Method, simulation method, constraint method and device for generating magnetic configuration of magnetic material

Abstract

The present invention discloses a method, a simulation method, a constraint method and a device for generating a magnetic configuration of a magnetic material, belonging to the fields of materials science and computer technology. The generation method includes: obtaining an initial magnetic configuration and an initial atomic configuration of the magnetic material, and randomly generating a variety of random configuration combinations; selecting one of them as the target magnetic configuration and the target atomic configuration, and according to the Hamiltonian of the magnetic configuration, obtaining the constraint constant and wave function of the target magnetic configuration by means of a self-consistent iteration method; determining the magnetic system energy, temperature and pressure of the target magnetic configuration and storing them; returning to select a new random configuration combination, and after all are completed, obtaining the initial magnetic configuration and the initial atomic configuration of a new different magnetic material and repeating the above steps; using a deep learning model to obtain a magnetic material with a specified phase transition temperature through a magnetic configuration library. The present invention can realize the simulation of the target magnetic configuration, form a magnetic configuration library, and guide the synthesis of the target magnetic material.

Description

Technical Field

[0001] The present invention belongs to the fields of materials science and computer technology, and particularly relates to a method, a simulation method, a constraint method and a device for generating a magnetic configuration of a magnetic material. Background Art

[0002] The properties of magnetic materials are usually generated by the combined action of magnetic configurations and atomic configurations. A magnetic configuration often corresponds to multiple possible atomic configurations because different combinations of atoms or ions and their interaction modes can produce the same magnetic behavior. Conversely, if the atomic configuration is known, we cannot directly determine its magnetic configuration because factors such as the spin and orbital motion of electrons and their superexchange interactions need to be considered.

[0003] Magnetic constraint refers to obtaining a magnetic configuration that deviates from the magnetic ground state in a magnetic material and using a local potential field to constrain the local distribution of electrons in the magnetic material. The basic principle of magnetic constraint is that electrons with different spin directions are distributed differently in space. However, the current magnetic constraint methods have poor constraint effects, can only constrain the atomic magnetic moments in the collinear representation; can only constrain either the orientation or the modulus length of the atomic magnetic moments, and cannot accurately constrain both simultaneously; the constraint algorithm depends on initial parameters, lacks adaptability to new systems and new configurations, and it is difficult to converge for some systems; it only achieves constraint through electron iteration under the energy functional, with a slow convergence speed and low accuracy; there is no method integrated with the conventional plane wave first-principles calculation software, and its high-performance algorithms and high-precision pseudopotential libraries cannot be utilized, making it difficult to use and promote; it is impossible to selectively apply constraints to certain atoms, and it is easy to introduce unreasonable restrictions on non-magnetic atoms. Summary of the Invention

[0004] This application aims to solve at least one of the technical problems in the related art to some extent.

[0005] To this end, the first object of this application is to form a magnetic configuration library, use the magnetic configuration library to guide the synthesis of magnetic materials through a deep learning model, and improve the existing magnetic constraint methods during the process of forming the magnetic configuration library. Specifically, this application measures a magnetic material using a magneto-optical Kerr measurement system to obtain the initial magnetic configuration of the magnetic material, uses an X-ray instrument to perform diffraction on the magnetic material to obtain the initial atomic configuration, uses a superconducting quantum interference device to measure the phase transition temperature of a known magnetic material, and uses the phase transition temperature, magnetic configuration, atomic configuration, magnetic system energy, temperature, and pressure of the known magnetic material to train the deep learning model, and uses the trained deep learning model to obtain a magnetic material with a specified phase transition temperature through the magnetic configuration library.

[0006] The second object of the present application is to propose a method for simulating magnetic configurations. According to the theoretically initial magnetic configurations, the constraint constants and wave functions of the target magnetic configurations are obtained by the magnetic constraint method based on the first principles, so as to realize the simulation and analysis of the target magnetic configurations.

[0007] The third object of the present application is to propose a magnetic constraint method and device based on the first principles to solve the technical problem of poor constraint effect of the current magnetic constraint methods.

