Method and system for judging hardening degree of pulverized solid hydrogen storage alloy material

Abstract

The invention relates to the technical field of solid hydrogen storage, in particular to a method and system for judging the hardening degree of a pulverized solid hydrogen storage alloy material. The method comprises the following steps of: observing a morphology picture of the lower part of the device in a cyclic hydrogen charging and discharging experiment reaction process, simultaneously detecting strain at a connecting line of centers of two integral flanges in a symmetrical first shooting unit, and finding that when a newly added microcrack appears, a corresponding strain sudden increase point 1; when the newly added microcracks are further thickened, a strain sudden increase point 2 is formed; when the cracks are completely closed after the primary circulating hydrogen charging and discharging experiment, the particles are further hardened, and at the moment, the corresponding strain sudden increase point 3 is obtained; therefore, an obvious corresponding relation between fracture evolution and strain growth can be found, and the strain accumulation condition of the reaction bed can be accurately evaluated by monitoring fracture change.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of solid-state hydrogen storage technology, and in particular to a method and system for judging the degree of solidification of hydrogen storage alloy material after pulverization. BACKGROUND

[0002] The information disclosed in this Background section is only for the purpose of increasing an understanding of the general background of the application and does not necessarily constitute an admission or a recognition that the information forms part of the prior art already known to a person of ordinary skill in the art.

[0003] The development and utilization of hydrogen energy can effectively alleviate the problems of fossil energy shortage and environmental pollution. The key to hydrogen energy application lies in hydrogen storage. Solid-state hydrogen storage technology in hydrogen storage technology has broad application prospects in fixed hydrogen storage due to its advantages of no high pressure, no high energy consumption, good safety and high volume hydrogen storage density.

[0004] Among the many hydrogen storage materials, hydrogen storage alloy has been widely concerned as the core active material of hydrogen storage device due to its mild operating conditions and high hydrogen storage capacity. Hydrogen storage alloy usually stores hydrogen in the tetrahedral or octahedral interstitial sites of its crystal lattice. However, during repeated hydrogen absorption and desorption, hydrogen can cause large lattice stress in the alloy, leading to pulverization and formation of hydrogen storage alloy fine powder. Under the action of gravity, gas flow and periodic compression, hydrogen storage alloy fine powder accumulates and consolidates in one direction to form agglomerates, causing local stress to accumulate rapidly, and even leading to plastic deformation and structure rupture of the reaction bed. This not only threatens the safety and service life of the device, but also significantly increases the heat and mass transfer resistance, blocks the diffusion of hydrogen, and seriously degrades the hydrogen storage performance.

[0005] To solve the problem of hydrogen storage alloy pulverization and solidification, some studies have attempted to alleviate it through structural modification and mechanical intervention, which can improve the mechanical and heat transfer performance of the hydrogen storage reactor to some extent, but still has limitations such as monitoring lag, state zero, and passive intervention means. The evolution process of hydrogen storage alloy pulverization, solidification and crack in the reactor is difficult to capture in real time, which leads to abnormality of the device when the stress accumulates to the limit or the hydrogen transport is severely blocked, making it difficult to take effective measures in time. SUMMARY

[0006] In order to overcome the above problems, the present application provides a method and system for judging the degree of solidification of solid-state hydrogen storage alloy material after pulverization.

[0007] In order to achieve the above technical purpose, the present application adopts the following technical solutions: The first aspect of the present application provides a method for judging the degree of solidification of solid-state hydrogen storage alloy material after pulverization, comprising the following steps: (1) continuously performing hydrogen charging and discharging experiments on the solid-state hydrogen storage alloy material in a cyclic hydrogen charging and discharging experimental device until the solid-state hydrogen storage alloy material is consolidated, obtaining a morphology photo of the lower part of the device after the cyclic hydrogen charging and discharging experiment, marking the cracks on the morphology photo after pretreatment, and establishing a training sample set; (2) training the convolutional neural network model using the training sample set to obtain the trained convolutional neural network model; (3) obtaining a morphology photo of the lower part of the device in real time during the reaction process of the cyclic hydrogen charging and discharging experiment, inputting the pretreated morphology photo into the trained convolutional neural network model, identifying the cracks and their widths, and determining the consolidation degree of the solid-state hydrogen storage alloy material after pulverization according to a preset determination basis.

