Hydrogen injection valve high-frequency pulse characteristic and flow self-correction testing device and method

Abstract

This invention discloses a testing device and method for the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve. The method includes the following steps: setting initial parameters and a target flow rate for the high-frequency pulse of the hydrogen injection valve, and constructing a nonlinear mapping model between the initial pulse duty cycle and the hydrogen output flow rate; performing a multi-gradient pulse frequency injection test; acquiring dynamic hydrogen flow rate data in real time, and simultaneously acquiring flow data during the pulse injection process; calculating the deviation between the actual flow rate and the target flow rate, and constructing a dynamic deviation evaluation model based on the stoichiometric ratio of the hydrogen fuel cell; iteratively correcting the pulse width parameter based on the deviation value, and applying a fuzzy PID control algorithm to achieve adaptive parameter optimization; outputting a calibration parameter set that meets the accuracy threshold, and generating a final parameter set containing a frequency-pulse width two-dimensional calibration matrix. This invention effectively solves the problems of flow deviation and response lag that easily occur in the hydrogen supply of the hydrogen injection valve under high-frequency pulse conditions in traditional hydrogen fuel cell systems.

Description

Technical Field

[0001] This invention belongs to the field of hydrogen supply control technology for hydrogen fuel cell systems, specifically relating to a test device and method for testing the high-frequency pulse characteristics and flow self-calibration of hydrogen injection valves. Background Technology

[0002] Hydrogen fuel cells are one of the core technologies of current clean new energy power systems. The precise supply of hydrogen at the anode directly determines the energy conversion efficiency, dynamic response speed and operational safety of the system. As a key actuator in the anode hydrogen supply chain, the hydrogen injection valve's high-frequency pulse injection characteristics and flow control precision are the core guarantee for hydrogen fuel cells to adapt to complex operating conditions such as cold start and high-frequency load changes.

[0003] Current traditional hydrogen injection valve testing technologies primarily focus on static flow testing under low-frequency steady-state pulse conditions. They typically employ conventional volumetric flow meters to collect hydrogen flow data, using open-loop fixed pulse width regulation control logic. This fails to consider the impact of real-time environmental fluctuations in hydrogen temperature and pressure on gas density, making hydrogen flow measurements susceptible to interference from temperature and pressure changes. Furthermore, it cannot eliminate systematic errors caused by changes in gas density, resulting in insufficient measurement accuracy. Additionally, it neglects the timing synchronization requirements of data acquisition and control signals, and lacks specific suppression measures for electromagnetic and random noise. This leads to significant timing deviations between flow data and control signals, prominent noise interference, and difficulty in guaranteeing data fidelity. Moreover, offline manual calibration for parameter configuration results in delayed parameter updates, making it impossible to dynamically adapt to complex operating conditions. Ultimately, traditional technologies are ill-suited to meet the real-time testing requirements under dynamic operating conditions.

[0004] Solving the aforementioned technical problems is a research direction that those skilled in the art are dedicated to. Summary of the Invention

[0005] The first objective of this invention is to provide a test method for the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve, in order to solve the problems of flow deviation and response lag that easily occur in the hydrogen supply of hydrogen injection valves under high-frequency pulse conditions in existing hydrogen fuel cell systems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve, comprising the following steps: Step S1: Set the initial parameters of the high-frequency pulse of the hydrogen injection valve and the target flow rate, and construct a nonlinear mapping model between the initial pulse duty cycle and the hydrogen output flow rate. Step S2: Perform a multi-gradient pulse frequency injection test; Step S3: Real-time acquisition of dynamic hydrogen flow rate data, and synchronous acquisition of flow data during the pulse jet process via a thermal mass flow meter; Step S4: Calculate the deviation between the actual flow rate and the target flow rate, and construct a dynamic deviation assessment model based on the stoichiometric ratio of the chemical reaction in the hydrogen fuel cell. Step S5: Iteratively correct the pulse width parameter based on the deviation value, and apply the fuzzy PID control algorithm to achieve adaptive parameter optimization; Step S6: Output the calibration parameter set that meets the accuracy threshold, and generate the final parameter set containing the frequency-pulse width two-dimensional correction matrix.

[0007] As a specific implementation method, in step S1, the initial parameters of the high-frequency pulse of the hydrogen injection valve and the target flow rate are set through the following steps: Based on the rated power of the fuel cell stack and the tolerance range of hydrogen pressure on the anode side, the target value of hydrogen flow rate actually required by the fuel cell system is calculated. Then, according to the valve response parameters provided by the hydrogen injection valve manufacturer, the target value of hydrogen flow rate is converted into a control signal that the valve can recognize, and the initial pulse frequency and pulse width range are determined. During this process, the real-time ambient temperature feedback from the temperature sensor in the fuel cell system is collected to correct the volume change of hydrogen caused by temperature changes. Then, the duty cycle of the pulse is matched with the actual output hydrogen flow rate to establish an initial matching model.

[0008] As a specific implementation method, the specific process of step S2 is as follows: A segmented incremental pulse frequency excitation strategy is adopted. Within a complete test time window, the pulse signal is controlled to start from the initial frequency and increase stepwise with a fixed step size until the upper limit of the frequency is reached. Each independent frequency band needs to continuously output N complete pulse cycles, where N is a natural number greater than 10, and the pressure fluctuation of the fuel cell anode manifold is monitored simultaneously during this process; When the test frequency enters the high-frequency region, the pulse width reduction compensation mechanism is activated simultaneously. By reducing the on-time of a single pulse, the average current and heat accumulation of the hydrogen injection valve solenoid coil are controlled. The high-frequency region is the region where the frequency reaches 3 / 4 of the upper frequency limit.

[0009] As a specific implementation method, the specific process of step S3 is as follows: In the anode circulation loop of the fuel cell, a temperature-pressure composite sensor is deployed to eliminate measurement errors caused by changes in gas density; an FIR digital filter is used to suppress noise generated by the opening and closing of the solenoid valve; a timestamp alignment mechanism is established to ensure that the timing deviation between the flow data and the pulse control signal is ≤10μs; for the fuel cell system, background flow interference caused by the circulating hydrogen pump is subtracted in the acquisition algorithm to obtain a clean injection flow waveform.

