Visual calibration method for hydrogen-sensitive material sensor based on schlieren imaging

Abstract

The invention provides a hydrogen-sensitive material sensor visual calibration method based on schlieren imaging. The hydrogen-sensitive material sensor visual calibration method comprises the following steps: (1) selecting a sensor to be calibrated and configuring experimental parameters; (2) constructing a visual calibration experiment bench; (3) executing a calibration experiment; (4) carrying out gray scale-concentration calibration; (5) calibrating a response time error; (6) scoring the performance of the sensor; and (7) correcting the sensor. According to the schlieren imaging-based hydrogen-sensitive material sensor visual calibration method provided by the invention, a reflective parallel light schlieren system and sensor collaborative experiment bench is constructed, so that time-space synchronous visual dynamic monitoring of a hydrogen diffusion process and sensor output is realized, and the defect that traditional calibration is invisible is overcome; a visual concentration curve is extracted by using a gray-concentration calibration technology, and optimization of a sensor compensation circuit, a structure and an algorithm is guided by combining time and concentration dimension error quantization and a comprehensive scoring mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of gas sensor calibration, and specifically relates to a visualization calibration method for a hydrogen-sensitive material sensor based on schlieren imaging. Background Technique

[0002] Currently, the hydrogen energy storage and transportation system generally adopts a high-pressure storage method of 35 - 70 MPa. The characteristics of high-pressure hydrogen, such as easy leakage, wide flammable range (4% - 75% vol), and low ignition energy (0.017 mJ), make the accuracy of leakage detection sensors a key technical bottleneck in the hydrogen safety system.

[0003] The core of the existing calibration methods is to improve the sensor performance through multi-dimensional compensation (temperature, non-linearity, noise) and dynamic calibration (standard gas, real-time algorithm). For example, temperature compensation technology, which includes hardware compensation, designing a temperature compensation circuit (such as the temperature compensation circuit of Zhang Qi et al.), directly correcting the temperature drift error in the sensor output. Integrating a temperature sensor (such as the platinum temperature sensor in YuKun's research), and real-time monitoring of the ambient temperature. There is also an algorithm compensation that adjusts the baseline value of the sensor output according to the temperature change or combines the thermodynamic model of hydrogen adsorption (such as the hydrogen dissolution characteristics of PdNi alloy), and dynamically corrects the non-linear temperature influence (the improved algorithm proposed by YuKun).

[0004] These traditional gas sensor calibration methods mainly have three major technical defects: relying on complex compensation circuit design, and needing to optimize parameters through multiple rounds of iteration; lacking visualization monitoring means during the calibration process, and being unable to directly obtain the dynamic process of gas diffusion; existing laboratory calibrations mostly adopt static concentration field calibration, and being unable to truly reflect the dynamic response characteristics during the actual leakage process. These defects result in the calibration error of the sensor response time generally exceeding 200 ms, and it is difficult to meet the fast response requirements of high-pressure hydrogen system leakage.

[0005] Therefore, a visualization calibration method for a hydrogen-sensitive material sensor based on schlieren imaging is proposed. Summary of the Invention

[0006] The purpose of the present invention is to solve the problems of non-visualization, low accuracy, and lack of dynamic response calibration in traditional sensor calibration technology in the background technique. A visualization calibration method for a hydrogen-sensitive material sensor based on schlieren imaging is provided. Through high-precision spatio-temporal synchronous monitoring and quantitative analysis, the response speed and accuracy of the sensor are optimized to meet the hydrogen energy safety monitoring requirements, and it is especially suitable for the accuracy calibration of leakage detection sensors in hydrogen energy field storage and transportation devices.

[0007] The specific technical solution is as follows:

[0008] A visualization calibration method for a hydrogen-sensitive material sensor based on schlieren imaging, comprising the following steps:

[0009] (1) Select the sensor to be calibrated and configure the experimental parameters: Configure the concentration of the hydrogen-air mixture gas according to the maximum range Cmax of the sensor, adjust the size of the nozzle and the vertical distance from the sensor, so that the hydrogen-sensitive material is completely covered by the jet, and adjust the frame rate of the high-speed camera based on the nominal response speed of the sensor;

[0010] (2) Construct a visualization calibration experimental bench: including a hydrogen injection device, a reflective parallel light schlieren imaging system and a sensor installation device. The schlieren imaging system includes a schlieren instrument, a high-speed camera and a data acquisition board. The sensor is fixed directly above the injection port and connected to the data acquisition system;

