A real-time online detection method and device for proton membrane

Abstract

The present invention discloses a real-time online detection method and device for proton membranes. The method comprises: responding to an encoder pulse signal, using a line-scan smart camera to detect surface defects on the proton membrane and determine the location of the surface defects; wherein the line-scan smart camera and a coaxial light source are installed at the first work section of the real-time online detection device for proton membranes; synchronously connecting the encoder pulse signal to a white light interference sensor, using the white light interference sensor to detect the membrane thickness of the proton membrane and determine the location of abnormal membrane thickness; wherein the white light interference sensor is installed at the second work section of the real-time online detection device for proton membranes, and the first work section is located at the front end of the second work section. Using this solution, the line-scan smart camera and the white light interference sensor are installed in the same device, and the synchronous pulse signal is triggered by the encoder to achieve synchronous real-time online detection of the surface defects and membrane thickness of the proton membrane.

Description

Technical Field

[0001] The present invention relates to the technical field of online detection, and in particular to a real-time online detection method and equipment for a proton membrane. Background Art

[0002] As global demand for clean energy continues to grow, hydrogen energy, as a clean and efficient energy source, has attracted much attention. In the hydrogen energy industry, proton membranes are core components of fuel cells and water electrolysis hydrogen production equipment, and their thickness quality inspection and airtightness testing technologies are crucial.

[0003] When it comes to proton membrane thickness quality testing, traditional mechanical contact measurement techniques have limited accuracy and may damage the membrane surface. Electronic thickness gauges, while highly accurate, are mostly used for offline laboratory testing. Advances in materials science have led to the emergence of new proton membrane materials, posing new challenges to thickness measurement. Simultaneously, industry development has placed higher demands on the accuracy and efficiency of proton membrane thickness quality testing. The airtightness of proton membranes is directly dependent on surface defects during the production process, which can have numerous negative impacts on the product. For example, defects such as bubbles and cracks can reduce the tensile strength of the proton membrane, making it more susceptible to rupture under differential pressure and shortening its service life. Defects such as pinholes and micropores can compromise its integrity, reducing its gas barrier properties and allowing cross-permeation of gases like hydrogen, resulting in hydrogen waste and reduced battery open-circuit voltage. Current visual inspection methods often face challenges in visually inspecting products. Defects vary significantly between different defect types, making it difficult to balance accuracy and speed. Curved surfaces and unusual shapes can affect the inspection field of view and imaging, and transparent or translucent materials can complicate light transmission, making it difficult to distinguish defects from normal areas.

[0004] Currently, online thin film thickness testing mostly relies on radiographic online thickness gauges. This method suffers from low accuracy, high maintenance costs, and inability to detect the thickness of composite materials and multi-layer composite materials. Line scan smart cameras are commonly used for online thin film defect detection. This technology typically requires a separate mounting frame for the camera and light source, often constructed from aluminum profiles. This takes up a lot of space, has poor long-term structural stability, and requires long on-site installation and deployment cycles, requiring high technical skills from the installation and commissioning personnel. Proton membrane thickness and surface defect testing typically require separate measurements, making online testing impossible. Summary of the Invention

[0005] The present invention provides a real-time online detection method and device for proton membranes to solve the problem in the prior art that the surface defects and membrane thickness of the proton membrane need to be detected separately, and online detection cannot be achieved.

[0006] According to one aspect of the present invention, a real-time online detection method for proton membrane is provided, the method comprising:

[0007] In response to the encoder pulse signal, a line scan smart camera is used to detect surface defects on the proton membrane and determine the location of the surface defects; wherein the line scan smart camera and the coaxial light source are installed at the first working section of the real-time online detection equipment of the proton membrane;

[0008] The encoder pulse signal is synchronously connected to the white light interference sensor, and the white light interference sensor is used to detect the membrane thickness of the proton membrane and determine the abnormal position of the membrane thickness; wherein, the white light interference sensor is installed at the second work section position of the proton membrane real-time online detection equipment, and the first work section position is located at the front end of the second work section position.

[0009] According to another aspect of the present invention, a real-time online detection device for proton membrane is provided, the device comprising a line scan intelligent camera, a coaxial light source, and a white light interferometer sensor; wherein:

[0010] The line scan smart camera is used to respond to the encoder pulse signal to detect surface defects on the proton membrane and determine the location of the surface defects; wherein the line scan smart camera and the coaxial light source are installed at the first working section of the real-time online detection equipment of the proton membrane;

[0011] The white light interference sensor is used to respond to the encoder synchronization pulse signal, detect the membrane thickness of the proton membrane and determine the location of abnormal membrane thickness; wherein, the white light interference sensor is installed at the second work section position of the proton membrane real-time online detection equipment, and the first work section position is located at the front end of the second work section position.

[0012] The technical solution of the embodiment of the present invention is to detect surface defects of the proton membrane and determine the location of surface defects by using a line scan intelligent camera in response to an encoder pulse signal; wherein the line scan intelligent camera and the coaxial light source are installed at the first section position of the real-time online detection equipment of the proton membrane; the encoder pulse signal is synchronously connected to the white light interference sensor, and the white light interference sensor is used to detect the membrane thickness of the proton membrane and determine the location of abnormal membrane thickness; wherein the white light interference sensor is installed at the second section position of the real-time online detection equipment of the proton membrane, and the first section position is located at the front end of the second section position. This solves the problem in the prior art that the surface defects and membrane thickness of the proton membrane need to be detected separately, and at the same time, online detection cannot be achieved. Synchronous real-time online detection of proton membrane surface defects and membrane thickness is achieved, thereby improving the comprehensiveness and timeliness of proton membrane detection.

