Lithium battery element distribution lossless imaging system and method

Abstract

The invention relates to the technical field of imaging analysis, and discloses a lithium battery element distribution lossless imaging system and method. The system comprises a high-energy negative muon beam source module which is used for outputting momentum-adjustable negative muon beams to irradiate a lithium battery to be detected in a packaging state so as to induce characteristic X-rays of lithium battery elements; the encoding device module is used for performing spatial modulation on the characteristic X-rays of the lithium battery element through the encoding aperture; the X-ray detector module is used for collecting the characteristic X-rays of the lithium battery elements after spatial modulation to form a penumbra imaging coding graph with a reversible coding characteristic; and the image reconstruction module is used for decoding the penumbra imaging coding graph with the reversible coding characteristic by adopting a reconstruction algorithm to obtain a lithium battery element two-dimensional space distribution image. According to the invention, lossless and high-resolution element space imaging of the lithium battery in a packaging state is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of imaging analysis, in particular to a lithium battery element distribution nondestructive imaging system and method. BACKGROUND

[0002] Lithium ion batteries, as the core components of the new generation of clean energy systems, are widely used in electric vehicles, portable electronic devices, energy storage systems and other fields. With the development of high energy density, miniaturization and modularization trends, battery thermal runaway, capacity attenuation and safety accidents frequently occur in actual operation, which exposes the serious shortage of current in-service battery internal structure and key element distribution detection means. Especially in the battery packaging state, traditional methods cannot realize nondestructive detection of internal failure mechanism, which greatly restricts the design and quality control ability of high safety and high reliability lithium batteries.

[0003] The current mainstream imaging analysis techniques include X-ray fluorescence (XRF), scanning electron microscope (SEM), transmission electron microscope (TEM), atomic force microscope (AFM), electron spin resonance (ESR) and nuclear magnetic resonance (NMR) and the like. Although these methods have been widely used in surface or local structure analysis, they generally have the following limitations: (1) insufficient penetration ability: such as XRF, synchrotron X-ray, which is difficult to penetrate the battery shell and high-density layer, and it is difficult to obtain deep information; (2) destructive to samples: such as SEM, TEM, AFM, which need to open the battery, destroy the original structure, and cannot meet the nondestructive detection requirements of in-service; (3) limited sensitivity and resolution: such as NMR and ESR, which have weak signal response to light elements (especially lithium), and cannot realize element identification under high spatial resolution; (4) difficult to realize quantitative and deep layered imaging: most of the techniques cannot realize depth-adjustable detection of element distribution, and do not have in-situ quantitative ability.

[0004] Muon induced X-ray emission (MIXE) is a new non-destructive spectroscopic analysis technique, which uses the formation of muonium atom after the combination of negative muon and atomic nucleus, and releases characteristic X-ray signals through cascade transition. It has the following unique advantages: (1) high penetration ability: the mass of muon is 207 times that of electron, and the induced X-ray energy is much higher than that of conventional XRF, which can penetrate packaging structure of more than millimeter; (2) strong element selectivity: it can accurately distinguish key elements including lithium, cobalt, nickel, manganese, etc., especially the detection sensitivity of light elements is better than that of traditional methods; (3) depth-controllable detection: different detection depths can be selected by adjusting the incident momentum of muon, which is suitable for layered imaging; (4) strong non-destructive detection ability: without opening the battery, the internal element distribution information under the packaging state can be obtained.

[0005] Previous studies have demonstrated the feasibility of MIXE in elemental analysis of lithium batteries, for example, the J-PARC experiment realized the non-destructive discrimination of metallic lithium and ionic lithium, and the GIANT system at PSI in Switzerland can achieve a concentration resolution of 1 at.%. However, the main bottleneck of current MIXE technology is that its spatial resolution generally stays at the millimeter level (5-10 mm), which cannot meet the imaging requirements of the microstructure (<10 μm) inside lithium batteries.

[0006] To overcome the above-mentioned deficiencies, the international community has begun to explore the combination of MIXE and imaging technology, such as the joint use of pinhole imaging and multi-pixel detection array; the team of Xidian University also proposed a simulation study based on MURA encoder. However, there is still a lack of an imaging scheme that combines high penetration, strong element resolution and micron-level spatial resolution, especially for in-situ non-destructive elemental imaging of lithium batteries in a packaged state.

[0007] Therefore, it is urgent to propose a new high-resolution imaging scheme based on MIXE to realize micron-scale, deep, non-destructive and high-sensitivity imaging and detection of the element distribution inside the battery, to support the material mechanism research, safety performance evaluation and manufacturing process control of lithium batteries. SUMMARY

[0008] In view of the significant deficiencies of the existing lithium battery imaging and elemental analysis technology in spatial resolution, penetration depth, element recognition ability and non-destructive in-service, the present application provides a lithium battery elemental distribution non-destructive imaging system and method.