[0010] As a specific implementation method, in step S3, the specific process of deploying a temperature-pressure composite sensor in the fuel cell anode circulation loop to eliminate measurement errors caused by gas density changes is as follows: The instantaneous mass flow rate Q in the anode circuit is measured using a thermal mass flow meter. m (t), whose original signal S raw (t) contains the actual jet flow rate Q inject (t), background flow rate Q generated by the circulating hydrogen pump background And electromagnetic noise N introduced by the opening and closing of the solenoid valve em (t) and random noise N r (t), that is: S raw (t)=Q inject (t)+Q background +N em (t)+N r (t), To eliminate gas temperature T g With pressure P g The measurement error caused by the change in gas density is compensated in real time by deploying a temperature-pressure composite sensor near the measuring point of the thermal mass flow meter. The actual volumetric flow rate Q is calculated based on the ideal gas law. v (t) Mass flow rate Q corrected to standard conditions m (t), the correction formula is: , Where P0 and T0 represent standard pressure and standard temperature, respectively. Here, R is the molar mass of hydrogen, and R is the universal gas constant. The above formula is used to convert the volumetric flow rate under actual working conditions to the standard state, thus eliminating the measurement error introduced by the change of gas state. To suppress the electromagnetic noise N generated by the opening and closing of the solenoid valve em (t), the fuel cell system is processed using a finite-length unit impulse response digital filter based on the Hanning window design, and its difference equation is expressed as: , Where x[ni] is the sampled value of the original signal at time ni; y[n] is the filtered output sequence; and h[i] is the signal based on the cutoff frequency f. c The calculated filter coefficients, where N is the filter order, are used to attenuate noise energy in a specific frequency band through convolution operations, thereby obtaining a smooth flow signal S. filtered (t); For fuel cell systems, the specific process of deducting background flow interference from the circulating hydrogen pump in the acquisition algorithm is as follows: The average background flow rate was measured during the quiescent period when the hydrogen injection valve was closed. The jet flow waveform is then subtracted from the dynamic signal containing the jet, thereby extracting the pure jet flow waveform. This ensures that the objects targeted in subsequent analyses are entirely generated by the pulse action of the hydrogen injection valve.

[0011] As a specific implementation method, the specific process of calculating the deviation between the actual flow rate and the target flow rate in step S4 is as follows: The absolute error between the integral flow rate and the target value within a single pulse cycle is taken, and the response overshoot at the rising / falling edge is superimposed. A weighted moving average algorithm is used to smooth instantaneous fluctuations. Considering the scenario of sudden changes in fuel cell load, the derivative of the deviation value is calculated as a dynamic performance index. When the deviation is detected to continuously exceed the safety threshold, an emergency correction mechanism is triggered to prevent damage to the fuel cell stack. To obtain the pure dynamic injection flow waveform Q of the hydrogen injection valve ingect After (t), it is compared with the target hydrogen flow rate Q calculated in real time by the fuel cell system based on the current load. target (t) is compared, and a dynamic deviation evaluation model based on the stoichiometry of the hydrogen fuel cell chemical reaction is constructed to evaluate each independent pulse cycle T. P For accurate flow tracking, first calculate the periodic integral flow deviation: , Where k is the period number; Q target (k) represents the average target flow rate within k periods; Qinject(t) represents the actual dynamic hydrogen injection flow rate of the hydrogen injection valve at time t, which is a continuous flow signal that varies with time; Tp represents the duration of a single hydrogen injection valve pulse period, which is the reciprocal of the pulse frequency, Tp=1 / f; To quantify the dynamic response defects of valves, the overshoot Q of the flow waveform at the rising and falling edges is identified and extracted. vershoot with under-adjustment U ndershoot This is added as a penalty term to the integral deviation to form the original instantaneous deviation: , In the formula, α and β are weighting coefficients for the effects of overshoot and undershoot. Due to pulse injection and fluid pulsation, E raw (k) In the case of high-frequency fluctuations, a weighted moving average algorithm is needed to smooth the fluctuations in order to extract the trend term. The smoothed deviation E smooth (k) From the formula: , Where M is the width of the sliding window, w i The weighting coefficients are allocated according to the exponential decay law, satisfying... While preserving the true deviation trend, it filters out meaningless fluctuations caused by measurement noise. To address the severe operating conditions of rapid load changes in fuel cells, the deviation change rate is introduced as a key dynamic performance indicator: , This indicator directly reflects the severity of the hydrogen supply lag; The entire deviation assessment model outputs two key criteria: smoothing deviation E. smooth (k) and its rate of change R dynamic (k); To ensure the safety of the fuel cell stack, two safety thresholds are preset: when E smooth (k) Continuously exceeds the warning threshold Th warning Reaching N cycles, or R dynamic (k) Instantly exceeds the emergency threshold Th critical In this case, the fuel cell system immediately triggers an emergency correction mechanism.

[0012] As a specific implementation method, the specific process of step S5 is as follows: The flow deviation and the rate of change of deviation are used as input variables. The linguistic variables, including positive large, positive medium, and negative small, are quantified by the membership function. The output is converted into the pulse width correction coefficient. The iterative process follows the golden section search strategy and prioritizes multiple convergence calculations in the high-efficiency working area of ​​the fuel cell. After each iteration, the pressure difference across the membrane electrode is verified to prevent water flooding failure. After obtaining the smoothed flow deviation and its rate of change, the fuel cell system enters the pulse width parameter iterative correction stage based on the fuzzy proportional-integral-derivative (FID) control algorithm. The fuzzy proportional-integral-derivative (FID) control algorithm uses the smoothing deviation and the rate of change of deviation as two precise input variables. First, it quantizes them into seven standard linguistic variables through predefined triangular and trapezoidal membership functions. The fuzzification process determines the specific degree to which the input quantity belongs to each linguistic value. Then, the fuel cell system inference output is a fuzzy set of pulse width correction coefficients. Then, a precise correction coefficient value is obtained by defuzzification using the centroid method. This correction coefficient will be directly applied to the current pulse width parameter at the current test frequency point to generate a candidate new pulse width value.

[0013] As a specific implementation method, the specific process of step S6 is as follows: several sets of optimized parameters are stored at a resolution of 10Hz, with an additional temperature compensation coefficient. The parameters are written to the EEPROM storage area of ​​the fuel cell controller via the CAN bus, and a check code is generated to ensure data integrity. Before the parameter set is enabled, the actual metering ratio fluctuation range is verified by a hydrogen concentration sensor. The final output file contains a dynamic response curve spectrum for failure mode analysis of the on-board diagnostic system.