[0013] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0015] Figure 1 This is a flow chart of a real-time online detection method for proton membrane according to Example 1 of the present invention;

[0016] Figure 2 This is a flow chart of a real-time online detection method for proton membrane according to the second embodiment of the present invention;

[0017] Figure 3 This is a schematic diagram showing a heat map of proton membrane detection results provided in accordance with the second embodiment of the present invention;

[0018] Figure 4 A schematic structural diagram of a proton membrane real-time online detection device provided in Example 3 of the present invention;

[0019] Figure 5 A schematic structural diagram of another proton membrane real-time online detection device provided in Example 3 of the present invention;

[0020] Figure 6 This is a structural schematic diagram of another proton membrane real-time online detection device provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0022] Among them, the acquisition, storage, use and processing of data in the technical solution of this application are in compliance with the relevant provisions of laws and regulations. It should be noted that the terms "first", "second", "target", "original", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including", "etc." and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] Example 1

[0024] Figure 1 A flowchart of a proton membrane real-time online detection method is provided for the first embodiment of the present invention. This embodiment is applicable to the case where the surface defect detection and membrane thickness detection of the proton membrane are performed simultaneously in the same proton membrane real-time online detection device. The method can be executed by the proton membrane real-time online detection device. The proton membrane real-time online detection device can be implemented in the form of hardware and / or software. The proton membrane real-time online detection device can be configured in any electronic device with network communication function. Figure 1 As shown, the method includes:

[0025] S110 , in response to the encoder pulse signal, using a line scan smart camera to detect surface defects on the proton membrane and determine the locations of the surface defects.

[0026] Among them, the proton membrane may refer to a proton exchange membrane, which is a material that plays a key role in fuel cells and other electrochemical devices. Since performance parameters such as gas permeability and tensile strength are directly related to the efficiency, stability and durability of fuel cells and water electrolysis hydrogen production equipment, it is necessary to strictly control the air tightness of the proton membrane; and the air tightness of the proton membrane directly depends on the surface defects in the production process, which has many adverse effects on the proton membrane. In the proton membrane water electrolysis technology, the air tightness of the membrane electrode is extremely important. Once a problem occurs, it will bring great safety risks, and the existing membrane electrode air tightness testing method also has many shortcomings. Therefore, an embodiment of the present invention provides a real-time online detection method for proton membranes, which responds to encoder pulse signals and uses a line scan intelligent camera to detect surface defects on the proton membrane and determine the location of surface defects to perform air tightness detection on the proton membrane.

[0027] The encoder pulse signal may refer to a pulse sampling signal triggered by an encoder, and the encoder pulse signal is used to trigger a line scan smart camera, so as to detect surface defects of the proton membrane using the line scan smart camera.

[0028] The line scan smart camera may refer to a camera that utilizes line scan technology. A line scan smart camera has only one or three rows of photosensitive image elements and continuously scans moving objects at a high line speed. The line scan smart camera integrates advanced image processing technology. In embodiments of the present invention, the line scan smart camera is used to detect surface defects on the proton membrane based on the membrane's direction of travel and to determine the location of surface defects. The line scan smart camera possesses high resolution, enabling detection of small surface defects on the proton membrane, improving the accuracy of surface defect detection. Furthermore, the line scan smart camera possesses fast image capture and precise image processing capabilities, enabling real-time online detection.

[0029] Surface defect detection can refer to detecting defects on the membrane surface, including but not limited to bubbles and cracks. Surface defects can reduce the membrane's tensile strength, making it more susceptible to rupture when subjected to differential pressure, and shortening its service life. Online surface defect detection provides real-time feedback for timely adjustments, reducing production waste, shortening production cycles, and improving overall production efficiency.

[0030] Optionally, the line-scanning smart camera and the coaxial light source are installed at the first work section position of the real-time online detection equipment for proton membranes. The real-time online detection equipment for proton membranes may refer to a real-time online detection device that performs surface defect detection and model thickness anomaly detection on proton membranes; the first work section position may refer to the corresponding position when the proton membrane is fed in for real-time online detection. For example, the real-time online detection equipment for proton membranes includes a feed flattening roller and a discharge flattening roller. The proton membrane enters the real-time online detection equipment for proton membranes from the feed flattening roller and is output from the discharge flattening roller. The first work section position may be the corresponding position near the feed flattening roller. A line-scanning smart camera and a corresponding coaxial light source are arranged near the feed flattening roller. The proton membrane is flattened on the feed flattening roller to facilitate the line-scanning smart camera to capture images and perform detection. At the same time, the coaxial light source provides light for the line-scanning smart camera, facilitating the line-scanning smart camera to perform surface defect detection.

[0031] S120 , synchronously connecting the encoder pulse signal to a white light interferometer sensor, using the white light interferometer sensor to detect the thickness of the proton membrane and determine the location of abnormal membrane thickness.

[0032] The thickness of the proton membrane is also directly related to the efficiency, stability and durability of fuel cells and water electrolysis hydrogen production equipment. In the embodiment of the present invention, a white light interferometer sensor is used to detect the thickness of the proton membrane and determine the location of abnormal membrane thickness.

[0033] In this embodiment of the present invention, the encoder pulse signal that triggers the line scan smart camera is synchronously connected to the white light interferometer sensor to detect the membrane thickness of the proton membrane. The encoder pulse signal is synchronously connected to the white light interferometer sensor via a servo motor, enabling simultaneous detection of surface defects and membrane thickness of the proton membrane. The white light interferometer sensor detects the membrane thickness of the proton membrane based on pulse triggering, and its pulse firing interval is configured and output by the servo motor controller.

[0034] The white light interference sensor is a technical device that uses the principle of white light interference to perform high-precision non-contact measurement. The white light interference sensor forms a reference light path and a detection light path through different optical elements. Any change in the optical path difference between the two coherent light beams will very sensitively cause the movement of the interference fringes. The change in the optical path of a certain beam of coherent light is caused by the change in the geometric path it passes through or the refractive index of the medium. Therefore, the geometric length or refractive index can be measured by the movement change of the interference fringes, thereby realizing displacement measurement. In the embodiment of the present invention, the white light interference sensor is used to perform non-contact detection of the thickness of the proton membrane, reducing the potential damage to the proton membrane that may be caused by manual operation, reducing labor costs and human errors, and improving the degree of production automation and efficiency. In the embodiment of the present invention, there is no specific limitation on the sensor for detecting the thickness of the proton membrane layer, and it also includes but is not limited to infrared interference sensors and multi-spectral confocal sensors.