[0008] In one or more embodiments, in step (1), the determination basis for the consolidation of the solid-state hydrogen storage alloy material is that, taking the morphology photo after the first cycle of hydrogen charging and discharging as a reference, when the cracks change significantly, it can be determined that the solid-state hydrogen storage alloy material is consolidated; the significant change includes: appearance of new cracks, crack expansion, and closure of reference cracks; crack expansion refers to an increase in the length, width or area of the cracks. The appearance of new cracks is a characteristic of consolidation of loose particles, crack expansion is a characteristic of further consolidation of a single consolidated body, and closure of reference cracks is a characteristic of further consolidation between consolidated bodies. The reference crack is a crack in the morphology photo after the first cycle of hydrogen charging and discharging.

[0009] In one or more embodiments, in step (1), the cyclic hydrogen charging and discharging experimental device comprises a cylindrical body, and two groups of shooting units are symmetrically arranged at the lower part of the cylindrical body. The shooting unit comprises, in sequence, a light shield, an LED ring-shaped light source and an industrial high-definition camera; and the light shield is connected to the cylindrical body through an integral flange.

[0010] Preferably, the shooting unit comprises a first shooting unit and a second shooting unit, and the first shooting unit is located above the second shooting unit.

[0011] Preferably, a hydrogen inlet and outlet are arranged at the top of the cylindrical body.

[0012] Preferably, the length-diameter ratio of the cylindrical body is 8-10.

[0013] Preferably, a protective shell is arranged outside the shooting unit.

[0014] Preferably, a rotatable polarizing plate is arranged at the front end of the lens of the industrial high-definition camera.

[0015] Preferably, the light shield, the LED ring-shaped light source and the industrial high-definition camera are concentrically fixed.

[0016] Preferably, a hydrogen inlet and outlet are arranged at the top of the cylindrical body.

[0017] Preferably, the inner wall of the whole flange is fixed with a sapphire window.

[0018] In one or more embodiments, in step (1), the solid-state hydrogen storage alloy material comprises one of: a LaNi5 alloy, a TiFe alloy, a TiMn2 alloy, or a Mg-based hydrogen storage material.

[0019] In one or more embodiments, in steps (1) and (3), the pre-processing method comprises: S1, super-resolution processing; S2, adjusting brightness and contrast; S3, global noise reduction and local noise reduction in noise areas; S4, threshold segmentation, screening cracks as dark and solid-state hydrogen storage alloy materials as bright; S5, removing noise points.

[0020] In one or more embodiments, in step (3), the method for determining the degree of solidification of the solid-state hydrogen storage alloy material after pulverization according to the predetermined determination basis comprises: Obtaining the appearance photos of the lower part of the device after the first three cycle hydrogen charging and discharging experiments, inputting the pre-processed appearance photos into the trained convolutional neural network model, obtaining the cracks and their widths, and setting the smallest width in the cracks as W min . When the real-time crack width is first expanded from a fine crack to greater than or equal to 3 times W min , it is confirmed that the degree of solidification is in the first warning period; When the real-time reference crack width is first less than or equal to W min , and the real-time reference crack width before 2 cycles is greater than W min , it is confirmed that the degree of solidification is in the second warning period; the reference crack is the crack in the appearance photo after the first cycle hydrogen charging and discharging.

