CN-122024945-A - Method, device, equipment, medium and vehicle for reverse design of anode material

Abstract

The application provides a reverse design method, a device, equipment, a medium and a vehicle of a cathode material, which comprise the steps of responding to a reverse design request for a lithium-rich manganese-based cathode material, carrying out high-throughput calculation on a plurality of candidate element doping combinations by utilizing a machine learning potential function based on target performance parameters, determining an optimal element formula and an atomic scale key parameter meeting target performance parameter constraint, carrying out phase field simulation by adopting an anisotropic lattice strain model containing Jahn-Teller distortion correction based on the optimal element formula and the atomic scale key parameter, determining bulk phase gradient structural characteristics and surface coating layer configuration parameters, carrying out process inversion by calling a process and structure mapping model stored in a preset knowledge map based on the bulk phase gradient structural characteristics and the surface coating layer configuration parameters, and determining a candidate coating material system and a corresponding synthesis process path thereof.

Inventors

• LI YANTAO
• Request for anonymity
• Request for anonymity
• Request for anonymity

Assignees

• 晶核能源(嘉兴)有限公司

Dates

Publication Date

20260512

Application Date

20251230

Claims (17)

1. 1. A method of reverse engineering a positive electrode material, comprising: responding to a reverse design request for a lithium-rich manganese-based positive electrode material, and based on target performance parameters appointed in the reverse design request, performing high-flux calculation on a plurality of candidate element doping combinations by utilizing a pre-trained machine learning potential function, and determining an optimal element formula meeting the constraint of the target performance parameters and corresponding atomic scale key parameters; Based on the optimal element formula and the atomic scale key parameters, adopting an anisotropic lattice strain model containing Jahn-Teller distortion correction to perform phase field simulation, and determining mesoscale bulk phase gradient structural characteristics and surface coating configuration parameters; and calling a mapping model of a process and a structure stored in a preset knowledge graph to perform process inversion based on the bulk phase gradient structural characteristics and the surface coating layer configuration parameters, and determining an executable process scheme of a macro scale, wherein the executable process scheme comprises a candidate coating material system and a corresponding synthesis process path.
2. 2. The method according to claim 1, wherein the method further comprises: inputting the executable process scheme into a high-flux parallel reaction device for automatic experiment verification, and obtaining actual performance data obtained by experiments; Generating a process parameter optimization suggestion based on the deviation of the actual performance data and the target performance parameter; and carrying out iterative optimization on the executable process scheme based on the process parameter optimization suggestion until an optimal process scheme meeting a preset convergence condition is determined.
3. 3. The method of claim 1, wherein the determining, in response to a reverse design request for a lithium-rich manganese-based cathode material, a plurality of candidate element doping combinations for high-throughput computation using a pre-trained machine learning potential function based on target performance parameters specified in the reverse design request, an optimal element recipe and corresponding atomic scale key parameters that satisfy the target performance parameter constraints, comprises: receiving a preset performance parameter threshold value in target performance parameters in response to a reverse design request for the lithium-rich manganese-based anode material; Determining a plurality of candidate element doping combinations to be screened based on a preset doping element library; Performing high-throughput simulation calculation on the plurality of candidate element doping combinations by utilizing a pre-trained machine learning potential function, and predicting at least one key property parameter corresponding to each candidate element doping combination; And comparing the predicted at least one key property parameter with the preset performance parameter threshold, screening out candidate element doping combinations meeting all constraint conditions as the optimal element formula, and outputting atomic scale key parameters corresponding to the optimal element formula.
4. 4. The method of claim 1, wherein the determining mesoscale bulk gradient structural features and surface cladding layer configuration parameters based on the optimal elemental formulation and the atomic scale key parameters using a phase field simulation using an anisotropic lattice strain model comprising Jahn-Teller distortion correction comprises: constructing an anisotropic strain energy model comprising a Jahn-Teller lattice distortion effect based on the optimal element formula and the atomic scale key parameters; Performing phase field simulation by using the anisotropic strain energy model to minimize lattice distortion energy of the lithium-rich manganese-based positive electrode material in an electrochemical cycle process, determining sequential order distribution characteristics of a bulk phase gradient structure in a direction from a core to an outer shell based on a phase field simulation result to obtain bulk phase gradient structure characteristics, and Coupling an oxygen vacancy generation and diffusion process in the phase field simulation, and determining a minimum oxygen diffusion blocking rate required for realizing oxygen loss control based on a comparison result of the oxygen loss rate obtained through the simulation and a preset voltage attenuation target value; And converting the minimum oxygen diffusion blocking rate into core performance requirements of the surface coating layer to obtain the configuration parameters of the surface coating layer.
5. 5. The method of claim 4, wherein the bulk gradient structural feature comprises an inner core region being an ordered layered structure having a first range of order, an outer shell region being an unordered spinel structure having a second range of order, and a continuous transition region between the inner core and the outer shell, the first order being greater than the second order.
6. 6. The method of claim 4, wherein said converting said minimum oxygen diffusion barrier to core performance requirements for a surface coating to obtain surface coating configuration parameters comprises: Based on the minimum oxygen diffusion blocking rate, determining a theoretical minimum thickness required by the candidate coating material meeting the blocking rate requirement by combining the oxygen diffusion coefficient of the candidate coating material obtained by inquiring from the preset knowledge graph; The theoretical minimum thickness is taken as the thickness requirement in the surface coating configuration parameters.
