Hydrogen fuel cell stack water balance detection device

Abstract

The invention discloses a hydrogen fuel cell stack water balance detection device, and belongs to the technical field of hydrogen fuel cell stacks. A hydrogen fuel cell stack water balance detection device comprises a detection platform and a cell stack module, an insulation detection box is arranged above the detection platform, a sliding seat is arranged in the insulation detection box, and a guide rail is arranged below the sliding seat. In order to solve the problems that the existing fuel cell stack water balance monitoring method is mainly used for condensing and collecting water from one side of an anode or a cathode of a cell, is long in time consumption, is not suitable for real-time use, and is easily influenced by environment temperature and working temperature in a detection process, so that a detection result has an error, the invention provides a water balance monitoring method for a fuel cell stack. The topological structure adopted by the invention has the characteristics of small input current ripple, high efficiency, high voltage gain ratio, compact structure and high redundancy, the service life of the fuel cell stack can be prolonged under higher switching frequency, and meanwhile, online measurement of high-frequency impedance is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrogen fuel cell stacks, and particularly to a hydrogen fuel cell stack water balance detection device. Background Art

[0002] A fuel cell is a power generation device that converts chemical energy into electrical energy. During the operation of a fuel cell, sufficient hydration of its proton exchange membrane is a necessary condition for maintaining conductivity and minimizing ohmic losses. At the same time, removing the water generated at the cathode is crucial for avoiding flooding of the catalyst layer and the porous carbon paper diffusion layer. In addition, local drying of the proton exchange membrane may also occur, leading to excessive ohmic heating effects, and ultimately resulting in the formation of pinholes on the membrane, which becomes a key problem affecting the durability of the fuel cell stack. Therefore, improper water management not only affects the performance of the fuel cell, but also causes rapid degradation of the proton exchange membrane electrode assembly; on-site monitoring of the water balance is crucial for understanding the real-time characteristics of the fuel cell stack, optimizing the structural design, and stable operation.

[0003] The existing fuel cell stack water balance monitoring methods mainly condense and collect the water output from one side of the anode or cathode of the battery. This method takes a long time and is not suitable for real-time use. It is easily affected by the ambient temperature and the working temperature during the detection process, resulting in errors in the detection results. Moreover, the existing fuel cell stack water balance monitoring devices usually also have limitations such as insufficient structural compactness and low redundancy. Summary of the Invention

[0004] The purpose of the present invention is to provide a hydrogen fuel cell stack water balance detection device. The adopted topological structure has the characteristics of small input current ripple, high efficiency, high voltage gain ratio, compact structure, and high redundancy. It can extend the life of the fuel cell stack at a higher switching frequency and simultaneously realize on-line measurement of high-frequency impedance, which can solve the problems in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A hydrogen fuel cell stack water balance detection device includes a detection platform and a battery stack module. An insulation detection box is arranged above the detection platform. Among them, a sliding seat is arranged inside the insulation detection box, and a guide rail is arranged below the sliding seat. The sliding seat is slidably connected to the detection platform through the guide rail. A battery compartment board is arranged above the sliding seat, and the battery stack module is installed inside the battery compartment board. A signal generator is arranged at one end of the battery compartment board, and the signal generator is electrically connected to the battery stack module. A side hydraulic component is arranged at the other end of the battery compartment board. The battery stack module is placed inside the battery compartment board, and then the sliding seat drives the battery stack module to move into the insulation detection box. After that, the staff connects and tests between the signal generator and the battery stack module.

[0006] Further, a shaft bracket is provided above the battery compartment board. The shaft bracket is connected to the detection platform through a bracket. An upper hydraulic component is provided above the shaft bracket. One end of the upper hydraulic component is provided with a sensing pressure plate, and the sensing pressure plate is located directly above the battery compartment board. The upper hydraulic component can control the sensing pressure plate to descend above the battery stack module. At the same time, the side hydraulic component controls the flow partition plate to extend into the detection window and contact the flow sensing element inside the battery stack module.

[0007] Further, the sensing pressure plate is telescopically connected to the shaft bracket through a guide rod. A circulation pipe valve is provided on one side above the sensing pressure plate, and the circulation pipe valve extends through to the bottom of the sensing pressure plate. Among them, the circulation pipe valve is connected to the battery stack module.

[0008] Further, one end of the side hydraulic component is provided with a flow partition plate, and the flow partition plate is telescopically connected. One side inside the battery compartment board is provided with a detection window, and the flow partition plate is connected to the battery compartment board through the detection window. A power supply box is provided on the top of the insulation detection box.