[0021] In a second aspect of the present application, a system for determining the degree of solidification of a solid-state hydrogen storage alloy material after pulverization is provided, comprising: A training sample set module for continuously performing hydrogen charging and discharging experiments on a solid-state hydrogen storage alloy material in a cycle hydrogen charging and discharging experimental device until the solid-state hydrogen storage alloy material is solidified, obtaining appearance photos of the lower part of the device after the cycle hydrogen charging and discharging experiments, pre-processing, marking cracks on the appearance photos, and establishing a training sample set; A model training module for training a convolutional neural network model using the training sample set, and obtaining a trained convolutional neural network model; The degree of hardening judgment module is used for acquiring the appearance photos of the lower part of the device in the cyclic hydrogen charging and discharging experiment in real time, inputting the preprocessed appearance photos into the trained convolutional neural network model, identifying the crack width, and determining the degree of hardening of the pulverized solid-state hydrogen storage alloy material according to the preset judgment basis.

[0022] The present application has the advantages that: In the present application, by observing the appearance photos of the lower part of the device in the cyclic hydrogen charging and discharging experiment and simultaneously detecting the strain at the connecting line of the centers of the two overall flanges in the symmetrical first shooting unit, it is found that when new microcracks appear, the corresponding strain increases sharply at point 1; when the new microcracks further thicken and the reference crack reduces, strain increases sharply at point 2; and when the reference crack is completely closed and the particles further harden, strain increases sharply at point 3; thus it can be seen that there is a clear corresponding relationship between crack evolution and strain growth, and the reaction bed strain accumulation condition can be accurately evaluated by monitoring the crack change. Therefore, the preprocessed appearance photos are input into the trained convolutional neural network model to identify the crack width, and the degree of hardening of the pulverized solid-state hydrogen storage alloy material is determined according to the preset judgment basis. When the degree of hardening is between the first and second warning periods, local measures are taken to prevent the solid-state hydrogen storage alloy material from hardening, and the negative effects of the hardening of the pulverized solid-state hydrogen storage alloy material on the safety of the reaction device and the hydrogen storage performance of the material are eliminated. BRIEF DESCRIPTION OF DRAWINGS

[0023] The accompanying drawings, which form a part of the present application, are used to provide a further understanding of the present application, and the illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute improper limitations on the present application.

[0024] Figure 1 Fig. 1 is a structural schematic diagram of a cyclic hydrogen charging and discharging experiment device, wherein a is an elevation view, b is an elevation sectional view, and c is a side view; Figure 2 Fig. 2 is an inclined sectional view of the cyclic hydrogen charging and discharging experiment device; Figure 3 Fig. 3 is a structural schematic diagram of a light shield; Figure 4 Fig. 4 is a structural schematic diagram of an LED ring light source; Figure 5 Fig. 5 is a structural schematic diagram of an industrial high-definition camera; Figure 6The relationship between the morphology photos taken at the same first shooting unit after the continuous hydrogen charging and discharging experiment of the solid-state hydrogen storage alloy material and the cycle number and strain value, wherein (a) is the morphology photo taken at the same first shooting unit after the continuous hydrogen charging and discharging experiment of the solid-state hydrogen storage alloy material in the cycle hydrogen charging and discharging experimental device, and (b) is the relationship between the cycle number and the strain value; Figures 1-5 In the figure, 1 is a cylindrical body, 2 is an integral flange, 3 is a light shield, 4 is an LED ring light source, 5 is an industrial high-definition camera, 6 is a signal line 6, 7 is a protective shell, 8 is a hydrogen inlet and outlet, 9 is a first shooting unit, 10 is a second shooting unit, 11 is a sapphire window installation groove, 12 is a first flange, 13 is a second flange, 14 is a connecting plate, and 15 is a cover body. DETAILED DESCRIPTION

[0025] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0026] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and furthermore, it should be understood that when the terms "comprise" and / or "include" are used in the specification, there is a reference to the presence of a feature, step, operation, device, component, and / or combinations thereof.

[0027] In order to enable those skilled in the art to more clearly understand the technical solutions of the present application, the technical solutions of the present application will be described in detail below with specific examples.

[0028] Example 1 Figure 1 is a structural schematic diagram of the cycle hydrogen charging and discharging experimental device, and Figure 1 The cycle hydrogen charging and discharging experimental device comprises a cylindrical body 1, and two groups of shooting units are symmetrically arranged on the lower part of the cylindrical body 1. The shooting unit comprises a light shield 3, an LED ring light source 4 and an industrial high-definition camera 5 connected in sequence; the light shield 3 is connected with the cylindrical body 1 through an integral flange 2.

