Positive electrode material reverse design method, device, equipment, medium and vehicle

Abstract

The invention provides a positive electrode material reverse design method, device and equipment, a medium and a vehicle, and the method comprises the steps: responding to a reverse design request for a lithium-rich manganese-based positive electrode material, and carrying out the high-throughput calculation of a plurality of candidate element doping combinations based on a target performance parameter through a machine learning potential function, determining an optimal element formula and atomic scale key parameters meeting target performance parameter constraints; based on the optimal element formula and the atomic scale key parameters, performing phase field simulation by adopting an anisotropic lattice strain model containing Jahn-Teller distortion correction, and determining bulk phase gradient structure features and surface coating layer configuration parameters; and based on the bulk phase gradient structure characteristics and the surface coating layer configuration parameters, calling a mapping model of a process and a structure stored in a preset knowledge graph to perform process inversion, and determining a candidate coating material system and a corresponding synthesis process path thereof, so that the success efficiency and research and development capability of the positive electrode material design are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of battery material preparation and optimization technology, and in particular to a method, apparatus, equipment, medium, and vehicle for reverse design of cathode materials. Background Technology

[0002] Today, with the urgent need for high-energy-density, low-cost lithium-ion batteries in electric vehicles and large-scale energy storage, lithium-rich manganese-based cathode materials have attracted much attention due to their theoretical specific capacity exceeding that of traditional layered materials. The high capacity of these materials is partly due to reversible redox reactions involving anions, and they are considered one of the key candidates for achieving breakthroughs in next-generation battery technology.

[0003] Currently, the development of lithium-rich manganese-based cathode materials mainly relies on trial-and-error experiments. This involves adjusting process parameters such as element types and sintering temperatures, combined with surface coating methods, to explore performance improvements. However, this trial-and-error development approach has two main problems: first, it depends on experience, resulting in long development cycles and high costs; second, the disconnect between material composition, structural design, and fabrication processes leads to significant deviations between the final product and the design goals, making it difficult to meet high performance requirements. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, medium, and vehicle for reverse design of cathode materials. By integrating machine learning computation, physical simulation, and process knowledge base, it transforms the traditional experience-based trial-and-error process into an engineering process driven by precise performance parameters and optimized across multiple scales, thereby improving the success rate and R&D capabilities of material design.

[0005] In a first aspect, embodiments of this application provide a reverse design method for cathode materials, the method comprising:

[0006] In response to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, high-throughput calculations are performed on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the constraints of the target performance parameters.

[0007] Based on the optimal elemental formulation and the key atomic-scale parameters, an anisotropic lattice strain model including Jahn-Teller distortion correction was used to perform phase-field simulation to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscale.

[0008] Based on the bulk gradient structure characteristics and the surface coating layer configuration parameters, the process inversion is performed by calling the process-structure mapping model stored in the preset knowledge graph to determine the executable process scheme at the macro scale; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths.

[0009] Secondly, embodiments of this application also provide a reverse design apparatus for positive electrode materials, the apparatus comprising:

[0010] The atomic parameter determination module is used to respond to a reverse design request for lithium-rich manganese-based cathode materials. Based on the target performance parameters specified in the reverse design request, it uses a pre-trained machine learning potential function to perform high-throughput calculations on multiple candidate element doping combinations to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the target performance parameter constraints.

[0011] The mesoscopic parameter determination module is used to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscopic scale by performing phase field simulation using an anisotropic lattice strain model including Jahn-Teller distortion correction based on the optimal element formulation and the atomic-scale key parameters.

[0012] The macroscopic scheme determination module is used to determine the executable process scheme at the macroscopic scale by calling the process-structure mapping model stored in the preset knowledge graph based on the bulk gradient structure characteristics and the surface coating layer configuration parameters; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths.

[0013] Thirdly, embodiments of this application also provide an electronic device, which includes:

[0014] One or more processors;

[0015] Storage device for storing one or more programs.

[0016] When one or more programs are executed by one or more processors, the one or more processors implement a reverse design method for cathode materials as described in any of the embodiments of this application.

[0017] Fourthly, embodiments of this application also provide a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a reverse design method for a cathode material as described in any of the embodiments of this application.

[0018] Fifthly, embodiments of this application also provide a vehicle, the vehicle including a power battery, wherein the positive electrode of the power battery is a lithium-rich manganese-based positive electrode material designed and prepared using the reverse design method of the positive electrode material of embodiments of this application.

[0019] This application provides a reverse design method for cathode materials. The method includes: responding to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, using a pre-trained machine learning potential function to perform high-throughput calculations on multiple candidate element doping combinations to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the target performance parameter constraints; then, based on the optimal element formulation and atomic-scale key parameters, using an anisotropic lattice strain model including Jahn-Teller distortion correction to perform phase-field simulation to determine the bulk phase gradient structure characteristics and surface coating layer configuration parameters at the mesoscale; and then, based on the bulk phase gradient structure characteristics and surface coating layer configuration parameters, calling a process-structure mapping model stored in a preset knowledge graph to perform process inversion and determine an executable process scheme at the macroscale; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths. The technical solution provided in this embodiment constructs a reverse design system that takes the final performance target as the starting point and drives the microscopic design of materials and the formulation of macroscopic processes. This method integrates machine learning calculation, physical simulation and process knowledge base to transform the traditional trial-and-error process that relies on experience into an engineering process that is precisely driven by performance parameters and optimized across multiple scales. This results in the efficient and accurate output of manufacturing solutions that match the design blueprint, thereby improving the success rate of material design and R&D capabilities. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the exemplary embodiments of this application, the accompanying drawings used in describing the embodiments are briefly introduced below. Obviously, the accompanying drawings described are only a portion of the embodiments to be described in this application, and not all of them. For those skilled in the art, other drawings can be obtained from these drawings without any creative effort.

[0021] Figure 1 A schematic flowchart illustrating a reverse design method for cathode materials provided in an embodiment of this application;

[0022] Figure 2 A flowchart illustrating another reverse design method for cathode materials provided in an embodiment of this application;

[0023] Figure 3 A flowchart illustrating another reverse design method for cathode materials provided in this application embodiment;

[0024] Figure 4 This is a logic diagram of the reverse design method for the cathode material involved in this embodiment;

[0025] Figure 5This is a schematic diagram of a reverse design device for a positive electrode material provided in an embodiment of this application;

[0026] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0027] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present application, not the entire structure.

[0028] Before introducing the technical solutions provided in the embodiments of this application, the application scenarios of the solutions can be described first. This embodiment is applicable to various scenarios that require efficient and accurate reverse engineering of lithium-rich manganese-based cathode materials. Currently, although empirical trial-and-error material development methods are widely used in materials exploration, traditional methods have obvious limitations. In practical applications, due to the complex characteristics of lithium-rich manganese-based material systems, such as irreversible anion oxidation, structural stress accumulation, and interfacial side reactions, and the need to consider multi-scale coupling issues such as elemental doping, structural design, and process implementation during the research and development process, traditional trial-and-error methods are difficult to achieve optimal performance design at the system level, which can easily lead to problems such as long research and development cycles, high costs, and large deviations between product performance and design goals. Therefore, there is an urgent need for a design method that can coordinate atomic-scale formulation, mesoscopic structural features, and macroscopic process paths, and comprehensively consider multi-physics coupling and complex performance constraints, in order to improve the accuracy and manufacturability of material design.

[0029] Example 1

[0030] Figure 1 This is a flowchart illustrating a reverse design method for cathode materials provided in an embodiment of this application. This embodiment is applicable to various situations requiring efficient and accurate reverse design of lithium-rich manganese-based cathode materials. The method can be executed by a cathode material reverse design device, which can be implemented in the form of software and / or hardware. The hardware can be a controller, such as a mobile terminal, PC, or server.

[0031] like Figure 1 As shown, the reverse design method for cathode materials provided in this embodiment of the invention includes the following steps:

[0032] S110. In response to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, high-throughput calculations are performed on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation that satisfies the target performance parameter constraints and the corresponding atomic-scale key parameters.

[0033] Among them, lithium-rich manganese-based cathode materials refer to layered transition metal oxide materials that are used in electrochemical energy storage devices, with manganese as the main structural element and containing an excess of lithium in stoichiometry.

[0034] In this context, a reverse design request refers to a top-down materials design task instruction oriented towards target performance. This request aims to trigger an automated design and calculation process that starts from macroscopic objectives and reverse-engineers scale by scale to the atomic composition and microstructure of the material.

[0035] The target performance parameters are the specific performance indicators and numerical constraints that are desired to be achieved by the final material, as specified in the reverse engineering request. The machine learning potential function is a mathematical function trained using machine learning methods that can approximately describe interatomic interactions. Candidate element doping combinations refer to the set of possible schemes for introducing one or more foreign elements into the base composition to modulate the material's properties.