[0009] Further, the signal generator includes a signal processing module, a data acquisition module, an EIS module, and a data fitting module. Among them, the signal generator is used to send a perturbation current into the battery stack module;

[0010] The signal processing module, the data acquisition module, the EIS module, and the data fitting module are built into the battery stack module.

[0011] Further, the EIS module interacts with the signal processing module and the data fitting module respectively. Among them, the EIS module further includes an impedance compensation unit, and the impedance compensation unit is used to calculate the real-time AC impedance information, and then estimate the water content in the stack according to the equivalent model;

[0012] The EIS module is used to obtain the relationship between the impedance or admittance of the electrochemical system and the frequency. Among them, the impedance and admittance are complex numbers, including real and imaginary parts, corresponding to resistance and capacitance characteristics respectively, and are plotted into an impedance spectrum or an admittance spectrum through complex data;

[0013] The impedance compensation unit includes hardware compensation and software compensation;

[0014] Among them, the hardware compensation consists of temperature compensation and parameter compensation. The temperature compensation is through integrated temperature sensing: thermocouples / thermistors are deployed on the surface or inside of the battery to collect temperature data in real time, and then a reference impedance curve at different temperatures is established through experiments to dynamically correct the measured values.

[0015] For parameter compensation, the current injection and voltage detection circuits are separated to eliminate the influence of wire resistance, and shielded cables and common-mode suppression circuits are used to reduce electromagnetic interference (EMI).

[0016] Software compensation includes:

[0017] Dynamic baseline calibration, recording the impedance baseline in a known healthy state (such as the factory state), performing offset compensation by comparing real-time data, predicting the current impedance theoretical value using historical data, and correcting the measured deviation;

[0018] Multi-frequency point collaborative correction, analyzing impedance components (such as ohmic impedance, charge transfer impedance, etc.) at different frequencies, separating the interference frequency band, extracting the key features of the impedance spectrum, and reducing noise sensitivity;

[0019] Closed-loop feedback compensation, dynamically adjusting the amplitude / frequency of the injected signal according to the error to maintain measurement stability, such as the LMS (Least Mean Square) algorithm, and suppressing noise in specific frequency bands in real time;

[0020] Subsequently, a high-frequency impedance decomposition model is established:

[0021]

[0022] Compensate for the non-linear error after fitting, where Z is the impedance, R ohm is the resistance value, j is the imaginary unit, ω is the angular frequency, R ct is the resistance generated by the charge transfer process of the solid electrolyte interface film on the electrode surface, C dl is the capacitance value.

[0023] Furthermore, there is a two-way interaction between the signal processing module and the data acquisition module. Among them, the signal processing module is used to receive and analyze the current and voltage signals inside the battery stack module, and convert the selected electrical signals from the time domain to the complex frequency domain;

[0024] The data acquisition module is used to select the current and voltage signals parsed by the acquisition signal processing module.

[0025] Furthermore, the data fitting module includes an equivalent circuit unit. Among them, the equivalent circuit unit is used to form a feedback map after data fitting analysis by the EIS module.

[0026] Furthermore, the battery stack module includes a flow rate monitoring module and a voltage and current module. Among them, the flow rate monitoring module includes a water inlet monitoring unit and a water outlet monitoring unit. The water inlet monitoring unit and the water outlet monitoring unit are respectively used to monitor the flow rate values of the anodic and cathodic reduced water in the battery stack module, detect the water volume information generated by the reduction reaction between the anode and the cathode through the flow rate monitoring module, and record the inlet and outlet volume values of the anode and the cathode respectively. The voltage and current module is used to record the discharge amount when the current battery stack module is in the working state. Then, the values detected by the voltage and current module are fitted with the water volume balance difference values monitored by the flow rate monitoring module, and further judge the current and voltage under different water volumes;

[0027] The voltage and current module is used to monitor the current and voltage values generated by the reduction reaction inside the battery stack module;

[0028] The water inflow monitoring unit and the water outflow monitoring unit interact with the loss compensation unit, and the loss compensation unit is used to calculate the loss of water due to the reaction inside the battery stack and perform numerical compensation for the loss.

[0029] Furthermore, the flow monitoring module interacts with the balance difference dynamic module, and the voltage and current module interacts with the fluctuation signal unit. Among them, the balance difference dynamic module is used to calculate the balance difference between the water inflow and outflow of the anode and cathode;

[0030] The fluctuation signal unit is used to record the numerical values and differences of the current and voltage fluctuations of the voltage and current module in the working state;

[0031] Both the balance difference dynamic module and the fluctuation signal unit interact with the node frequency module. Among them, the node frequency module is used to compare the water balance value obtained by the balance difference dynamic module with the voltage and current fluctuation values recorded by the fluctuation signal unit, and match the fluctuation time nodes between the two.