[0029] Among them, the shooting unit comprises a first shooting unit 9 and a second shooting unit 10, and the first shooting unit 9 is located above the second shooting unit 10.

[0030] In order to protect the shooting unit, a protective shell 7 is arranged outside the shooting unit.

[0031] Referring to Figure 2 , the integral flange 2 comprises a first flange 12 and a second flange 13; wherein the first flange is integrally formed with the cylindrical body 1, and the first flange 12 is uniformly provided with first through holes, and the second flange 13 is uniformly provided with second through holes.

[0032] Referring to Figure 3 , the light shield 3 comprises a shield body 15, and the two sides of the shield body 15 are detachably connected with connecting plates 14, and the connecting plates 14 are uniformly provided with third through holes, and the first through holes, the second through holes and the third through holes are opposite to each other in position. The second flange 13 is located in the shield body 15, and the first bolt passes through the second through hole, the third through hole and the first through hole in sequence to connect the light shield 3 with the cylindrical body 1.

[0033] Referring to Figure 4 , the outer wall of the LED ring light source 4 is uniformly provided with bolt holes, the bolt holes are opposite to the third through holes on the connecting plates 14 away from the integral flange 2, and the bolt holes and the third through holes are fixed by the second bolt.

[0034] In order to facilitate the entry and exit of hydrogen, the top of the cylindrical body 1 is provided with a hydrogen inlet and outlet 8. The length-diameter ratio of the cylindrical body 1 is 8-10.

[0035] In order to clearly obtain the morphology photos of the solid-state hydrogen storage alloy cyclic hydrogen charging and discharging experiment, referring to Figure 1 , the light shield, the LED ring light source and the industrial high-definition camera are fixed concentrically; referring to Figure 2 , a sapphire window setting groove 11 is arranged on the inner side of the first flange, and a sapphire window is placed in the sapphire window setting groove 11; a diffusion sheet and a polarizing sheet are arranged in sequence at the light emitting end of the LED ring light source 4; referring to Figure 5 , a rotatable polarizing sheet 16 is arranged at the front end of the lens of the industrial high-definition camera 5.

[0036] In order to transmit the morphology photos, the industrial high-definition camera 5 is connected with a signal line 6.

[0037] Embodiment 2 The cyclic hydrogen charging and discharging experiment is carried out by using the cyclic hydrogen charging and discharging experiment device provided in embodiment 1: The specific process includes: ① 100 g of LaNi 3.9 Co 0.6 Mn 0.3 Al 0.4The solid hydrogen storage alloy material was activated in a cyclic hydrogen charge-discharge experimental apparatus. It was charged with hydrogen at 80 °C and 1.5 MPa for 60 min, followed by vacuuming for 30 min, and this process was repeated until the solid hydrogen storage alloy material stably absorbed 1.1 wt% of hydrogen. ② After activation, the hydrogen storage alloy underwent a cyclic hydrogen charge-discharge experiment: the alloy absorbed hydrogen at 1.0 MPa for 30 min, then was heated to 100 °C to release hydrogen for 60 min. After hydrogen release, the cracks remained stable, and images were captured using an industrial high-definition camera. The cyclic hydrogen charge-discharge experiment was repeated according to the above steps.

[0038] During the gradual hydrogen absorption and desorption cycle, the hydrogen storage alloy material pulverizes and then agglomerates unidirectionally under the influence of gravity settling, airflow transport, and cyclic compression effects, forming agglomerates of varying sizes. The agglomerates exhibit "breathing" behavior as they absorb and desorb hydrogen. When the hydrogen storage alloy material releases hydrogen, it contracts, creating cracks between the agglomerates. When the hydrogen storage alloy material absorbs hydrogen, it expands and compresses, closing the cracks and further compacting the agglomerates.