[0011] (3) Perform the calibration experiment: Synchronously trigger hydrogen leakage, schlieren imaging and sensor data acquisition, record the gas diffusion process and the sensor concentration-time curve g(t), and intercept the instantaneous frame when the gas touches the sensor to determine the accurate time point;

[0012] (4) Gray-scale - concentration calibration: Convert the schlieren video frame by frame into grayscale images, extract the relative gray-scale difference ΔG of the selected area, establish a quantitative relationship between the gray-scale value and the hydrogen concentration, and generate a visualization concentration-time curve f(t);

[0013] (5) Response time error calibration: Extract the time dimension characteristic quantities T10, T90, ΔT10, ΔT90, and the concentration dimension characteristic quantities Δc10, Δc90, Δcs based on f(t) and g(t);

[0014] (6) Sensor performance scoring: Calculate the comprehensive score of the sensor through the comprehensive weight formula

[0015] S Q = αS ΔT10 + βS ΔT90 + γS Δc10 + δS Δc90 + ∈S Δcs where the weight coefficients are determined by the subjective and objective weighting method;

[0016] (7) Sensor correction: Adjust the sensor compensation circuit, geometric structure or signal processing algorithm according to the scoring results, and repeat the calibration until the performance meets the standard.

[0017] For the above visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging, in step (1), the frame rate setting of the high-speed camera needs to meet that the time interval of each frame is 1-2 orders of magnitude smaller than the response speed of the sensor.

[0018] For the above visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging, the gray-scale - concentration calibration in step (4) specifically includes:

[0019] (4.1) Convert the color video frame by frame into grayscale images;

[0020] (4.2) Select multiple frames of images for grayscale averaging processing, and extract the relative grayscale difference ΔG of the sensor probe area;

[0021] (4.3) Establish the mapping relationship between ΔG and hydrogen concentration through function fitting.

[0022] The above visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging, wherein the time dimension feature quantity in the step (5) is defined as:

[0023] T10 = t2 - t1, where t1 is the time when the sensor starts to show a reading, and t2 is the time when the sensor output reaches 10% of the stable value;

[0024] ΔT10 = T10′ - T10, where T10′ is the time difference calculated based on the visualization curve.

[0025] The above visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging, wherein the concentration dimension feature quantity in the step (5) includes:

[0026] Δc10 = f(t2′) - g(t2′), where t2′ is the moment when the visualization curve reaches 10% concentration;

[0027] Characterize the comprehensive response accuracy of the sensor.

[0028] The above visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging, wherein the weight coefficient in the step (6) is determined by the analytic hierarchy process, including:

[0029] Construct an index hierarchy, including the response speed ΔT and the response accuracy Δc;

[0030] Combine expert scoring and principal component analysis, calculate the subjective and objective weights and then perform weighted fusion.

[0031] The above visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging, wherein the layout requirements of the reflective parallel light schlieren system in the step (2) are:

[0032] The hydrogen nozzle is located at the center position below the field of view of the schlieren instrument;

[0033] The position of the schlieren instrument, the nozzle and the lighting conditions remain constant during the experiment.

[0034] Among them, the above visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging is applicable to the dynamic response calibration of the leakage detection sensor of the hydrogen energy storage and transportation device, and is compatible with any gas and sensor principle.

[0035] The above visualization calibration method for a hydrogen-sensitive material sensor based on schlieren imaging, wherein the sensor correction in step (7) includes:

[0036] Adjust the parameters of the temperature compensation circuit;

[0037] Optimize the gas diffusion path or surface area of the sensor;

[0038] Improve the signal processing algorithm to reduce the dynamic response error.

[0039] Among them, the visualization calibration method for the hydrogen-sensitive material sensor based on schlieren imaging controls the calibration error of the sensor response time within 10 ms through visualization dynamic calibration.