[0036] The optimal element formulation refers to the specific types and proportions of elements that best meet the constraints of the target performance parameters, obtained through calculation and screening. Atomic-scale key parameters refer to the core physical quantities that characterize the intrinsic properties of materials at the atomic level, and their values ​​are determined by the optimal element formulation.

[0037] Specifically, when a reverse design request is received targeting lithium-rich manganese-based cathode materials with specific performance indicators, a pre-trained machine learning potential function can be invoked. Without relying on experimental trial and error, atomic-level precision energy and property calculations can be performed on multiple candidate element doping combinations. By comparing the matching degree between the calculation results and the target performance parameters, the element doping formula that simultaneously meets all performance constraints and has the best overall performance is selected as the optimal element formula, and the atomic-scale key parameters corresponding to the formula are output simultaneously.

[0038] In this embodiment, optionally, the specific implementation steps of using a pre-trained machine learning potential function to perform high-throughput calculations on multiple candidate element doping combinations to determine the optimal element formulation that satisfies the target performance parameter constraints and the corresponding atomic-scale key parameters may include:

[0039] (1) In response to a reverse design request for lithium-rich manganese-based cathode material, receive a preset performance parameter threshold from the target performance parameters.

[0040] The preset performance parameter threshold refers to the numerical boundary conditions set in advance for the target performance parameter and used for screening.

[0041] Specifically, when a reverse design instruction targeting lithium-rich manganese-based cathode material is detected, a pre-set performance parameter threshold can be extracted from the target performance parameters carried by the instruction. This threshold is used as a rigid boundary condition that must be met and not exceeded in all subsequent calculations, screening, and optimization processes. This ensures that any candidate element doping combination and its corresponding atomic-scale key parameters are considered valid solutions and proceed to the next mesoscale simulation only when they are simultaneously not lower than or do not exceed the pre-set performance parameter threshold.

[0042] (2) Based on the pre-set doping element library, determine multiple candidate element doping combinations to be screened.

[0043] The dopant element library refers to a pre-set set of foreign elements that can be selected to adjust the properties of materials.

[0044] Specifically, all available dopant elements and their concentration ranges can be read from the dopant element library that has been pre-installed in the computing environment. These elements are then arranged on the main lattice sites of the lithium-rich manganese-based cathode material according to predetermined combination rules, generating multiple candidate element doping combinations that cover the entire chemical space and are mutually exclusive. This forms a complete candidate list required for subsequent high-throughput simulation calculations, ensuring that no potentially feasible formulations are overlooked.

[0045] (3) Using a pre-trained machine learning potential function, perform high-throughput simulation calculations on multiple candidate element doping combinations to predict at least one key property parameter corresponding to each candidate element doping combination.

[0046] Among them, at least one key property parameter refers to one or more core physical or chemical property indicators used to evaluate the performance of candidate materials, which are predicted by machine learning potential function simulation calculations at the atomic scale.

[0047] Specifically, a machine learning potential function that has been trained on massive amounts of known material data and can efficiently approximate interatomic interactions can be used to perform rapid and automated batch simulations of a large number of different element doping schemes, thereby numerically predicting one or more core physical or chemical properties that each candidate element doping combination may possess.

[0048] (4) Compare the predicted key property parameter with the preset performance parameter threshold, select the candidate element doping combination that meets all constraints as the optimal element formulation, and output the atomic-scale key parameters corresponding to the optimal element formulation.

[0049] In this embodiment, at least one key property parameter corresponding to each candidate element doping combination obtained through the calculation model can be compared and judged with a preset performance parameter threshold. The candidate element doping combination in which all predicted performance indicators meet or exceed the preset standard can be selected and determined as the final selected optimal element formula. At the same time, a series of atomic-scale physical quantities corresponding to the optimal element formula are provided as atomic-scale key parameters.

[0050] S120. Based on the optimal elemental formulation and key atomic-scale parameters, an anisotropic lattice strain model including Jahn-Teller distortion correction was used to perform phase-field simulation to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscale.

[0051] Among them, the Jahn-Teller distortion-corrected anisotropic lattice strain model is a physical model used to calculate and describe the asymmetric geometric distortion of the lattice under a specific electronic state, and the resulting direction-dependent internal stress. This model takes into account the contribution of lattice strain anisotropy caused by the Jahn-Teller effect when simulating the evolution of the microstructure of materials.

[0052] Among them, the bulk gradient structure characteristics refer to the set of key parameters that are used to quantitatively describe the non-uniform and continuously changing microscopic properties of material particles, such as crystal structure type, atomic arrangement order, or chemical composition, distributed along the spatial position from the internal core to the external surface.

[0053] Among them, the surface coating configuration parameters refer to the key specifications used to define the physical and chemical properties of the functional coating applied to the surface of material particles, including but not limited to the core design information such as the material composition, thickness, structural morphology and interface characteristics of the coating with the matrix.

[0054] In this embodiment, the selected optimal elemental formulation and its corresponding atomic-scale key parameters are used as input. The Jahn-Teller distortion-corrected anisotropic lattice strain model is adopted. By solving the free energy evolution equation that considers the coupling of chemical, electrochemical, mechanical and defect fields, the spatiotemporal evolution process of lattice distortion, phase separation, defect rearrangement and stress release is reproduced at the mesoscale. This accurately captures the composition and structural gradient formed by the difference in element distribution in different regions inside the grain. At the same time, the lattice matching, stress transmission and interface orientation relationship between the surface coating layer and the bulk phase are analyzed. Finally, the bulk phase gradient structural characteristics and surface coating layer configuration parameters are output.

[0055] S130. Based on the bulk gradient structure characteristics and surface coating layer configuration parameters, the process-structure mapping model stored in the preset knowledge graph is called to perform process inversion and determine the executable process scheme at the macro scale.

[0056] The pre-built knowledge graph refers to a large database and reasoning model that is pre-constructed and stored before execution, containing structured information on various entities in the field of materials science, such as elements, processes, structures, properties, and their complex relationships. The mapping model between processes and structures refers to a computational or relational model stored in the pre-built knowledge graph that describes the quantitative or qualitative correspondence between material synthesis or processing parameters and the resulting microstructural features.

[0057] Here, an executable process scheme refers to one or more verifiable and complete manufacturing process plans proposed to prepare materials that meet design requirements. An executable process scheme includes candidate coating material systems and their corresponding synthesis process routes. A candidate coating material system refers to one or more specific substances and combinations thereof that can be selected for forming a functional layer on the surface of material particles. A synthesis process route refers to the specific manufacturing methods, steps, and key control parameter sequences for achieving a specific configuration of the candidate coating material system on the particle surface.

[0058] Specifically, the obtained bulk gradient structure features and surface coating layer configuration parameters can be used as query conditions. The mapping model between process and structure can be activated in a pre-built and continuously updated knowledge graph. Through semantic association and weighted path search between graph nodes, all macroscopic process variable combinations required to reproduce the target mesoscopic structure can be derived in reverse. Then, a macroscopic executable process scheme covering candidate coating material systems and their corresponding synthesis process paths can be output.

[0059] This application provides a reverse design method for cathode materials. The method includes: responding to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, using a pre-trained machine learning potential function to perform high-throughput calculations on multiple candidate element doping combinations to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the target performance parameter constraints; then, based on the optimal element formulation and atomic-scale key parameters, using an anisotropic lattice strain model including Jahn-Teller distortion correction to perform phase-field simulation to determine the bulk phase gradient structure characteristics and surface coating layer configuration parameters at the mesoscale; and then, based on the bulk phase gradient structure characteristics and surface coating layer configuration parameters, calling a process-structure mapping model stored in a preset knowledge graph to perform process inversion and determine an executable process scheme at the macroscale; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths. The technical solution provided in this embodiment constructs a reverse design system that takes the final performance target as the starting point and drives the microscopic design of materials and the formulation of macroscopic processes. This method integrates machine learning calculation, physical simulation and process knowledge base to transform the traditional trial-and-error process that relies on experience into an engineering process that is precisely driven by performance parameters and optimized across multiple scales. This results in the efficient and accurate output of manufacturing solutions that match the design blueprint, thereby improving the success rate of material design and R&D capabilities.

[0060] Based on the above embodiments, optionally, after obtaining the executable process scheme, the executable process scheme can be optimized through automated experimental verification and closed-loop optimization to determine the optimal process scheme. The specific implementation method may include the following steps:

[0061] (1) Input the executable process scheme into the high-throughput parallel reactor for automated experimental verification and obtain the actual performance data obtained from the experiment.

[0062] High-throughput parallel reaction apparatus refers to an integrated experimental equipment system capable of simultaneously and independently executing multiple material synthesis or processing experiments to achieve rapid, batch preparation and testing. Actual performance data refers to one or more measurement results obtained directly from physical testing or chemical characterization of material samples prepared according to the executable process scheme, which objectively reflect the true performance of the material.