[0032] Furthermore, the battery stack module includes:

[0033] The prediction model construction module is used to establish an attenuation prediction model through the analysis of historical data and real-time monitoring data;

[0034] The attenuation prediction module uses the survival analysis method and combines machine learning algorithms to predict the pinhole formation time of the proton exchange membrane and the active attenuation rate of the catalyst according to the material properties, working conditions, voltage change rules, and current change rules of each component by using the attenuation prediction model;

[0035] The life evaluation module comprehensively analyzes the water balance state and the change trend of performance parameters of the battery stack module on the basis of predicting the active attenuation rate of the proton exchange membrane, and combines the operating conditions and maintenance records to evaluate the life of each component of the hydrogen fuel cell system by using the integrated learning method to obtain the remaining service life of each component;

[0036] The maintenance module is used to judge the components whose remaining service life is no more than a preset time threshold from the end of the life according to the remaining service life of the components, and send a prompt message to prompt the operator to arrange the replacement or repair of the components in advance.

[0037] Furthermore, the battery stack module includes:

[0038] The working temperature monitoring module is used to measure the working temperature of the hydrogen fuel cell stack and convert the working temperature into Kelvin temperature;

[0039] An electrode degradation evaluation module, which is used to calculate the electrode degradation rate of each reaction by using the following formula:

[0040]

[0041] wherein, represents the electrode degradation rate of the th reaction; represents a correction factor, and its value is determined according to the water quality, electrode material and proton exchange membrane material of the battery stack module; represents the electron transfer coefficient between electrodes, taking ; represents the number of electrons transferred in the th reaction; represents the Faraday constant, taking ; represents the activation overpotential between electrodes; represents the ideal gas constant, that is ; represents the working temperature, unit: Kelvin;

[0042] A degradation function generation module, which is used to draw an electrode degradation rate curve based on the electrode degradation rate of each reaction, perform smoothing processing, and obtain an electrode degradation function by performing function fitting on the smoothed electrode degradation rate curve;

[0043] An operating benefit evaluation module, which is used to perform integral operation on the electrode degradation function, evaluate the degree of electrode degradation according to the integral operation result; perform operating benefit analysis on the battery stack module according to the degree of electrode degradation, and if the operating benefit is lower than the set benefit threshold, an alarm prompt is issued.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] 1. The topology structure adopted by the present invention has the characteristics of small input current ripple, high efficiency, high voltage gain ratio, compact structure and high redundancy;

[0046] 2. The present invention establishes an attenuation prediction model, which is trained to predict the active attenuation speed of the proton exchange membrane of the battery stack module. By using the material working temperature, voltage and current of the proton exchange membrane as the input data of the attenuation prediction model, the pinhole formation time of the proton exchange membrane and the active attenuation speed of the catalyst are output after being processed by the attenuation prediction model; on this basis, combined with the life evaluation of each different component, the remaining service life of different components is obtained, and referring to the preset duration threshold, information on component maintenance requirements can be provided to the operator, thereby ensuring the long-term and stable operation of the equipment, reducing the failure rate and maintenance cost;

[0047] 3. In the present invention, the water quantity information generated by the reduction reaction between the anode and the cathode is detected by the flow rate monitoring module, and the inflow and outflow values of the anode and the cathode are respectively recorded. The voltage and current module is used to record the discharge capacity of the current battery stack module in the working state. Then, the values detected by the voltage and current module are fitted with the water quantity balance difference values monitored by the flow rate monitoring module, so as to judge the current and voltage under different water quantities. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is the overall front view of the present invention;

[0049] Figure 2 is the overall side view of the present invention;

[0050] Figure 3 is the schematic diagram of the battery compartment board structure of the present invention;

[0051] Figure 4 is the schematic diagram of the EIS turbulence flow process of the present invention;

[0052] Figure 5 is the schematic diagram of the flow rate monitoring process of the present invention.

[0053] In the figure: 1, detection platform; 2, insulation detection box; 3, battery stack module; 101, guide rail; 102, sliding seat; 103, shaft frame; 104, upper hydraulic component; 105, signal generator; 106, battery compartment board; 107, side hydraulic component; 1031, sensing pressure plate; 1032, circulation pipe valve; 1051, signal processing module; 1052, data acquisition module; 1053, impedance compensation unit; 1054, EIS module; 1055, data fitting module; 1056, equivalent circuit unit; 1061, detection window; 1071, flow partition board; 201, power supply box; 301, flow rate monitoring module; 302, voltage and current module; 303, node frequency module; 304, balance difference dynamic module; 305, loss compensation unit; 306, fluctuation signal unit; 3011, water inflow monitoring unit; 3012, water outflow monitoring unit. DETAILED DESCRIPTION OF THE INVENTION