[0009] In order to achieve the above-mentioned application purposes, the technical scheme adopted by the present application is: In a first aspect, the present application provides a lithium battery elemental distribution non-destructive imaging system, comprising: A high-energy negative muon beam source module for outputting momentum-adjustable negative muon beams to irradiate a lithium battery in a packaged state to induce characteristic X-rays of lithium battery elements; An encoder module for spatially modulating the characteristic X-rays of lithium battery elements through an encoding aperture; An X-ray detector module for collecting the spatially modulated characteristic X-rays of lithium battery elements to form a penumbra imaging code graph with reversible coding characteristics; An image reconstruction module for decoding the penumbra imaging code graph with reversible coding characteristics using a reconstruction algorithm to obtain a two-dimensional spatial distribution image of lithium battery elements.

[0010] Further, according to the design beam parameters of the high-energy negative muon beam source and the typical multi-layer structure and element distribution of the lithium battery, a Monte Carlo simulation model of the interaction between muons and the lithium battery is established; the transport process of muons with different momenta in the multi-phase material of the lithium battery, the probability of muon atom formation and the spatial distribution of X-ray yield are simulated, and the X-ray excitation efficiency of the target element is maximized while the penetration depth of the lithium battery package is controlled by optimizing the incident momentum of the muon.

[0011] Further, the encoder module is made of a spherical array encoder, a MURA quasi-random hole array or an aperture multi-element array structure made of high-Z material.

[0012] Further, the influence of the encoder material, size and object image light path arrangement on the imaging resolution and signal-to-noise ratio is simulated to quantize the modulation performance of the encoder for characteristic X-rays, and the mapping relationship between the penumbra imaging encoding graph and the element distribution is established.

[0013] Further, the reconstruction algorithm is mainly RL deconvolution iteration, supplemented by Wiener filtering, regularization and blind deconvolution, and the iteration number is optimized by the histogram difference function to make the reconstructed image close to the original image.

[0014] Further, the reconstruction algorithm expands the penumbra imaging encoding graph into a polar coordinate graph by iradon, filtered back projection and iterative reconstruction, derives along the radial direction, and integrates by iradon to reconstruct the distribution of each point on the light source plane.

[0015] Further, the reconstruction algorithm generates an image pattern by a deep convolutional neural network and uses random initialization noise as input to optimize the initial reconstructed image; the deep convolutional neural network updates the network parameters by minimizing a composite loss function including the projection domain and the image domain, and the loss function includes measurement loss, structural similarity loss and total variation loss.

[0016] Further, the high-energy negative muon beam source module induces characteristic X-rays of elements of the lithium battery in different directions by multi-angle irradiation and rotating the sample; The X-ray detector module collects characteristic X-rays of elements of the lithium battery in different directions to generate penumbra imaging encoding graphs in different directions; The image reconstruction module decodes the penumbra imaging encoding graphs in different directions by using a reconstruction algorithm to obtain a three-dimensional spatial distribution image of the elements of the lithium battery.

[0017] Further, the image reconstruction module identifies and quantitatively analyzes different elements according to X-ray spectrum information.

[0018] In a second aspect, the present application provides a non-destructive imaging method for the element distribution of a lithium battery, comprising the following steps: The characteristic X-rays of the lithium battery elements are induced by using a high-energy negative muon beam source module to irradiate the lithium battery to be tested in a packaged state with a momentum-adjustable negative muon beam; The characteristic X-rays of the lithium battery elements are spatially modulated by using an encoding device module through an encoding aperture; The characteristic X-rays of the lithium battery elements after spatial modulation are collected by using an X-ray detector module to form a penumbra imaging encoding graph with reversible encoding characteristics; The penumbra imaging encoding graph with reversible encoding characteristics is decoded by using an image reconstruction module to adopt a reconstruction algorithm to obtain a two-dimensional spatial distribution image of the lithium battery elements.

[0019] The present application has the following beneficial effects: On the basis of maintaining the basic advantages of MIXE technology such as non-destructive, strong element selectivity and depth adjustability, the present application introduces a high-resolution encoding imaging structure and an image reconstruction algorithm, and systematically breaks through the limitations of traditional MIXE and X-ray imaging methods in spatial resolution and light element sensitivity, and realizes non-destructive, high-resolution element spatial imaging of lithium batteries in a packaged state. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 It is a structural schematic diagram of a lithium battery element distribution non-destructive imaging system; Figure 2 It is a schematic diagram of simulating muon irradiation of a carbon element sphere to induce X-rays; Figure 3 It is a schematic diagram of the process of simulating muon beam irradiation of a carbon element sphere to induce X-rays; Figure 4 It is a schematic diagram of the reconstruction process using the iradon algorithm; Figure 5 It is a schematic diagram of a ball encoding imaging system and reconstruction using visible light as a light source. DETAILED DESCRIPTION

[0021] The specific embodiments of the present application are described below to facilitate understanding of the present application by those skilled in the art, but it should be clear that the present application is not limited to the scope of the specific embodiments, and for those skilled in the art, it is obvious that various changes are within the spirit and scope of the present application defined and determined by the appended claims, and all applications utilizing the concept of the present application are within the scope of protection.