[0040] The present invention has the following beneficial effects:

[0041] The visualization calibration method for the hydrogen-sensitive material sensor based on schlieren imaging provided by the present invention realizes the spatio-temporal synchronous visualization dynamic monitoring of the hydrogen diffusion process and the sensor output by constructing a reflective parallel light schlieren system and a sensor collaborative test bench, solves the defect of non-visualization of traditional calibration, extracts the visualization concentration curve by using the gray-scale - concentration calibration technology, and combines the error quantification and comprehensive scoring mechanism in the time and concentration dimensions to guide the optimization of the sensor compensation circuit, structure and algorithm. This method improves the calibration accuracy to the 10 ms level, significantly improves the reliability of hydrogen leakage monitoring, and is applicable to the field of hydrogen energy safety. Description of the Drawings

[0042] Figure 1 It is a flowchart of the visualization calibration method for the hydrogen-sensitive material sensor based on schlieren imaging provided by the embodiment of the present invention;

[0043] Figure 2 It is a comparison curve of the sensor response time error in the visualization calibration method for the hydrogen-sensitive material sensor based on schlieren imaging provided by the embodiment of the present invention Figure 1 ;

[0044] Figure 3 It is a comparison curve of the sensor response time error in the visualization calibration method for the hydrogen-sensitive material sensor based on schlieren imaging provided by the embodiment of the present invention Figure 2 。 Detailed Embodiments

[0045] The technical solution of the present invention will be further described below with reference to the drawings and through specific embodiments.

[0046] Among them, the attached drawings are only for illustrative purposes, showing only schematic diagrams rather than actual physical diagrams, and should not be construed as a limitation on this patent; in order to better illustrate the embodiments of the present invention, some components in the attached drawings will be omitted, enlarged or reduced, which does not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the attached drawings may be omitted.

[0047] In the attached drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components; in the description of the present invention, it should be understood that if terms such as "upper", "lower", "left", "right", "inner", "outer", etc. are used to indicate the orientation or positional relationship, it is based on the orientation or positional relationship shown in the attached drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms used to describe the positional relationship in the attached drawings are only for illustrative purposes and should not be construed as a limitation on this patent. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0048] In the description of the present invention, unless otherwise clearly specified and defined, if terms such as "connection" are used to indicate the connection relationship between components, this term should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] Embodiment

[0050] A visualization calibration method for a hydrogen-sensitive material sensor based on schlieren imaging provided by this embodiment is as Figure 1 - Figure 3 shown, and includes the following steps:

[0051] (1) Select the sensor to be calibrated and configure the experimental parameters: configure the concentration of the hydrogen-air mixture gas according to the maximum range Cmax of the sensor, adjust the size of the nozzle and the vertical distance from the sensor so that the hydrogen-sensitive material is completely covered by the jet, and adjust the frame rate of the high-speed camera based on the nominal response speed of the sensor;

[0052] (2) Construct a visualization calibration experimental bench: including a hydrogen injection device, a reflective parallel light schlieren imaging system and a sensor installation device. The schlieren imaging system includes a schlieren instrument, a high-speed camera and a data acquisition board. The sensor is fixed directly above the injection port and connected to the data acquisition system;

[0053] (3) Conduct a calibration experiment: Synchronously trigger hydrogen leakage, schlieren imaging, and sensor data acquisition, record the gas diffusion process and the sensor concentration-time curve g(t), and capture the instantaneous frame when the gas touches the sensor to determine the precise time point;

[0054] (4) Gray-scale - concentration calibration: Convert the schlieren video frame by frame into grayscale images, extract the relative gray-scale difference ΔG of the selected area, establish a quantitative relationship between the gray-scale value and the hydrogen concentration, and generate a visualized concentration-time curve f(t);

[0055] (5) Response time error calibration: Based on f(t) and g(t), extract the characteristic quantities T10, T90, ΔT10, ΔT90 in the time dimension, and the characteristic quantities Δc10, Δc90, Δcs in the concentration dimension;

[0056] (6) Sensor performance scoring: Calculate the comprehensive score of the sensor through the comprehensive weight formula

[0057] S Q = αS ΔT10 + βS ΔT90 + γS Δc10 + δS Δc90 + ∈S Δcs where the weight coefficients are determined by the subjective and objective weighting method;

[0058] (7) Sensor correction: Adjust the sensor compensation circuit, geometric structure, or signal processing algorithm according to the scoring results, and repeat the calibration until the performance meets the standard.

[0059] Through the spatio-temporal synchronous monitoring and gray-scale - concentration calibration method, the calibration accuracy is improved to the 10 ms level, solving the problems of non-visualization and large errors in traditional calibration, and supporting the analysis of dynamic response characteristics.

[0060] Among them, in step (1), the frame rate of the high-speed camera needs to be set so that the time interval between each frame is 1 - 2 orders of magnitude smaller than the sensor response speed, ensuring that the time resolution is higher than the sensor response speed, accurately capturing the transient gas diffusion process, and avoiding calibration errors in the time dimension.