[0035] Since the proton membrane is composed of a substrate and a coating, the membrane thickness detection includes substrate thickness detection, coating thickness detection, and overall membrane thickness detection. The white light interferometer sensor returns rays of varying lengths to provide feedback on the thickness of the three membrane layers, allowing for detection of the substrate thickness, coating thickness, and overall membrane thickness. In the embodiments of the present invention, membrane thickness detection of the proton membrane allows for timely identification and resolution of quality issues, reducing rework and scrap due to quality issues and lowering production costs.

[0036] Wherein, the white light interference sensor is installed at the second section position of the real-time online detection equipment for proton membranes, and the first section position is located at the front end of the second section position. The second section position may refer to the corresponding position when the proton membrane is subjected to real-time online detection and discharge. For example, the real-time online detection equipment for proton membranes includes a feed flattening roller and a discharge flattening roller. The proton membrane enters the real-time online detection equipment for proton membranes from the feed flattening roller and is output from the discharge flattening roller. The second section position may be the corresponding position near the discharge flattening roller. The white light interference sensor is installed near the discharge flattening roller. The proton membrane is flattened on the discharge flattening roller, which can reduce the problem of membrane thickness detection error caused by the proton membrane not being flattened, thereby improving the accuracy of membrane thickness detection.

[0037] In an optional solution of an embodiment of the present invention, the proton membrane layer thickness detection result and the abnormal membrane layer thickness position are verified based on the proton membrane surface defect detection result and the surface defect position. For example, if the proton membrane surface defect detection result shows that a crack appears at position A, the membrane layer thickness will decrease due to the crack on the proton membrane surface; if the proton membrane layer thickness detection result shows that the membrane layer thickness at position A exceeds the preset thickness threshold, it can be determined that the proton membrane layer thickness detection result is abnormal. By verifying the proton membrane layer thickness detection result and the abnormal membrane layer thickness position based on the proton membrane surface defect detection result and the surface defect position, the accuracy and reliability of the proton membrane layer thickness detection can be improved, providing more accurate data support for quality control.

[0038] An embodiment of the present invention provides a real-time online detection method for proton membranes. In response to encoder pulse signals, a line-scan smart camera is used to detect surface defects on the proton membrane and determine the location of the surface defects. The line-scan smart camera and a coaxial light source are installed at the first workstation of the real-time online detection equipment for proton membranes. The encoder pulse signal is synchronously connected to a white-light interferometer sensor, which is used to detect the membrane thickness of the proton membrane and determine the location of abnormal membrane thickness. The white-light interferometer sensor is installed at the second workstation of the real-time online detection equipment for proton membranes, with the first workstation located in front of the second workstation. Using the technical solution of the embodiment of the present invention, the line-scan smart camera and the white-light interferometer sensor are installed in the same device, and the encoder pulse signal triggers a synchronous detection signal for the line-scan smart camera and the white-light interferometer sensor to achieve simultaneous real-time online detection of surface defects and membrane thickness of the proton membrane. This changes the previous situation where separate detection may be required or real-time online detection is impossible, thereby improving the comprehensiveness and timeliness of proton membrane detection.

[0039] Example 2

[0040] Figure 2This is a flow chart of a real-time online detection method for proton membrane provided by the second embodiment of the present invention. The embodiment of the present invention further optimizes the above embodiment on the basis of the above embodiment. The embodiment of the present invention can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the method includes:

[0041] S210 , in response to the encoder pulse signal, using a line scan smart camera to detect surface defects on the proton membrane and determine the locations of the surface defects.

[0042] The line scan smart cameras correspond one-to-one with the coaxial light source. Due to the wide width of the proton membrane, using a single line scan smart camera may not be enough to fully inspect the membrane. Therefore, in the embodiments of the present invention, at least two sets of line scan smart cameras and coaxial light sources are used to inspect the proton membrane for surface defects. The specific method for surface defect inspection is not specifically limited in the embodiments of the present invention.

[0043] S220 , synchronously connecting the encoder pulse signal to a white light interferometer sensor, and using the white light interferometer sensor to detect the thickness of the proton membrane.

[0044] The proton membrane is composed of a substrate and a coating, and the membrane thickness detection includes substrate thickness detection, coating thickness detection and proton membrane overall thickness detection.

[0045] Because the detection principle of white light interferometry sensors is to calculate thickness by analyzing interference fringes derived from optical path differences, and because multiple optical path differences are superimposed and detected by the same photosensitive component, it is impossible to directly determine which film layer produces the interference spectrum generated by the superposition. Furthermore, when multiple films interfere, optical path differences may overlap, and the interaction of light of different wavelengths in each film layer is complex, making the interference spectrum difficult to interpret and the thickness of each layer difficult to clearly distinguish. Therefore, an embodiment of the present invention proposes a method for processing the interlayer correspondence relationship of multi-layer thickness data. This method includes preprocessing the film thickness data, setting a film thickness range, and mapping the relationship between the film thickness data and the film thickness range.

[0046] First, the acquired film thickness data is preprocessed, including but not limited to data denoising and data distribution analysis. Data denoising can involve using filtering algorithms to eliminate sensor noise and abnormal jitter; filtering algorithms include but are not limited to Gaussian filtering and median filtering. For example, sudden changes in the values ​​of consecutive film thickness measurements can be smoothed. Data distribution analysis can involve plotting a histogram or density distribution curve of the film thickness data to observe clustering characteristics of the film thickness data.

[0047] Secondly, set the membrane thickness interval range, and the membrane thickness interval range setting method includes but is not limited to threshold division based on prior knowledge and unsupervised cluster analysis. The threshold division based on prior knowledge can refer to setting the membrane thickness interval according to the membrane design thickness; for example, the design value Layer1=50um±5um, Layer2=150um±10um, Layer3=300um±20um, then the corresponding membrane thickness interval is Layer1:45-55um, Layer2:140-160um, Layer3:280-320um. Among them, Layer1 is the thickness of the first membrane layer, Layer2 is the thickness of the second membrane layer, and Layer3 is the thickness of the third membrane layer. The first membrane layer is located at the top layer of the proton membrane, the second membrane layer is located at the next membrane layer of the first membrane layer, and the third membrane layer is located at the next membrane layer of the second membrane layer. Optionally, after determining the film thickness interval, a dynamic threshold correction can be performed on the film thickness interval based on the measured film thickness data. For example, the standard deviation σ of the measured film thickness data can be calculated, and the film thickness interval boundaries can be dynamically adjusted according to μ±2σ. Where μ represents the film thickness interval boundary. The unsupervised cluster analysis includes, but is not limited to, the K-means algorithm and is not specifically limited in the present embodiments.