[0063] Specifically, the executable process can be broken down into independent parallel experimental tasks and deployed to a high-throughput parallel reaction device with multiple independent reaction chambers for synchronous execution. During the experimental verification process, in-situ monitoring technology is used to collect interfacial reaction data and element valence state change data in real time, thereby obtaining actual performance data based on the material characterization data after the experiment.

[0064] (2) Based on the deviation between the actual performance data and the target performance parameters, generate process parameter optimization suggestions.

[0065] Among them, process parameter optimization suggestions refer to guiding instructions for making specific adjustments to one or more control parameters in the currently executable process scheme.

[0066] Specifically, the actual performance data can be automatically compared with the target performance parameters specified in the reverse design request to calculate the performance deviation. Then, based on the performance deviation and combined with the process parameter-performance association rules stored in the dynamic knowledge graph, quantitative adjustment suggestions for one or more process parameters in the executable process plan can be generated.

[0067] (3) Based on the optimization suggestions of process parameters, the executable process scheme is iteratively optimized until the optimal process scheme that meets the preset convergence conditions is determined.

[0068] The preset convergence condition refers to the performance achievement standard or upper limit of the number of iterations that is pre-set during the iterative optimization process to determine whether to terminate the optimization loop. The optimal process scheme refers to the final executable process scheme determined after multiple rounds of iterative optimization, in which the actual performance data of the corresponding material has met the preset convergence condition.

[0069] Specifically, based on the process parameter optimization suggestions, the process parameters in the executable process scheme are updated to form an optimized executable process scheme. Then, the optimized executable process scheme is re-input into the high-throughput parallel reactor for a new round of automated experimental verification. On this basis, the update and verification steps are repeated until the actual performance data corresponding to the optimized executable process scheme meets the preset convergence conditions, and the scheme that meets the conditions is determined as the optimal process scheme.

[0070] Based on the above embodiments, the executable process scheme, actual performance data, performance deviation and corresponding process parameter optimization suggestions obtained from each experimental verification can be stored back into the dynamic knowledge graph as new sample data, which is used to update the mapping model between the process and the structure and the process parameter-performance association rules.

[0071] Example 2

[0072] Figure 2 This is a schematic diagram of a reverse design method for a cathode material provided in this application embodiment. Based on the foregoing embodiments, this embodiment provides a more detailed description of step S120. For specific implementation details, please refer to the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here.

[0073] like Figure 2As shown, the method specifically includes the following steps:

[0074] S210. In response to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, high-throughput calculations are performed on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation that satisfies the target performance parameter constraints and the corresponding atomic-scale key parameters.

[0075] S220, based on the optimal elemental formulation and key atomic-scale parameters, constructs an anisotropic strain energy model that includes the Jahn-Teller lattice distortion effect.

[0076] In this embodiment, a physical calculation model can be established based on the optimal elemental composition determined at the atomic scale and its corresponding key atomic scale parameters. This model is specifically used to describe and calculate the direction-dependent lattice strain and its corresponding energy changes caused by the Jahn-Teller effect, namely, the anisotropic strain energy model.

[0077] S230. Phase field simulation is performed using an anisotropic strain energy model to minimize the lattice distortion energy of lithium-rich manganese-based cathode material during electrochemical cycling. Based on the phase field simulation results, the continuous distribution characteristics of the order degree of the bulk gradient structure in the direction from the core to the shell are determined, and the bulk gradient structure characteristics are obtained.

[0078] Lattice distortion energy refers to the elastic potential energy stored within the crystal lattice when a material's crystal structure undergoes geometric deformation that deviates from ideal symmetry. Phase-field simulation results refer to the spatiotemporal distribution dataset of field variables (such as order parameters, strain field, and composition field) obtained through numerical simulation calculations of the microstructural evolution of materials using the phase-field method. Continuous distribution characteristics of order degree refer to key parameters used to quantitatively describe the uninterrupted, gradual spatial variation of the degree of regularity of atomic arrangement (i.e., order degree) from the core to the surface in a material.

[0079] Specifically, a phase field numerical calculation can be performed using a model that describes anisotropic strain energy. The goal is to reduce the elastic energy accumulated by the material due to structural distortion during charging and discharging. From the microstructure evolution data obtained by the simulation calculation, a set of key parameters can be extracted to show the continuous and gradual change in the regularity of the atomic arrangement from the center to the edge of the particle, thereby obtaining a specific description of the gradient structure.

[0080] In this embodiment, optionally, the bulk gradient structure features include: a core region being an ordered layered structure with a first degree of order, an outer shell region being a disordered spinel structure with a second degree of order, and a continuous transition region between the core and the outer shell, wherein the first degree of order is greater than the second degree of order. This can be understood as follows: in the bulk gradient structure features, the core region maintains a highly regular crystallographic arrangement, has very few defects in the layered sequence, and the cation occupation order is in a high numerical range; the outer shell region, due to lattice distortion and cation mixing, loses long-range order in the layered sequence, exhibits localized spinel-type stacking, and the degree of order is in a low numerical range; the continuous transition region is located between the two and gradually decreases in order along the radial direction, so that the overall structure forms a gradient distribution with a seamless transition from high to low degree of order in the direction from the core to the outer shell.

[0081] S240. In phase-field simulation, the generation and diffusion process of oxygen vacancies are coupled. Based on the comparison between the oxygen loss rate obtained from the simulation and the preset voltage decay target value, the minimum oxygen diffusion blocking rate required to achieve oxygen loss control is determined.

[0082] Oxygen loss rate refers to the average rate or proportion of irreversible oxygen loss in the crystal lattice of lithium-rich manganese-based cathode materials during electrochemical cycling. Voltage decay target value refers to the pre-set upper limit, intended to control battery performance degradation, the maximum allowable decrease in the average discharge voltage of the cathode material within a specific cycle period. Minimum oxygen diffusion blocking rate refers to the minimum suppression efficiency against outward migration of oxygen atoms required by the functional layer applied to the material surface, derived from simulation analysis of the oxygen loss process, to achieve the aforementioned voltage decay target value.

[0083] In this embodiment, oxygen vacancy formation energy, migration barrier, and concentration gradient-driven diffusion equations can be simultaneously introduced into the established anisotropic strain energy phase field framework. This allows lattice distortion, cation rearrangement, and oxygen vacancy evolution to interact in real time. The oxygen loss rate of the material during cycling is directly output through coupled calculations, and this oxygen loss rate is compared with a preset voltage decay target value. If the oxygen loss rate is too high and the voltage decay exceeds the tolerance limit, the surface barrier strength is adjusted iteratively in reverse until the minimum oxygen diffusion blocking rate that just reduces the oxygen loss rate to the target value is found. This provides a quantitative performance benchmark for subsequent coating layer design.

[0084] S250. The minimum oxygen diffusion blocking rate is transformed into the core performance requirement of the surface coating layer, and the surface coating layer configuration parameters are obtained.

[0085] Specifically, the calculated minimum oxygen diffusion blocking rate, which is necessary to achieve the voltage decay control target, is transformed into key design constraints on the composition, structure, or thickness of the coating layer to be applied to the material surface, thereby forming a set of clear and quantifiable surface coating configuration parameters.

[0086] Based on the above embodiments, optionally, the minimum oxygen diffusion blocking rate can be transformed into a core performance requirement for the surface coating layer, and the specific implementation methods of obtaining the surface coating layer configuration parameters may include:

[0087] Based on the minimum oxygen diffusion blocking rate, and combined with the oxygen diffusion coefficient of the candidate coating materials obtained from the preset knowledge graph, the theoretical minimum thickness required for the candidate coating materials that meet the blocking rate requirement is determined; the theoretical minimum thickness is used as the thickness requirement in the surface coating layer configuration parameters.

[0088] Specifically, based on the performance index of the minimum oxygen diffusion blocking rate required to achieve oxygen loss control, and by calling the oxygen diffusion coefficient data of different candidate coating materials stored in the preset knowledge graph, the physical relationship between the blocking rate, diffusion coefficient and coating thickness can be established to calculate the theoretical minimum coating thickness required to achieve the minimum oxygen diffusion blocking rate for each candidate coating material. This calculated theoretical minimum thickness value is then set as the core design requirement for the thickness of the final surface coating configuration parameters.

[0089] Based on the above embodiments, optionally, the surface coating configuration parameters may also include the maximum allowable thickness determined based on thermomechanical performance constraints and / or the compensation layer thickness determined based on the initial coulombic efficiency target.