[0054] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0055] To solve the problem that the existing fuel cell stack water balance monitoring method mainly condenses and collects the water output from one side of the anode or cathode of the battery. This method takes a long time, is not suitable for real-time use, and is easily affected by the ambient temperature and working temperature during the detection process, resulting in errors in the detection results; please refer to Figures 1-5 , the following technical solutions are provided in this embodiment:

[0056] Embodiment 1:

[0057] Refer to Figures 1-3 , a hydrogen fuel cell stack water balance detection device, including a detection platform 1 and a battery stack module 3. An insulation detection box 2 is arranged above the detection platform 1. Among them, a sliding seat 102 is arranged inside the insulation detection box 2, a guide rail 101 is arranged below the sliding seat 102, and the sliding seat 102 is slidably connected to the detection platform 1 through the guide rail 101. A battery compartment plate 106 is arranged above the sliding seat 102, and the battery stack module 3 is installed inside the battery compartment plate 106. A signal generator 105 is arranged at one end of the battery compartment plate 106, and the signal generator 105 is electrically connected to the battery stack module 3. A side hydraulic component 107 is arranged at the other end of the battery compartment plate 106. The battery stack module 3 is placed inside the battery compartment plate 106, and then the sliding seat 102 drives the battery stack module 3 to move into the insulation detection box 2. After that, the staff connects and tests between the signal generator 105 and the battery stack module 3;

[0058] An axle frame 103 is arranged above the battery compartment plate 106. The axle frame 103 is connected to the detection platform 1 through a bracket. An upper hydraulic component 104 is arranged above the axle frame 103. One end of the upper hydraulic component 104 is provided with a sensing pressure plate 1031. The sensing pressure plate 1031 is located directly above the battery compartment plate 106. A sensing element is arranged inside the sensing pressure plate 1031 to detect the operating state inside the battery stack. The sensing pressure plate 1031 is telescopically connected to the axle frame 103 through a guide rod. A circulation pipe valve 1032 is arranged on one side above the sensing pressure plate 1031. The circulation pipe valve 1032 extends through to the bottom of the sensing pressure plate 1031. Among them, the circulation pipe valve 1032 is connected to the battery stack module 3. One end of the side hydraulic component 107 is provided with a flow partition plate 1071. The flow partition plate 1071 is telescopically connected. A detection window 1061 is arranged on one side inside the battery compartment plate 106. The flow partition plate 1071 is connected to the battery compartment plate 106 through the detection window 1061. A power supply box 201 is arranged on the top of the insulation detection box 2;

[0059] The upper hydraulic component 104 can control the sensing pressure plate 1031 to descend above the battery stack module 3. At the same time, the side hydraulic component 107 controls the flow partition plate 1071 to extend into the detection window 1061 and contact the flow sensing element inside the battery stack module 3.

[0060] Embodiment 2:

[0061] Referring to Figure 4 , the signal generator 105 includes a signal processing module 1051, a data acquisition module 1052, an EIS module 1054, and a data fitting module 1055. Among them, the signal generator 105 is used to send a disturbance current into the battery stack module 3;

[0062] The signal processing module 1051, the data acquisition module 1052, the EIS module 1054, and the data fitting module 1055 are built into the battery stack module 3;

[0063] There is two-way interaction between the signal processing module 1051 and the data acquisition module 1052. Among them, the signal processing module 1051 is used to receive and analyze the current and voltage signals inside the battery stack module 3, and convert the selected electrical signals from the time domain to the complex frequency domain;

[0064] The data acquisition module 1052 is used to select the current and voltage signals parsed by the signal processing module 1051;

[0065] The EIS module 1054 interacts with the signal processing module 1051 and the data fitting module 1055 respectively. Among them, the EIS module 1054 further includes an impedance compensation unit 1053. The impedance compensation unit 1053 is used to calculate the real-time AC impedance information, and then estimate the water content in the stack according to the equivalent model;

[0066] The EIS module 1054 is used to obtain the relationship between the impedance or admittance of the electrochemical system and the frequency. Among them, the impedance and admittance are complex numbers, including real and imaginary parts, corresponding to resistance and capacitance characteristics respectively, and are plotted into an impedance spectrum or an admittance spectrum through complex data.