[0048] Finally, the acquired membrane thickness data is mapped against the set membrane thickness range. Membrane thicknesses are in ascending order, and the thickness of the lower membrane layer must be ≥ the cumulative thickness of the upper membrane layer. For example, if the proton membrane consists of a coating layer and a substrate layer, the coating thickness must be less than the substrate thickness. Clustering the acquired membrane thickness data yields three cluster centers: [52 μm, 148 μm, 305 μm], corresponding to Layer 1 → Layer 2 → Layer 3.

[0049] Optionally, a buffer zone or membership function can be used to perform fuzzy classification on the thickness data of interlayer transitions. For example, a ±3 μm overlap zone can be set to map the acquired film thickness data to the film thickness range. For example, if the film thickness range for Layer 1 is 45-55 μm and for Layer 2 is 140-160 μm, and the acquired film thickness data is 57 μm, then the 57 μm ± 3 μm interval is determined to be the film thickness range that meets the requirements, namely Layer 1.

[0050] Optionally, if the acquired film thickness data contains missing layers, the interference signal intensity feature is used to assist in the judgment, or an alarm is triggered when the amount of thickness data of a certain layer is less than 5% of the total.

[0051] This embodiment of the present invention effectively addresses the challenge of analyzing the overlapping interference spectra of multilayer membranes by combining a white-light interferometer with dynamic thresholding or unsupervised cluster analysis, enabling precise separation of the thicknesses of the substrate, coating, and overall membrane. Dynamic threshold correction is supported for membrane thickness intervals, adapting to fluctuations in membrane thickness across batches and ensuring classification stability. To address the issue of overlapping optical path differences between layers, fuzzy classification of transition zones and missing layer alarm logic are proposed to reduce misjudgment rates.

[0052] 230. Determine a membrane thickness detection result, compare the membrane thickness detection result with a preset membrane thickness threshold, determine whether the proton membrane has a membrane thickness abnormality, and determine the location of the membrane thickness abnormality if the proton membrane has a membrane thickness abnormality.

[0053] After determining the proton membrane thickness test result, the membrane thickness test result is compared with a preset membrane thickness threshold to determine whether the proton membrane has a membrane thickness abnormality. The predicted membrane thickness threshold is the set membrane thickness interval range.

[0054] As an optional but non-limiting implementation, determining the membrane thickness detection result, comparing the membrane thickness detection result with a preset membrane thickness threshold, determining whether the proton membrane has a membrane thickness abnormality, and determining the location of the membrane thickness abnormality when the proton membrane has a membrane thickness abnormality, includes but is not limited to steps B1-B3:

[0055] Step B3: Determine the overall proton membrane thickness detection result, and compare the overall proton membrane thickness detection result with a preset proton membrane overall thickness threshold to determine whether the overall proton membrane thickness is abnormal; if the overall proton membrane thickness is abnormal, determine the abnormal position of the overall proton membrane thickness based on the position information obtained by the white light interferometer sensor.

[0056] Step B2: Determine the substrate thickness detection result of the proton membrane, and compare the substrate thickness detection result with a preset substrate thickness threshold to determine whether the substrate thickness of the proton membrane is abnormal; if the substrate thickness of the proton membrane is abnormal, determine the abnormal position of the substrate thickness of the proton membrane based on the position information obtained by the white light interferometer sensor.

[0057] Step B3: Determine the coating thickness detection result of the proton membrane, and compare the coating thickness detection result with a preset coating thickness threshold to determine whether the coating thickness of the proton membrane is abnormal; if the coating thickness of the proton membrane is abnormal, determine the abnormal position of the coating thickness of the proton membrane based on the position information obtained by the white light interferometer sensor.

[0058] Among them, the preset membrane thickness threshold includes a preset proton membrane overall thickness threshold, a preset substrate thickness threshold and a preset coating thickness threshold; the membrane thickness abnormal position includes a proton membrane overall thickness abnormal position, a substrate thickness abnormal position and a coating thickness abnormal position.

[0059] The acquired membrane thickness data is compared with a membrane thickness threshold to determine whether the proton membrane has an abnormal membrane thickness problem; if the membrane thickness is determined to be abnormal, the location of the abnormal membrane thickness is determined. Optionally, when determining whether the proton membrane has an abnormal membrane thickness, the overall thickness of the proton membrane may be compared first. If the overall thickness of the proton membrane is determined to be abnormal, the substrate thickness and coating thickness of the proton membrane may not be compared.

[0060] The embodiment of the present invention detects and compares the substrate thickness, coating thickness and overall thickness of the proton membrane, thereby determining the specific membrane layer where the proton membrane has an abnormality and the location of the abnormal membrane thickness, thereby improving the accuracy of proton membrane layer thickness detection.

[0061] S230. Synchronize the first detection information with the second detection information in real time via a transmission control protocol, display the detected surface defect locations and / or film thickness abnormality locations via a heat map, and use different icon symbols to represent different surface defect types.

[0062] Among them, the first detection information includes the proton membrane surface defect detection results and surface defect location information, and the second detection information includes the proton membrane layer thickness detection results and membrane layer thickness abnormality location information. In the embodiment of the present invention, the line scan smart camera is installed in the front section of the white light interferometer sensor, and all proton membrane surface defects detected by the line scan smart camera can be located at the detailed coordinate position on the proton membrane. The line scan smart camera surface defect detection system is synchronized with the membrane layer thickness detection system through the transmission control protocol. During the proton membrane production process, all defect types and coordinate information detected are synchronized to the membrane layer thickness detection system in real time, and displayed in the membrane layer thickness detection system through a thermal map, and different icon symbols are used to represent different surface defect types.