[0090] Thermomechanical performance constraints refer to the limitations imposed on the interfacial thermal expansion mismatch stress and mechanical strain to maintain structural integrity during temperature changes and electrochemical cycling. Maximum permissible thickness refers to the upper limit of the maximum physical thickness of the surface coating layer without compromising the overall electrochemical performance of the battery (such as interfacial resistance and ion conduction). The initial coulombic efficiency target refers to the pre-set minimum efficiency value that the ratio of reversible capacity to initial charge capacity of the cathode material is expected to achieve during the first charge-discharge cycle. The compensation layer thickness refers to the necessary physical thickness of the functional coating designed and applied to the material surface to provide additional active lithium to compensate for irreversible capacity loss in order to achieve the initial coulombic efficiency target.

[0091] In this embodiment, the design specifications for the surface coating layer, in addition to including the basic components and structural parameters, also need to include two types of thickness parameters that may be superimposed: one is the upper limit of the coating layer thickness calculated to suppress thermomechanical failure during cycling, and the other is the thickness of the surface functional layer required to compensate for irreversible lithium loss, calculated to meet the first charge-discharge efficiency target.

[0092] Based on the above embodiments, the specific implementation of the maximum allowable thickness determined by thermomechanical performance constraints may include:

[0093] Based on the interfacial strain parameters between the cathode material and the electrolyte obtained from phase field simulation during cycling, and combined with the mechanical property parameters of candidate coating materials obtained from a pre-defined knowledge graph, the strain buffer thickness of the coating layer required to avoid interfacial failure is determined; the strain buffer thickness of the coating layer is taken as the maximum allowable thickness.

[0094] Interfacial strain parameters refer to the strain values ​​that vary with time or cycles in the interfacial region where the material contacts the electrolyte or coating layer, due to differences in thermal expansion coefficients or electrochemical volume changes. Mechanical property parameters are inherent properties describing the mechanical behavior and strength of a material, such as elastic modulus and yield strength. The coating layer strain buffer thickness refers to the physical thickness of the surface coating layer calculated to effectively alleviate and accommodate the stress or deformation described by the interfacial strain parameters, preventing interfacial failure.

[0095] Specifically, firstly, the strain data that may be generated at the interface between the cathode material and the electrolyte during charge and discharge cycles is obtained through phase field simulation calculation. At the same time, the strength, elasticity and other mechanical property data of the candidate coating materials are retrieved from the preset knowledge graph. Combining these two types of data, the necessary thickness of the coating layer to buffer the interface strain and prevent mechanical failure is calculated through mechanical model analysis. The strain buffer thickness calculated in this way is set as the upper limit of the coating layer thickness design, that is, the maximum allowable thickness.

[0096] Based on the above embodiments, the specific implementation method of the compensation layer thickness determined based on the initial Coulomb efficiency target may include the following steps:

[0097] (1) Receive the first coulombic efficiency target value for lithium-rich manganese-based cathode materials.

[0098] In this embodiment, the minimum performance requirement value of the ratio of reversible capacity to initial charge capacity that must be achieved in the first charge-discharge cycle can be obtained by the designer or the upper system for the designed lithium-rich manganese-based cathode material, which is the target value of the initial coulombic efficiency.

[0099] (2) Based on the bulk gradient structure characteristics and electrochemical reaction model, the amount of irreversible lithium consumed on the particle surface during the first charging process is determined.

[0100] Electrochemical reaction models refer to mathematical or computational models used to describe the microscopic reaction mechanisms and kinetic processes of lithium-ion insertion / extraction, electron transfer, and accompanying phase transitions during the charging and discharging of cathode materials. Irreversible lithium consumption refers to the number of lithium ions that cannot be reversibly returned during subsequent discharges due to surface side reactions, structural reconstruction, or lithium-ion capture during the first charge, quantified in the molar ratio of lithium to transition metals or other relevant units.

[0101] Specifically, based on the previously determined gradient structural characteristics parameters of the material from the core to the surface, combined with the electrochemical reaction model describing the laws of lithium ion insertion / extraction and charge transfer, the amount of lithium loss that cannot be recovered during discharge due to irreversible chemical reactions or structural changes in the surface region of the material particles during the first charging stage can be predicted through simulation calculations.

[0102] (3) Based on the irreversible lithium consumption and the lithium compensation capability parameters of the candidate compensation layer materials obtained from the preset knowledge graph, determine the minimum thickness of the surface compensation layer required.

[0103] Here, candidate compensation layer materials refer to one or more selectable coating materials for forming on the surface of cathode materials, designed to provide additional active lithium to compensate for irreversible losses during the first cycle. The lithium compensation capability parameter refers to a key characteristic describing the effective lithium content (typically expressed as lithium equivalent per unit mass or unit volume) that the candidate compensation layer material can provide for compensation. The minimum surface compensation layer thickness refers to the minimum physical coating thickness that must be formed on the particle surface when using a specific candidate compensation layer material to compensate for the calculated irreversible lithium consumption.

[0104] In this embodiment, the calculated irreversible lithium consumption on the particle surface can be used as a basis to query and extract the lithium compensation data per unit volume or per unit area that each candidate compensation layer material can provide from a preset knowledge graph. The minimum physical coating thickness required to fully compensate for the consumption and to cover the particle surface for each candidate compensation layer material can then be calculated.

[0105] (4) The minimum thickness of the surface compensation layer is used as the compensation layer thickness in the surface functional layer configuration parameters.

[0106] In this embodiment, the minimum thickness of the surface compensation layer required to meet the first coulomb efficiency target, calculated above, can be set and integrated into the final set of design requirements parameters for the functional coating on the material surface, serving as the core design specification index for the thickness of the compensation layer.

[0107] It should be noted that the volume gradient structure simulation described in S220 to S230 and the surface performance constraint analysis involved in S240 to S250 can be processed in parallel in terms of calculation process, and there is no necessary sequential dependency.

[0108] S260. Based on the bulk gradient structure characteristics and surface coating layer configuration parameters, the process-structure mapping model stored in the preset knowledge graph is called to perform process inversion and determine the executable process scheme at the macro scale.

[0109] Among them, the executable process schemes include candidate coating material systems and their corresponding synthesis process routes.

[0110] The technical solution of this application, when determining the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscale, uses atomic-scale elemental formulations and key parameters as inputs. It employs a physical model incorporating the Jahn-Teller lattice distortion effect to perform mesoscale simulations, theoretically optimizing the strain energy distribution of the material during cycling. This results in the design of a bulk structure with a continuous gradient of internal order, fundamentally improving the material's structural stability and cycle life. Simultaneously, by coupling the oxygen loss kinetics process in the simulation, a quantitative relationship between oxygen loss rate and macroscopic voltage decay is established. Furthermore, the minimum oxygen blocking performance requirements that the surface coating must possess to achieve voltage stability are derived in reverse. This precisely transforms the macroscopic performance decay control target into a specific and quantifiable design specification for the coating, providing a clear and scientific basis for subsequent process implementation.

[0111] Example 3

[0112] Figure 3 This is a schematic diagram of a reverse design method for a cathode material provided in this application embodiment. Based on the foregoing embodiments, this embodiment provides a more detailed description of step S130. For specific implementation details, please refer to the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here.

[0113] like Figure 3 As shown, the method specifically includes the following steps:

[0114] S310. In response to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, high-throughput calculations are performed on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation that satisfies the target performance parameter constraints and the corresponding atomic-scale key parameters.

[0115] S320, based on the optimal elemental formulation and key atomic-scale parameters, uses an anisotropic lattice strain model including Jahn-Teller distortion correction to perform phase-field simulation, and determines the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscale.

[0116] S330. Based on the target order gradient distribution defined in the bulk gradient structure characteristics, call the segmented sintering process and structure mapping model stored in the preset knowledge graph, and determine one or more segmented sintering process parameter sets for synthesizing bulk materials with target gradient structures through inversion optimization algorithm, and obtain an executable process scheme.

[0117] The target order gradient distribution refers to the numerical change curve of the desired order degree from the particle core to the surface, determined by mesoscale design. The segmented sintering process and structure mapping model refers to a computational or correlation model stored in a dynamic knowledge graph, used to describe the quantitative relationship between sintering process parameters such as staged temperature and atmosphere control and the internal order degree distribution of the final material. The bulk material with the target gradient structure refers to a lithium-rich manganese-based cathode material possessing the aforementioned target order gradient distribution. The segmented sintering process parameter set refers to the set of key control parameters, such as temperature, time, atmosphere type, and switching procedures, required at each stage of the segmented sintering process to achieve the target gradient structure.

[0118] Specifically, the target curve of the order gradient distribution obtained from the mesoscale design can be used as input conditions. The mapping model described in the knowledge graph that describes the causal relationship between the segmented sintering process and the formation of the material structure can be retrieved and applied. An optimization algorithm designed to solve the reverse engineering problem is used to calculate and derive one or more complete segmented sintering process parameter combinations that can prepare bulk phase materials that meet the target gradient structure. This forms an executable process scheme that can be directly used for experimental execution.