[0067] A sinusoidal electrical signal (voltage or current) X with an angular frequency of w is used as an excitation signal (also called a disturbance signal in electrochemistry terms) to input into the system. Then, a sinusoidal electrical signal (current or voltage) Y with the same angular frequency w is output from the system. Y is the response signal, and the relationship between Y and X can be expressed by the following formula:

[0068] Y = G(w) X

[0069] If the disturbance signal Y is a sinusoidal voltage signal and X is a sinusoidal current signal, then G is called the impedance of the system M. If the disturbance signal X is a sinusoidal voltage signal and Y is a sinusoidal current signal, then G is called the admittance of the system M;

[0070] According to the definition of impedance (admittance), for a stable linear system, when there is a unique causality between the response and the perturbation, both the impedance Z and the admittance Y are determined by the internal structure of the system and reflect the frequency response characteristics of the system. Therefore, there is a unique correspondence between Z and Y:

[0071] Z = 1 / Y

[0072] G is a vector that varies with frequency and is represented by a complex function with the variable being the frequency f or its angular frequency ω. Therefore, the general expression of G can be written as:

[0073] G(ω) = G'(ω) + jG"(ω)

[0074] where: j = ; G' = the real part of the impedance-admittance, G" = the imaginary part of the impedance-admittance;

[0075] If G is impedance, then: Z = Z' + jZ"

[0076] The modulus of the impedance Z: |Z| = 2 Z' 2

[0077] The phase angle φ of the impedance: tanφ =

[0078] The data fitting module 1055 includes an equivalent circuit unit 1056. Among them, the equivalent circuit unit 1056 is used to form a feedback spectrum after the EIS module 1054 performs data fitting analysis.

[0079] A perturbation signal is generated using a DC / DC converter, and a signal conditioning circuit and a signal acquisition device are designed to extract the AC component in the signal. Finally, through a real-time calculation module for AC impedance, online measurement of AC impedance is achieved. The ripple current frequency generated when the DC-DC converter operates is the same as the switching frequency of the switching semiconductor. Then, a two-phase DC / DC converter is used to decouple the power demand and impedance measurement. Two different DC-DC converters are used to be responsible for power regulation and generating the perturbation signal respectively. For the two-phase DC-DC converter topology, this circuit has two operating modes, namely the normal operating mode (only DC operates) and the diagnostic mode (DC and DC operate simultaneously), and impedance testing only occurs when DC operates, ensuring a small current and voltage ripple rate for DC, and controlling the operating time of DC, effectively solving the problem that the excessive ripple current during impedance measurement affects the durability of the fuel cell stack. When a four-phase DC boost converter is used, in a high-power fuel cell stack, a multi-phase interleaved structure DC-DC converter is often used to withstand a larger current. For the N-phase interleaved parallel DC converter integrated with an anti-coupled inductor, the adopted topology has the characteristics of small input current ripple, high efficiency, high voltage gain ratio, compact structure, and high redundancy, and can extend the life of the fuel cell stack at a higher switching frequency, while realizing online measurement of high-frequency impedance.

[0080] Embodiment 3:

[0081] Refer to Figure 5 , the fuel cell stack module 3 includes a flow rate monitoring module 301 and a voltage and current module 302. Among them, the flow rate monitoring module 301 includes a water inlet monitoring unit 3011 and a water outlet monitoring unit 3012. The water inlet monitoring unit 3011 and the water outlet monitoring unit 3012 are respectively used to monitor the flow rate values of the anode and cathode reduced water of the fuel cell stack module 3;

[0082] The voltage and current module 302 is used to monitor the current and voltage values generated by the reduction reaction inside the fuel cell stack module 3;

[0083] The water inlet monitoring unit 3011 and the water outlet monitoring unit 3012 interact with the loss compensation unit 305. The loss compensation unit 305 is used to calculate the loss of water due to the reaction inside the fuel cell stack and perform numerical compensation for the loss;

[0084] The flow rate monitoring module 301 interacts with the balance difference dynamic module 304, and the voltage and current module 302 interacts with the fluctuation signal unit 306. Among them, the balance difference dynamic module 304 is used to calculate the balance difference between the water inlet and outlet amounts of the anode and cathode;

[0085] The fluctuation signal unit 306 is used to record the numerical values and differences of the current and voltage fluctuations of the voltage and current module 302 during the working state;

[0086] Both the balance difference dynamic module 304 and the fluctuation signal unit 306 interact with the node frequency module 303. Among them, the node frequency module 303 is used to compare the water balance output value obtained by the balance difference dynamic module 304 with the voltage and current fluctuation values recorded by the fluctuation signal unit 306, and match the fluctuation time nodes between the two. When the node frequency module 303 selects nodes, it only targets the high or low time nodes outside the balance value, which can reduce the occupation of algorithm resources and improve the operation efficiency;

[0087] The flow monitoring module 301 detects the water volume information generated by the reduction reaction between the anode and the cathode, and respectively records the inflow and outflow values of the anode and the cathode. The voltage and current module 302 is used to record the discharge amount under the working state of the current fuel cell stack module 3. Then, the value detected by the voltage and current module 302 is fitted with the water volume balance difference value monitored by the flow monitoring module 301, so as to judge the current and voltage under different water volumes.