[0039] Figure 6 Image (a) shows the morphological photographs of the solid hydrogen storage alloy material taken at the same first imaging unit after continuous hydrogen charging and discharging experiments in the cyclic hydrogen charging and discharging experimental device. Simultaneously, the strain at the midpoint of the line connecting the centers of the first flanges was measured, obtaining a graph showing the relationship between the number of cycles and the strain value. Figure 6 As shown in (b), when crack 2 appears as a microcrack after 5 cycles, this corresponds to the strain surge point 1. Subsequently, after 8 cycles, crack 2 expands into a coarse crack, while crack 1 shrinks into a microcrack, corresponding to the strain surge point 2. Continuing the hydrogen absorption and desorption cycle, after 13 cycles, crack 1 completely closes, crack 2 further expands, and the particles further agglomerate, corresponding to the strain surge point 3. This demonstrates a clear correlation between crack evolution and strain growth, allowing for precise assessment of strain accumulation in the reaction bed by monitoring crack changes. The aforementioned strain surge point is a node with a significant and rapid increase in strain compared to the previous strain increase. After the strain surge point, the strain will accumulate steadily and gradually decrease until the particles further agglomerate, at which point the strain surges again.

[0040] Example 3 Based on the clear correlation between crack evolution and strain growth in Example 2, a method for judging the degree of caking after pulverization of solid hydrogen storage alloy materials is provided, including the following steps: (1) The solid hydrogen storage alloy material was continuously charged and discharged in the hydrogen cyclic charging and discharging experimental device until the solid hydrogen storage alloy material was caking. The morphological photos of the lower part of the device after the hydrogen cyclic charging and discharging experiment were obtained. After preprocessing, crack markings were made on the morphological photos to establish a training sample set. (2) Train the convolutional neural network model using the training sample set to obtain the trained convolutional neural network model; (3) Real-time acquisition of the morphology photo of the lower part of the device during the cyclic hydrogen charging and discharging experiment, inputting the pre-processed morphology photo into the trained convolutional neural network model, identifying the crack width, and determining the current solid-state hydrogen storage alloy material powder agglomeration degree according to the preset judgment basis.

[0041] In step (1), the judgment basis for the solid-state hydrogen storage alloy material agglomeration in step (1) is: taking the morphology photo after the first cycle of hydrogen charging and discharging as the reference, when the crack changes obviously, it can be judged that the solid-state hydrogen storage alloy material is agglomerated; the obvious change includes: new crack appears, crack expansion and reference crack closure; crack expansion is the increase of the length or width or area of the crack. The appearance of new cracks is a characteristic of loose particle agglomeration, crack expansion is a characteristic of single agglomeration body further agglomeration, and reference crack closure is a characteristic of further agglomeration between agglomeration bodies. The reference crack is the crack in the morphology photo after the first cycle of hydrogen charging and discharging.

[0042] In steps (1) and (3), the pre-processing method includes: S1, super-resolution processing; S2, adjusting brightness and contrast; S3, global noise reduction and local noise reduction in noisy areas; S4, threshold segmentation, screening cracks as dark and solid-state hydrogen storage alloy materials as bright; S5, removing outliers.

[0043] Specifically, in S1, the Waifu2x network is used for super-resolution processing to improve image clarity, and the original image resolution is enhanced by two times; In S2, Fiji software is used to adjust image brightness and contrast; In S3, Topaz DeNoise AI (https: / / www.topazlabs.com / denoise-ai) software is used for global noise reduction and local noise reduction in noisy areas; In S4, Fiji software is used for threshold segmentation, screening cracks as dark and solid-state hydrogen storage alloy materials as bright, adjusting the Minimum / Maximum parameters to display pixels below Minimum as pure black and pixels above Maximum as pure white, and obtaining a clean image; In S5, Fiji software is used to remove outliers. First, the median filter is used to smooth isolated noise points; then the RemoveOutliers algorithm is used to remove bright / dark outliers; then small area noise points are eliminated and the crack edge is smoothed; finally, all small area isolated particles are automatically deleted using the AnalyzeParticles function to obtain a relatively clean crack binary image.