[0063] For example, see Figure 3 , the heat map contains the width and longitudinal length information of the proton membrane, so it can more intuitively show the surface defects and the thickness of the proton membrane at the same time. Figure 3 It can be seen that the current coating thickness threshold is 19um-21um, the standard coating thickness is 20um, and the solid line is the measured coating thickness data obtained; Figure 3, confirming that the current coating thickness is normal and within the coating thickness threshold. Surface defects exist at a length of 263m and a width of 375mm on the proton membrane, and the type of surface defect is indicated by different icons; for example, ● indicates a bubble, and ◆ indicates a crack.

[0064] This embodiment of the present invention uses heatmap overlay technology to fuse surface defect coordinates with the location of thickness anomalies, visually demonstrating the defect-thickness correlation. It also generates a rotatable view that fuses the surface image, thickness distribution, and defect annotations, producing a 3D interactive report and allowing for easy navigation to the anomaly location.

[0065] As an optional but non-limiting implementation, after determining that the proton membrane is abnormal, the method further includes but is not limited to steps C1-C3:

[0066] Step C1: If it is determined that the proton membrane has surface defects or the membrane thickness is abnormal, the proton membrane abnormality level is marked as level one abnormality and a yellow warning is issued.

[0067] Step C2: If it is determined that the proton membrane has surface defects and the membrane thickness abnormality exceeds a first preset thickness threshold, the proton membrane abnormality level is marked as a level 2 abnormality, a red warning is issued, and a pop-up window prompts whether the operation needs to be stopped for maintenance.

[0068] Step C3: If it is determined that the proton membrane has surface defects and the membrane thickness abnormality exceeds a second preset thickness threshold, the proton membrane abnormality level is marked as level three abnormality, the system is shut down for maintenance, and a self-check program is started to automatically mark the abnormal location;

[0069] Among them, the abnormality level of the third-level abnormality of the proton membrane is greater than the abnormality level of the second-level abnormality, and the abnormality level of the second-level abnormality is greater than the abnormality level of the first-level abnormality; the second preset thickness threshold is greater than the first preset thickness threshold.

[0070] When a proton membrane surface defect and membrane thickness abnormality show an abnormal correlation in spatial location, a graded warning can be issued. Refer to Table 1 for different warning trigger conditions. The proton membrane real-time online detection equipment issues different responses.

[0071] Table 1 Graded warning

[0072] Abnormal Level Trigger Conditions Early warning response Level 1 anomaly Surface defects or abnormal film thickness Record data and issue a yellow warning (continue production) Secondary anomaly Surface defects + thickness fluctuation exceeds 3σ Red alarm (manual confirmation window pops up) Level 3 abnormality Surface defects + thickness deviation > 10% Stop the production line + automatically mark the defective section (start the self-test program)

[0073] If a single anomaly (surface defects or membrane thickness abnormalities only) occurs, it is considered a Level 1 anomaly and a yellow alert is issued, but proton membrane production will not be halted. If a surface defect and the membrane thickness abnormality exceed the first preset thickness threshold, it is considered a Level 2 anomaly and a red alert is issued, with a pop-up window prompting whether to stop production for maintenance. If a surface defect and the membrane thickness abnormality exceed the second preset thickness threshold, it is considered a Level 3 anomaly and requires a maintenance stop, with a self-check program automatically marking the abnormal location.

[0074] Optionally, after determining that the proton membrane is abnormal, a thickness-defect correlation matrix can be established and used to adjust the proton membrane manufacturing process parameters. (See Table 2.) Based on the thickness-defect correlation matrix, the coating machine pressure or solution concentration can be adjusted in real time to reduce process fluctuations.

[0075] Table 2 Thickness-defect correlation matrix

[0076] Defect type Thickness deviation characteristics Process parameter adjustment suggestions pinhole Local thickness drop Increase the concentration of the precursor solution by 2% wrinkles Wave thickness Adjust the tension roller pressure to ±5N pollution particles spike-like protrusions Improve environmental cleanliness level

[0077] Optionally, in an embodiment of the present invention, an evidence report including three-dimensional coordinates, defect type, thickness deviation characteristics and process parameter adjustment suggestions can be generated to improve the efficiency of proton membrane quality root cause analysis.

[0078] An embodiment of the present invention provides a real-time online detection method for proton membranes. By performing real-time online detection of surface defects and membrane thickness of proton membranes and providing graded warnings based on the detection results, continuous detection enables refined monitoring of the production process, optimizes process parameters based on subtle changes, stabilizes product quality, reduces production waste, shortens production cycles, and improves overall production efficiency.

[0079] Example 3

[0080] Figure 4 This is a schematic diagram of the structure of a proton membrane real-time online detection device provided in Example 3 of the present invention. Figure 4 As shown, the device includes a line scan smart camera, a coaxial light source, and a white light interferometer sensor; wherein:

[0081] The line scan smart camera is used to respond to the encoder pulse signal to detect surface defects on the proton membrane and determine the location of the surface defects; wherein the line scan smart camera and the coaxial light source are installed at the first working section of the real-time online detection equipment of the proton membrane;

[0082] The white light interference sensor is used to respond to the encoder synchronization pulse signal, detect the membrane thickness of the proton membrane and determine the location of abnormal membrane thickness; wherein, the white light interference sensor is installed at the second work section position of the proton membrane real-time online detection equipment, and the first work section position is located at the front end of the second work section position.

[0083] Among them, see Figure 4 The real-time online detection equipment for proton membrane includes a line-scanning intelligent camera 101, a feed flattening roller 102, a coaxial light source 103, a linear guide positioning main beam 104, a white light interference sensor 105, a discharge flattening roller 106, a flattening roller assembly positioning subbeam 107, and a fixed support wall panel 108. The line-scanning intelligent camera and the coaxial light source are installed at the first working section position near the feed flattening roller, and the white light interference sensor is installed at the second working section position near the discharge flattening roller. At the same time, the proton membrane is synchronously detected based on the encoder pulse signal. In the embodiment of the present invention, a line-scanning intelligent camera is installed at the front working section position of the white light interference sensor membrane thickness detection to assist in improving the data reliability of the membrane thickness detection. The drive mode of servo motor + synchronous pulley + synchronous belt is adopted to ensure the lateral scanning positioning accuracy of the white light interference sensor, while reducing the overall space size of the equipment and improving the scanning speed of the white light interference sensor.