[0088] On the basis of the foregoing embodiments, the fuel cell stack module 3 includes:

[0089] A prediction model construction module, which is used to establish a decay prediction model through the analysis of historical data and real-time monitoring data;

[0090] A decay prediction module, which uses the survival analysis method and combines machine learning algorithms to predict the pinhole formation time of the proton exchange membrane and the activity decay rate of the catalyst according to the material characteristics, working conditions, voltage change rules and current change rules of each component by using the decay prediction model;

[0091] A life evaluation module, based on the prediction of the activity decay rate of the proton exchange membrane, comprehensively analyzes the water balance state and the change trend of performance parameters of the fuel cell stack module, combines the operating conditions and maintenance records, and uses a life evaluation model constructed based on neural networks and completed learning and training. Using the operating conditions and maintenance records as input data, it evaluates the life of each component of the hydrogen fuel cell system to obtain the remaining service life of each component;

[0092] A maintenance module, which is used to judge the components whose remaining service life is not more than a preset time threshold from the end of the life according to the remaining service life of the components, and send a prompt message to prompt the operator to arrange the replacement or repair of the components in advance.

[0093] Working principle and technical effect: The proton exchange membrane is a key component of the device. In this solution, an attenuation prediction model is established and used to predict the active attenuation rate of the proton exchange membrane of the fuel cell stack module after training. By taking the working temperature, voltage, and current of the proton exchange membrane material as the input data of the attenuation prediction model, the attenuation prediction model outputs the pinhole formation time of the proton exchange membrane and the active attenuation rate of the catalyst after processing; on this basis, combined with the life assessment of each different component, the remaining service life of different components is obtained, and referring to the preset duration threshold, component maintenance requirement information can be provided to the operator, thereby ensuring the long-term and stable operation of the equipment, reducing the failure rate and maintenance cost.

[0094] Based on the foregoing embodiments, the fuel cell stack module 3 includes:

[0095] A working temperature monitoring module for measuring the working temperature of the hydrogen fuel cell stack and converting the working temperature into Kelvin temperature;

[0096] An electrode degradation evaluation module for calculating the electrode degradation rate of each reaction by using the following formula:

[0097]

[0098] Where, represents the electrode degradation rate of the th reaction; represents the correction factor, and the value is determined according to the water quality, electrode material, and proton exchange membrane material of the fuel cell stack module; represents the electron transfer coefficient between electrodes, taking ; represents the th number of electrons transferred in the reaction; represents the Faraday constant, taking ; represents the activation overpotential between electrodes; represents the ideal gas constant, that is ; represents the working temperature, unit: Kelvin;

[0099] A degradation function generation module for drawing an electrode degradation rate curve according to the electrode degradation rate of each reaction, performing smoothing processing, and obtaining an electrode degradation function by function fitting on the smoothed electrode degradation rate curve;

[0100] The operation benefit evaluation module is used to perform integral operation on the electrode degradation function, and the result of the integral operation on the electrode degradation function is used as the evaluation index of the electrode degradation degree; retrieve the pre-stored comparison table of electrode degradation evaluation index and operation benefit obtained through experimental tests, query the comparison table of evaluation index and operation benefit according to the evaluation index of the electrode degradation degree, obtain the current operation benefit, and if the operation benefit is lower than the set benefit threshold, an alarm prompt is issued.

[0101] Working principle and technical effect: In this solution, by monitoring the working temperature and using the above-mentioned electrode degradation rate calculation formula, the electrode degradation rate of each reaction is calculated, and thus an electrode degradation rate curve is generated; on the basis of the smoothing process of the electrode degradation rate curve, function fitting is implemented to obtain the electrode degradation function; further integral operation is performed on the electrode degradation function to evaluate the electrode degradation degree; retrieve the pre-stored comparison table of electrode degradation evaluation index and operation benefit obtained through experimental tests, query the comparison table of evaluation index and operation benefit according to the evaluation index of the electrode degradation degree, obtain the current operation benefit. If the operation benefit is lower than the set benefit threshold, an alarm prompt is issued. Among them, the operation benefit refers to the ratio of the electric energy obtained by the device operation to the corresponding hydrogen fuel consumed under a certain electrode degradation evaluation index; this solution provides a quantitative analysis method for electrode degradation and correlates electrode degradation with the operation benefit of the device. If the electrode degradation degree is too large, resulting in low electric energy conversion efficiency and a decrease in the output ratio, and the operation benefit of the device cannot be guaranteed, an alarm prompt is issued, which can provide objective decision-making reference data and avoid the long-term phenomenon of input-output inversion.

