to the comparison of the random number and the probability amplitude; according to the number of bits corresponding to the window size and the fitting order respectively, divide the binary encoding into 2, and convert them into decimal respectively, so as to obtain the window size and the fitting order of an individual.

[0052] Specifically, a random number rand is compared with cos(θ) (i.e., α). If the random number at this position is greater than or equal to cos(θ), then this position is encoded as 1; otherwise, it is 0. In this way, a set of binary codes containing quantum information can be obtained. Then, according to the rules of binary coding, the binary codes are converted into the window size and the fitting order. For example, if the window size is set to the first 5 bits and the fitting order is set to the last 5 bits, then the first 5-bit binary code is converted into a decimal number as the window size, and the decimal number converted from the last 5-bit binary code is used as the fitting order.

[0053] Based on different random numbers, the window sizes and fitting orders of several individuals are obtained.

[0054] S213. Quality determination: According to the preset fitness function, calculate the quality of the window size and fitting order of each individual; the higher the fitness value, the higher the quality of the individual.

[0055] In this embodiment, the fitness function is defined

[0056] In the formula, fitness is the fitness, U0 is the ideal reference value of the sampling voltage, U SG,k is the voltage value after SG filtering at the k-th moment, and T is the sampling time.

[0057] Substitute the window size and fitting order of each individual into the SG filtering algorithm, perform SG filtering on the voltages of all single cells after analog-to-digital conversion to obtain the voltage value after SG filtering, and then substitute it into the fitness function for calculation to obtain the fitness function of the individual.

[0058] S214. Iteration: Update the population by adjusting the quantum rotation gate, and maintain the diversity of the population by changing the quantum state probability through the rotation adjustment of the quantum gate, so that the individuals move in the direction of higher fitness.

[0059] Then repeat S212 and S213 for iteration until the preset number of iterations is reached.

[0060] The original rotation angle of the quantum rotation gate is fixed. During the operation of the algorithm, it may cross the optimal solution, causing the result to continuously take values near the optimal solution. Moreover, the rotation angle cannot be too small, otherwise the algorithm will converge too slowly and may fall into a local optimum. The rotation angle of the quantum rotation gate can be adaptively determined according to the velocity update formula in the particle swarm algorithm. The velocity update formula has a weight parameter, namely the inertia weight value. A larger inertia weight value is required for global search in the early stage of the algorithm, while a smaller weight is needed for local search in the later stage. In this embodiment, the weight parameter is dynamically adjusted based on the average fitness, maximum fitness, and minimum fitness of the current population.

[0061] Furthermore, in order to reduce the possibility of the algorithm falling into a local optimum, a mutation and catastrophe mechanism is added.

[0062] Specifically, the population is updated by mutating random numbers using the Hadamard gate. In this embodiment, the mutation probability is set to 0.01. When the random number is less than 0.01, mutation is performed according to a preset mutation formula. When the highest fitness values of three consecutive iterations are the same, the catastrophe mechanism is adopted. According to the principle of survival of the fittest, after the third iteration, a part of the individuals with the lowest fitness values are initialized; the part is a preset percentage, such as ten percent.

[0063] S215. Output the window size and fitting order corresponding to the highest final fitness value as the window size of the SG filter and the fitting order of the polynomial.

[0064] S22. Use the voltage values of the corresponding battery collected as data points, and several data points form a data sequence; for any data point in the data sequence, its window contains (M - 1) / 2 data points before and after plus the data point itself.

[0065] In this embodiment, 100 data points form a data sequence. The data sequence of the voltage values of Battery No. 1 contains 100 voltage value points collected from Battery No. 1, and the data sequence of the voltage values of Battery No. 2 contains 100 voltage value points collected from Battery No. 2. For each point in each data sequence, a window centered on this point is selected. If the window size is M, then this window will contain (M - 1) / 2 data points before and after plus the center point itself.

[0066] S23. In the selected window, fit a polynomial to the data points according to the fitting order P of the polynomial and solve; for each data point in the window, calculate the fitting value of the polynomial at the center of the window, and use the fitting value as the smoothed result. The specific steps of S23 include: S231. In the selected window, use the least squares method to perform polynomial fitting on the data points; the purpose is to find a polynomial function such that the sum of the squared differences between the function and all data points within the window is minimized.

[0067] The form of the polynomial is: , where a0, a1, ..., ap are undetermined coefficients, which are solved by the least squares method.