[0008] To achieve the above object, the present application proposes a method for generating magnetic configurations of magnetic materials, and the method includes:

[0009] Step 1: Use a magneto-optical Kerr measurement system to measure a known magnetic material and obtain various initial magnetic configurations of the magnetic material. Turn off the magneto-optical Kerr measurement system, and then use an X-ray instrument to perform diffraction on the magnetic material to obtain various initial atomic configurations. Turn off the X-ray instrument;

[0010] Step 2: Randomly generate A random magnetic configurations and B random atomic configurations for each initial magnetic configuration and initial atomic configuration, so as to generate A×B combinations of random configurations; where both A and B are positive integers;

[0011] Step 3: Select one combination of random configurations in Step 2 as the target magnetic configuration and the target atomic configuration. According to the Hamiltonian of the magnetic configuration, use the self-consistent iteration method to obtain the constraint constants and wave functions of the target magnetic configuration;

[0012] Step 4: Determine the magnetic system energy, temperature and pressure of the target magnetic configuration according to the target magnetic configuration, the target atomic configuration, the constraint constants and wave functions of the target magnetic configuration, and store the target atomic configuration, the target magnetic configuration, the constraint constants, wave functions, magnetic system energy, temperature and pressure of the target magnetic configuration in the magnetic configuration library;

[0013] Step 5: Return to Step 3, select a new combination of random configurations in Step 2 as the target magnetic configuration and the target atomic configuration, and repeat Steps 4 to 5 until all combinations of random configurations have been selected as the target magnetic configuration and the target atomic configuration, then execute Step 6;

[0014] Step 6: After updating the magnetic material, restart the magneto-optical Kerr measurement system and the X-ray instrument, and execute Steps 1 to 5; when the number of magnetic configurations and atomic configurations in the magnetic configuration library reaches a preset value, execute Step 7;

[0015] Step 7: Measure the phase transition temperature of a known magnetic material using a superconducting quantum interference device, and train a deep learning model using the phase transition temperature, magnetic configuration, atomic configuration, and magnetic system energy, temperature, and pressure of the known magnetic material. Use the trained deep learning model to obtain magnetic materials within a specified phase transition temperature range from a magnetic configuration library.

[0016] Preferably, in step 3, a self-consistent iterative method is used to obtain the constraint constant and wave function of the target magnetic configuration, which specifically includes the following steps:

[0017] Step 331: Initialize the constraint constant v αa to 0, and obtain the target magnetic moment according to the target magnetic configuration;

[0018] Step 332: On the premise of the known constraint constant v αa , use the Hamiltonian of the initial magnetic configuration to solve the KS equation to obtain the corresponding wave function ψ as the electron density, and specifically determine the wave function ψ according to the following formula:

[0019] {H KS +H v}||ψ nσ >= ∈ n |ψ nσ >;

[0020] where H KS is the KS equation, H V is the Hamiltonian of the magnetic system, ψ nσ is the σ component of the spinor wave function, ∈ n is the eigenvalue, and n is the nth electron energy level;

[0021] Step 333: According to the wave function ψ and the constraint constant v αa , use the function M(Δv αa ) to update the magnetic moment M of the atom; where,

[0022] The function M(Δv αa ) is specifically determined according to the following formula:

[0023]

[0024] {H KS +H v +ΔH v}|ψ nσ >= ∈ n |ψ nσ >

[0025]

[0026] M αa ←∫d 3rm α (r)Θ(r cut -|r - r a |);

[0027] Among them, H V (r) is the Hamiltonian of the magnetic system, v αa is the constraint constant, is the Pauli matrix in the α direction, θ(·) is the truncation function, r cut is the truncation radius, r is the real - space coordinate, r a is the real - space coordinate of atom a, M αa is the magnetic moment of atom a in the α direction; m α (r) is the spin density in the α direction, f n is the occupancy number on orbital n, ψ nσ is the σ component of the spinor wave function, ψ nσ’ is the σ′ component of the spinor wave function; d 3 r is the three - dimensional differential symbol;