[0119] Based on the above embodiments, optionally, the specific implementation method of determining one or more segmented sintering process parameter sets through inversion optimization algorithm may include: taking minimizing the overall deviation between the order distribution predicted by the model and the target order gradient distribution as the optimization objective, iteratively optimizing the process parameters, including the temperature, time and atmosphere type of each sintering stage, until the preset convergence condition is met, and outputting the segmented sintering process parameter set.

[0120] Specifically, the goal of computational optimization can be to reduce the overall difference between the final material order distribution predicted by the process-structure mapping model and the preset target order gradient distribution. The temperature parameters, time parameters, and atmosphere type parameters of each stage of the segmented sintering process are cyclically adjusted and evaluated until the deviation between the model prediction result and the target reaches an acceptable preset standard or the number of optimizations reaches the upper limit. At this point, the optimization process is terminated and the optimal or satisfactory set of segmented sintering process parameters is output.

[0121] In this embodiment, optionally, based on the bulk gradient structure characteristics and surface coating configuration parameters, the process-structure mapping model stored in the preset knowledge graph is invoked to perform process inversion and determine the executable process scheme at the macroscopic scale. This may also include the following steps:

[0122] (1) Based on the target coating thickness and material type defined in the surface coating configuration parameters, at least one feasible coating process path is matched from the preset knowledge graph.

[0123] Among them, the target coating thickness and material type refer to the specific thickness value and the type of coating material selected for the functional coating applied to the surface of the cathode material, which are determined according to the mesoscale design.

[0124] The coating process route refers to a complete process technology route that can be selected to coat the surface of material particles with a coating of a specific material type at a target thickness. This route usually includes specific preparation methods (such as vapor deposition and solid-phase reaction), operation procedures and their theoretical basis.

[0125] Specifically, the surface coating configuration parameters output by the mesoscale design, which clearly specify the coating thickness and specific material type, can be used as query conditions. In the process-material-performance association network stored in the preset knowledge graph, one or more verified or theoretically feasible complete process technology routes that can achieve the specific combination of coating material and thickness can be retrieved and screened.

[0126] Based on the above embodiments, optionally, the coating process route includes at least one of the following: vapor phase deposition, solid-state reaction, or ion exchange. Vapor phase deposition refers to a process route that forms a coating layer by causing a gaseous or vaporous precursor to undergo a chemical reaction or physical condensation on the material surface. Solid-state reaction refers to a process route that generates a coating layer by causing a direct chemical reaction between a solid material and the surface of the cathode substrate at high temperature. Ion exchange refers to a process route that forms or modifies a coating layer by immersing the material in a specific solution and utilizing the displacement reaction between ions in the solution and ions on the material surface.

[0127] In detail, when the coating process is based on vapor deposition, the key process control parameters include the concentration of reactant gases, deposition temperature, and deposition time. In other words, when using vapor deposition to prepare the surface coating, to achieve the target thickness and material type, the core operating variables that need to be determined through model inversion or process rules are the content of the reactant gases in the atmosphere, the substrate or ambient temperature required for the deposition process, and the duration of the deposition process.

[0128] When the coating process follows a solid-state reaction pathway, key process control parameters include sintering temperature and ambient oxygen partial pressure. When using a solid-state reaction pathway to prepare the surface coating, to achieve the target thickness and material type, the core operating variables that need to be determined through model inversion or process rules are the heat treatment temperature that induces the solid-state reaction, and the chemical potential or partial pressure of oxygen in the reaction environment.

[0129] When the coating process uses an ion exchange pathway, key process control parameters include the concentration of the reaction solution, the reaction temperature, and the reaction time. When using an ion exchange pathway to prepare or modify a surface coating, the core operational variables that need to be determined through model inversion or process rules to achieve the target thickness and material type are the content of specific ions in the reaction solution used for ion exchange, the solution temperature maintained during the ion exchange reaction, and the duration of the reaction process.

[0130] (2) Call the process dynamics or thermodynamics model corresponding to the selected coating process path, and combine the target coating thickness and material type to invert and calculate at least one set of key process control parameters required to achieve the target, thus forming the synthesis process path.

[0131] Based on the above embodiments, for each feasible coating process route matched from the knowledge graph, the associated mathematical calculation model describing the reaction rate or spontaneous direction of the route can be invoked. The specific target coating thickness and material type are input into the model as known quantities. Through reverse calculation, one or more specific core process operation parameter numerical combinations necessary to achieve this target are derived. These parameter sets constitute the specific synthesis process path of the coating process.

[0132] The technical solution of this application, based on the bulk gradient structure characteristics and surface coating layer configuration parameters, calls the process-structure mapping model stored in the preset knowledge graph to perform process inversion and determine the executable process scheme at the macro scale. By taking the target order gradient distribution theoretically designed at the mesoscale as input, and using the segment sintering process-structure mapping model stored in the knowledge graph, and with the help of the inversion optimization algorithm, one or more executable segmented sintering process parameter sets that can accurately realize the gradient structure are derived. In this way, the abstract structural design concept is directly transformed into specific and operable macro-manufacturing instructions, ensuring the accurate and efficient connection from micro-structure design to macro-synthesis process, and providing a process scheme that can be verified and optimized, which greatly shortens the research and development cycle of materials from design to preparation.

[0133] Next, three specific examples will be used to illustrate the reverse design method for cathode materials provided in the embodiments of the present invention. Figure 4 This is a logic diagram of the reverse design method for cathode materials provided in this embodiment.

[0134] Example 1: High voltage stability design (oxygen loss suppression)

[0135] Step 1: Atomic-scale modeling:

[0136] (1) Input target value: The target performance requirement is voltage decay ≤ 0.8% over 100 cycles. The system uses the dynamic knowledge graph to determine the voltage decay, oxygen loss, and other parameters. The causal chain model interprets this macroscopic objective as a microscopic design constraint: the anion redox energy barrier. need (Ensure reversible oxidation-reduction of oxygen);

[0137] (2) Calling DeePMD to machine learning potential functions. First, a supercell model of the doping candidate set is constructed based on the typical unit cell structure in the knowledge graph. DeePMD performs high-throughput energy calculations on the doping configurations of various elements such as Ta and Sb and their combinations at Mn or Li sites. By comparing the oxygen vacancy formation energies under different configurations ( (Computational core), quickly filter out those that can significantly improve The elemental combination. Simulation yields the Ta doping amount. Sb doping amount hour Optimal.

[0138] Step 2, Mesoscale Optimization:

[0139] (1) Input atomic modeling output Value (0.28 eV);

[0140] (2) Gradient structure design: based on the inclusion of Jahn-Teller correction terms Phase field simulation was performed using an anisotropic strain model. This was achieved through scanning... The value (0.1-0.5) was used to simulate the cyclic process, and it was found that when... At that time, the average lattice distortion energy accumulated inside the particle is the lowest. Based on this, a continuous gradient distribution of order (0.85 to 0.25) from the core to the outer shell was optimized, which most effectively buffers cyclic stress. The cladding parameters were calculated using phase-field simulation coupled with an oxygen vacancy diffusion module. Simulations show that to achieve the voltage decay target, the surface barrier needs to block oxygen diffusion at a rate ≥63%. This is based on a one-dimensional oxygen diffusion model. Combined with data retrieved from the knowledge graph The oxygen diffusion coefficient Dcoat is used to calculate the composition of the coating layer. Theoretical minimum coating thickness .

[0141] Step 3: Macroscopic process inversion:

[0142] (1) Input the target order gradient curve and the coating thickness. ;

[0143] (2) Step A: Gradient structure sintering process inversion: The system uses the target gradient curve as input and calls the "multi-segment order evolution model" for inversion. This model is based on the formula:

[0144] ;

[0145] Among them, key parameters (Atmosphere-related pre-existing factors) and (Location-dependent activation energy) is predicted by a dynamic knowledge graph based on the current material composition. A global optimization algorithm is used to solve for the optimal n-segment process parameters. , , This makes the final ordered distribution predicted by the model... The error with the target curve is minimized. Inversion output: The optimal three-stage sintering program is: Stage 1: Air atmosphere, 850℃, 10h (promoting high bulk phase ordering); Stage 2: Nitrogen atmosphere, 2h (transition zone control); Third segment: (Controllable disordering / spinelization of the surface).

[0146] Step B: According to the fluidized bed kinetic equations:

[0147] ;

[0148] Inversion, when concentration ,temperature deposition rate The thickness reached after 25 minutes of deposition was 4.2 nm.

[0149] (3) Output process parameters: precursor pretreatment ( → Fluidized bed deposition ( concentration );

[0150] The final output of step three: a complete two-step process package: 1. Bulk gradient structure sintering procedure: to to 2. Surface Coating deposition parameters: precursor pretreatment ( ) to fluidized bed deposition ( ).