[0061] Among them, the gray-scale - concentration calibration in step (4) specifically includes:

[0062] (4.1) Convert the color video frame by frame into grayscale images;

[0063] (4.2) Select multiple frames of images for gray-scale averaging processing, and extract the relative gray-scale difference ΔG of the sensor probe area;

[0064] (4.3) Establish a mapping relationship between ΔG and the hydrogen concentration through function fitting.

[0065] Eliminate the interference of light and noise, establish the quantitative relationship between gray scale and concentration, and improve the concentration calibration accuracy to within ±1% vol.

[0066] Among them, the time dimension feature quantity in step (5) is defined as:

[0067] T10 = t2 - t1, where t1 is the time when the sensor starts to show a reading, and t2 is the time when the sensor output reaches 10% of the stable value;

[0068] ΔT10 = T10' - T10, where T10' is the time difference calculated based on the visualization curve.

[0069] Quantify the transient response delay of the sensor and provide a direct basis for optimizing the signal processing algorithm.

[0070] Among them, the concentration dimension feature quantities in step (5) include:

[0071] Δc10 = f(t2') - g(t2'), where t2' is the moment when the visualization curve reaches 10% concentration;

[0072] Characterize the comprehensive response accuracy of the sensor.

[0073] Evaluate the sensor concentration measurement error from the transient and steady-state dimensions and guide the optimization of the compensation circuit.

[0074] Among them, the weight coefficients in step (6) are determined by the analytic hierarchy process and include:

[0075] Construct an index hierarchy, including the response speed ΔT and the response accuracy Δc;

[0076] Combine expert scoring and principal component analysis, calculate the subjective and objective weights, and then perform weighted fusion.

[0077] Through the weight assignment that combines subjectivity and objectivity, avoid the limitations of the single weighting method, and ensure the scientificity and applicability of the scoring system.

[0078] Among them, the layout requirements of the reflective parallel light schlieren system in step (2) are:

[0079] The hydrogen nozzle is located at the center position below the field of view of the schlieren instrument;

[0080] The positions of the schlieren instrument, the nozzle, and the lighting conditions remain constant during the experiment.

[0081] Eliminate the influence of experimental environment fluctuations on the imaging quality and ensure the repeatability and reliability of the gray scale - concentration calibration results.

[0082] Among them, the visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging is applicable to the dynamic response calibration of the leakage detection sensor of the hydrogen energy storage and transportation device, and is compatible with any gas and sensor principle. It expands the application scope of technology and supports the calibration requirements of multiple types of sensors (such as electrochemistry type, optical type) and gases (such as methane, CO2).

[0083] Among them, the sensor correction in step (7) includes:

[0084] Adjust the parameters of the temperature compensation circuit;

[0085] Optimize the gas diffusion path or surface area of the sensor;

[0086] Improve the signal processing algorithm to reduce the dynamic response error.

[0087] Through the multi-dimensional correction strategy, specifically improve the response speed and accuracy of the sensor, and shorten the iterative optimization cycle.

[0088] Among them, the visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging controls the calibration error of the sensor response time within 10 ms through visualization dynamic calibration, meets the fast response requirements of high-pressure hydrogen system leakage detection, and improves the safety protection level of the hydrogen energy storage and transportation device.

[0089] In summary, the visualization calibration method of the hydrogen-sensitive material sensor based on schlieren imaging provided in this embodiment has the following main solutions:

[0090] 1. Experimental parameter configuration

[0091] Configure the hydrogen-air mixed gas concentration according to the maximum range Cmax of the sensor;

[0092] Adjust the size of the nozzle and the vertical distance from the sensor to ensure that the hydrogen-sensitive material is completely covered by the jet;

[0093] Based on the nominal response speed of the sensor, set the frame rate of the high-speed camera (1-2 orders of magnitude faster than the sensor response).

[0094] 2. Construction of the visualization calibration test bench

[0095] Gas injection device: including a standard hydrogen mixed gas source, solenoid valve, pressure reducing valve and nozzle;

[0096] Schlieren imaging system: a reflective parallel light schlieren instrument, a high-speed camera (frame rate ≥ 1000 fps), and a data acquisition board;

[0097] Sensor installation device: The sensor to be calibrated is fixed directly above the nozzle to ensure that the jet covers the probe area and is connected to the data acquisition system in real time.

[0098] 3. Execution of synchronous calibration experiment

[0099] Trigger hydrogen leakage by controlling the solenoid valve through the host computer (record the starting time t0).