[0044] The morphology photos of the solid-state hydrogen storage alloy material in Example 2 after the continuous 1-20 cycle hydrogen charging and discharging experiments in the cycle hydrogen charging and discharging experimental device are taken at the same first shooting unit, the images are cut into four equal parts, the cut images are rotated by 30°, 45°, 60°, 90° or flipped as a whole respectively, 400 random image data sets are obtained, and manual crack marking is performed, and the original image and the marked image are used as a training sample set.

[0045] The training sample set is input into the U-Net convolutional neural network to autonomously learn the mapping relationship between the marked image and the original image, train the U-Net network model, and obtain the trained convolutional neural network model. (3) Real-time acquisition of the morphology photos of the lower part of the device during the reaction process of the cycle hydrogen charging and discharging experiment, input of the preprocessed morphology photos into the trained convolutional neural network model, identification of the crack width, and determination of the degree of solidification of the solid-state hydrogen storage alloy material after pulverization according to the preset determination basis.

[0046] In step (3), the method for determining the degree of solidification of the solid-state hydrogen storage alloy material after pulverization according to the preset determination basis comprises: Acquisition of the morphology photos of the lower part of the device after the first three cycle hydrogen charging and discharging experiments, input of the preprocessed morphology photos into the trained convolutional neural network model, acquisition of the cracks and their widths, and setting of the smallest width in the cracks as W min ; When the real-time crack width is first expanded from a fine crack to greater than or equal to 3 times W min , it is confirmed that the degree of solidification is in the first warning period. When the real-time reference crack width is first less than or equal to W min , and the real-time reference crack width before 2 cycles is greater than W min , it is confirmed that the degree of solidification is in the second warning period; the reference crack is the crack in the morphology photo after the first cycle hydrogen charging and discharging.

[0047] When the degree of solidification is between the first warning period and the second warning period, local measures are taken to prevent the solid-state hydrogen storage alloy material from solidifying, and to eliminate the negative effects of the solidification of the solid-state hydrogen storage alloy material after pulverization on the safety of the reaction device and the hydrogen storage performance of the material.

[0048] Example 4 A system for judging the degree of solidification of a solid-state hydrogen storage alloy material after pulverization, comprising: A training sample set module for continuously performing hydrogen charging and discharging experiments on a solid-state hydrogen storage alloy material in a cycle hydrogen charging and discharging experimental device until the solid-state hydrogen storage alloy material solidifies, acquiring the morphology photos of the lower part of the device after the cycle hydrogen charging and discharging experiments, preprocessing, marking cracks on the morphology photos, and establishing a training sample set. a model training module configured to train the convolutional neural network model by using a training sample set, and obtain a trained convolutional neural network model; a degree of hardening determination module configured to acquire a morphology photo of a lower part of the device in a cyclic hydrogen charging and discharging experiment in real time, input the preprocessed morphology photo into the trained convolutional neural network model, identify a crack width, and determine a degree of hardening of the pulverized solid-state hydrogen storage alloy material according to a preset determination basis.

[0049] The preferred embodiments of the present application have been described above with the aid of drawing; however, for the person skilled in the art, various modifications and changes can be made thereto without departing from the spirit and principles of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall fall within the scope of the present application.

Claims

1. A method for determining the degree of caking after pulverization of a solid hydrogen storage alloy material, characterized in that, Includes the following steps: (1) The solid hydrogen storage alloy material was continuously charged and discharged in the hydrogen cyclic charging and discharging experimental device until the solid hydrogen storage alloy material was caking. The morphological photos of the lower part of the device after the hydrogen cyclic charging and discharging experiment were obtained. After preprocessing, crack markings were made on the morphological photos to establish a training sample set. (2) Train the convolutional neural network model using the training sample set to obtain the trained convolutional neural network model; (3) Real-time acquisition of morphological photos of the lower part of the device during the cyclic hydrogen charging and discharging experiment reaction, inputting the pre-processed morphological photos into the trained convolutional neural network model, identifying the crack width, and determining the degree of pulverization and caking of the current solid hydrogen storage alloy material according to the preset judgment criteria.