[0084] Among them, the real-time online detection equipment of the proton membrane integrates a surface defect detection system and a membrane thickness detection system. The real-time online detection equipment of the proton membrane is installed at the middle position after the proton membrane leaves the oven and before the membrane roll is slitting and rewinding through the equipment anchor bolts to perform surface defect detection and membrane thickness detection on the proton membrane; wherein, the surface defect detection system is used to perform surface defect detection on the proton membrane, and the membrane thickness detection system is used to perform membrane thickness detection on the proton membrane.

[0085] The real-time online detection equipment for proton membranes provided in an embodiment of the present invention is used to assemble a proton membrane surface defect detection system and a proton membrane layer thickness detection system. The white light interferometer sensor for proton membrane thickness detection and the line scan intelligent camera for surface defect detection are fixed on the same rigid structure to reduce the on-site deployment and debugging time of the equipment. The equipment can be directly installed on the on-site production line, and real-time online non-contact detection of the proton membrane can be achieved without complex adjustments.

[0086] The need for real-time proton membrane inspection is to address the challenges of detecting surface defects and thickness variations during the coating process during proton membrane production. Based on the actual installation layout of the coating machine production line, the real-time proton membrane inspection equipment provided by the present invention is precisely installed at a critical location after the proton membrane exits the oven and before the membrane roll is slitting and rewinding. This position ensures that defects and thickness variations arising during the production process are detected promptly without interfering with the subsequent slitting and rewinding processes.

[0087] Optionally, the proton membrane real-time online detection device further includes a feed flattening roller and a discharge flattening roller; wherein:

[0088] The feed flattening roller and the discharge flattening roller are used to stretch and flatten the proton membrane.

[0089] Among them, the real-time online detection equipment for proton membrane provided by an embodiment of the present invention is equipped with two flattening rollers, which can flatten the proton membrane, reduce the detection impact caused by the jitter and warping of the proton membrane, improve the stability of surface defect detection of the line scan intelligent camera, and ensure the accuracy of membrane thickness detection of the white light interference sensor.

[0090] Optionally, the proton membrane real-time online detection device further includes a linear guide positioning main beam, a linear guide slider 110, a multi-link height adjustment arm 111, a multi-link depth adjustment guide roller 112, a multi-link rotating component 113, and a multi-link angle adjustment component 114; wherein:

[0091] The linear guide rail positioning main beam is used to position the linear guide rail so as to keep the white light interference sensor parallel to the proton membrane during lateral scanning;

[0092] The linear guide slider is used to connect at least two multi-link components to assemble the white light interference sensor on the linear guide slider through the multi-link components; wherein the at least two multi-link components include a multi-link height adjustment arm, a multi-link depth adjustment guide roller, a multi-link rotating component, and a multi-link angle adjustment component;

[0093] The multi-link height adjustment arm is used to adjust the height of the white light interference sensor relative to the proton membrane;

[0094] The multi-link depth adjustment guide roller is used to adjust the depth distance between the white light interference sensor and the discharge flattening roller, so that the detection focus distance of the white light interference sensor is close to the wrap angle of the discharge flattening roller;

[0095] The multi-link rotating component is used to adjust the rotation angle of the white light interferometer sensor so that the white light interferometer sensor is perpendicular to the proton membrane being measured when the white light interferometer sensor detects the membrane thickness of the proton membrane;

[0096] The multi-link angle adjustment component is used to adjust the angle of the white light interference sensor relative to the flattening plane of the discharge flattening roller, so that the white light interference sensor is perpendicular to the flattening plane of the discharge flattening roller.

[0097] Among them, see Figure 5The white light interference sensor is assembled on the linear guide slider through a multi-link component, which is used for multi-degree-of-freedom adjustment of the white light interference sensor in terms of up, down, front, back, left and right twisting. The linear guide positioning main beam enables the white light interference sensor to maintain the same straightness with the proton membrane during horizontal scanning. The linear guide slider is used to connect the multi-link height adjustment arm, the multi-link depth adjustment guide roller, the multi-link rotating component and the multi-link angle adjustment component. The multi-link height adjustment arm is used to adjust the height of the white light interference sensor relative to the proton membrane. The multi-link depth adjustment guide roller is used to adjust the depth distance between the white light interference sensor and the flattening roller. The closer the detection focus of the white light interference sensor is to the wrap angle of the flattening roller, the better the warping and flattening effect of the proton membrane. The multi-link rotating component is used to adjust the Y-axis rotation angle of the white light interference sensor so that the detection angle of the white light interference sensor is perpendicular to the membrane surface of the proton membrane being measured.

[0098] Optionally, the device further includes a sub-beam end surface fixing bolt 115, a sub-beam height adjustment fastening bolt 116, and a sub-beam positioning pin 117; wherein:

[0099] The sub-beam height adjustment fastening bolts are used to adjust the flattening rollers so that the flattening plane of the flattening rollers remains parallel to the linear guide rails; wherein the flattening rollers include a feed flattening roller and a discharge flattening roller;

[0100] The auxiliary beam end face fixing bolts and the auxiliary beam positioning pins are used to fix the flattening roller assembly positioning auxiliary beam.

[0101] Among them, see Figure 6 The linear guide rail positioning beam of the proton membrane real-time online detection system serves as the positioning beam, while the flattening roller assembly positioning subbeam serves as the adjustment beam. The flattening roller assembly positioning subbeam is used to adjust the flattening roller's flattening plane to maintain parallelism with the linear guide rail. After the positioning and adjustment beams are aligned, four tapered pins are driven into their end faces to prevent loosening during transportation and operation.

[0102] Optionally, the real-time online detection equipment for proton membranes proposed in the embodiment of the present invention has a very high degree of integration. During the equipment production, debugging and assembly stage, key installation and debugging parameters such as the focal length, scanning line and light source angle of the line scan smart camera have been accurately adjusted. At the same time, the white light interference sensor also adjusts all installation indicators to the appropriate range based on the parameters of the flattening roller. Therefore, on site, it is only necessary to smoothly insert the equipment into the designated position in the coating machine production line, use the equipment anchor bolts to quickly level the equipment flattening roller and the tape membrane on the production line, and the basic installation work can be completed, which greatly shortens the on-site installation time and reduces the impact on the normal operation of the production line.