[0100] Synchronously start the high-speed camera to record the jet diffusion process and collect the sensor concentration-time curve g(t).

[0101] Capture the instantaneous frame when the gas touches the sensor (time point t1′) as the visual accurate time reference.

[0102] 4. Gray-scale - Concentration Calibration

[0103] Convert the schlieren video frame by frame into grayscale images, and select the probe area for grayscale averaging processing.

[0104] Extract the relative gray-scale difference ΔG, establish the quantitative relationship between ΔG and concentration through function fitting, and generate the visual concentration curve f(t).

[0105] 5. Response Error Quantification and Performance Scoring

[0106] Time dimension: Calculate T10 = t2 - t1 (the time difference when the sensor reaches 10% of the stable value) and ΔT10 = T10′ - T10 (the time difference between visualization and the sensor).

[0107] Concentration dimension: Calculate Δc10 = f(t2′) - g(t2′) (the error at the 10% concentration point) and the integral error

[0108] Comprehensive scoring: Calculate the sensor performance score through the formula S Q = αS ΔT10 + βS ΔT90 + γS Δc10 + δS Δc90 + ∈S Δcs The weight coefficients are determined based on the Analytic Hierarchy Process (AHP) and Principal Component Analysis (PCA).

[0109] 6. Sensor Correction and Iterative Optimization

[0110] Adjust the compensation circuit parameters, optimize the gas diffusion path, or improve the signal processing algorithm according to the score.

[0111] Repeat the calibration until the sensor response time error ≤ 10 ms and the comprehensive score SQ ≥ 90.

[0112] Technical Effects

[0113] 1. Visual Dynamic Monitoring: Real-time capture the gas diffusion process through schlieren imaging, solving the defect of invisibility in traditional calibration.

[0114] 2. High-precision calibration: Through spatio-temporal synchronous comparison, the response time error is reduced from the 200 ms level to the 10 ms level;

[0115] 3. Strong versatility: Applicable to any gas type and sensor principle (such as thermal conductivity type, electrochemistry type);

[0116] 4. Quantitative evaluation system: A comprehensive scoring mechanism combining time and concentration dimensions provides a scientific basis for sensor optimization.

[0117] The specific process of the calibration method for the hydrogen-sensitive sensor based on schlieren imaging is as follows:

[0118] First: Select the sensor to be calibrated and experimental parameters

[0119] The calibration method proposed in the present invention is applicable to hydrogen-sensitive sensors of various principles. First, some experimental parameters need to be determined according to the sensor to be calibrated. According to the maximum range C max of the sensor, configure the concentration of the hydrogen-air mixed gas for calibration, for example, make the mixed gas concentration approximately equal to C max ; According to the area of the hydrogen-sensitive material, select an appropriate nozzle size and the appropriate vertical distance between the sensor and the nozzle so that the hydrogen-sensitive material is completely covered by the jet; Considering the nominal response speed of the sensor, adjust the frame rate of the high-speed camera so that the time interval between each frame is at least 1-2 orders of magnitude smaller than the sensor response speed. In addition, it is also necessary to estimate the general performance of the sensor, determine the maximum tolerance value of the sensor index, and the maximum recording time of the experimental data.

[0120] Second: The layout scheme of the sensor visualization calibration experimental bench and the gray-scale - concentration calibration method

[0121] Design and construct a sensor visualization calibration experimental platform for low-pressure gas leakage, including a gas injection device, including standard hydrogen mixed gas, pipelines, valves, etc.; a schlieren imaging system, including a schlieren instrument, a high-speed camera, and a data acquisition board, etc.; a sensor installation device, including the sensor to be calibrated, a sensor fixing bracket, etc. Install a reflective parallel light schlieren system, connect the hydrogen gas source, set the injection pressure to low pressure, place the hydrogen nozzle below and centered in the field of view of the schlieren instrument, and fix the thermal conductivity sensor directly above the injection port to ensure that the gas diffusion range covers the sensor, connect the data acquisition system, and record the sensor signal in real time. According to the experimental parameters selected in the first part, arrange the test system and adjust the lens focal length, position, and angle, set the resolution and exposure time through the camera software to ensure that the field of view brightness is moderate and located in the center of the screen. And keep the schlieren instrument, nozzle position, and lighting conditions unchanged during the experiment.