[0068] S232. Construct a Vandermonde matrix A, whose number of rows is the window size M and the number of columns is the fitting order P of the polynomial. Each row of the Vandermonde matrix A corresponds to a data point within the window, and each column corresponds to a power of the polynomial.

[0069] S233. Use the least squares method to solve the coefficients of the fitted polynomial.

[0070] S234. For each data point in the window, use the polynomial with the solved coefficients to calculate the fitted value at the center of the window, and take the fitted value as the smoothed result.

[0071] S24. Move the window to the next data point and repeat S23 until the entire data sequence is traversed; take all the smoothed results as the voltages of all single cells after filtering.

[0072] In the present invention, the improved quantum genetic algorithm (IQGA) is used to optimize the key parameters (window size and fitting order) of the Savitzky - Golay (SG) filtering algorithm. After determining the optimal parameter combination, the voltage data of the single cells collected is filtered to eliminate noise and retain the voltage change trend. Finally, the inspection system feeds the processed accurate voltage data back to the fuel cell controller in real time to achieve efficient and accurate monitoring and management of the cell voltages.

[0073] As the second aspect of the present invention, the present invention provides a hydrogen fuel cell stack inspection system, as Figure 1 and Figure 3 shown, including: Differential amplifiers. Among them, the hydrogen fuel cell stack is divided into several groups, and each group is composed of several single cells connected in series. The number of differential amplifiers is the same as the number of groups of the hydrogen fuel cell stack.

[0074] A plurality of opto - isolator switches. One end of the opto - isolator switch is sequentially connected to the connection nodes and endpoints of the single cells of each group of hydrogen fuel cell stacks, and the other end of the opto - isolator switch is connected to the input port of the differential amplifier corresponding to this group of hydrogen fuel cell stacks; the on - off of the opto - isolator switch enables only one single cell to be connected to the input port of the differential amplifier at a time; the on - off of the said opto - isolator switch is controlled by an encoder.

[0075] An analog-to-digital conversion unit, connected to each differential amplifier, for performing analog-to-digital conversion on the voltages of all single cells collected by the differential amplifier.

[0076] A data processing unit, for performing SG filtering on the voltages of all single cells after analog-to-digital conversion to obtain the voltages of all single cells after filtering; wherein the window size of the SG filtering and the fitting order of the polynomial are determined by an improved quantum genetic algorithm. Its specific algorithm is the same as S2 in the above method part and will not be repeated here.

[0077] Through the innovative inspection architecture and digital processing technology, the present invention solves the problems of slow inspection speed and poor anti-electromagnetic interference ability existing in the existing fuel cell stack inspection methods, and provides a reliable guarantee for the stable operation of the fuel cell system.

[0078] It should be understood that those of ordinary skill in the art can make improvements or transformations according to the above description, and all such improvements and transformations should fall within the protection scope of the appended claims of the present invention.

Claims

1. A method for inspecting a hydrogen fuel cell stack, characterized in that: The method includes: S1. Divide the hydrogen fuel cell stack into several groups, each group consisting of several single cells connected in series, with each group corresponding to a differential amplifier, and the differential amplifiers of several groups work simultaneously for inspection; During inspection, each group uses an optocoupler switch to connect the positive and negative electrodes of each single cell in this group to the corresponding differential amplifier respectively, and by controlling the on / off of the optocoupler switch terminal, the differential amplifier inspects the voltages of all single cells in this group; Perform analog-to-digital conversion on the voltages of all single cells inspected; S2. Perform SG filtering on the voltages of all single cells after analog-to-digital conversion to obtain the voltages of all single cells after filtering; where the window size and the fitting order of the polynomial of SG filtering are determined by an improved quantum genetic algorithm.

2. The hydrogen fuel cell stack inspection method according to claim 1, characterized in that: The specific content of S2 includes: S21. Determine the window size M and the fitting order P of the polynomial according to the improved quantum genetic algorithm; S22. Take the voltage value of the corresponding battery collected as a data point, and several data points form a data sequence; for any data point in the data sequence, its window contains (M - 1) / 2 data points before and after plus this data point itself; S23. Within the selected window, fit a polynomial to the data points according to the fitting order P of the polynomial and solve; for each data point in the window, calculate the fitting value of the polynomial at the center of the window, and take the fitting value as the smoothed result; S24. Move the window to the next data point and repeat S23 until the entire data sequence is traversed; take all the smoothed results as the voltages of all single cells after filtering.