[0028] Step 334, determine whether the difference between the magnetic moment of the atom and the target magnetic moment is greater than the set threshold; if it is less than or equal, execute Step 335, if it is greater, update the constraint constant v according to the following formula αa , and return to Step 332, using the new constraint constant:

[0029]

[0030] v αa ← αa +Δv αa ;

[0031] Among them is the target magnetic moment; M is the magnetic moment of the atom, v αa is the constraint constant, is the Lagrangian coefficient optimization for M and ;

[0032] Step 335, output the constraint constant v αa and the wave function ψ at this time as the constraint constant and wave function of the target magnetic configuration.

[0033] Preferably, the specific methods for obtaining the energy, temperature, and pressure of the magnetic system are as follows:

[0034] Obtain the force on the magnetic moment according to the constraint constant, obtain the energy of the magnetic system according to the wave function and the constraint constant or according to the force on the magnetic moment; obtain the temperature and pressure of the magnetic configuration according to the force on the magnetic moment, the magnetic moment, and the atomic configuration.

[0035] Preferably, in step 7, the phase transition temperature, magnetic configuration, atomic configuration, magnetic system energy, temperature and pressure of the known magnetic material are used to train the deep learning model. Specifically:

[0036] The magnetic configuration, atomic configuration, magnetic system energy, temperature and pressure of the known magnetic material are used as inputs, and the phase transition temperature of the known magnetic material is used as the output to train the deep learning model.

[0037] Preferably, step 7 further includes: determining the synthesis process of the obtained magnetic material according to the magnetic configuration and phase transition temperature range of the obtained magnetic material.

[0038] The present invention also discloses a magnetic configuration simulation method for generating magnetic materials, including the following steps:

[0039] Step 310, obtaining an initial magnetic configuration and an initial atomic configuration;

[0040] Step 320, setting a target magnetic configuration and a target atomic configuration; obtaining a target magnetic moment according to the target magnetic configuration;

[0041] Step 330, optimizing the wave function and the constraint constant in a self-consistent iterative manner according to the Hamiltonian of the magnetic configuration, specifically including the following steps:

[0042] Step 3301, initializing the constraint constant v αa to 0, and obtaining the target magnetic moment according to the target magnetic configuration;

[0043] Step 3302, on the premise of the known constraint constant v αa , using the Hamiltonian of the initial magnetic configuration to solve the KS equation to obtain the corresponding wave function ψ as the electron density, and specifically determining the wave function ψ according to the following formula:

[0044] {H KS +H v}|ψ nσ >=∈ n |ψ nσ >;

[0045] wherein, H KS is the KS equation, H V is the Hamiltonian of the magnetic system, ψ nσ is the σ component of the spinor wave function, ∈ n is the eigenvalue, and n is the nth electron energy level;

[0046] Step 3303, according to the wave function ψ and the constraint constant v αa , using the function M(△v αa ) to update the magnetic moment M of the atom; wherein,

[0047] The function M(Δv αa ) is specifically determined according to the following formula:

[0048]

[0049] {H KS +H v +ΔH v}|ψ nσ >=∈ n |ψ nσ >

[0050]

[0051] M αa ←∫d 3 rm α (r)Θ(r cut -|r - r a |);

[0052] Among them, H V (r) is the Hamiltonian of the magnetic system, v αa is the constraint constant, is the Pauli matrix in the α direction, θ(·) is the truncation function, r cut is the truncation radius, r is the real space coordinate, r a is the real space coordinate of atom a, M αa is the magnetic moment of atom a in the α direction; m α (r) is the spin density in the α direction, f n is the occupancy number on orbital n, ψ nσ is the σ component of the spinor wave function, ψ nσ’ is the σ′ component of the spinor wave function; d 3 r is the three-dimensional differential symbol;

[0053] Step 3304, determine whether the difference between the magnetic moment of the atom and the target magnetic moment is greater than the set threshold; if it is less than or equal, execute step 3305, if it is greater, update the constraint constant v αa according to the following formula, and return to step 3302, using the new constraint constant:

[0054]

[0055] v αa ←v αa +Δv αa ;

[0056] Among them is the target magnetic moment; v αa is the constraint constant, is for M and Perform Lagrangian coefficient optimization, where M is the magnetic moment of the atom;

[0057] Step 3305, output the constraint constant v at this time αa and the wave function ψ as the constraint constant and wave function of the target magnetic configuration;

[0058] Step 340, implement the magnetic configuration simulation of the magnetic material:

[0059] Based on the constraint constant and the wave function, determine the spatial positions of all the atoms corresponding to the magnetic configuration after constraint, the final magnetic moment of each atom, the magnetic system energy of the magnetic system after constraint, the atomic force received by each atom, the magnetic interaction force received by each atom, and the full-degree-of-freedom constraint field, so as to implement the simulation of the target magnetic configuration.

[0060] Preferably, the step 330 further includes:

[0061] Add a linear energy contribution to the Hamiltonian. The linear energy contribution includes the constraint constant and the difference between the target magnetic moment and the magnetic moment of the atom. And the dimension of the constraint constant is three times the number of atoms, and the constraint constants of any atom in different directions are different;

[0062] Fix the spatial position of at least one atom in the initial magnetic configuration, or fix the magnetic moment of at least one atom in the initial magnetic configuration, so as to obtain the relaxation process of the magnetic moments of the remaining atoms.

[0063] The present invention also discloses a first-principles magnetic constraint method, which is characterized in that it includes:

[0064] Obtain the magnetic configuration and the target magnetic moment of the initial magnetic system; including obtaining the spatial positions of all atoms corresponding to the magnetic configuration of the initial magnetic system and the magnetic moment of each atom;

[0065] Based on the first principles, optimize the electron density and the constraint constant of the initial magnetic system according to the system parameters and the target magnetic moment, so as to minimize the energy of the initial magnetic system; including: optimizing the electron density and the constraint constant based on self-consistent iteration; during the self-consistent iteration, the change amount introduced in the situation area is proportional to the constraint constant, during the optimization of the electron density, the constraint constant remains unchanged; during the optimization of the constraint constant, the electron density remains unchanged;

[0066] Determine the magnetic configuration of the constrained magnetic system based on the optimized electron density and the optimized constraint constant, and determine the spatial positions of all the constrained atoms corresponding to the magnetic configuration of the constrained magnetic system, the final magnetic moment of each atom, the system energy of the constrained magnetic system, the atomic forces exerted on each atom, the magnetic interaction forces exerted on each atom, and the full-degree-of-freedom constraint field.

[0067] Preferably, during the process of optimizing the electron density and the constraint constant of the initial magnetic system, a linear energy contribution is added to the Hamiltonian. The linear energy contribution includes the constraint constant and the difference between the target magnetic moment and the magnetic moment of the atom. In the linear energy contribution, the dimension of the constraint constant is three times the number of atoms, and the constraint constants of any atom in different directions are different.

[0068] The present invention also discloses a magnetic constraint device for a magnetic constraint method for first principles, including:

[0069] An acquisition module, configured to acquire the magnetic configuration of the initial magnetic system and the target magnetic moment;

[0070] An optimization module, configured to optimize the electron density and the constraint constant of the initial magnetic system based on first principles according to the system parameters and the target magnetic moment, so as to minimize the energy of the initial magnetic system;

[0071] A determination module, configured to determine the magnetic configuration of the constrained magnetic system based on the optimized electron density and the optimized constraint constant.

[0072] In summary, the technical solution provided by the present invention at least brings the following beneficial effects:

[0073] 1) The present invention forms a magnetic configuration library. By analyzing and learning existing magnetic materials, the magnetic configuration library can be used to guide the development of target magnetic materials.

[0074] 2) Based on the acquired initial magnetic configuration and atomic configuration, the present invention can obtain the target magnetic configuration based on constraints, realizing the simulation of the target magnetic configuration.