[0122] The purpose of gray-scale - concentration calibration is to extract hydrogen concentration data at a certain position by using the obtained schlieren video. This data processing process is divided into three steps: First, convert the original color video file frame by frame into grayscale images; Second, select several frames with better effects from thousands of frames of video for each concentration, perform gray-scale averaging processing, and select the gray-scale acquisition area (in this experiment, it is the area close to the hydrogen-sensitive material of the sensor); Finally, by comparing the grayscale images at different concentrations with the background image, extract the relative grayscale difference ΔG between the jet image and the background image in the selected area. According to the visualization principle, there is a functional relationship between ΔG and hydrogen concentration. Therefore, a quantitative relationship between the grayscale value and hydrogen concentration can be established by using the function fitting method, so as to associate the visualization result (grayscale value) of the selected area with the hydrogen concentration (volume fraction).

[0123] Third: Experimental procedure

[0124] First step, control the solenoid valve to open through the upper computer to start the hydrogen leakage experiment, and record this moment as t0. Record the jet diffusion process of the hydrogen mixture through the upper computer control software of the high-speed camera. At the same time, with the help of the data acquisition board, start to synchronously record the sensor concentration - time data to obtain the sensor concentration change curve g(t). The start recording points of the two groups of data need to be controlled at the same moment until the maximum recording time t4 selected in the first part is reached.

[0125] Second step, based on the visualization video, intercept the instant frame when the gas touches the sensor, and record this moment as t1'. Since the visualization image has real-time and intuitiveness, this time point is regarded as the exact time point when the sensor should detect the concentration change. Then, based on the sensor concentration change curve, record the time when the sensor starts to show a reading as t1. Record the sensor response time ΔT = t1 - t1'.

[0126] Third step, export the leakage images at each moment collected by the visualization system. Through the aforementioned gray-scale - concentration calibration technology, collect gray-scale data at the fixed sensor probe position point at each moment and identify it as concentration data to obtain the gas concentration - time data points at the sensor probe based on schlieren imaging, and then obtain the visualization concentration change curve graph f(t).

[0127] So far, the accurate hydrogen concentration - time change curve f(t) based on the visualization image and the hydrogen concentration - time change curve g(t) based on the sensor collection in the same time period have been obtained. Perform data difference according to the same time scale so that the data points at any moment on the two curves can correspond one by one.

[0128] Fourth: Calibration method for the error of sensor response time based on schlieren imaging visualization technology

[0129] The concentration-time curve images obtained by the sensor at different hydrogen concentrations and the concentration-time curve images obtained by visualization are plotted in the same coordinate system, as shown in Figure 2 and Figure 3 .

[0130] 4.1 Extraction of Feature Quantities in the Time Dimension

[0131] In the fields of sensor performance calibration and gas detection, T10 and T90 are two important technical indicators used to describe the response speed of the sensor. They respectively represent the time it takes for the sensor to reach a specific concentration reading from detecting the target gas.

[0132] T10 is the time when the sensor reaches 10% of the final stable value from contacting the target gas. Denote the time when it stabilizes at 10% of the final value as t2. According to the concentration-time curve obtained from the sensor signal, T10 = t2 - t1. According to the concentration-time curve obtained from the visualization image, T10' = t2' - t1'.

[0133] T90 is the time when the sensor reaches 90% of the final stable value from point X contacting the target gas. Denote the time when it stabilizes at 90% of the final value as t3. According to the concentration-time curve obtained from the sensor signal, T90 = t3 - t1. According to the concentration-time curve obtained from the visualization image, T90' = t3' - t1'.

[0134] Plot the two curves on the same coordinate axis and mark the key time points such as t0, t1, t2, t3, t1', t2', t3', etc. Calculate the difference in response time between the visualization curve and the sensor curve using ΔT10 = T10' - T10 and ΔT90 = T90' - T90. ΔT10 and ΔT90 characterize the performance of the sensor in terms of transient response speed and long-term response speed. The shorter ΔT10 and Δt90 are, the better the response speed of the sensor.

[0135] To simply and clearly judge the sensor performance, define the concentration-time curve obtained by visualization as the standard curve and make S X = H(X), where H(X) is a function with a value range of [0, 100]. For example, for S ΔT10 = H(ΔT10), assign 100 points when ΔT10 = 0, and assign 0 points when ΔT10 = ΔT10 max (ΔT10 max is the maximum tolerable range of ΔT10). Thus, the score for the sensor in the concentration dimension can be obtained as S ΔT10 and S ΔT90 .