[0014] The second objective of this invention is to provide a high-frequency pulse characteristic and flow self-calibration testing device for a hydrogen injection valve, comprising a support frame, a storage cabinet, a support plate disposed above the support frame and the storage cabinet, a control panel and a set of testing instruments mounted on the support plate, and further comprising hardware components, connecting components and a storage medium. The hardware components include a hydrogen supply system housed in the storage cabinet, which includes a hydrogen supply pipeline and a hydrogen injection valve installed on the pipeline. The detection instrument group integrates a thermal mass flow meter for collecting hydrogen flow upstream or downstream of the hydrogen injection valve and a temperature-pressure composite sensor for collecting hydrogen temperature and pressure. The control panel integrates a fuel cell controller for controlling the opening and closing of the hydrogen injection valve and receiving data from the temperature-pressure composite sensor. A host computer for performing algorithm calculations and logic processing is installed on one side of the control panel. The connection component includes a signal / pipeline connection module, one end of which is connected to the host computer and the fuel cell controller; another end is connected to the outlet of the hydrogen supply system, the inlet and outlet of the hydrogen injection valve and the input end of the return pipeline; and the third end is connected to the signal end of the thermal mass flow meter and the temperature-pressure composite sensor. The storage medium is placed in the memory of the host computer, and the device executes the program in the storage medium through the host computer to perform high-frequency pulse characteristics and flow self-calibration test of the hydrogen injection valve.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1) This invention utilizes a temperature-pressure composite sensor for real-time density compensation and employs an FIR digital filter to suppress solenoid valve switching noise, fundamentally eliminating the impact of gas state changes and electrical noise on flow measurement accuracy. The established timestamp alignment mechanism controls timing deviations to the microsecond level, ensuring strict synchronization between control signals and flow data. More importantly, by subtracting background flow interference from the hydrogen circulation pump through an algorithm, a pure flow waveform characterizing the hydrogen injection valve's action is successfully extracted. The calibration test method of this invention provides highly faithful and time-accurate raw data, ensuring that all subsequent analysis and optimization are based on reliable data, greatly enhancing the credibility of the entire testing and calibration process. 2) The dynamic deviation assessment model constructed in this scheme combines the integral flow error of a single pulse cycle, the overshoot / undershoot of the response edge, and the deviation change rate, thereby comprehensively quantifying the static deviation and dynamic lag risk of flow control; the weighted moving average algorithm is used to smooth instantaneous fluctuations, effectively separating noise from the true trend; by setting two-level safety thresholds for smoothing deviation and its change rate, and triggering an emergency correction mechanism accordingly, the model can proactively intervene in the early stages when a severe shortage of hydrogen supply is about to occur. Its technical effect is to achieve proactive prevention of "hydrogen starvation" faults, significantly enhancing the safety and reliability of fuel cell system operation; 3) This solution utilizes an iterative algorithm that integrates fuzzy PID intelligent decision-making with the golden section efficient search to automatically optimize pulse width parameters at each frequency point within the key operating frequency band. Each iteration undergoes mandatory membrane electrode pressure difference safety verification, eliminating the risk of flooding failures due to improper parameters. The resulting frequency-pulse width two-dimensional correction matrix covers the entire operating range with a 10Hz resolution and includes a temperature compensation coefficient. It is written to the controller via the CAN bus and complies with ASIL-C functional safety requirements. The ultimate technical benefit is a plug-and-play, safe, and reliable high-precision hydrogen injection valve control parameter solution. This enables the fuel cell system to obtain rapid, accurate, and safe hydrogen supply under various operating conditions, including cold start, variable load, and high-efficiency zones, directly improving the performance, efficiency, and lifespan of the fuel cell stack. Attached Figure Description

[0016] Figure 1 This is a perspective view of the hydrogen injection valve high-frequency pulse characteristics and flow self-calibration test device described in this invention; Figure 2 This is a timing diagram of the multi-gradient pulse frequency test in step S2 of the hydrogen injection valve high-frequency pulse characteristics and flow self-calibration test method described in this invention; Figure 3 This is a diagram of the dynamic hydrogen flow rate data acquisition and processing architecture in step S3 of the hydrogen injection valve high-frequency pulse characteristics and flow self-calibration test method described in this invention. Figure 4 This is a logic diagram of dynamic flow deviation calculation and safety threshold evaluation in step S4 of the hydrogen injection valve high-frequency pulse characteristics and flow self-calibration test method described in this invention; Figure 5 This is a flowchart of the fuzzy PID parameter iteration and safety verification process in step S5 of the hydrogen injection valve high-frequency pulse characteristics and flow self-calibration test method described in this invention. The components include: 1. Support frame; 2. Storage cabinet; 3. Support plate; 4. Host computer; 5. Control panel; 6. Detection instrument group; 7. Hydrogen supply pipeline; and 8. Signal / pipeline connection module. Detailed Implementation

[0017] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0018] This invention discloses a test device for the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve, see [link to relevant documentation]. Figure 1As shown, the device includes a support frame 1, a storage cabinet 2, a support plate 3 mounted above the support frame 1 and the storage cabinet 2, a control panel 5 and a detection instrument group 6 mounted on the support plate 3. The support frame 1 serves as the main load-bearing structure, and the storage cabinet 2 is mounted on the side of the support frame 1. The device also includes hardware components, connecting components, and a storage medium. The hardware components include a hydrogen supply system placed inside the storage cabinet 2. The hydrogen supply system includes a hydrogen supply pipeline 7 and a hydrogen injection valve installed on the hydrogen supply pipeline 7. The detection instrument group 6 integrates a thermal mass flow meter for collecting hydrogen flow upstream or downstream of the hydrogen injection valve and a temperature-pressure composite sensor for collecting hydrogen temperature and pressure. The control panel 5 integrates a fuel cell controller for controlling the opening and closing of the hydrogen injection valve and receiving data from the temperature-pressure composite sensor. A host computer 4 for performing algorithm calculations and logic processing is installed on one side of the control panel 5. The connection components include a signal / pipeline connection module 8, which is a centralized connection interface for signal lines and hydrogen supply / return pipelines. It is located in the middle area of ​​the device support frame 1, near the intersection of the hydrogen supply pipeline 7 and the electrical control module. One end of the signal / pipeline connection module 8 is electrically connected to the host computer 4 and the fuel cell controller through a control signal line interface; the other end is connected to the signal terminals of the hydrogen injection valve, the thermal mass flow meter, and the temperature-pressure composite sensor through a dedicated interface; and the other end is connected to the outlet of the hydrogen supply system, the inlet and outlet of the hydrogen injection valve, and the input terminal of the return pipeline through a pipeline interface, forming a centralized transfer hub for control signals and hydrogen pathways. The storage medium is placed in the memory of the host computer 4. The device executes the program in the storage medium through the host computer 4 to perform high-frequency pulse characteristics and flow self-calibration test of the hydrogen injection valve.