[0075] 3) The present invention is applicable to atomic magnetic moments under non-linear representation; and has high constraint accuracy, and can improve the calculation accuracy of atomic magnetic moments to 10 -8 μB and the calculation accuracy of magnetic energy to 10 -9 eV magnitude.

[0076] 4) The present invention has a wide range of applications. For different magnetic systems, only a small number of tests are required to complete the constraint, and most parameters are self-adaptive during the constraint process. Therefore, it can effectively converge even for magnetic excited states with very high energies.

[0077] 5) In addition to the conventional first-principles calculations outputs, such as the final magnetic moment and the corresponding system energy, the present invention can also give the constraint fields with full degrees of freedom corresponding to the magnetic configurations.

[0078] 6) The present invention can perform selective constraints on some atoms or some magnetic moment components, and is used to study the relaxation process of other atomic magnetic moments during the process of fixing the magnetic moments of certain atoms.

[0079] Additional aspects and advantages of the present invention will be given in part in the following description, will become apparent in part from the following description, or will be learned through the practice of the present application. Description of the Drawings

[0080] The above-mentioned and / or additional aspects and advantages of the present application will become apparent and easy to understand from the following description of the embodiments in conjunction with the drawings, where:

[0081] Figure 1 is a flowchart of a first-principles magnetic constraint method provided by an embodiment of the present application;

[0082] Figure 2 is a schematic diagram of the distribution of energy in the material NiO provided by an embodiment of the present application with respect to the direction of the magnetic moment of the first Ni atom;

[0083] Figure 3 is a schematic diagram of the distribution of energy in the material NiO provided by an embodiment of the present application with respect to the direction of the magnetic moment of the second Ni atom;

[0084] Figure 4 is a schematic structural diagram of a first-principles magnetic constraint device provided by an embodiment of the present application;

[0085] Figure 5 is a flowchart of a method for generating a magnetic configuration of a magnetic material provided by an embodiment of the present application. Detailed Embodiments

[0086] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present application, and should not be construed as a limitation of the present application. On the contrary, the embodiments of the present application include all changes, modifications, and equivalents that fall within the spirit and connotation of the appended claims.

[0087] The present application will be described in detail below in conjunction with specific embodiments.

[0088] Figure 1 It is a flowchart of a first-principles magnetic confinement method provided by an embodiment of the present application.

[0089] As Figure 1 shown, a first-principles magnetic confinement method provided by an embodiment of the present application includes the following steps:

[0090] Step 110, obtaining the magnetic configuration and the target magnetic moment of the initial magnetic system.

[0091] Step 120, based on first principles, optimizing the electron density and the confinement constant of the initial magnetic system according to the system parameters and the target magnetic moment, so as to minimize the energy of the initial magnetic system.

[0092] Step 130, determining the magnetic configuration of the confined magnetic system based on the optimized electron density and the optimized confinement constant.

[0093] In an embodiment of the present application, obtaining the magnetic configuration of the initial magnetic system includes:

[0094] Obtaining the spatial positions of all atoms corresponding to the magnetic configuration of the initial magnetic system, and the magnetic moment of each atom.

[0095] In an embodiment of the present application, optimizing the electron density and the confinement constant of the initial magnetic system according to the system parameters and the target magnetic moment includes:

[0096] Based on self-consistent iteration, optimizing the electron density and the confinement constant.

[0097] During the self-consistent iteration process, the change amount introduced in the local situation is proportional to the confinement constant.

[0098] In an embodiment of the present application, based on self-consistent iteration, optimizing the electron density and the confinement constant includes:

[0099] During the process of optimizing the electron density, the confinement constant remains unchanged.

[0100] During the process of optimizing the confinement constant, the electron density remains unchanged.

[0101] According to some embodiments, during the process of optimizing the electron density and the confinement constant, it specifically includes the following steps:

[0102] Step 210, initializing the confinement constant v αa to 0; determining the target magnetic moment.