[0136] 4.2 Extraction of Feature Quantities in the Concentration Dimension

[0137] In this experiment, the visual schlieren imaging system is regarded as a completely accurate sensor. We take the time when the visual curve reaches 10%c and 90%c as the accurate stabilization time. Take t2' on the visual image and the sensor image, and calculate the concentration difference between the two curves at this point, that is, Δc10 = f(t2') - g(t2'); take t3' on the visual image and the sensor image, and calculate the concentration difference between the two curves at this point, that is, Δc90 = f(t3') - g(t3'). Δc10 and Δc90 characterize the performance of the sensor in terms of transient response accuracy and long-term response accuracy. According to the integral formula Calculate the area between the two curves. Δcs characterizes the performance of the sensor in terms of comprehensive response accuracy. The smaller Δc10, Δc90, and Δcs are, the better the response accuracy of the sensor represents.

[0138] Similarly, the concentration-time curve obtained by visualization is defined as the standard curve. For S Δc10 = H(Δc10). Therefore, when Δc10 = 0, it is assigned a score of 100. Let the concentration when the image tends to be stable be represented by c, and assign a score of 0 when Δc10 = Δc10 max (Δc10 max is the maximum tolerable range of Δc10). Thus, the sensor can be scored in the concentration dimension as S Δc10 , S Δc90 and S Δcs .

[0139] 4.3 Sensor scoring mechanism

[0140] According to the above two methods for calibrating the accuracy of the sensor, the calibration formula for evaluating the performance of the sensor can be written

[0141] S Q = α * S ΔT10 + β * S ΔT90 + γ * S Δc10 + δ * S Δc90 + ∈ * S Δcs

[0142] In this formula, α, β, γ, δ, and ε are weight coefficients, and α + β + γ + δ + ε = 1. The coefficients should be determined by the uses of different sensors. The subjective weighting method based on expert experience and the objective weighting method based on data-driven can be used to determine the weights. For example, the process of using the analytic hierarchy process (objective weighting) is as follows: ①Construct the index hierarchy: goal layer - sensor comprehensive performance S Q ; criterion layer - response speed (ΔT), response accuracy (Δc); sub-criterion layer S ΔT10 , S ΔT90 , S Δc10 , S Δc90 , S Δcs. ②Construct a judgment matrix through expert scoring. ③Calculate the weights of each index by the eigenvector method. Finally, the subjective weight is calculated.

[0143] Meanwhile, subjective weighting methods such as principal component analysis can be used to reduce the dimension and extract the main influencing factors, and the weights are allocated according to the variance contribution rate. Finally, the objective weight is calculated. Combining the subjective weighting method and the objective weighting method, the comprehensive weight is finally obtained. Where k is the balance factor (0 ≤ k ≤ 1).

[0144] Fifth: Sensor correction method based on visualization calibration and scoring

[0145] According to the concentration-time curve obtained by visualization and the sensor output concentration curve, combined with the scoring results S ΔT10 、S ΔT90 、S Δc10 、S Δc90 、S Δcs and S Q , by adjusting the sensor compensation circuit, improving the geometric structure of the sensor (such as increasing the surface area, improving the gas diffusion path), optimizing the signal processing algorithm, etc., make the sensor curve fit the visualization curve better and improve the comprehensive performance score of the sensor. The improved sensor will go through the full process of the visualization calibration experiment again until the sensor performance meets the expected requirements. The complete flow chart is as Figure 1 shown.

[0146] An example of the calibration method of the schlieren imaging-based hydrogen sensor is as follows:

[0147] 1. Experimental parameters

[0148] The sensor range Cmax = 5% vol, configure a 4.8% hydrogen-air mixture;

[0149] The nozzle diameter is 2 mm, and the vertical distance of the sensor is 10 mm;

[0150] The frame rate of the high-speed camera is set to 2000 fps (the nominal response time of the sensor is 20 ms)

[0151] 2. Calibration process

[0152] After triggering the hydrogen injection, the schlieren system records the jet diffusion ( Figure 1 ), and the sensor outputs the curve g(t);

[0153] Intercept t1' = 15.2 ms (visualization touch time), the sensor response time t1 = 16.5 ms, and the error ΔT = 1.3 ms;

[0154] Generate f(t) through grayscale-concentration calibration, calculate Δcs=0.8%·s, and the comprehensive score S Q =92.5.

[0155] 3. Sensor correction

[0156] After optimizing the signal processing algorithm, retest ΔT = 0.5ms, S Q =96.7, meeting the performance requirements.