[0151] Step 4: Closed-loop verification of materials and chips:

[0152] Procedure: The above process package was imported into a high-throughput system for execution. First, gradient sintering was completed in a 128-cavity microreactor. Rapid sampling and offline XRD refinement verified that the ordered distribution of the obtained particle cross-section matched the target curve with a degree of >95%. Subsequently, in-situ coating deposition was performed on the same platform, and 1Hz rapid XAS monitoring confirmed complete surface reaction. Electrochemical testing showed that the voltage decay over 100 cycles was [missing information]. To achieve the goal.

[0153] Feedback: The system automatically records the entire data chain from "composition, gradient sintering process, coating process to performance" and incorporates it as a high-quality sample into the dynamic knowledge graph to optimize relevant model parameters.

[0154] Example 2: Optimization of the all-solid-state battery interface

[0155] Step 1: Atomic-scale modeling:

[0156] (1) The target is to react with sulfide electrolytes interfacial impedance ². The knowledge graph interprets it as lattice mismatch. atomic-scale constraints;

[0157] (2) DFT screening of Na-doped stable O2-type structures (ABCB stacking), when Na content Time positive lattice To mismatch ;

[0158] (3) Output element formula ;

[0159] Step 2, Mesoscale Optimization:

[0160] (1) Input the lattice parameters output by atomic modeling ;

[0161] (2) Run the phase-field model of interfacial stress-electrochemical coupling to simulate the interaction between the cathode and electrolyte. Thermomechanical strain behavior during cycling. Quantitative analysis shows that introducing a layer... The composite coating can compensate approximately Strain mismatch. To achieve a long-term stable interface, the coating thickness must meet the following synergistic optimization criteria: Stress buffering criterion (determining the minimum thickness dmin): To ensure that the coating itself does not crack or peel under cyclic stress, its thickness must meet the following requirements:

[0162] ;

[0163] Substitute the mechanical parameters (Young's modulus) of the composite material obtained from the dynamic knowledge graph. ,tensile strength (etc.), calculated .

[0164] Ion transport criterion (determining maximum thickness dmax): To ensure that interface impedance does not affect battery rate performance, the sheet resistance introduced by the coating layer must be lower than the threshold Rmax, i.e. ,in The ionic conductivity of the composite layer is calculated. .

[0165] Collaborative optimization output: In Within the feasible range, the model optimizes with the dual objectives of minimizing the peak interfacial stress and minimizing the lithium-ion migration barrier, and determines the global optimal solution as follows: .

[0166] (3) The thickness of the output interface coating layer is 120nm (non-double layer).

[0167] Step 3: Macroscopic process inversion:

[0168] (1) The thickness of the input phase field optimization output is 120nm;

[0169] (2) According to the solid-state reaction thermodynamic model:

[0170] ;

[0171] Verification at The reaction can proceed spontaneously under oxygen partial pressure. Furthermore, based on the growth kinetic data of this reaction in the knowledge graph, the growth rate of the coating layer under this condition is determined to be approximately... To achieve a thickness of 120nm, the reaction time needs to be 100 minutes.

[0172] (3) Output process parameters: ball milling and mixing ( ) → Solid-state reaction ( ).

[0173] Step 4: Closed-loop verification of materials and chips:

[0174] This process was performed in a high-throughput sintered chip, and the interface impedance was monitored using in-situ impedance spectroscopy. Experiments confirmed that the interface impedance was... The objective is met. Process parameters " "It was reinforced and stored in the knowledge graph as a success story."

[0175] Example 3: Ultra-high capacity layered cathode

[0176] Step 1: Atomic-scale modeling:

[0177] Input target: target capacity In achieving Under the premise of theoretical capacity, the reversibility of its first cycle is simultaneously optimized. Based on data analysis of similar ultra-high capacity lithium-rich systems using dynamic knowledge graphs, if the first-cycle coulombic efficiency (CE) is lower than... If excessive active lithium is lost, the actual usable cycle capacity will decrease sharply, failing to achieve the capacity target. Therefore, the system's co-optimization objective for this stage is: to improve the first efficiency to [value missing] while maintaining high tetrahedral lithium site activity. .

[0178] Step 2, Mesoscale Optimization:

[0179] (1) Start " "Diffusion-Side Reaction Coupled Phase-Field Model" for simulation During the initial charge and discharge cycle, diffusion competition occurs between tetrahedral and octahedral sites. Simulations confirm that although a high tetrahedral percentage brings capacity potential, a large amount of lithium embeds into deep tetrahedral sites and encounters surface side reactions during the first charge, resulting in significant irreversible capacity loss; the initial first efficiency is only [missing information]. Unable to meet The goal.

[0180] (2) Simulations show that the most effective strategy is to construct a lithium-rich layer with a gradually changing chemical potential gradient on the particle surface, providing "sacrificial lithium" in advance to compensate for irreversible loss. To achieve this... The first-effect target ( According to the first-effect model:

[0181] ;

[0182] Permissible surface irreversible lithium loss Strictly limited. (Based on interface parameters) And using the mass balance formula:

[0183] ;

[0184] Calculations show that a thickness of approximately [missing information] needs to be constructed. Surface lithium content increased to (i.e., local lithium increment) The gradient lithium-rich layer has a corresponding lithium increment of approximately .

[0185] (3) To achieve the synergistic goal of capacity and first-efficiency, a clear surface treatment scheme is proposed: using ion exchange method, on P3 type Construct a matrix particle surface A thick, gradient lithium-rich layer corresponds to a lithium increment of approximately .

[0186] Step 3: Macroscopic process inversion:

[0187] (1) Input mesoscale optimization of output 5nm thick gradient lithium-rich layer and Li increment ;

[0188] (2) Applying the ion diffusion-reaction coupling equation:

[0189] ;

[0190] Inversion is performed. Using the target Li increment as the boundary condition, and combining it with the diffusion coefficient of LiBr in ethanol from the knowledge graph... And the reaction constant k, the optimal process parameters are obtained by numerical solution;

[0191] (3) Output process parameters precursor immersion ethanol solution .

[0192] Step 4: Material Chip Closed-Loop Verification: Verification is performed on a parallel liquid phase processing chip, with precise measurement of the Li content increment using ICP-OES. The measured Li increment is... A lithium-rich layer with a thickness of 5.1 nm was formed, improving the first-stage efficiency to [missing value]. The "component-process-result" data pair is fed back to the knowledge graph as a new sample.

[0193] Example 4

[0194] Figure 5 This is a schematic diagram of a reverse design device for a positive electrode material provided in an embodiment of this application. The device includes:

[0195] The atomic parameter determination module 410 is used to respond to a reverse design request for lithium-rich manganese-based cathode materials, and based on the target performance parameters specified in the reverse design request, to perform high-throughput calculations on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the target performance parameter constraints.

[0196] The mesoscopic parameter determination module 420 is used to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscopic scale by performing phase-field simulation using an anisotropic lattice strain model including Jahn-Teller distortion correction, based on the optimal elemental formulation and the atomic-scale key parameters.

[0197] The macroscopic scheme determination module 430 is used to determine the executable process scheme at the macroscopic scale by calling the process-structure mapping model stored in the preset knowledge graph based on the bulk gradient structure characteristics and the surface coating layer configuration parameters; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths.

[0198] This application provides a reverse design apparatus for cathode materials. In application, in response to a reverse design request for lithium-rich manganese-based cathode materials, the apparatus performs high-throughput calculations on various candidate element doping combinations based on the target performance parameters specified in the reverse design request, using a pre-trained machine learning potential function. This determines the optimal element formulation and corresponding atomic-scale key parameters that satisfy the target performance parameter constraints. Then, based on the optimal element formulation and atomic-scale key parameters, a phase-field simulation is performed using an anisotropic lattice strain model including Jahn-Teller distortion correction to determine the mesoscopic-scale bulk phase gradient structure characteristics and surface coating layer configuration parameters. Finally, based on the bulk phase gradient structure characteristics and surface coating layer configuration parameters, a process-structure mapping model stored in a pre-defined knowledge graph is invoked for process inversion to determine a macroscopic-scale executable process scheme. The executable process scheme includes candidate coating material systems and their corresponding synthesis process paths. The technical solution provided in this embodiment constructs a reverse design system that takes the final performance target as the starting point and drives the microscopic design of materials and the formulation of macroscopic processes. This method integrates machine learning calculation, physical simulation and process knowledge base to transform the traditional trial-and-error process that relies on experience into an engineering process that is precisely driven by performance parameters and optimized across multiple scales. This results in the efficient and accurate output of manufacturing solutions that match the design blueprint, thereby improving the success rate of material design and R&D capabilities.