[0019] This document also discloses a test method for the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve, including the following steps: Step S1: Set the initial parameters of the high-frequency pulse of the hydrogen injection valve and the target flow rate, and construct a nonlinear mapping model between the initial pulse duty cycle and the hydrogen output flow rate. Step S2: Perform a multi-gradient pulse frequency injection test; Step S3: Real-time acquisition of dynamic hydrogen flow rate data, and synchronous acquisition of flow data during the pulse jet process via a thermal mass flow meter; Step S4: Calculate the deviation between the actual flow rate and the target flow rate, and construct a dynamic deviation assessment model based on the stoichiometric ratio of the chemical reaction in the hydrogen fuel cell. Step S5: Iteratively correct the pulse width parameter based on the deviation value, and apply the fuzzy PID control algorithm to achieve adaptive parameter optimization; Step S6: Output the calibration parameter set that meets the accuracy threshold, and generate the final parameter set containing the frequency-pulse width two-dimensional correction matrix.

[0020] Specifically, in step S1, the initial parameters of the high-frequency pulse and the target flow rate of the hydrogen injection valve are set through the following steps: When initially setting the high-frequency pulse control parameters of the hydrogen injection valve during the cold start phase of a hydrogen fuel cell system, theoretical calculations must first be performed based on the rated power of the fuel cell stack and the tolerance range of the hydrogen pressure on the anode side. Taking an 80kW fuel cell stack as an example, its anode hydrogen working pressure typically needs to be strictly controlled between 0.15MPa and 0.25MPa. Based on this, the theoretical hydrogen mass flow rate required by the system can be derived, typically around 120 SLPM. The core task of initial parameter setting is to convert this theoretical flow rate into an executable control signal for the hydrogen injection valve. This requires determining the initial pulse frequency range and basic pulse width based on the dynamic response characteristic curve provided by the valve manufacturer. For example, the frequency range can be set to 50Hz to 200Hz, and the basic pulse width range can be set to 0.5ms to 2ms. During this process, special attention must be paid to integrating real-time ambient temperature feedback from the system temperature sensor and synchronously compensating for the gas expansion coefficient of hydrogen due to temperature changes to ensure the physical accuracy of the flow rate reference. The key output and core objective of this step is to construct an initial nonlinear mapping model between the pulse duty cycle and the hydrogen output flow rate. This model will serve as the sole reference benchmark for all subsequent multi-gradient pulse frequency injection tests and parameter iterative optimizations. The accuracy and reliability of this model directly determine the convergence speed of the entire flow self-calibration process and the upper limit of the accuracy of the final calibration parameters.

[0021] Step S2 involves performing a multi-gradient pulse frequency injection test. This test is conducted to accurately verify and optimize the dynamic response characteristics of the hydrogen injection valve under varying load conditions in the fuel cell system, and to overcome potential hydrogen supply lag issues during acceleration. The specific process is as follows: A segmented incremental pulse frequency excitation strategy is adopted. Within a complete test time window, the pulse signal is controlled to start from the initial frequency and increase stepwise with a fixed step size until the upper limit of the frequency is reached. Each independent frequency band needs to continuously output N complete pulse cycles, where N is a natural number greater than 10, and the pressure fluctuation of the fuel cell anode manifold is monitored simultaneously during this process; When the test frequency enters the high-frequency region, the pulse width reduction compensation mechanism is activated simultaneously. By reducing the on-time of a single pulse, the average current and heat accumulation of the hydrogen injection valve solenoid coil are controlled. The high-frequency region is the region where the frequency reaches 3 / 4 of the upper frequency limit.

[0022] For example, see Figure 2As shown, within a complete 10ms test time window, the control pulse signal starts at a frequency of 50Hz and increases in steps of 25Hz until it reaches the upper frequency limit of 200Hz. To ensure that stable and reliable data can be collected at each test frequency point, each independent frequency band needs to maintain stable injection for a sufficient duration, specifically by continuously outputting 20 complete pulse cycles, while simultaneously monitoring the pressure fluctuations of the fuel cell anode manifold.

[0023] When the test frequency enters the high-frequency region above 150Hz, the pulse width reduction compensation mechanism must be activated simultaneously. By appropriately reducing the on-time of a single pulse, the average current and heat accumulation of the hydrogen injection valve's solenoid coil can be effectively controlled, thereby preventing valve performance degradation or hardware damage caused by coil overheating. By covering the excitation and observation of the entire operating frequency band of the hydrogen injection valve, its flow response characteristics, especially its rapid dynamic response capability, can be accurately characterized, thus providing a key data foundation for subsequent parameter calibration. Its direct effect is to significantly improve the transient response speed and following accuracy of hydrogen supply in fuel cell systems when facing power step demands.

[0024] The specific process of step S3 is as follows: In the anode circulation loop of the fuel cell, a temperature-pressure composite sensor is deployed to eliminate measurement errors caused by changes in gas density; an FIR digital filter is used to suppress noise generated by the opening and closing of the solenoid valve; a timestamp alignment mechanism is established to ensure that the timing deviation between the flow data and the pulse control signal is ≤10μs; for the fuel cell system, background flow interference caused by the circulating hydrogen pump is subtracted from the acquisition algorithm to obtain a clean injection flow waveform. The dynamic hydrogen flow rate data acquisition and processing architecture diagram is shown below. Figure 3 As shown.