3. The hydrogen fuel cell stack inspection method according to claim 2, wherein: The specific content of S21 includes: S211. Quantum chromosome initialization: Form the individuals in the population in the form of binary encoding for the combination of the window size and the fitting order, preset the population size, and determine the number of bits of the binary encoding and the probability amplitude; S212. Binary encoding conversion: Determine the value of each bit of the binary encoding according to the comparison of the random number and the probability amplitude; according to the number of bits corresponding to the window size and the fitting order respectively, divide the binary encoding into 2, and convert them into decimal respectively, so as to obtain the window size and the fitting order of an individual; Obtain the window sizes and fitting orders of several individuals according to different random numbers; S213. Quality determination: According to the preset fitness function, calculate the quality of the window size and the fitting order of each individual; the higher the fitness value, the higher the quality of the individual; S214. Iteration: Update the population through quantum rotation gate adjustment, and then repeat S212 and S213 for iteration until the preset number of iterations is reached; Output the window size and the fitting order of the polynomial corresponding to the highest fitness value finally obtained as the window size of SG filtering and the fitting order of the polynomial.

4. The hydrogen fuel cell stack inspection method according to claim 3, characterized in that: In S214 described above, the rotation angle of the quantum rotation gate is adaptively valued according to the velocity update algorithm in the particle swarm algorithm.

5. The hydrogen fuel cell stack inspection method according to claim 4, wherein: There is a weight parameter in the velocity update algorithm, and the weight parameter is dynamically adjusted according to the average fitness of the current population.

6. The hydrogen fuel cell stack inspection method according to claim 3, characterized in that: In S214 described above, the population is also updated by using the Hadamard gate to mutate the random number.

7. The hydrogen fuel cell stack inspection method according to claim 3, characterized in that: In S214, when the highest fitness values of three consecutive iterations are the same, a part of the individuals with the lowest fitness values are initialized after the third iteration; the part is a preset percentage.

8. The hydrogen fuel cell stack inspection method according to claim 2, wherein: S23 specifically includes: S231. Within the selected window, use the least squares method to perform polynomial fitting on the data points; S232. Construct a Vandermonde matrix, the number of rows of which is the window size, and the number of columns is the fitting order of the polynomial. Each row of the Vandermonde matrix corresponds to a data point within the window, and each column corresponds to a power of the polynomial; S233. Use the least squares method to solve the coefficients of the fitted polynomial; S234. For each data point in the window, use the polynomial with the solved coefficients to calculate the fitted value at the center of the window, and use the fitted value as the smoothed result.

9. The method for inspecting a hydrogen fuel cell stack according to claim 1, characterized in that: S1 specifically includes: Divide the hydrogen fuel cell stack into several groups, and each group is formed by connecting several single cells in series; In each group, several opto-coupler switches are sequentially arranged at the connection nodes and endpoints of the single cells; Control the adjacent opto-coupler switches to close in sequence through the encoder, and open the other opto-coupler switches, so that only one single cell is connected to the input port of the differential amplifier at a time; The differential amplifier transmits the voltage of the single cell obtained by the inspection to the analog-to-digital conversion unit for analog-to-digital conversion.

10. A hydrogen fuel cell stack inspection system, characterized in that: It includes: Differential amplifiers, where the hydrogen fuel cell stack is divided into several groups, and each group is formed by connecting several single cells in series. The number of differential amplifiers is the same as the number of groups of the hydrogen fuel cell stack; Several opto-coupler switches, one end of the opto-coupler switch is sequentially connected to the connection nodes and endpoints of the single cells of each group of hydrogen fuel cell stacks, and the other end of the opto-coupler switch is connected to the input port of the differential amplifier corresponding to this group of hydrogen fuel cell stacks; the on / off of the opto-coupler switch makes only one single cell connected to the input port of the differential amplifier at a time; An analog-to-digital conversion unit, connected to each differential amplifier, for performing analog-to-digital conversion on the voltages of all single cells collected by the differential amplifier; A data processing unit, for performing SG filtering on the voltages of all single cells after analog-to-digital conversion to obtain the voltages of all single cells after filtering; where the window size of the SG filtering and the fitting order of the polynomial are determined by an improved quantum genetic algorithm.

Images