[0102] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0103] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

Claims

1. A hydrogen fuel cell stack water balance detection device, characterized in that, It includes a detection platform (1) and a battery stack module (3). An insulation detection box (2) is arranged above the detection platform (1). Among them, a sliding seat (102) is arranged inside the insulation detection box (2), and a guide rail (101) is arranged below the sliding seat (102). The sliding seat (102) is slidably connected to the detection platform (1) through the guide rail (101). A battery compartment board (106) is arranged above the sliding seat (102), and the battery stack module (3) is installed inside the battery compartment board (106). A signal generator (105) is arranged at one end of the battery compartment board (106), and the signal generator (105) is electrically connected to the battery stack module (3). The signal generator (105) includes a signal processing module (1051), a data acquisition module (1052), an EIS module (1054), and a data fitting module (1055). Among them, the signal generator (105) is used to send a disturbance current into the battery stack module (3).

2. The horizontal balance detection device for a hydrogen fuel cell stack according to claim 1, wherein: A side hydraulic component (107) is arranged at the other end of the battery compartment board (106). An axle frame (103) is arranged above the battery compartment board (106), and the axle frame (103) is connected to the detection platform (1) through a bracket. An upper hydraulic component (104) is arranged above the axle frame (103). A sensing pressure plate (1031) is arranged at one end of the upper hydraulic component (104), and the sensing pressure plate (1031) is located directly above the battery compartment board (106). The sensing pressure plate (1031) is telescopically connected to the axle frame (103) through a guide rod. A circulation pipe valve (1032) is arranged on one side above the sensing pressure plate (1031), and the circulation pipe valve (1032) extends through to the bottom of the sensing pressure plate (1031). Among them, the circulation pipe valve (1032) is in contact connection with the battery stack module (3).

3. The horizontal balance detection device for a hydrogen fuel cell stack according to claim 1, characterized in that: A flow partition plate (1071) is arranged at one end of the side hydraulic component (107), and the flow partition plate (1071) is telescopically connected to the side hydraulic component (107). A detection window (1061) is arranged on one side inside the battery compartment board (106), and the flow partition plate (1071) is connected to the battery compartment board (106) through the detection window (1061). A power supply machine box (201) is arranged at the top of the insulation detection box (2).

4. The horizontal balance detection device for a hydrogen fuel cell stack according to claim 1, characterized in that: The signal processing module (1051), the data acquisition module (1052), the EIS module (1054), and the data fitting module (1055) are built into the battery stack module (3); There is two-way interaction between the signal processing module (1051) and the data acquisition module (1052). Among them, the signal processing module (1051) is used to receive and analyze the current and voltage signals inside the battery stack module (3), and convert the selected electrical signals from the time domain to the complex frequency domain; The data acquisition module (1052) is used to select the current and voltage signals analyzed by the signal processing module (1051).

5. The horizontal balance detection device for a hydrogen fuel cell stack according to claim 4, characterized in that: The EIS module (1054) interacts with the signal processing module (1051) and the data fitting module (1055) respectively. Among them, the EIS module (1054) further includes an impedance compensation unit (1053), and the impedance compensation unit (1053) is used to calculate the real-time AC impedance information and estimate the in-pile water content according to the equivalent model. The EIS module (1054) is used to obtain the variation relationship of the impedance or admittance of the electrochemical system with frequency. Among them, the impedance and admittance are complex numbers, including real and imaginary parts. Among them, the impedance corresponds to resistance, and the admittance corresponds to capacitance characteristics. The EIS module (1054) then plots the impedance spectrum or admittance spectrum through complex data. The impedance compensation unit (1053) includes hardware compensation and software compensation. Among them, the hardware compensation consists of temperature compensation and parameter compensation. The temperature compensation is through integrated temperature sensing: deploy thermocouples / thermistors on the battery surface or inside to collect temperature data in real time, and then establish a reference impedance curve at different temperatures through experiments to dynamically correct the measured value. For parameter compensation, separate the current injection and voltage detection circuits to eliminate the influence of wire resistance, and use shielded cables and common-mode rejection circuits to reduce electromagnetic interference (EMI). The software compensation includes: Dynamic baseline calibration, record the impedance baseline in the known health state (such as the factory state), perform offset compensation by comparing real-time data, predict the current impedance theoretical value using historical data, and correct the measured deviation. Multi-frequency point collaborative correction, analyze the impedance components (ohmic impedance, charge transfer impedance, etc.) at different frequencies, separate the interference frequency bands, extract the key features of the impedance spectrum, and reduce the noise sensitivity. Closed-loop feedback compensation, dynamically adjust the amplitude / frequency of the injected signal according to the error to maintain the measurement stability, such as the LMS (least mean square) algorithm, and suppress the noise in specific frequency bands in real time. Subsequently, establish a high-frequency impedance decomposition model: Compensate for the non - linear error after fitting, where Z is the impedance, R ohm is the resistance value, j is the imaginary unit, ω is the angular frequency, R ct is the resistance generated by the charge transfer process of the solid electrolyte interface film on the electrode surface, C dl is the capacitance value.