[0103] After the proton membrane real-time online inspection equipment is powered on, it seamlessly connects to the customer's manufacturing execution system via a network cable. Once connected, the equipment automatically retrieves production work orders from the production line and automatically creates product work orders based on the work order information. Once preparatory work is complete, the equipment immediately begins real-time online inspection of the membrane's surface defects and thickness, achieving efficient collaboration between production and inspection, eliminating the need for excessive human intervention and improving production efficiency and inspection accuracy.

[0104] During the proton membrane real-time online detection equipment inspection process, the system comprehensively and completely records the surface defects and membrane thickness data of the proton membrane production process. Once a quality defect in the proton membrane is detected, the system will quickly issue an alarm prompt and, combined with built-in algorithms and a rich experience database, provide targeted treatment suggestions. It is worth mentioning that as the cumulative operating time of the proton membrane real-time online detection equipment increases, the system continuously optimizes the detection model based on big data analysis and machine learning technology. The false detection rate shows a significant downward trend, and the defect treatment suggestions provided are increasingly close to practical and feasible methods, providing strong support for customers to continuously improve product quality.

[0105] The proton membrane real-time online detection equipment provided in the embodiments of the present invention can execute the proton membrane real-time online detection method provided in any of the above embodiments of the present invention, and has the corresponding functions and beneficial effects of executing the proton membrane real-time online detection method. For detailed process, please refer to the relevant operations of the proton membrane real-time online detection method in the above embodiments.

Claims

1. A real-time online detection method for proton membrane, characterized in that: The method comprises: In response to the encoder pulse signal, a line scan smart camera is used to detect surface defects on the proton membrane and determine the location of the surface defects; wherein the line scan smart camera and the coaxial light source are installed at the first working section position of the real-time online detection equipment of the proton membrane, and the first working section position refers to the position corresponding to the feed flattening roller when the proton membrane is subjected to surface defect detection; The encoder pulse signal is synchronously connected to the white light interference sensor, and the white light interference sensor is used to detect the membrane thickness of the proton membrane and determine the abnormal position of the membrane thickness; wherein, the white light interference sensor is installed at the second section position of the proton membrane real-time online detection equipment, the first section position is located at the front end of the second section position, and the second section position refers to the position corresponding to the discharge flattening roller when the proton membrane is subjected to membrane thickness detection; the membrane thickness detection uses rays of different lengths returned by the white light interference sensor to feedback the thickness of the three membrane layers, so as to detect the substrate thickness, coating thickness and overall thickness of the proton membrane; the proton membrane real-time online detection equipment is installed at the middle position after the proton membrane exits the oven and before the membrane roll is slitting and winding through the equipment anchor bolts to perform surface defect detection and membrane thickness detection on the proton membrane; The encoder pulse signal is synchronously connected to a white light interferometer sensor, and the white light interferometer sensor is used to detect the thickness of the proton membrane, including: The encoder pulse signal is synchronously connected to the white light interferometer sensor, and the white light interferometer sensor is used to acquire and preprocess the membrane thickness data of the proton membrane; a membrane thickness range is set, and the acquired membrane thickness data is mapped to the set membrane thickness range; wherein the membrane thickness data includes substrate thickness data, coating thickness data and overall proton membrane thickness data; for the membrane thickness data of interlayer transition, a buffer zone can be set or a membership function can be introduced for fuzzy classification; if there is a missing layer in the acquired membrane thickness data, the interference signal intensity feature is used to assist in judgment, or an alarm is triggered when the amount of thickness data of a certain layer is less than 5% of the total.

2. The method according to claim 1, characterized in that The step of synchronously connecting the encoder pulse signal to a white light interferometer sensor, and using the white light interferometer sensor to detect the thickness of the proton membrane and determine the location of abnormal thickness of the membrane comprises: Synchronously connecting the encoder pulse signal to a white light interferometer sensor, and using the white light interferometer sensor to detect the membrane thickness of the proton membrane; wherein the proton membrane is composed of a substrate and a coating, and the membrane thickness detection includes substrate thickness detection, coating thickness detection, and overall proton membrane thickness detection; Determine a membrane thickness detection result, compare the membrane thickness detection result with a preset membrane thickness threshold, determine whether the proton membrane has a membrane thickness abnormality, and determine the location of the membrane thickness abnormality if the proton membrane has a membrane thickness abnormality.

3. The method according to claim 2, characterized in that The determining of the membrane thickness detection result, comparing the membrane thickness detection result with a preset membrane thickness threshold, determining whether the proton membrane has a membrane thickness abnormality, and determining the location of the membrane thickness abnormality when the proton membrane has a membrane thickness abnormality, includes: Determine a proton membrane overall thickness detection result, and compare the proton membrane overall thickness detection result with a preset proton membrane overall thickness threshold to determine whether the proton membrane overall thickness is abnormal; if the proton membrane overall thickness is abnormal, determine the abnormal position of the proton membrane overall thickness based on the position information obtained by the white light interferometer sensor; Determine a result of a substrate thickness detection of the proton membrane, and compare the substrate thickness detection result with a preset substrate thickness threshold to determine whether the substrate thickness of the proton membrane is abnormal; if the substrate thickness of the proton membrane is abnormal, determine a position of the abnormal substrate thickness of the proton membrane based on position information obtained by the white light interferometer sensor; Determine a coating thickness detection result of the proton membrane, and compare the coating thickness detection result with a preset coating thickness threshold to determine whether the coating thickness of the proton membrane is abnormal; if the coating thickness of the proton membrane is abnormal, determine the abnormal position of the coating thickness of the proton membrane based on the position information obtained by the white light interferometer sensor; Among them, the preset membrane thickness threshold includes a preset proton membrane overall thickness threshold, a preset substrate thickness threshold and a preset coating thickness threshold; the membrane thickness abnormal position includes a proton membrane overall thickness abnormal position, a substrate thickness abnormal position and a coating thickness abnormal position.