[0103] Step 220: Solve the KS equation according to the Hamiltonian of the system to obtain the corresponding wave function ψ, and specifically determine the wave function ψ according to the following formula:

[0104] {H KS +H v}|ψ nσ >=∈ n |ψ nσ >;

[0105] where H KS is the KS equation, H V is the Hamiltonian of the magnetic system, ψ nσ is the wave function, and ∈ n is the eigenvalue.

[0106] Step 230: Fix the wave function ψ and update the magnetic moment M of the atom using the function M(△v αa ) under the condition of fixing the constraint constant v αa ; specifically, determine the function M(△v αa ) according to the following formula:

[0107] The function M(△v αa ) is specifically determined according to the following formula:

[0108]

[0109] {H KS +H v +ΔH v}|ψ nσ >=∈ n |ψ nσ >

[0110]

[0111] M αa ←∫d 3 rm α (r)Θ(r cut -|r-r a |);

[0112] where H V (r) is the Hamiltonian of the magnetic system, v αa is the constraint constant, is the Pauli matrix in the α direction, θ(·) is the truncation function, r cut is the truncation radius, r is the real space coordinate, r a is the real space coordinate of atom a, and M αa is the magnetic moment of atom a in the α direction; m α (r) is the spin density in the α direction, f n is the occupancy number of orbital n, and ψnσ is the σ component of the spinor wave function, ψ nσ’ is the σ′ component of the spinor wave function; d 3 r is the three-dimensional differential symbol.

[0113] Step 240: Determine whether the difference between the magnetic moment of the atom and the target magnetic moment is greater than a set threshold; if not, output the magnetic moment of the atom and the constraint constant at this time; if greater, update v according to the following formula αa and repeat Step 230:

[0114]

[0115] v αa ← v αa + Δν αa ;

[0116] where is the target magnetic moment; v αa is the constraint constant, is the Lagrangian coefficient optimization for M and M is the magnetic moment of the atom.

[0117] In the embodiments of the present application, optimizing the electron density and the constraint constant of the initial magnetic system according to the system parameters and the target magnetic moment includes:

[0118] In the process of optimizing the electron density and the constraint constant, a linear energy contribution is added to the Hamiltonian, and the linear energy contribution includes the constraint constant and the difference between the target magnetic moment and the magnetic moment of the atom.

[0119] According to some embodiments, the Hamiltonian is a direct expression of the system interaction, and is derived by reusing the variational method for the energy functional with a magnetic penalty term, and is first obtained in the density functional theory (DFT) of the first principle.

[0120] In some embodiments, the system energy of the magnetic system is secondarily obtained in DFT and can be calculated by using the Harris-Foulkes functional. During the calculation process, terms such as double-counting need to be calculated.

[0121] In the embodiments of the present application, in the process of optimizing the electron density and the constraint constant, a linear energy contribution is added to the Hamiltonian, including:

[0122] In the linear energy contribution, the dimension of the constraint constant is three times the number of atoms, and the constraint constants of any atom in different directions are different.

[0123] In the embodiments of the present application, optimizing the electron density and the constraint constant of the initial magnetic system according to the system parameters and the target magnetic moment includes:

[0124] In the process of optimizing the electron density and the constraint constant of the initial magnetic system, fixing the spatial positions of at least one atom in the initial magnetic system, or fixing the magnetic moment of at least one atom in the initial magnetic system, so as to obtain the relaxation process of the magnetic moments of the remaining atoms.

[0125] In the embodiments of the present application, determining the magnetic configuration of the constrained magnetic system based on the optimized electron density and the optimized constraint constant includes:

[0126] Determining the spatial positions of all the constrained atoms corresponding to the magnetic configuration of the constrained magnetic system, the final magnetic moment of each atom, the system energy of the constrained magnetic system, the atomic force received by each atom, the magnetic interaction force received by each atom, and the full-degree-of-freedom constraint field based on the optimized electron density and the optimized constraint constant.