[0022] The present application proposes a new type of lithium battery micron-level element distribution non-destructive imaging method and system combining muon-induced X-ray emission (MIXE) and penumbra encoding imaging, aiming to solve the following key technical problems: (1) Spatial resolution limitation: Existing MIXE imaging systems are limited by beam spot size and detector geometry, with a spatial resolution typically at the millimeter level, which is difficult to resolve the microstructure and subtle element migration behavior inside lithium batteries.

[0023] (2) Difficulty in imaging encapsulated batteries in service: Traditional techniques cannot perform deep three-dimensional element imaging without disassembling the battery, limiting its application in diagnosis and quality control under real service conditions.

[0024] (3) Low sensitivity and difficulty in quantifying light elements (such as lithium): Conventional X-ray or proton-induced X-ray techniques have difficulty in identifying low atomic number elements (such as lithium), and often require sample destruction, making it difficult to meet the needs of fine battery research.

[0025] (4) Conflict between system detection efficiency and imaging accuracy: Existing imaging systems often sacrifice signal-to-noise ratio or measurement efficiency when improving spatial resolution, and cannot meet the requirements of high resolution and high throughput imaging.

[0026] The technical effects achieved by the present application are as follows: (1) A muon-induced X-ray based spherical (or uniform redundant array) coded penumbra imaging method is provided, which breaks through the resolution bottleneck of existing MIXE systems and realizes micron-level imaging capability with a spatial resolution better than 10 μm; (2) A MIXE coded imaging system suitable for lithium battery samples in encapsulated state is constructed, which has non-invasive, in-service and fully encapsulated applicability, meeting the actual application needs; (3) High sensitivity (<1at.%) identification and quantitative imaging of elements such as lithium, lithium ion, cobalt, nickel, manganese, etc. inside lithium batteries are realized, which is suitable for battery failure analysis, material mechanism research and quality evaluation; (4) Corresponding image reconstruction algorithms (such as filtered back projection, maximum likelihood expectation maximization, deep neural network, etc.) are used to improve the interpretability and accuracy of imaging data; (5) An experimental platform and engineering scheme for lithium battery non-destructive imaging based on accelerator muon source (such as CiADS, HIAF or CSNS, etc.) are provided, promoting the practicalization and localization development of muon imaging technology in the field of energy materials.

[0027] The present application first systematically introduces efficient coded imaging and image reconstruction technology on the basis of maintaining the advantages of MIXE such as high penetration, strong element selectivity and depth adjustability, realizes the unification of high resolution, high sensitivity, strong adaptability and in-situ availability, and will provide key support means for failure prediction, safety evaluation and quality control of lithium battery design, production and service throughout the whole cycle.

[0028] Example 1 The application provides a lithium battery micron-level element distribution imaging system based on the combination of Muon Induced X-ray Emission (MIXE) and penumbra coding imaging technology, aiming to realize non-destructive and high-resolution element spatial imaging of a lithium battery in a packaged state, and comprising the following steps: a high-energy negative muon beam source module, which is used for outputting a negative muon beam with adjustable momentum to irradiate a lithium battery to be measured in a packaged state, so as to induce characteristic X-rays of elements of the lithium battery; a coding device module, which is used for spatially modulating the characteristic X-rays of elements of the lithium battery through a coding aperture; an X-ray detector module, which is used for collecting the characteristic X-rays of elements of the lithium battery after spatial modulation, so as to form a penumbra imaging coding graph with reversible coding characteristics; an image reconstruction module, which is used for decoding the penumbra imaging coding graph with reversible coding characteristics by using a reconstruction algorithm, so as to obtain a two-dimensional spatial distribution image of elements of the lithium battery.

[0029] The application combines MIXE and penumbra imaging methods, bypasses the limitation that a muon beam spot is difficult to focus to a micron scale, and realizes micron-level imaging of element distribution through X-ray modulation and image reconstruction. When a muon beam current irradiates a lithium battery sample, the muons interact with each atom inside the battery, and induce characteristic X-rays. After the X-rays are modulated through a coding aperture (such as a spherical or circular hole) made of high-Z material (such as tungsten or lead), they reach an energy-resolved pixel-type X-ray detector (such as cadmium zinc telluride CdZnTe or cadmium telluride CdTe), forming a penumbra imaging coding graph. The detector selects the coding graph of characteristic X-rays from target elements (such as lithium) according to energy, and due to the geometric magnification effect of the imaging system, micron-level structures on the battery sample can be projected to tens of microns of pixels on the detector, realizing the improvement of spatial resolution. Through reconstruction algorithms such as deconvolution, back-projection and deep learning, the coding graph is decoded to obtain a two-dimensional spatial distribution image of elements of the battery. Combined with the rotation of the battery sample and CT reconstruction technology, a three-dimensional element distribution image can be constructed.