[0025] Specifically, the process of deploying a temperature-pressure composite sensor in the anode loop of a fuel cell to eliminate measurement errors caused by changes in gas density is as follows: The instantaneous mass flow rate Q in the anode circuit is measured using a thermal mass flow meter. m (t), whose original signal S raw (t) contains the actual jet flow rate Q inject (t), background flow rate Q generated by the circulating hydrogen pump background And electromagnetic noise N introduced by the opening and closing of the solenoid valve em (t) and random noise N r (t), that is: S raw (t)=Q inject (t)+Q background +N em (t)+N r (t), To eliminate gas temperature T gWith pressure P g The measurement error caused by the change in gas density is compensated in real time by deploying a temperature-pressure composite sensor near the measuring point of the thermal mass flow meter. The actual volumetric flow rate Q is calculated based on the ideal gas law. v (t) Mass flow rate Q corrected to standard conditions m (t), the correction formula is: , Where P0 and T0 represent standard pressure and standard temperature, respectively. Where is the molar mass of hydrogen and R is the universal gas constant. The above formula is used to convert the volumetric flow rate under actual working conditions to the standard state, thereby ensuring the comparability of flow rate data under different temperature and pressure conditions and eliminating the measurement error introduced by the change of gas state. To suppress the electromagnetic noise N generated by the opening and closing of the solenoid valve em (t), the fuel cell system is processed using a finite-length unit impulse response digital filter based on the Hanning window design, and its difference equation is expressed as: , Where x[ni] is the sampled value of the original signal at time ni; y[n] is the filtered output sequence; and h[i] is the signal based on the cutoff frequency f. c The calculated filter coefficients, where N is the filter order, are used to attenuate noise energy in a specific frequency band through convolution operations, thereby obtaining a smooth flow signal S. filtered (t); In order to achieve precise time correlation between flow data and pulse control signal, a strict timestamp alignment mechanism must be established. Through hardware triggering and high-precision clock synchronization, the timing deviation Δt between the two must not exceed 10μs. For fuel cell systems, the specific process of deducting background flow interference from the circulating hydrogen pump in the acquisition algorithm is as follows: The average background flow rate was measured during the quiescent period when the hydrogen injection valve was closed. The jet flow waveform is then subtracted from the dynamic signal containing the jet, thereby extracting the pure jet flow waveform. This ensures that the objects targeted in subsequent analyses are entirely generated by the pulse action of the hydrogen injection valve. After the above precise compensation, filtering and correction processes, the final dynamic hydrogen flow rate data has high fidelity and time consistency, providing a reliable data foundation for the next step of accurately calculating the dynamic deviation between the actual flow rate and the target flow rate.

[0026] The processing of flow data in step S3 does not rely solely on the FIR filter, but rather constructs an integrated data processing architecture that integrates multiple techniques such as real-time temperature and pressure compensation, customized FIR filtering, microsecond-level timestamp alignment, and background flow deduction. Among these techniques, the FIR filter, as a key component for noise suppression, is not a universal standard filter.

[0027] Here, in step S4, the deviation between the actual flow rate and the target flow rate is calculated. The specific process of constructing a dynamic deviation evaluation model based on the stoichiometric ratio of the hydrogen fuel cell is as follows: The absolute error between the integral flow rate and the target value within a single pulse cycle is taken, and the response overshoot at the rising / falling edge is superimposed. A weighted moving average algorithm is used to smooth the instantaneous fluctuations. Considering the scenario of sudden changes in fuel cell load, the derivative of the deviation value is calculated as a dynamic performance index. When the deviation is detected to continuously exceed the safety threshold, an emergency correction mechanism is triggered to prevent "hydrogen starvation" damage to the fuel cell stack. To obtain the pure dynamic injection flow waveform Q of the hydrogen injection valve ingect After (t), it is compared with the target hydrogen flow rate Q calculated in real time by the fuel cell system based on the current load. target (t) is compared, and a dynamic deviation evaluation model based on the stoichiometry of the chemical reaction in hydrogen fuel cells is constructed. The core of the deviation calculation is to evaluate the flow tracking accuracy within each independent pulse cycle Tp. First, the cycle integral flow deviation is calculated: , Where k is the period number; Q target (k) represents the average target flow rate within k periods; Qinject(t) represents the actual dynamic hydrogen injection flow rate of the hydrogen injection valve at time t, which is a continuous flow signal that varies with time; Tp represents the duration of a single hydrogen injection valve pulse period, which is the reciprocal of the pulse frequency, Tp=1 / f; To quantify the dynamic response defects of valves, the overshoot Q of the flow waveform at the rising and falling edges is identified and extracted. vershoot with under-adjustment U ndershoot This is added as a penalty term to the integral deviation to form the original instantaneous deviation: , In the formula, α and β are weighting coefficients for the effects of overshoot and undershoot, and their values ​​depend on the degree of harm they cause to the stack performance. Because pulse jets and fluid pulsations can lead to E... raw (k) In the case of high-frequency fluctuations, a weighted moving average algorithm is needed to smooth the fluctuations in order to extract the trend term. The smoothed deviation E smooth (k) From the formula: , Where M is the width of the sliding window, w iThe weighting coefficients are allocated according to the exponential decay law, satisfying... While preserving the true deviation trend, it filters out meaningless fluctuations caused by measurement noise; For the severe operating conditions of fuel cell load surges, the absolute value of the deviation itself is insufficient to reflect the risk; therefore, the deviation change rate must be introduced as a key dynamic performance indicator. , This indicator directly reflects the severity of the hydrogen supply lag; The entire deviation assessment model outputs two key criteria: smoothing deviation E. smooth (k) and its rate of change R dynamic (k); To ensure the safety of the fuel cell stack, two safety thresholds are preset: when E smooth (k) Continuously exceeds the warning threshold Th warning Reaching N cycles, or R dynamic (k) Instantly exceeds the emergency threshold Th critical At this time, its actions include, but are not limited to, switching to a preset safe pulse width spectrum and increasing the pulse frequency, with the aim of forcibly pulling back the flow to avoid irreversible damage to the battery stack caused by "hydrogen starvation." The precise deviation signal calculated and determined in this step will become a reliable input for the next step of driving the fuzzy PID controller to perform iterative correction of the pulse width parameters. The logic diagram for dynamic flow deviation calculation and safety threshold assessment is shown in [link to logic diagram]. Figure 4 As shown.