[0199] Based on the above-mentioned device, optionally, the cathode material reverse design device further includes: a process scheme optimization module, used to input the executable process scheme into a high-throughput parallel reactor for automated experimental verification and obtain the actual performance data obtained from the experiment; generate process parameter optimization suggestions based on the deviation between the actual performance data and the target performance parameters; and iteratively optimize the executable process scheme based on the process parameter optimization suggestions until the optimal process scheme that meets the preset convergence conditions is determined.

[0200] Based on the above-mentioned device, optionally, the atomic parameter determination module 410 is specifically used to respond to a reverse design request for lithium-rich manganese-based cathode materials, receive a preset performance parameter threshold from the target performance parameters; determine multiple candidate element doping combinations to be screened based on a preset doping element library; perform high-throughput simulation calculations on the multiple candidate element doping combinations using a pre-trained machine learning potential function to predict at least one key property parameter corresponding to each candidate element doping combination; compare the predicted at least one key property parameter with the preset performance parameter threshold, screen out the candidate element doping combination that satisfies all constraints as the optimal element formulation, and output the atomic-scale key parameters corresponding to the optimal element formulation.

[0201] Based on the above-described device, optionally, the mesoscopic parameter determination module 420 includes:

[0202] The model building unit is used to construct an anisotropic strain energy model that includes the Jahn-Teller lattice distortion effect based on the optimal elemental formulation and the atomic-scale key parameters.

[0203] The bulk gradient feature determination unit is used to perform phase field simulation using the anisotropic strain energy model to minimize the lattice distortion energy of the lithium-rich manganese-based cathode material during electrochemical cycling, and based on the phase field simulation results, to determine the continuous distribution characteristics of the order degree of the bulk gradient structure in the direction from the core to the shell, thereby obtaining the bulk gradient structure features.

[0204] The oxygen blocking rate determination unit is used to couple the generation and diffusion process of oxygen vacancies in the phase field simulation. Based on the comparison between the oxygen loss rate obtained from the simulation and the preset voltage decay target value, it determines the minimum oxygen diffusion blocking rate required to achieve oxygen loss control.

[0205] The coating layer parameter determination unit is used to convert the minimum oxygen diffusion blocking rate into the core performance requirements of the surface coating layer, and obtain the surface coating layer configuration parameters.

[0206] Based on the above-mentioned device, optionally, the bulk gradient structure features include: the core region is an ordered layered structure with a first degree of order, the outer shell region is a disordered spinel structure with a second degree of order, and a continuous transition region between the core and the outer shell, wherein the first degree of order is greater than the second degree of order.

[0207] Based on the above device, optionally, a coating layer parameter determination unit is used to determine the theoretical minimum thickness required for a candidate coating material that meets the blocking rate requirement, based on the minimum oxygen diffusion blocking rate and combined with the oxygen diffusion coefficient of the candidate coating material obtained from the preset knowledge graph; and to use the theoretical minimum thickness as the thickness requirement in the surface coating layer configuration parameters.

[0208] Based on the above-mentioned device, optionally, the surface coating configuration parameters may also include the maximum allowable thickness determined based on thermomechanical performance constraints and / or the compensation layer thickness determined based on the initial coulombic efficiency target.

[0209] Based on the above-mentioned device, optionally, the coating layer parameter determination unit is further used to determine the coating layer strain buffer thickness required to avoid interface failure by combining the interfacial strain parameters between the cathode material and the electrolyte obtained from phase field simulation during the cycling process with the mechanical property parameters of the candidate coating materials obtained from the preset knowledge graph; and to use the coating layer strain buffer thickness as the maximum allowable thickness.

[0210] Based on the above-mentioned device, optionally, the coating layer parameter determination unit is further configured to receive the initial coulombic efficiency target value for the lithium-rich manganese-based cathode material; determine the irreversible lithium consumption on the particle surface during the first charging process based on the bulk gradient structure characteristics and electrochemical reaction model; determine the required minimum thickness of the surface compensation layer based on the irreversible lithium consumption and the lithium compensation capability parameters of the candidate compensation layer materials obtained from the preset knowledge graph; and use the minimum thickness of the surface compensation layer as the compensation layer thickness in the surface functional layer configuration parameters.

[0211] Based on the above-mentioned device, optionally, the macroscopic scheme determination module 430 is specifically used to, based on the target order gradient distribution defined in the bulk gradient structure characteristics, call the segmented sintering process and structure mapping model stored in the preset knowledge graph, and determine one or more segmented sintering process parameter sets for synthesizing bulk materials with the target gradient structure through an inversion optimization algorithm, so as to obtain an executable process scheme.

[0212] Based on the above-mentioned device, optionally, the macroscopic scheme determination module 430 is more specifically used to optimize the process parameters, including temperature, time and atmosphere type of each sintering stage, with the optimization objective of minimizing the overall deviation between the order distribution predicted by the model and the target order gradient distribution, until the preset convergence condition is met, and output the segmented sintering process parameter set.

[0213] Based on the above-mentioned device, optionally, the macroscopic scheme determination module 430 is further used to match at least one feasible coating process path from the preset knowledge graph based on the target coating layer thickness and material type defined in the surface coating layer configuration parameters; call the process dynamics or thermodynamic model corresponding to the selected coating process path, and in combination with the target coating layer thickness and material type, calculate at least one set of key process control parameters required to achieve the target, thereby constituting the synthesis process path.

[0214] Based on the above-mentioned apparatus, optionally, the coating process path includes at least one of a vapor phase deposition path, a solid-state reaction path, or an ion exchange path; when the coating process path is a vapor phase deposition path, the key process control parameters include the reactant gas concentration, deposition temperature, and deposition time; when the coating process path is a solid-state reaction path, the key process control parameters include the sintering temperature and the ambient oxygen partial pressure; when the coating process path is an ion exchange path, the key process control parameters include the reaction solution concentration, reaction temperature, and reaction time.

[0215] The cathode material reverse design apparatus provided in this application can execute the cathode material reverse design method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects of the method.

[0216] It is worth noting that the various units and modules included in the above system are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of this application.

[0217] Example 5

[0218] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 6 A block diagram is shown of an exemplary electronic device 50 suitable for implementing embodiments of the present application. Figure 6 The electronic device 50 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0219] like Figure 6As shown, the electronic device 50 is represented in the form of a general-purpose computing device. The components of the electronic device 50 may include, but are not limited to: one or more processors or processing units 501, system memory 502, and bus 503 connecting different system components (including system memory 502 and processing unit 501).

[0220] Bus 503 represents one or more of several bus architectures, including memory buses or memory electronics, peripheral buses, graphics acceleration ports, processors, or local buses using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0221] Electronic device 50 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 50, including volatile and non-volatile media, removable and non-removable media.

[0222] System memory 502 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 504 and / or cache memory 505. Electronic device 50 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 506 may be used to read and write non-removable, non-volatile magnetic media (… Figure 6 Not shown; usually referred to as a "hard drive"). Although Figure 6 As not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 503 via one or more data media interfaces. Memory 502 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0223] A program / utility 508 having a set (at least one) of program modules 507 may be stored, for example, in memory 502. Such program modules 507 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 507 typically perform the functions and / or methods described in the embodiments of this application.

[0224] Electronic device 50 can also communicate with one or more external devices 509 (e.g., keyboard, pointing device, display 510, etc.), and with one or more devices that enable a user to interact with electronic device 50, and / or with any device that enables electronic device 50 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 511. Furthermore, electronic device 50 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 512. As shown, network adapter 512 communicates with other modules of electronic device 50 via bus 503. It should be understood that, although... Figure 6 As not shown, other hardware and / or software modules may be used in conjunction with electronic device 50, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0225] The processing unit 501 executes various functional applications and page processing by running programs stored in the system memory 502, such as implementing the reverse design method for cathode materials provided in the embodiments of this application.

[0226] Example 6

[0227] This application embodiment also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a reverse design method for a cathode material, the method comprising:

[0228] In response to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, high-throughput calculations are performed on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the constraints of the target performance parameters.

[0229] Based on the optimal elemental formulation and the key atomic-scale parameters, an anisotropic lattice strain model including Jahn-Teller distortion correction was used to perform phase-field simulation to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscale.

[0230] Based on the bulk gradient structure characteristics and the surface coating layer configuration parameters, the process inversion is performed by calling the process-structure mapping model stored in the preset knowledge graph to determine the executable process scheme at the macro scale; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths.

[0231] The computer storage medium in this application embodiment can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0232] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0233] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0234] Computer program code for performing the operations of the embodiments of this application can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as C or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0235] Example 7

[0236] This application also provides a vehicle including a power battery. The positive electrode of the power battery is a lithium-rich manganese-based positive electrode material designed and prepared using the following reverse design method for positive electrode materials, the method comprising:

[0237] In response to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, high-throughput calculations are performed on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the constraints of the target performance parameters.

[0238] Based on the optimal elemental formulation and the key atomic-scale parameters, an anisotropic lattice strain model including Jahn-Teller distortion correction was used to perform phase-field simulation to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscale.