[0030] In an optional embodiment of the application, the high-energy negative muon beam source module can rely on an existing or to-be-built high-current accelerator platform (such as CiADS, HIAF or CSNS, etc.), output a negative muon beam with adjustable momentum (such as 25-50 MeV / c), and be used for detection of different depth levels.

[0031] The measured sample targeted by the embodiment is a lithium battery or its components in a packaged state, which is directly placed in the irradiation area of the muon beam. Without opening or special pretreatment, the authenticity under in-service conditions is ensured.

[0032] This work, based on the designed beam parameters (spot size, muon momentum, and flux) of the CiADS / HIAF muon source and combined with the typical multilayer structure and element distribution of lithium-ion batteries, develops a Monte Carlo simulation model of muon-battery interactions. Geant4 simulations are used to investigate muon transport within the multiphase materials of lithium-ion batteries, the probability of muon atom formation, and the spatial distribution of deexcitation X-ray yield. Simulations optimize the muon incident energy to ensure penetration of the battery encapsulation while maximizing the X-ray yield of core elements (lithium, cobalt, nickel, manganese, iron, phosphorus, and carbon).

[0033] This example uses Geant4 to establish a Monte Carlo model of the interaction between a muon beam and a multilayer lithium battery structure. The model investigates the transport, capture mechanisms, and de-excitation X-ray yields of muons of varying momenta within the core elements (including lithium, cobalt, nickel, manganese, iron, phosphorus, and carbon) of each lithium battery material layer. The QBBC physics list is used to simulate muon nuclear capture, muon atomic de-excitation, X-ray emission, muon decay, and related electromagnetic processes within the sample. By optimizing the incident muon momentum, the excitation efficiency is maximized for elements at target depths (e.g., the electrode-electrolyte interface). The characteristic X-ray energy spectra, spatial distribution, and angular information of the MIXEs for each element are obtained.

[0034] In this example, a carbon spherical shell with an outer radius of 1 mm and a thickness of 7 μm is constructed as a quick verification of the MIXE simulation. The carbon spherical shell is irradiated with a parallel negative muon beam of 0.5 MeV / c to 2.5 MeV / c and a beam spot diameter of 2 mm. Figure 2 As shown; the upper left picture is the experimental geometric arrangement, the yellow line in the upper right picture is the negative muon beam, the green line is the generated X-ray (MIXE), and the lower left picture is the carbon spherical shell modeling, in which the three letters ABC are hollowed out for easy observation. The lower middle picture is the X-ray emission position, which clearly and correctly reflects the spatial distribution of the carbon spherical shell. The lower right picture shows the process of X-rays passing through the circular coded hole to reach the flat-panel detector. Obvious 72.55keV (μC-Kα) and 13.85keV (μC-Lα) X-ray characteristic peaks are generated, as shown Figure 3 As shown; the left picture is the energy spectrum distribution of X-rays, which shows the MIXE carbon element characteristic peak; the right picture is the position distribution of X-rays, which clearly shows the spatial distribution of carbon elements.

[0035] In an optional embodiment of the present invention, the encoding device module utilizes a spherical encoder (ball array), a Modified Uniformly Redundant Array (MURA), or an aperture multi-element array structure, made of a high-Z material (such as tungsten or lead). Its function is to modulate the X-ray signal, projecting a specific encoding pattern onto the detector, thereby improving the spatial resolution of the imaging.

[0036] The embodiment is designed based on a ball / circular hole coded aperture of high-Z material (such as tungsten alloy, lead), and the influence of the coded aperture material, size, and object-image light path arrangement on the imaging resolution and signal-to-noise ratio is studied. In combination with Geant4 simulation, the modulation performance of the coded aperture on characteristic X-rays is quantified, the mapping relationship between the coded image and the element distribution is established, and the modulation distortion caused by the multi-layer structure, signal superposition, and scattering is solved.

[0037] The embodiment adopts a spherical coded aperture array or a MURA (Modified Uniformly Redundant Array) structure made of high-Z material, modulates the muon-induced X-rays in space, and forms a penumbra pattern with reversible coding characteristics. Through the shape, arrangement mode, aperture size of the coding structure, and the design of the detector, the system has a micron-level imaging resolution.

[0038] In an optional embodiment of the application, the X-ray detector module adopts a high-energy resolution and high-pixel density detector, such as a CdTe or Ge semiconductor detector array, a HEXITEC type pixelated array, a CdTe-DSD (Double-sided Strip Detector), and the like, to collect atomic characteristic X-rays induced by muons.