4. The method according to claim 1, wherein The method further comprises: The first detection information is synchronized with the second detection information in real time through the transmission control protocol, and the detected surface defect positions and / or membrane thickness abnormality positions are displayed through a thermal map, and different icon symbols are used to represent different surface defect types; wherein, the first detection information includes the proton membrane surface defect detection results and surface defect position information, and the second detection information includes the proton membrane layer thickness detection results and membrane layer thickness abnormality position information.

5. The method according to claim 4, characterized in that After determining that the proton membrane is abnormal, the method further includes: If it is determined that the proton membrane has surface defects or abnormal membrane thickness, the proton membrane abnormality level is marked as level one abnormality and a yellow warning is issued; If it is determined that the proton membrane has surface defects and the membrane thickness abnormality exceeds a first preset thickness threshold, the proton membrane abnormality level is marked as a level 2 abnormality, a red warning is issued, and a pop-up window prompts whether the operation needs to be stopped for maintenance; If it is determined that the proton membrane has surface defects and the membrane thickness abnormality exceeds a second preset thickness threshold, the proton membrane abnormality level is marked as level three abnormality, the operation is stopped for maintenance, and a self-check program is started to automatically mark the abnormal location; Among them, the abnormality level of the third-level abnormality of the proton membrane is greater than the abnormality level of the second-level abnormality, and the abnormality level of the second-level abnormality is greater than the abnormality level of the first-level abnormality; the second preset thickness threshold is greater than the first preset thickness threshold.

6. A real-time online detection device for proton membrane, characterized in that: The device includes a line scan smart camera, a coaxial light source, and a white light interferometer sensor; wherein: The line scan smart camera is used to respond to the encoder pulse signal to detect surface defects on the proton membrane and determine the location of the surface defects; wherein the line scan smart camera and the coaxial light source are installed at the first working section position of the real-time online detection equipment of the proton membrane, and the first working section position refers to the position corresponding to the feed flattening roller when the proton membrane is subjected to surface defect detection; The white light interference sensor is used to respond to the encoder synchronization pulse signal to detect the membrane thickness of the proton membrane and determine the position of abnormal membrane thickness; wherein, the white light interference sensor is installed at the second section position of the proton membrane real-time online detection equipment, the first section position is located at the front end of the second section position, and the second section position refers to the position corresponding to the discharge flattening roller when the proton membrane is subjected to membrane thickness detection; the membrane thickness detection uses rays of different lengths returned by the white light interference sensor to feedback the thickness of the three membrane layers, so as to detect the substrate thickness, coating thickness and overall thickness of the proton membrane; the proton membrane real-time online detection equipment is installed at the middle position after the proton membrane exits the oven and before the membrane roll is slitting and winding through the equipment anchor bolts to perform surface defect detection and membrane thickness detection on the proton membrane; Wherein, the white light interference sensor is specifically used for: The encoder pulse signal is synchronously connected to the white light interferometer sensor, and the white light interferometer sensor is used to acquire and preprocess the membrane thickness data of the proton membrane; a membrane thickness range is set, and the acquired membrane thickness data is mapped to the set membrane thickness range; wherein the membrane thickness data includes substrate thickness data, coating thickness data and overall proton membrane thickness data; for the membrane thickness data of interlayer transition, a buffer zone can be set or a membership function can be introduced for fuzzy classification; if there is a missing layer in the acquired membrane thickness data, the interference signal intensity feature is used to assist in judgment, or an alarm is triggered when the amount of thickness data of a certain layer is less than 5% of the total.

7. The device according to claim 6, characterized in that The real-time online detection equipment for proton membranes integrates a surface defect detection system and a membrane thickness detection system; wherein, the surface defect detection system is used to perform surface defect detection on the proton membrane, and the membrane thickness detection system is used to perform membrane thickness detection on the proton membrane.

8. The device according to claim 6, characterized in that The proton membrane real-time online detection device further includes a feed flattening roller and a discharge flattening roller; wherein: The feed flattening roller and the discharge flattening roller are used to stretch and flatten the proton membrane.

9. The device according to claim 6, characterized in that The proton membrane real-time online detection equipment further includes a linear guide rail positioning main beam, a linear guide rail slider, a multi-link height adjustment arm, a multi-link depth adjustment guide roller, a multi-link rotating component, and a multi-link angle adjustment component; wherein: The linear guide rail positioning main beam is used to position the linear guide rail so as to keep the white light interference sensor parallel to the proton membrane during lateral scanning; The linear guide slider is used to connect at least two multi-link components to assemble the white light interference sensor on the linear guide slider through the multi-link components; wherein the at least two multi-link components include a multi-link height adjustment arm, a multi-link depth adjustment guide roller, a multi-link rotating component, and a multi-link angle adjustment component; The multi-link height adjustment arm is used to adjust the height of the white light interference sensor relative to the proton membrane; The multi-link depth adjustment guide roller is used to adjust the depth distance between the white light interference sensor and the discharge flattening roller, so that the detection focus distance of the white light interference sensor is close to the wrap angle of the discharge flattening roller; The multi-link rotating component is used to adjust the rotation angle of the white light interferometer sensor so that the white light interferometer sensor is perpendicular to the proton membrane being measured when the white light interferometer sensor detects the membrane thickness of the proton membrane; The multi-link angle adjustment component is used to adjust the angle of the white light interference sensor relative to the flattening plane of the discharge flattening roller, so that the white light interference sensor is perpendicular to the flattening plane of the discharge flattening roller.

10. The device according to claim 6, characterized in that The device also includes auxiliary beam end surface fixing bolts, auxiliary beam height adjustment fastening bolts and auxiliary beam positioning pins; wherein: The sub-beam height adjustment fastening bolts are used to adjust the flattening rollers so that the flattening plane of the flattening rollers remains parallel to the linear guide rails; wherein the flattening rollers include a feed flattening roller and a discharge flattening roller; The auxiliary beam end face fixing bolts and the auxiliary beam positioning pins are used to fix the flattening roller assembly positioning auxiliary beam.

Images