7. 7. The method of claim 6, wherein the surface cladding configuration parameters further comprise a maximum allowable thickness determined based on thermo-mechanical performance constraints and/or a compensation layer thickness determined based on a first coulombic efficiency target.
8. 8. The method of claim 7, wherein the maximum allowable thickness determined based on the thermo-mechanical performance constraint comprises: Based on interface strain parameters between the anode material and the electrolyte in the circulation process, which are obtained through phase field simulation, and combining mechanical property parameters of the candidate coating materials obtained from the preset knowledge graph, determining the anti-strain buffer thickness of the coating layer required for avoiding interface failure; and taking the anti-strain buffer thickness of the coating layer as the maximum allowable thickness.
9. 9. The method of claim 7, wherein the compensating layer thickness determined based on the first coulombic efficiency goal comprises: Receiving a first coulombic efficiency target value for the lithium-rich manganese-based positive electrode material; Based on the bulk phase gradient structural characteristics and an electrochemical reaction model, determining irreversible lithium consumption on the particle surface in the primary charging process; Determining the minimum thickness of the required surface compensation layer based on the irreversible lithium consumption and the lithium compensation capability parameter of the candidate compensation layer material obtained from the preset knowledge graph; and taking the minimum thickness of the surface compensation layer as the thickness of the compensation layer in the configuration parameters of the surface functional layer.
10. 10. The method of claim 1, wherein the invoking a mapping model of processes and structures stored in a preset knowledge graph for process inversion based on the bulk gradient structural features and the surface coating configuration parameters, determining a macro-scale executable process recipe, comprises: And calling a segmented sintering process and structure mapping model stored in a preset knowledge graph based on the target order gradient distribution defined in the bulk phase gradient structural features, and determining one or more segmented sintering process parameter sets for synthesizing the bulk phase material with the target gradient structure through an inversion optimization algorithm to obtain an executable process scheme.
11. 11. The method of claim 10, wherein determining one or more segmented sintering process parameter sets by an inversion optimization algorithm comprises: taking the overall deviation between the order degree distribution predicted by the minimized model and the target order degree gradient distribution as an optimization target, performing iterative optimization on 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.
12. 12. The method of claim 1, wherein the invoking the mapping model of the process and structure stored in the preset knowledge-graph to perform the process inversion based on the bulk gradient structural feature and the surface cladding layer configuration parameter, determines a macro-scale executable process recipe, further comprising: Based on the target coating thickness and the material type defined in the surface coating configuration parameters, at least one feasible coating process path is matched from the preset knowledge graph; And calling a process dynamics or thermodynamic model corresponding to the selected coating process path, and combining the thickness and the material type of the target coating layer, and carrying out inversion calculation to obtain at least one group of key process control parameters required for realizing the target to form the synthesis process path.
13. 13. The method of claim 12, wherein the coating process path comprises at least one of a vapor deposition path, a solid state reaction path, or an ion exchange path; When the cladding process path is a vapor deposition path, the critical process control parameters include reactant gas concentration, deposition temperature, and deposition time; when the coating process path is a solid state reaction path, the critical process control parameters include sintering temperature and ambient oxygen partial pressure; When the coating process path is an ion exchange path, the critical process control parameters include reaction solution concentration, reaction temperature, and reaction time.
14. 14. A positive electrode material reverse engineering apparatus, the apparatus comprising: The atomic parameter determining module is used for responding to a reverse design request for the lithium-rich manganese-based positive electrode material, performing high-flux calculation on a plurality of candidate element doping combinations by utilizing a pre-trained machine learning potential function based on target performance parameters appointed in the reverse design request, and determining an optimal element formula meeting the constraint of the target performance parameters and corresponding atomic scale key parameters; the mesoscopic parameter determining module is used for performing phase field simulation by adopting an anisotropic lattice strain model containing Jahn-Teller distortion correction based on the optimal element formula and the atomic scale key parameters to determine mesoscopic scale bulk phase gradient structural characteristics and surface coating layer configuration parameters; the macro-scale scheme determining module is used for calling a mapping model of a process and a structure stored in a preset knowledge graph to perform process inversion based on the bulk phase gradient structural characteristics and the surface coating layer configuration parameters, and determining a macro-scale executable process scheme, wherein the executable process scheme comprises a candidate coating material system and a corresponding synthesis process path.
15. 15. An electronic device, the electronic device comprising: one or more processors; storage means for storing one or more programs, When executed by the one or more processors, causes the one or more processors to implement the positive electrode material reverse design method of any one of claims 1-13.
16. 16. A storage medium containing computer executable instructions, which when executed by a computer processor are for performing the positive electrode material reverse design method of any one of claims 1-13.
17. 17. A vehicle comprising a power cell having a positive electrode that is a lithium-rich manganese-based positive electrode material designed and prepared by the method of any one of claims 1 to 13.