[0039] The MIXE-induced X-rays adopted in the embodiment form a penumbra image, i.e., a coded image, on the detector through the coded aperture. The penumbra imaging system can effectively enhance the signal-to-noise ratio of imaging, reduce the exposure time, improve the resolution, and maintain the quality of the reconstructed image. The spatial resolution calculation method of the penumbra imaging system is as follows:

[0040] where e is the resolution of the detector, M is the magnification, p is the distance (object distance) from the sample to the coded aperture, is the wavelength of the X-rays, L is a term related to the penetration length through the edge of the coded aperture, is a measure of the roundness of the coded aperture; compared with the diffraction term, L and are very small. The wavelength corresponding to the muon-induced lithium element characteristic Ka line 18.8 keV is 2-6.6 x 10-1 m, the object distance p can be set to 0.1 m (a smaller object distance can obtain better resolution); when the magnification is M=5 (a larger M can be taken to make the first term smaller), the spatial resolution of the spherical coded imaging system h≈10.32 mm, and when M=10, the spatial resolution of the system h≈5.62 um.

[0041] The embodiment constructs a complete model of muon source-lithium battery-encoding aperture-detector in Geant4, and collects penumbra images. In Matlab, the CSV data file (containing lithium battery MIXE information) output from Geant4 is read by using the csvread function to set a matrix O, an encoding aperture matrix A (a circular aperture or a spherical aperture) is defined, and an ideal penumbra imaging encoding graph is calculated wherein represents convolution, and N is noise. The penumbra images obtained by Geant4 and Matlab can be compared to identify influencing factors (such as detector edge effect, encoding aperture roundness error, etc.), and the design of the penumbra imaging system is optimized.

[0042] In an optional embodiment of the present application, the image reconstruction module includes a high-precision data acquisition circuit and image inversion software, and uses reconstruction algorithms such as filtered back projection (FBP), maximum likelihood expectation maximization (MLEM), and deep learning network (such as U-Net and GAN) to realize three-dimensional high-resolution element distribution image reconstruction.

[0043] The embodiment develops an image reconstruction method suitable for low statistics and high noise based on the signal inversion problem of the nonlinear modulation of the penumbra imaging. Based on the reconstruction algorithms such as Richardson-Lucy (RL) iterative deconvolution, iradon reconstruction, anti-filter projection (FBP), and iterative reconstruction (IR), the blur problem of the traditional back projection method under the inclined geometry is solved. A multi-view two-dimensional imaging and three-dimensional CT reconstruction strategy is constructed to realize micron-level three-dimensional element distribution reconstruction. A deep neural network is introduced to denoise and image enhance the low statistics signal, overcome the signal-to-noise ratio limitation caused by the insufficient brightness of the muon source, and further eliminate various artifacts.

[0044] The embodiment is to perform micron-level imaging on the element distribution of the lithium battery. After the penumbra imaging encoding graph obtained on the detector after the encoding aperture modulation in the previous step is filtered according to the characteristic energy, the encoding graph of a certain specific element is obtained. The encoding graph is reconstructed by using a series of algorithms, and the element distribution of the element in the lithium battery is obtained. The three types of reconstruction methods used in the embodiment are as follows: The deconvolution method mainly uses Richardson-Lucy (RL) iteration, supplemented by Wiener, Reg, and Blnd deconvolution algorithms. The RL algorithm is particularly suitable for image reconstruction with low count rate and Poisson statistical noise. The core idea is based on maximum likelihood estimation: wherein the superscript represents that the encoding matrix is flipped on each dimension, I is the penumbra encoding graph on the detector, O is the original image estimation (preliminary reconstruction result), * represents convolution, represents point-to-point element multiplication. The histogram difference function Optimize the number of iterations to make the reconstructed image closest to the original image (lithium battery MIXE source).

[0045] Back-projection algorithms include Iradon, filtered back projection (FBP), and iterative reconstruction (IR). When a point X-ray source, after spherical encoding, reaches a pixel detector, it appears as a circle with a dark center and bright edges. This is the projection of the point source illuminating the sphere. Each point X-ray source, after spherical encoding, forms several such projections on the detector. The combined effect creates a completely dark center, bright surroundings, and a penumbra region at the edge that gradually transitions from dark to bright, forming a penumbra image (similar to the principle of penumbra imaging during solar and lunar eclipses). The penumbra image encoding diagram can be expanded into polar coordinates, radially differentiated, and integrated using Iradon to reconstruct the distribution of each point in the light source plane.

[0046] The deep learning algorithm generates image patterns through a deep convolutional neural network (CNN), uses randomly initialized noise as input, and optimizes the initial reconstructed images obtained by the two methods mentioned above. The network updates network parameters and optimizes image quality by minimizing a composite loss function that includes the projection domain (the encoding image on the detector) and the image domain (the reconstructed image). The loss function includes: measurement loss (used to penalize the difference between the reconstructed image and the projection data), structural similarity loss (used to maintain the consistency of the image structure), and total variation loss (used to suppress image noise and promote image smoothness). By optimizing these steps, the network gradually generates high-quality reconstructed images. Compared with traditional back-projection algorithms, this deep generative regularization method performs better in detail preservation and background noise suppression, and is particularly suitable for experiments with low muon beam count rates.