[0028] The specific process of step S5 is as follows: Based on the iterative correction of pulse width parameters using deviation values, a fuzzy PID control algorithm is applied to achieve adaptive parameter optimization: Flow deviation and deviation change rate are used as input variables, and linguistic variables such as "positive large / positive medium / negative small" are quantified through membership functions; the output is converted into pulse width correction coefficients. The iterative process follows the golden section search strategy, prioritizing five convergence calculations in the high-efficiency operating range of the fuel cell, i.e., the 150-180Hz high-frequency band. After each iteration, the pressure difference across the membrane electrode is verified to prevent flooding failure. After obtaining the smoothed flow deviation and its rate of change, the system enters the pulse width parameter iterative correction stage based on the fuzzy proportional-integral-derivative (FIG) control algorithm. This algorithm uses the smoothed deviation and the rate of change as two precise input variables, first quantizing them into seven standard linguistic variables using predefined triangular and trapezoidal membership functions. The fuzzification process determines the specific degree to which the input quantity belongs to each linguistic value. Subsequently, the system performs inference based on a complete fuzzy rule base covering all input combinations, with rules in the form of conditional statements based on linguistic variables. The inference output is a fuzzy set of pulse width correction coefficients, which is then defuzzified using the centroid method to obtain a precise correction coefficient value. This correction coefficient is directly applied to the current pulse width parameter at the current test frequency point, generating a candidate new pulse width value. See the flowchart for the fuzzy PID parameter iteration and safety verification process. Figure 5 As shown.

[0029] The iterative optimization process specifically targets the key high-frequency band corresponding to the high-efficiency operating range of fuel cells and employs a golden section search strategy to accelerate convergence. Taking pulse width optimization at a specific frequency point as an example, an initial pulse width search interval containing the expected optimal solution is first determined. The first convergence calculation evaluates the system performance index corresponding to the two golden section points within the interval, which is weighted by the absolute value of the smoothing deviation and the absolute value of the rate of change of the deviation. After comparing the performance indexes corresponding to the two golden section points, the interval segment with the worse index is discarded, and a new search interval containing the better solution is retained. The second to fifth convergence calculations strictly repeat this evaluation and interval shrinkage process, with each iteration reducing the search interval to a fixed proportion of the previous one. After five precise calculations, the search interval has fully converged, and the midpoint of the interval determined after the fifth iteration is finally taken as the optimal pulse width value at that specific frequency point.

[0030] After each iteration calculates and obtains a new pulse width parameter, a safety verification is immediately performed. The system controls the hydrogen injection valve to operate briefly with the new parameters, while simultaneously monitoring the pressure difference data across the membrane electrode assembly (MEA) in real time. The verification standard is that the MEA pressure difference must be strictly stable within the pre-set allowable operating range. If the MEA pressure difference is detected to approach or exceed the safety limit, it is immediately determined that the pulse width parameter poses a potential risk of flooding, and the result of this iteration is marked as invalid. The system will automatically revert to the previously verified safe parameter and trigger a recalculation of the control algorithm. Only when the performance indicators meet the convergence requirements and the MEA pressure difference verification is confirmed to be successful is the pulse width parameter iteration at that frequency point considered a successful completion. This standardization process is executed sequentially at all specified frequency points within the target frequency band, ultimately generating an optimal pulse width parameter for each frequency point that has undergone dynamic optimization and safety verification. These complete parameters are temporarily stored by the system and are prepared to be integrated and output as a complete calibration parameter set that meets automotive-grade accuracy and safety requirements in the final step.

[0031] In step (5), taking parameter optimization at a frequency of 160Hz as an example, a five-fold convergence calculation process using the golden section search strategy is employed. The iteration process is shown in Table 1. This process aims to find a solution that optimizes performance indicators. Minimum optimal pulse width PW optimal The initial search interval is set to [1.0, 3.0] ms; Table 1

[0032]

[0033] In the golden section search, the performance index J corresponding to the pulse width at the golden section point is the sum of the absolute value of the weighted smoothed flow deviation and the absolute value of the deviation change rate. J a For example, the details are as follows:

[0034] k 1. k 2: This refers to the weighting coefficient, set based on the impact of smoothed flow deviation and deviation change rate on the performance and safety of the hydrogen fuel cell stack, to meet... k 1+ k 2 = 1; The absolute value of the flow deviation after smoothing by a weighted moving average algorithm at the golden section pulse width; The absolute value of the rate of change of flow deviation at the golden ratio pulse width; Assuming a frequency of 160Hz, the weighting coefficients are set. k 1 = 0.6 k 2=0.4, 1.764ms pulse width was acquired after security verification. =2.0, =1.625, then: J a =0.6×2.0+0.4×1.625=1.2+0.65=1.85 J b The calculation method and J a Similarly.

[0035] In each iterative calculation based on the golden section search, regardless of the golden section point PW a PW b Each evaluation is accompanied by a mandatory membrane electrode pressure differential safety verification. The system controls the hydrogen injection valve to perform a short injection test cycle using the candidate pulse width parameters to be evaluated, while simultaneously acquiring the pressure differential data ΔP across the membrane electrode at a high frequency. MEAThe core of the verification is to confirm ΔP. MEA Whether it remains stable within the predefined allowable operating range [ΔP] throughout the entire test cycle min ΔP max Within this range, it is particularly important to strictly monitor whether it approaches or exceeds the upper limit threshold ΔP, which characterizes the risk of flooding. max Only when the differential pressure response corresponding to a candidate parameter fully passes this verification is the performance index J of that parameter point allowed to participate in interval comparison and update decisions; otherwise, if the verification fails, the candidate parameter is immediately determined to be invalid, the search logic of this iteration will ignore this point, and may trigger a protective reset of the control algorithm. This safety verification mechanism is an integral part of parameter optimization, ensuring that the direction of each iteration simultaneously follows the dual criteria of optimal performance and lowest risk.

[0036] After the fifth iteration, the optimal pulse width PW optimal Take the midpoint of the final convergence interval [1.944, 2.124] ms, i.e., PW. optimal =2.034ms. This value has completed the membrane electrode differential pressure safety verification, and the corresponding performance index J has met the convergence requirements.

[0037] The table above clearly illustrates the shrinking of the search interval, the evaluation of computation points, the performance-based comparison decisions, and the ongoing safety verification throughout the five iterations, fully describing the closed-loop process of automatic parameter optimization and risk control. This optimal value will be used as the calibration parameter for that frequency point and incorporated into the final parameter set generated in step S6.