[0157] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging, characterized in that: The following steps are involved: (1) Select the sensor to be calibrated and configure the experimental parameters: configure the concentration of the hydrogen-air mixed gas according to the maximum range Cmax of the sensor, adjust the nozzle size and the vertical distance of the sensor so that the hydrogen-sensitive material is completely covered by the jet, and adjust the frame rate of the high-speed camera based on the nominal response speed of the sensor; (2) Constructing a visual calibration experimental bench: including a hydrogen injection device, a reflective parallel light schlieren imaging system and a sensor installation device. The schlieren imaging system includes a schlieren instrument, a high-speed camera and a data acquisition board. The sensor is fixed just above the injection port and connected to the data acquisition system. (3) Perform calibration experiments: synchronously trigger hydrogen leakage, schlieren imaging, and sensor data acquisition, record the gas diffusion process and sensor concentration-time curve g(t), and capture the instantaneous frame when the gas touches the sensor to determine the precise time point; (4) Grayscale-concentration calibration: convert the schlieren video frame by frame into a grayscale image, extract the relative grayscale difference ΔG of the selected area, establish a quantitative relationship between the grayscale value and the hydrogen concentration, and generate a visual concentration-time curve f(t); (5) Response time error calibration: Extract the time dimension features T10, T90, ΔT10, ΔT90, and the concentration dimension features Δc10, Δc90, Δcs based on f(t) and g(t); (6) Sensor performance scoring: through comprehensive weight formula S Q =αS ΔT10 +βS ΔT90 +γS Δ c 10 +δS Δ c 90 +∈S Δ cs calculates the comprehensive score of the sensor, where the weight coefficient is determined by the subjective and objective weighting method; (7) Sensor correction: Adjust the sensor compensation circuit, geometry, or signal processing algorithm based on the scoring results, and repeat the calibration until the performance meets the requirements.

2. The visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging according to claim 1 is characterized in that: In step (1), the frame rate of the high-speed camera must be set to ensure that the time interval between each frame is 1-2 orders of magnitude smaller than the sensor response speed.

3. The visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging according to claim 1 is characterized in that: The grayscale-concentration calibration in step (4) specifically includes: (4.1) Convert color video to grayscale image frame by frame; (4.2) Select multiple frames of images for grayscale averaging and extract the relative grayscale difference ΔG of the sensor probe area; (4.3) The mapping relationship between ΔG and hydrogen concentration is established through function fitting.

4. The visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging according to claim 1 is characterized in that: The time dimension feature quantity in step (5) is defined as: T10 = t2-t1, where t1 is the time when the sensor starts to show readings, and t2 is the time when the sensor output reaches 10% of the stable value; ΔT10=T10′-T10, where T10′ is the time difference calculated based on the visualization curve.

5. The visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging according to claim 1 is characterized in that: The concentration dimension feature quantity in step (5) includes: Δc10 = f(t2′) - g(t2′), where t2′ is the moment when the visualization curve reaches 10% concentration; Characterizes the comprehensive response accuracy of the sensor.

6. The visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging according to claim 1 is characterized in that: The weight coefficient in step (6) is determined by a hierarchical analysis method, including: Constructing a hierarchy of indicators, including response speed ΔT and response accuracy Δc; Combining expert scoring with principal component analysis, the subjective and objective weights are calculated and then weighted fusion is performed.

7. The visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging according to claim 1 is characterized in that: The arrangement requirements of the reflective parallel light schlieren system in step (2) are as follows: The hydrogen nozzle is located in the center below the field of view of the Schlieren instrument; The schlieren instrument, nozzle position and lighting conditions were kept constant during the experiment.

8. The visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging according to claim 1 is characterized in that: The method is suitable for the dynamic response calibration of leakage detection sensors for hydrogen energy storage and transportation devices, and is compatible with any gas and sensor principle.

9. The visual calibration method for hydrogen-sensitive material sensors based on Schlieren imaging according to claim 1 is characterized in that: The sensor correction in step (7) includes: Adjust the temperature compensation circuit parameters; Optimizing the sensor gas diffusion path or surface area; Improve signal processing algorithms to reduce dynamic response errors.

10. The visual calibration method for hydrogen-sensitive material sensor based on Schlieren imaging according to any one of claims 1 to 9, characterized in that: The method controls the calibration error of the sensor response time within 10 ms through visual dynamic calibration.

Images