[0047] In this embodiment, the RL deconvolution algorithm and the Iradon filter back projection algorithm are used in Matlab to preliminarily reconstruct the simulated spherical coded penumbra image. The obtained two-dimensional distribution map is relatively ideal, as shown in FIG. Figure 4 The figure shows the reconstruction process using the iradon algorithm. The top left image shows the two-dimensional map of the carbon element MIXE characteristic X-ray divergence position, simulated in Geanl4, imported into Mallab; the top middle image shows the circular coded aperture; the top right image shows the coded map obtained after the characteristic light source passes through the circular coded aperture; the bottom left image shows the polar coordinate expansion of the coded map; the bottom middle image shows the sino map obtained by taking the radial derivative of the polar coordinate expansion map; the bottom right image shows the preliminary reconstruction result.

[0048] This embodiment uses a visible light source to build a ball coding penumbra imaging system, and the obtained coding image is reconstructed by iradon algorithm, and sub-millimeter spatial resolution (the distance between three point visible light sources is less than 2mm) is obtained as shown in Figure 5. Although the reconstruction result has artifacts and noise, the distribution characteristics (three point light sources) of the visible light source can still be clearly distinguished. The middle figure is a ball coding imaging system, including a characteristic visible light source, a coding ball and a visible light CCD detector (diagonal line 43.2mm). The left figure is the size of the visible light source (<2mm). The right figure is the reconstruction process, and it can be clearly seen that the characteristics of the three point light source arrangement are successfully reconstructed.

[0049] The working principle of the above-mentioned lithium battery micron-level element distribution non-destructive imaging system based on muon-induced X-ray and penumbra coding imaging is as follows: The muon beam is decelerated after entering the sample and is captured by the atomic nucleus to form a muonic atom.

[0050] The muonic atom undergoes cascade transition and emits high-energy X-rays, which carry characteristic information of the corresponding element.

[0051] The emitted X-ray signal passes through the coding array to generate a penumbra pattern and forms a modulated two-dimensional image on the detector.

[0052] Multi-angle irradiation and rotating sample to obtain projection patterns in different directions.

[0053] Combined with the coding template, the two-dimensional or three-dimensional element distribution image of the target area is reconstructed by the inversion algorithm.

[0054] According to the X-ray energy spectrum information, different elements are identified and quantitatively analyzed.

[0055] The key parameters are designed as follows: Spatial resolution target: better than 10 μm (depending on the coding aperture size, object distance image distance ratio and decoding algorithm accuracy); Element sensitivity: ≤1 at.%; Detection depth: adjustable range up to hundreds of microns to several millimeters (with muon momentum change); Energy spectrum resolution: better than 1 keV (@ 50 keV); Data acquisition time: no more than 30 minutes for each angle, and the whole data acquisition time is ≤6 hours (suitable for beam experiment window); Applicable object: commercial lithium battery, sodium battery, solid-state battery, soft package battery, button cell, etc.

[0056] The present application has good adaptability and can be deployed on platforms including but not limited to the following: China Spallation Neutron Source (CSNS) MELODY beam line; Accelerator driven transmutation research device (CiADS) plan muon beam line; The negative muon beam station to be arranged in a heavy ion accelerator facility (HIAF); The muon source experiment platform of Shanghai SHINE, SULF light source and the like laser target.

[0057] When the system is deployed on the platform, the encoding device, the detection array and the sample support can be modularly connected, and the system is suitable for standard experiment cabin structure.

[0058] The lithium battery microelement distribution nondestructive imaging system based on muon induced X-ray and penumbra coding imaging provided by the application, on the basis of maintaining the basic advantages of MIXE technology such as nondestructive, strong element selectivity and depth adjustment, systematically breaks through the limitations of traditional MIXE and X-ray imaging methods in spatial resolution and light element sensitivity by introducing a high-resolution coding imaging structure and an image reconstruction algorithm. The key innovation points and corresponding technical effects are as follows: 1. Introducing a penumbra coding imaging structure to realize microscale spatial resolution The traditional MIXE system has a larger muon beam spot size and an uncontrolled X-ray propagation path, and the spatial resolution is generally in the millimeter level, so it is difficult to be used for battery internal microstructure characterization. The spherical array encoder (spherical array coding imaging) made of high-Z material or the MURA quasi-random hole array is adopted in the application, the spatial modulation of muon induced X-rays is realized, and the penumbra pattern with reversible characteristics is formed on the detector.

[0059] The spatial decoding algorithm (such as back projection or maximum likelihood) of the coding structure is combined, and the microscale target structure imaging result can be effectively obtained. Simulation and literature data show that the spatial resolution of the system can be better than 10 μm, which is much better than the traditional MIXE system (> 5 mm) and the synchrotron XRF (> 30 μm) imaging method.

[0060] The application first realizes the spatial imaging capability of the MIXE technology at the micron level, and meets the failure analysis requirements of the microstructure of the lithium battery electrode.