[0038] The specific process of step S6 is as follows: 200 sets of optimized parameters are stored at a resolution of 10Hz, with an additional temperature compensation coefficient. The parameters are written to the EEPROM storage area of ​​the fuel cell controller via the CAN bus, and a check code is generated to ensure data integrity. Before the parameter set is enabled, the actual metering ratio fluctuation range is verified by a hydrogen concentration sensor to ensure that the automotive-grade ASIL-C functional safety requirements are met. The final output file includes a dynamic response curve graph, which is used for failure mode analysis of the on-board diagnostic system.

[0039] The present invention relates to a test device and method for high-frequency pulse characteristics and flow self-calibration of hydrogen injection valves. Through initial parameter setting, multi-gradient pulse testing, dynamic flow acquisition, deviation calculation and safety assessment, pulse width iterative correction, and finally outputting a calibration parameter set, the device and method achieve accurate testing and flow self-calibration of the high-frequency pulse characteristics of hydrogen injection valves. This provides a reliable data foundation for subsequent analysis and optimization, improves the accuracy and dynamic response speed of hydrogen supply in hydrogen injection valves, and prevents faults such as "hydrogen starvation" and "flooding" through safety thresholds and other mechanisms. Ultimately, this ensures the safe, efficient, and stable operation of hydrogen fuel cell systems under cold start, variable load, and other operating conditions.

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used above are only some embodiments described in this invention. Obviously, those skilled in the art can obtain other drawings based on these drawings.

[0041] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve, characterized in that, Includes the following steps: Step S1: Set the initial parameters of the high-frequency pulse of the hydrogen injection valve and the target flow rate, and construct a nonlinear mapping model between the initial pulse duty cycle and the hydrogen output flow rate. Step S2: Perform a multi-gradient pulse frequency injection test; Step S3: Real-time acquisition of dynamic hydrogen flow rate data, and synchronous acquisition of flow data during the pulse jet process via a thermal mass flow meter; Step S4: Calculate the deviation between the actual flow rate and the target flow rate, and construct a dynamic deviation assessment model based on the stoichiometric ratio of the chemical reaction in the hydrogen fuel cell. Step S5: Iteratively correct the pulse width parameter based on the deviation value, and apply the fuzzy PID control algorithm to achieve adaptive parameter optimization; Step S6: Output the calibration parameter set that meets the accuracy threshold, and generate the final parameter set containing the frequency-pulse width two-dimensional correction matrix.

2. The method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve according to claim 1, characterized in that, In step S1, the initial parameters of the high-frequency pulse and the target flow rate of the hydrogen injection valve are set through the following steps: Based on the rated power of the fuel cell stack and the tolerance range of hydrogen pressure on the anode side, the target value of hydrogen flow rate actually required by the fuel cell system is calculated. Then, according to the valve response parameters provided by the hydrogen injection valve manufacturer, the target value of hydrogen flow rate is converted into a control signal that the valve can recognize, and the initial pulse frequency and pulse width range are determined. During this process, the real-time ambient temperature feedback from the temperature sensor in the fuel cell system is collected to correct the volume change of hydrogen caused by temperature changes. Then, the duty cycle of the pulse is matched with the actual output hydrogen flow rate to establish an initial matching model.

3. The method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve according to claim 1, characterized in that, The specific process of step S2 is as follows: A segmented incremental pulse frequency excitation strategy is adopted. Within a complete test time window, the pulse signal is controlled to start from the initial frequency and increase stepwise with a fixed step size until the upper limit of the frequency is reached. Each independent frequency band needs to continuously output N complete pulse cycles, where N is a natural number greater than 10, and the pressure fluctuation of the fuel cell anode manifold is monitored simultaneously during this process; When the test frequency enters the high-frequency region, the pulse width reduction compensation mechanism is activated simultaneously. By reducing the on-time of a single pulse, the average current and heat accumulation of the hydrogen injection valve solenoid coil are controlled. The high-frequency region is the region where the frequency reaches 3 / 4 of the upper frequency limit.

4. The method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve according to claim 1, characterized in that, The specific process of step S3 is as follows: In the anode circulation loop of the fuel cell, a temperature-pressure composite sensor is deployed to eliminate measurement errors caused by changes in gas density; an FIR digital filter is used to suppress noise generated by the opening and closing of the solenoid valve; a timestamp alignment mechanism is established to ensure that the timing deviation between the flow data and the pulse control signal is ≤10μs; for the fuel cell system, background flow interference caused by the circulating hydrogen pump is subtracted in the acquisition algorithm to obtain a clean injection flow waveform.

5. The method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve according to claim 4, characterized in that, In step S3, the specific process of deploying a temperature-pressure composite sensor in the fuel cell anode loop to eliminate measurement errors caused by changes in gas density is as follows: The instantaneous mass flow rate Q in the anode circuit is measured using a thermal mass flow meter. m (t), whose original signal S raw (t) contains the actual jet flow rate Q inject (t), background flow rate Q generated by the circulating hydrogen pump background And electromagnetic noise N introduced by the opening and closing of the solenoid valve em (t) and random noise N r (t), that is: S raw (t)=Q inject (t)+Q background +N em (t)+N r (t), To eliminate gas temperature T g With pressure P g The measurement error caused by the change in gas density is compensated in real time by deploying a temperature-pressure composite sensor near the measuring point of the thermal mass flow meter. The actual volumetric flow rate Q is calculated based on the ideal gas law. v (t) Mass flow rate Q corrected to standard conditions m (t), the correction formula is: ， Where P0 and T0 represent standard pressure and standard temperature, respectively. Here, R is the molar mass of hydrogen, and R is the universal gas constant. The above formula is used to convert the volumetric flow rate under actual working conditions to the standard state, thus eliminating the measurement error introduced by the change of gas state. To suppress the electromagnetic noise N generated by the opening and closing of the solenoid valve em (t), the fuel cell system is processed using a finite-length unit impulse response digital filter based on the Hanning window design, and its difference equation is expressed as: ， Where x[ni] is the sampled value of the original signal at time ni; y[n] is the filtered output sequence; and h[i] is the signal based on the cutoff frequency f. c The calculated filter coefficients, where N is the filter order, are used to attenuate noise energy in a specific frequency band through convolution operations, thereby obtaining a smooth flow signal S. filtered (t); For fuel cell systems, the specific process of deducting background flow interference from the circulating hydrogen pump in the acquisition algorithm is as follows: The average background flow rate was measured during the quiescent period when the hydrogen injection valve was closed. The jet flow waveform is then subtracted from the dynamic signal containing the jet, thereby extracting the pure jet flow waveform. This ensures that the objects targeted in subsequent analyses are entirely generated by the pulse action of the hydrogen injection valve.