Description

Method, device, equipment, medium and vehicle for reverse design of anode material Technical Field The embodiment of the invention relates to the technical field of battery material preparation optimization, in particular to a method, a device, equipment, a medium and a vehicle for reversely designing a positive electrode material. Background Today, with the urgent demands of electric vehicles and large-scale energy storage for high-energy density, low-cost lithium ion batteries, lithium-rich manganese-based cathode materials are attracting attention because they exceed the theoretical specific capacity of conventional layered materials. The high capacity of such materials, partly due to the reversible redox reactions in which anions participate, is considered as one of the key candidates for achieving the breakthrough of the next generation battery technology. At present, the lithium-rich manganese-based positive electrode material is mainly developed by means of trial-and-error experiments, namely, the performance improvement is explored by adjusting technological parameters such as element types, sintering temperature and the like and combining surface coating and other modes. However, the trial-and-error development method has two main problems that firstly, experience is relied on, the development period is long, the cost is high, and secondly, the material composition, the structural design and the preparation process links are disjointed, so that the deviation between a final product and a design target is large, and the performance is difficult to meet high requirements. Disclosure of Invention The application provides a reverse design method, a device, equipment, a medium and a vehicle of a positive electrode material, which are used for converting a traditional experience-dependent trial-and-error process into an engineering accurately driven by performance parameters and cooperatively optimized in a multi-scale manner through integrating machine learning calculation, physical simulation and a process knowledge base, so that the success efficiency and the research and development capability of material design are improved. In a first aspect, an embodiment of the present application provides a method for reverse designing a cathode material, the method including: responding to a reverse design request for a lithium-rich manganese-based positive electrode material, and based on target performance parameters appointed in the reverse design request, performing high-flux calculation on a plurality of candidate element doping combinations by utilizing a pre-trained machine learning potential function, and determining an optimal element formula meeting the constraint of the target performance parameters and corresponding atomic scale key parameters; Based on the optimal element formula and the atomic scale key parameters, adopting an anisotropic lattice strain model containing Jahn-Teller distortion correction to perform phase field simulation, and determining mesoscale bulk phase gradient structural characteristics and surface coating configuration parameters; and calling a mapping model of a process and a structure stored in a preset knowledge graph to perform process inversion based on the bulk phase gradient structural characteristics and the surface coating layer configuration parameters, and determining an executable process scheme of a macro scale, wherein the executable process scheme comprises a candidate coating material system and a corresponding synthesis process path. In a second aspect, embodiments of the present application further provide a positive electrode material reverse-engineering apparatus, the apparatus including: The atomic parameter determining module is used for responding to a reverse design request for the lithium-rich manganese-based positive electrode material, performing high-flux calculation on a plurality of candidate element doping combinations by utilizing a pre-trained machine learning potential function based on target performance parameters appointed in the reverse design request, and determining an optimal element formula meeting the constraint of the target performance parameters and corresponding atomic scale key parameters; the mesoscopic parameter determining module is used for performing phase field simulation by adopting an anisotropic lattice strain model containing Jahn-Teller distortion correction based on the optimal element formula and the atomic scale key parameters to determine mesoscopic scale bulk phase gradient structural characteristics and surface coating layer configuration parameters; the macro-scale scheme determining module is used for calling a mapping model of a process and a structure stored in a preset knowledge graph to perform process inversion based on the bulk phase gradient structural characteristics and the surface coating layer configuration parameters, and determining a macro-scale executable process scheme, wherein the executable