[0061] 2. In-situ nondestructive imaging in a packaged state without opening Traditional element analysis technologies, such as SEM, TEM, SIMS and the like, need to open or cut the sample, which destroys the original structure and makes it difficult to truly reflect the element migration and deposition behavior in a packaged state. The strong penetration ability (several millimeters to centimeters) of muons and high-energy X-rays is fully utilized, and the packaged lithium battery can be directly imaged.

[0062] In the system of the present invention, the sample to be tested does not need to be disassembled and can be directly placed on the muon beam irradiation path. By adjusting the incident muon momentum, tomographic detection at different depths can be achieved, and the multi-level element distribution inside the battery can be obtained in combination with the image inversion algorithm.

[0063] The present invention realizes in-situ, non-destructive microstructure imaging of lithium batteries under service conditions, providing a new means for life research, aging mechanism analysis and manufacturing quality monitoring.

[0064] 3. It has the ability to identify multiple elements, especially high sensitivity to light elements MIXE utilizes the characteristic high-energy X-rays (>100 keV) emitted by muon atomic transitions, significantly reducing the sample's self-absorption of radiation. In particular, because muons are much more massive than electrons, the X-ray energies corresponding to muon atomic energy transitions are much higher than those of conventional XRF, making it effective for detecting low-atomic-number elements such as lithium (Z=3), oxygen (Z=8), and sodium (Z=11).

[0065] Compared with traditional X-ray fluorescence imaging, this invention has significant advantages in detection sensitivity and signal-to-noise ratio of light elements. Experimental results show that it can achieve a concentration resolution of <1 at.% and distinguish changes in the oxidation state of lithium.

[0066] The present invention improves the imaging resolution of light elements (such as metallic lithium vs. ionic lithium), which is helpful for studying microscopic failure processes such as lithium dendrite formation and electrolyte decomposition products.

[0067] 4. Adjustable depth detection, supporting 3D element tomography reconstruction By carefully adjusting the incident muon momentum (e.g., 25–45 MeV / c), the present invention precisely controls the muon retention depth within the sample, creating a multi-layered X-ray emission source. Combining sample rotation with a multi-angle projection acquisition mechanism, a three-dimensional elemental distribution map within the sample can be reconstructed.

[0068] The present invention adopts a composite strategy of "depth adjustment + spatial decoding + energy spectrum recognition", which not only improves the imaging depth but also realizes three-dimensional tomography.

[0069] The present invention realizes non-destructive visualization of the three-dimensional element structure inside lithium batteries, providing a new tool for multi-layer electrode failure analysis and interface evolution research.

[0070] 5. Modular architecture, suitable for various muon source platforms and battery types The system design offers excellent platform compatibility and can be deployed at existing or under-construction muon sources, including CiADS, HIAF, CSNS, J-PARC, and PSI. System components (encoders, brackets, and detectors) are standardized modules for easy assembly and maintenance, and they accommodate various standard cell sizes (cylindrical, soft-pack, and button-type).

[0071] The application realizes rapid transfer of a technical platform and meets the needs of multiple scenes such as scientific research, detection, quality control and the like.

[0072] Embodiment 2 The application embodiment based on embodiment 1 proposes a muon-induced X-ray and penumbra coding imaging-based nondestructive imaging method for micron-level element distribution of a lithium battery, comprising the following steps: A high-energy negative muon beam source module is used to irradiate a lithium battery to be measured in a packaged state with a momentum-adjustable negative muon beam to induce characteristic X-rays of elements of the lithium battery; An encoder device module is used to modulate the characteristic X-rays of elements of the lithium battery in space through a coding aperture; An X-ray detector module is used to collect the characteristic X-rays of elements of the lithium battery after spatial modulation, and form a penumbra imaging coding graph with reversible coding characteristics; An image reconstruction module is used to decode the penumbra imaging coding graph with reversible coding characteristics by using a reconstruction algorithm, and obtain a two-dimensional spatial distribution image of elements of the lithium battery.

[0073] The application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0074] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including instruction devices, which implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0075] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide a product for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocksFigure 1 the functions specified in the one or more blocks.

[0076] The principles and implementations of the present application are described in the specific examples. The above examples are used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation and application range will be changed, and the above description should not be understood as a limitation of the present application.

[0077] Those skilled in the art will realize that the examples described herein are for the purpose of aiding the reader in understanding the principles of the present application and should be understood as not limiting the scope of the present application to such specific statements and examples. Those skilled in the art can make various other specific modifications and combinations according to the technical inspiration disclosed in the present application without departing from the essence of the present application, and these modifications and combinations are still within the scope of protection of the present application.