2. The determination method as described in claim 1, characterized in that, In step (1), the criterion for judging the solid hydrogen storage alloy material caking is: based on the morphological photos after the first cycle of hydrogen charging and discharging, when the cracks show obvious changes, it can be judged that the solid hydrogen storage alloy material is caking. The significant changes include: the appearance of new cracks, crack propagation, and closure of the baseline crack; crack propagation is an increase in the length, width, or area of ​​the crack.

3. The determination method as described in claim 1, characterized in that, In step (1), the hydrogen cyclic charging and discharging experimental device includes a cylindrical body, and two sets of imaging units are fixedly arranged symmetrically at the lower part of the cylindrical body; The shooting unit includes a light shield, an LED ring light source, and an industrial high-definition camera connected in sequence; the light shield is connected to the cylindrical body through an integral flange.

4. The determination method as described in claim 3, characterized in that, The shooting unit includes a first shooting unit and a second shooting unit, with the first shooting unit located above the second shooting unit.

5. The determination method as described in claim 3, characterized in that, The top of the cylindrical body is equipped with a hydrogen inlet and outlet; Alternatively, the cylindrical body has a length-to-diameter ratio of 8 to 10.

6. The determination method as described in claim 3, characterized in that, The camera unit is equipped with a protective shell on its outer side; Alternatively, the front end of the lens of the industrial high-definition camera may be provided with a rotatable polarizer.

7. The determination method as described in claim 3, characterized in that, The light shield, LED ring light source, and industrial high-definition camera are concentrically fixed together. Alternatively, a hydrogen inlet and outlet can be provided at the top of the cylindrical body; Alternatively, a sapphire window may be fixed to the inner wall of the integral flange.

8. The determination method as described in claim 1, characterized in that, In step (1), the solid hydrogen storage alloy material includes one of the following: LaNi5 series alloy, TiFe series alloy, TiMn2 series alloy or Mg-based alloy hydrogen storage material; Alternatively, in steps (1) and (3), the preprocessing methods include: S1, Super-resolution processing; S2, Adjust brightness and contrast; S3, global noise reduction and local noise reduction in noisy areas; S4. Threshold segmentation: the gaps in the sieve are dark, and the solid hydrogen storage alloy material is bright. S5, Remove noise.

9. The determination method as described in claim 1, characterized in that, In step (3), the method for determining the degree of condensation of the solid hydrogen storage alloy material after pulverization according to the preset judgment criteria includes: After the first three cycles of hydrogen charging and discharging experiments, morphological images of the lower part of the apparatus were obtained. These preprocessed images were then input into a trained convolutional neural network model to obtain the cracks and their widths. The minimum width of the crack was set to W. min ; When the real-time crack width first expands from a fine crack to greater than or equal to W min When the concentration is three times that of the target, the degree of compaction is confirmed to be in the first warning stage. When the real-time reference crack width is first less than or equal to W min And the real-time reference crack width before the second cycle is greater than W. min At that time, it was confirmed that the degree of compaction was in the second warning period; the reference crack was the crack in the morphology photo after the first cycle of hydrogen charging and discharging.

10. A system for judging the degree of caking after pulverization of solid hydrogen storage alloy materials, characterized in that, include: The training sample set module is used to continuously conduct hydrogen charging and discharging experiments on solid hydrogen storage alloy materials in a cyclic hydrogen charging and discharging experimental device until the solid hydrogen storage alloy materials are caking. After the cyclic hydrogen charging and discharging experiment, the morphology photos of the lower part of the device are obtained. After preprocessing, crack markings are made on the morphology photos to establish a training sample set. The model training module is used to train the convolutional neural network model using the training sample set and obtain the trained convolutional neural network model. The caking degree judgment module is used to acquire real-time morphological photos of the lower part of the device during the cyclic hydrogen charging and discharging experiment reaction. The pre-processed morphological photos are input into the trained convolutional neural network model to identify the crack width and determine the caking degree of the solid hydrogen storage alloy material after pulverization according to the preset judgment criteria.

Images