[0239] Based on the bulk gradient structure characteristics and the surface coating layer configuration parameters, the process inversion is performed by calling the process-structure mapping model stored in the preset knowledge graph to determine the executable process scheme at the macro scale; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths.

[0240] Note that the above description is merely a preferred embodiment and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments. Many other equivalent embodiments may be included without departing from the concept of this application, and the scope of this application is determined by the scope of the appended claims.

Claims

1. A reverse design method for cathode materials, characterized in that, include: In response to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, high-throughput calculations are performed on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the constraints of the target performance parameters. Based on the optimal elemental formulation and the key atomic-scale parameters, an anisotropic lattice strain model including Jahn-Teller distortion correction was used to perform phase-field simulation to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscale. Based on the bulk gradient structure characteristics and the surface coating layer configuration parameters, the process inversion is performed by calling the process-structure mapping model stored in the preset knowledge graph to determine the executable process scheme at the macro scale; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths.

2. The method according to claim 1, characterized in that, The method further includes: The executable process scheme is input into a high-throughput parallel reactor for automated experimental verification, and the actual performance data obtained from the experiment is obtained. Based on the deviation between the actual performance data and the target performance parameters, process parameter optimization suggestions are generated; Based on the proposed optimization of process parameters, the executable process scheme is iteratively optimized until the optimal process scheme that meets the preset convergence conditions is determined.

3. The method according to claim 1, characterized in that, In response to a reverse design request for lithium-rich manganese-based cathode materials, based on the target performance parameters specified in the reverse design request, high-throughput calculations are performed on multiple candidate element doping combinations using a pre-trained machine learning potential function to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the target performance parameter constraints, including: In response to a reverse design request for lithium-rich manganese-based cathode materials, a preset performance parameter threshold is received from the target performance parameters. Based on a pre-set doping element library, multiple candidate element doping combinations to be screened are determined; Using a pre-trained machine learning potential function, high-throughput simulation calculations are performed on the multiple candidate element doping combinations to predict at least one key property parameter corresponding to each candidate element doping combination. The predicted key property parameter is compared with the preset performance parameter threshold, and candidate element doping combinations that meet all constraints are selected as the optimal element formulation. The atomic-scale key parameters corresponding to the optimal element formulation are then output.

4. The method according to claim 1, characterized in that, Based on the optimal elemental formulation and the key atomic-scale parameters, an anisotropic lattice strain model incorporating Jahn-Teller distortion correction is used for phase-field simulation to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscale, including: Based on the optimal elemental formulation and the atomic-scale key parameters, an anisotropic strain energy model incorporating the Jahn-Teller lattice distortion effect is constructed. Phase-field simulations were performed using the anisotropic strain energy model to minimize the lattice distortion energy of the lithium-rich manganese-based cathode material during electrochemical cycling. Based on the phase-field simulation results, the continuous distribution characteristics of the order degree of the bulk gradient structure from the core to the shell were determined, thus obtaining the bulk gradient structure characteristics; and, In the phase field simulation, the generation and diffusion process of oxygen vacancies are coupled, and the minimum oxygen diffusion blocking rate required to achieve oxygen loss control is determined based on the comparison between the oxygen loss rate obtained from the simulation and the preset voltage decay target value. The minimum oxygen diffusion blocking rate is converted into the core performance requirements of the surface coating layer, and the surface coating layer configuration parameters are obtained.

5. The method according to claim 4, characterized in that, The bulk gradient structure features include: a core region being an ordered layered structure with a first degree of order, an outer shell region being a disordered spinel structure with a second degree of order, and a continuous transition region between the core and the outer shell, wherein the first degree of order is greater than the second degree of order.

6. The method according to claim 4, characterized in that, The process of converting the minimum oxygen diffusion blocking rate into a core performance requirement for the surface coating layer, resulting in surface coating layer configuration parameters, includes: Based on the minimum oxygen diffusion blocking rate, and combined with the oxygen diffusion coefficient of the candidate coating materials obtained from the preset knowledge graph, the theoretical minimum thickness required for the candidate coating materials that meet the blocking rate requirement is determined. The theoretical minimum thickness is used as the thickness requirement in the surface coating configuration parameters.

7. The method according to claim 6, characterized in that, The surface coating configuration parameters also include the maximum allowable thickness determined based on thermomechanical performance constraints and / or the compensation layer thickness determined based on the initial coulomb efficiency target.

8. The method according to claim 7, characterized in that, The maximum allowable thickness determined based on thermomechanical property constraints includes: Based on the interfacial strain parameters between the cathode material and the electrolyte obtained from phase field simulation during the cycling process, and combined with the mechanical property parameters of the candidate coating materials obtained from the preset knowledge graph, the strain buffer thickness of the coating layer required to avoid interfacial failure is determined. The strain-resistant buffer thickness of the coating layer is taken as the maximum allowable thickness.

9. The method according to claim 7, characterized in that, The compensation layer thickness determined based on the initial Coulomb efficiency target includes: Receive the first target coulombic efficiency value for the lithium-rich manganese-based cathode material; Based on the aforementioned bulk gradient structure characteristics and electrochemical reaction model, the amount of irreversible lithium consumed on the particle surface during the first charging process was determined. Based on the irreversible lithium consumption and the lithium compensation capability parameters of the candidate compensation layer materials obtained from the preset knowledge graph, the minimum required thickness of the surface compensation layer is determined. The minimum thickness of the surface compensation layer is used as the compensation layer thickness in the surface functional layer configuration parameters.

10. The method according to claim 1, characterized in that, The process of determining an executable process scheme at the macroscale by invoking a process-structure mapping model stored in a preset knowledge graph, based on the bulk gradient structure characteristics and the surface coating layer configuration parameters, includes: Based on the target order gradient distribution defined in the bulk gradient structure features, the segmented sintering process and structure mapping model stored in the preset knowledge graph is invoked. Through the inversion optimization algorithm, one or more segmented sintering process parameter sets for synthesizing bulk materials with the target gradient structure are determined, and an executable process scheme is obtained.

11. The method according to claim 10, characterized in that, The process of determining one or more segmented sintering process parameter sets through inversion optimization algorithms includes: With the goal of minimizing the overall deviation between the ordered distribution predicted by the model and the target ordered gradient distribution, the process parameters, including temperature, time and atmosphere type of each sintering stage, are iteratively optimized until the preset convergence condition is met, and the segmented sintering process parameter set is output.

12. The method according to claim 1, characterized in that, The step of performing process inversion based on the bulk gradient structure features and the surface coating layer configuration parameters, by calling the process-structure mapping model stored in a preset knowledge graph, to determine the executable process scheme at the macroscopic scale, further includes: Based on the target coating thickness and material type defined in the surface coating configuration parameters, at least one feasible coating process path is matched from the preset knowledge graph; By calling the process kinetics or thermodynamics model corresponding to the selected coating process path, and combining the target coating layer thickness and material type, at least one set of key process control parameters required to achieve the target are calculated in reverse, thus constituting the synthesis process path.

13. The method according to claim 12, characterized in that, The coating process path includes at least one of the following: vapor deposition path, solid-state reaction path, or ion exchange path. When the coating process is a vapor deposition process, the key process control parameters include the concentration of reactant gas, deposition temperature, and deposition time. When the coating process is a solid-state reaction process, the key process control parameters include sintering temperature and ambient oxygen partial pressure. When the coating process is an ion exchange process, the key process control parameters include the concentration of the reaction solution, the reaction temperature, and the reaction time.

14. A reverse design device for positive electrode materials, characterized in that, The device includes: The atomic parameter determination module is used to respond to a reverse design request for lithium-rich manganese-based cathode materials. Based on the target performance parameters specified in the reverse design request, it uses a pre-trained machine learning potential function to perform high-throughput calculations on multiple candidate element doping combinations to determine the optimal element formulation and corresponding atomic-scale key parameters that satisfy the target performance parameter constraints. The mesoscopic parameter determination module is used to determine the bulk phase gradient structure characteristics and surface coating configuration parameters at the mesoscopic scale by performing phase field simulation using an anisotropic lattice strain model including Jahn-Teller distortion correction based on the optimal element formulation and the atomic-scale key parameters. The macroscopic scheme determination module is used to determine the executable process scheme at the macroscopic scale by calling the process-structure mapping model stored in the preset knowledge graph based on the bulk gradient structure characteristics and the surface coating layer configuration parameters; wherein, the executable process scheme includes candidate coating material systems and their corresponding synthesis process paths.

15. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the reverse design method for cathode materials as described in any one of claims 1-13.

16. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the reverse design method for cathode materials as described in any one of claims 1-13.

17. A vehicle, characterized in that, The power battery includes a positive electrode made of a lithium-rich manganese-based positive electrode material designed and prepared by any one of claims 1 to 13.

Images