Claims

1. A non-destructive imaging system for elemental distribution of lithium batteries, characterized in that, The application relates to a high-energy negative muon beam source module for outputting a negative muon beam with adjustable momentum to irradiate a lithium battery in a packaging state to induce characteristic X-rays of lithium battery elements; an encoding device module for spatially modulating the characteristic X-rays of the lithium battery elements through an encoding aperture; an X-ray detector module for collecting the spatially modulated characteristic X-rays of the lithium battery elements to form a penumbra imaging encoding graph with reversible encoding characteristics; and an image reconstruction module for decoding the penumbra imaging encoding graph with reversible encoding characteristics by using a reconstruction algorithm to obtain a two-dimensional spatial distribution image of the lithium battery elements. According to the design beam current parameters of the high-energy negative muon beam source and the typical multi-layer structure and element distribution of the lithium battery, a Monte Carlo simulation model of the interaction between muons and the lithium battery is established; the transport process of muons with different momenta in the multi-phase material of the lithium battery, the spatial distribution law of the formation probability of muon atoms and the X-ray yield of de-excitation are simulated; and the incident momentum of the muons is optimized to maximize the X-ray excitation efficiency of the target elements while controlling the penetration depth of the lithium battery packaging. The encoding device module adopts a spherical array encoder made of high-Z material, a MURA quasi-random aperture array or an aperture multi-element array structure. The influence of the encoding device material, size and object image light path arrangement on the imaging resolution and signal-to-noise ratio is simulated to quantize the modulation performance of the encoding device on the characteristic X-rays and establish the mapping relationship between the penumbra imaging encoding graph and the element distribution. The reconstruction algorithm mainly uses RL deconvolution iteration, supplemented by Wiener filtering, regularization and blind deconvolution, and the iteration number is optimized by using a histogram difference function to make the reconstructed image close to the original image.

2. The system for non-destructive imaging of elemental distribution in lithium batteries of claim 1, wherein, The reconstruction algorithm expands the penumbra imaging encoding graph into a polar coordinate graph through iradon, filtered back projection and iterative reconstruction, derives along the radial direction and performs iradon integration, so that the distribution of each point on the light source plane can be reconstructed.

3. The system of claim 1, wherein the system is configured to perform non-destructive imaging of a lithium battery element distribution. The reconstruction algorithm generates an image pattern through a deep convolutional neural network and uses random initialization noise as input to optimize the initial reconstructed image; the deep convolutional neural network updates the network parameters by minimizing a composite loss function including the projection domain and the image domain, and the loss function includes a measurement loss, a structural similarity loss and a total variation loss.

4. The system of claim 1, wherein the system is configured to perform non-invasive imaging of the distribution of lithium in the lithium battery element. The high-energy negative muon beam source module induces characteristic X-rays of lithium battery elements in different directions through multi-angle irradiation and sample rotation; 5. The system of claim 1, wherein, The X-ray detector module collects the characteristic X-rays of the lithium battery elements in different directions to generate penumbra imaging encoding graphs in different directions; 6. The system of claim 1, wherein the system is a non-destructive imaging system for elemental distribution of lithium batteries. The image reconstruction module decodes the penumbra imaging encoding graphs in different directions by using a reconstruction algorithm to obtain a three-dimensional spatial distribution image of the lithium battery elements.

7. A system for non-destructive imaging of the distribution of elements in a lithium battery according to claim 5 or 6, characterized in that, The image reconstruction module identifies and quantitatively analyzes different elements according to X-ray spectrum information.

8. The system of claim 1, wherein the system is a non-destructive imaging system for elemental distribution of lithium batteries. The application relates to a high-energy negative muon beam source module for outputting a negative muon beam with adjustable momentum to irradiate a lithium battery in a packaging state to induce characteristic X-rays of lithium battery elements; an encoding device module for spatially modulating the characteristic X-rays of the lithium battery elements through an encoding aperture; an X-ray detector module for collecting the spatially modulated characteristic X-rays of the lithium battery elements to form a penumbra imaging encoding graph with reversible encoding characteristics; and an image reconstruction module for decoding the penumbra imaging encoding graph with reversible encoding characteristics by using a reconstruction algorithm to obtain a two-dimensional spatial distribution image of the lithium battery elements. The application relates to a high-energy negative muon beam source module for outputting a negative muon beam with adjustable momentum to irradiate a lithium battery in a packaging state to induce characteristic X-rays of lithium battery elements; an encoding device module for spatially modulating the characteristic X-rays of the lithium battery elements through an encoding aperture; an X-ray detector module for collecting the spatially modulated characteristic X-rays of the lithium battery elements to form a penumbra imaging encoding graph with reversible encoding characteristics; and an image reconstruction module for decoding the penumbra imaging encoding graph with reversible encoding characteristics by using a reconstruction algorithm to obtain a two-dimensional spatial distribution image of the lithium battery elements. ​ 9. The system of claim 1, wherein the system is a non-destructive imaging system for elemental distribution of lithium batteries. ​ 10. A method for non-destructive imaging of element distribution in lithium batteries, characterized in that: ​ ​ ​ ​ The image reconstruction module adopts a reconstruction algorithm to decode the penumbra imaging code map with reversible coding characteristics to obtain a two-dimensional spatial distribution image of lithium battery elements.

Images