6. The method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve according to claim 1, characterized in that, The specific process for calculating the deviation between the actual flow rate and the target flow rate in step S4 is as follows: The absolute error between the integral flow rate and the target value within a single pulse cycle is taken, and the response overshoot at the rising / falling edge is superimposed. A weighted moving average algorithm is used to smooth instantaneous fluctuations. Considering the scenario of sudden changes in fuel cell load, the derivative of the deviation value is calculated as a dynamic performance index. When the deviation is detected to continuously exceed the safety threshold, an emergency correction mechanism is triggered to prevent damage to the fuel cell stack. To obtain the pure dynamic injection flow waveform Q of the hydrogen injection valve ingect After (t), it is compared with the target hydrogen flow rate Q calculated in real time by the fuel cell system based on the current load. target (t) is compared, and a dynamic deviation evaluation model based on the stoichiometry of the hydrogen fuel cell chemical reaction is constructed to evaluate each independent pulse cycle T. P For accurate flow tracking, first calculate the periodic integral flow deviation: ， Where k is the period number; Q target (k) represents the average target flow rate within k periods; Qinject(t) represents the actual dynamic hydrogen injection flow rate of the hydrogen injection valve at time t, which is a continuous flow signal that varies with time; Tp represents the duration of a single hydrogen injection valve pulse period, which is the reciprocal of the pulse frequency, Tp=1 / f; To quantify the dynamic response defects of valves, the overshoot Q of the flow waveform at the rising and falling edges is identified and extracted. vershoot with under-adjustment U ndershoot This is added as a penalty term to the integral deviation to form the original instantaneous deviation: ， In the formula, α and β are weighting coefficients for the effects of overshoot and undershoot. Due to pulse injection and fluid pulsation, E raw (k) In the case of high-frequency fluctuations, a weighted moving average algorithm is needed to smooth the fluctuations in order to extract the trend term. The smoothed deviation E smooth (k) From the formula: ， Where M is the width of the sliding window, w i The weighting coefficients are allocated according to the exponential decay law, satisfying... While preserving the true deviation trend, it filters out meaningless fluctuations caused by measurement noise. To address the severe operating conditions of rapid load changes in fuel cells, the deviation change rate is introduced as a key dynamic performance indicator: , This indicator directly reflects the severity of the hydrogen supply lag; The entire deviation assessment model outputs two key criteria: smoothing deviation E. smooth (k) and its rate of change R dynamic (k); To ensure the safety of the fuel cell stack, two safety thresholds are preset: when E smooth (k) Continuously exceeds the warning threshold Th warning Reaching N cycles, or R dynamic (k) Instantly exceeds the emergency threshold Th critical In this case, the fuel cell system immediately triggers an emergency correction mechanism.

7. The method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve according to claim 1, characterized in that, The specific process of step S5 is as follows: The flow deviation and the rate of change of deviation are used as input variables. The linguistic variables, including positive large, positive medium, and negative small, are quantified by the membership function. The output is converted into the pulse width correction coefficient. The iterative process follows the golden section search strategy and prioritizes multiple convergence calculations in the high-efficiency working area of ​​the fuel cell. After each iteration, the pressure difference across the membrane electrode is verified to prevent water flooding failure. After obtaining the smoothed flow deviation and its rate of change, the fuel cell system enters the pulse width parameter iterative correction stage based on the fuzzy proportional-integral-derivative (FID) control algorithm. The fuzzy proportional-integral-derivative (FID) control algorithm uses the smoothing deviation and the rate of change of deviation as two precise input variables. First, it quantizes them into seven standard linguistic variables through predefined triangular and trapezoidal membership functions. The fuzzification process determines the specific degree to which the input quantity belongs to each linguistic value. Then, the fuel cell system inference output is a fuzzy set of pulse width correction coefficients. Then, a precise correction coefficient value is obtained by defuzzification using the centroid method. This correction coefficient will be directly applied to the current pulse width parameter at the current test frequency point to generate a candidate new pulse width value.

8. The method for testing the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve according to claim 1, characterized in that, The specific process of step S6 is as follows: Several sets of optimized parameters are stored at a resolution of 10Hz, with an additional temperature compensation coefficient. The parameters are written to the EEPROM storage area of ​​the fuel cell controller via the CAN bus, and a check code is generated to ensure data integrity. Before the parameter set is enabled, the actual metering ratio fluctuation range is verified by a hydrogen concentration sensor. The final output file contains a dynamic response curve spectrum for failure mode analysis of the on-board diagnostic system.

9. A testing device for the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve, used to implement the testing method for the high-frequency pulse characteristics and flow self-calibration of a hydrogen injection valve as described in any one of claims 1 to 8, characterized in that, It includes a support frame, storage cabinets, a support plate mounted on the support frame and storage cabinets, a control panel and a set of measuring instruments mounted on the support plate, as well as hardware components, connecting components and storage media. The hardware components include a hydrogen supply system housed in the storage cabinet, which includes a hydrogen supply pipeline and a hydrogen injection valve installed on the pipeline. The detection instrument group integrates a thermal mass flow meter for collecting hydrogen flow upstream or downstream of the hydrogen injection valve and a temperature-pressure composite sensor for collecting hydrogen temperature and pressure. The control panel integrates a fuel cell controller for controlling the opening and closing of the hydrogen injection valve and receiving data from the temperature-pressure composite sensor. A host computer for performing algorithm calculations and logic processing is installed on one side of the control panel. The connection component includes a signal / pipeline connection module, one end of which is connected to the host computer and the fuel cell controller; another end is connected to the outlet of the hydrogen supply system, the inlet and outlet of the hydrogen injection valve and the input end of the return pipeline; and the third end is connected to the signal end of the thermal mass flow meter and the temperature-pressure composite sensor. The storage medium is placed in the memory of the host computer, and the device executes the program in the storage medium through the host computer to perform high-frequency pulse characteristics and flow self-calibration test of the hydrogen injection valve.

Images