6. The horizontal balance detection device for a hydrogen fuel cell stack according to claim 5, characterized in that: The data fitting module (1055) includes an equivalent circuit unit (1056). Among them, the equivalent circuit unit (1056) is used to draw a circuit diagram for the feedback signal formed after the data fitting analysis of the EIS module (1054).

7. The horizontal balance detection device for a hydrogen fuel cell stack according to claim 1, characterized in that: The battery stack module (3) includes a flow monitoring module (301) and a voltage and current module (302). Among them, the flow monitoring module (301) includes a water inlet monitoring unit (3011) and a water outlet monitoring unit (3012). The water inlet monitoring unit (3011) and the water outlet monitoring unit (3012) are respectively used to monitor the flow values of the reduced water at the anode and cathode of the battery stack module (3). The voltage and current module (302) is used to monitor the current and voltage values generated by the reduction reaction inside the battery stack module (3). The water inlet monitoring unit (3011) and the water outlet monitoring unit (3012) interact with the loss compensation unit (305), and the loss compensation unit (305) is used to calculate the loss of water during the reaction inside the battery stack and perform numerical compensation for the loss.

8. An apparatus for detecting the water balance of a hydrogen fuel cell stack according to claim 7, characterized in that: The flow monitoring module (301) interacts with the differential dynamic module (304), and the voltage and current module (302) interacts with the fluctuation signal unit (306). Among them, the differential dynamic module (304) is used to calculate the balance difference between the anode and cathode water inflows and outflows; The fluctuation signal unit (306) is used to record the values and differences of the current and voltage fluctuations of the voltage and current module (302) in the working state; Both the differential dynamic module (304) and the fluctuation signal unit (306) interact with the node frequency module (303). Among them, the node frequency module (303) is used to compare the water balance production value obtained by the differential dynamic module (304) with the voltage and current fluctuation values recorded by the fluctuation signal unit (306), and match the fluctuation time nodes between the two.

9. The horizontal balance detection device for a hydrogen fuel cell stack according to claim 1, characterized in that: The battery stack module (3) includes: A prediction model construction module, which is used to establish a decay prediction model through the analysis of historical data and real-time monitoring data; A decay prediction module, which uses the survival analysis method and combines machine learning algorithms to predict the pinhole formation time of the proton exchange membrane and the activity decay rate of the catalyst according to the material characteristics, working conditions, voltage change rules and current change rules of each component, using the decay prediction model; A life evaluation module, which comprehensively analyzes the water balance state and the change trend of performance parameters of the battery stack module based on the prediction of the activity decay rate of the proton exchange membrane, combines the operating conditions and maintenance records, and uses the integrated learning method to evaluate the life of each component of the hydrogen fuel cell system to obtain the remaining service life of each component; A maintenance module, which is used to judge the components whose remaining service life is no more than a preset time threshold from the end of their life according to the remaining service life of the components, and send a prompt message to prompt the operator to arrange the replacement or repair of the components in advance.

10. The horizontal balance detection device for a hydrogen fuel cell stack according to claim 9, characterized in that: The battery stack module (3) includes: A working temperature monitoring module, which is used to measure the working temperature of the hydrogen fuel cell stack and convert the working temperature into Kelvin temperature; An electrode degradation evaluation module, which is used to calculate the electrode degradation rate of each reaction using the following formula: Among them, τ i represents the electrode degradation rate of the i-th reaction; k represents the correction factor, and its value is determined according to the water quality, electrode material, and proton exchange membrane material of the battery stack module; α represents the electron transfer coefficient between electrodes, taking α = 0.5; n i represents the number of electrons transferred in the i-th reaction; F represents the Faraday constant, taking F = 96485; v represents the activation overpotential between electrodes; R represents the ideal gas constant, that is, R = 8.314; T represents the working temperature, unit: Kelvin; A degradation function generation module, which is used to draw an electrode degradation rate curve according to the electrode degradation rate of each reaction, perform smoothing processing, and obtain an electrode degradation function by fitting a function to the smoothed electrode degradation rate curve; An operating benefit evaluation module, which is used to perform integral operation on the electrode degradation function, evaluate the degree of electrode degradation according to the integral operation result; perform an operating benefit analysis of the battery stack module according to the degree of electrode degradation. If the operating benefit is lower than the set benefit threshold, an alarm prompt is issued.

Images