[0127] According to some embodiments, the magnetic interaction force received by each atom is obtained by the following two methods:

[0128] The first method is obtained by differentiating the system energy of the magnetic system with respect to the magnetic moment, and is specifically obtained according to the following formula:

[0129]

[0130] where \(E_{\lambda}\) is the system energy, \(\lambda\) is the constraint constant, and \(\psi_{\lambda}\) is the wave function in the system under the constraint condition.

[0131] The second method is obtained by the magnetic interaction force received by each atom, that is, the constraint constant of each atom, and is specifically obtained according to the following formula.

[0132]

[0133] where is the target magnetic moment, is the energy of the system when the magnetic moment is constrained in the \(\alpha\) direction of atom \(a\), \(E\) KS [p] is the system energy solved by the KS equation when the electron density is \(\rho\), \(M\) αa [p] is the magnetic moment of atom \(\rho\) in the initial magnetic system, \(V\) αa is the constraint constant.

[0134] In the embodiments of the present application, after determining the magnetic configuration of the constrained magnetic system based on the optimized electron density and the optimized constraint constant, it further includes:

[0135] Perform high-throughput calculations on the spatial positions of all constrained atoms, the system energy of the constrained magnetic system, the atomic forces exerted on each atom, and the magnetic interaction forces exerted on each atom to obtain a machine learning dataset.

[0136] According to some embodiments, high-throughput calculation refers to a process of sampling a large number of inputs composed of atomic positions and target magnetic moments within a certain space, rapidly deploying calculations using modern computing clusters and task distribution systems, and obtaining a large number of outputs of energy, atomic forces, and magnetic interactions.

[0137] In some embodiments, the machine learning dataset obtained during high-throughput calculation includes the spatial positions of atoms, the magnetic moments of atoms, the system energy of the magnetic system, the atomic forces exerted on each atom, and the magnetic interaction forces exerted on each atom.

[0138] According to some embodiments, the method proposed in the embodiments of the present application can be integrated with the commonly used plane wave DFT software VASP. After patching its source code, it can be compiled in the same way; moreover, it is equipped with corresponding calculation examples and setting templates, is easy to use, has parameter settings compatible with VASP, and does not interfere with any functions of the original VASP.

[0139] Taking a scenario as an example, using the method proposed in the embodiments of the present application, the energy distribution of material Nio as a function of the magnetic moment directions of two Ni atoms is calculated, as Figure 2 and Figure 3 shown, where dark colors represent low energy and light colors represent high energy. It can be seen that the method provided in the embodiments of the present application has high constraint accuracy.

[0140] In summary, the method proposed in the embodiments of the present application obtains the magnetic configuration and target magnetic moment of the initial magnetic system; based on the first principles, optimizes the electron density and constraint constants of the initial magnetic system according to the system parameters and the target magnetic moment, thereby minimizing the energy of the initial magnetic system; determines the magnetic configuration of the constrained magnetic system based on the optimized electron density and the optimized constraint constants. Therefore, it can be applied to atomic magnetic moments in non-linear representations; and has high constraint accuracy, which can improve the calculation accuracy of atomic magnetic moments to 10 -8 uB and the calculation accuracy of magnetic energy to 10 -9 eV order of magnitude; has a wide range of applications, only requires a small number of tests for different magnetic systems to complete the constraint, and most parameters are adaptive during the constraint process. Therefore, it can effectively converge for magnetic excited states with very high energies; in addition to the conventional first-principles calculation outputs, such as the final magnetic moment and the corresponding system energy, it can also give the constraint fields with full degrees of freedom corresponding to the magnetic configuration; it can perform selective constraints on some atoms or some magnetic moment components to study the relaxation process of other atomic magnetic moments during the process of fixing certain atomic magnetic moments.

[0141] To implement the above embodiments, the present application also proposes a first-principles magnetic confinement device.

[0142] Figure 4 The structure diagram of a first-principles magnetic confinement device provided by an embodiment of the present application.

[0143] As Figure 4 shown, a first-principles magnetic confinement device 400 includes:

[0144] An acquisition module 410, configured to acquire the magnetic configuration and the target magnetic moment of the initial magnetic system.

[0145] An optimization module 420, configured to optimize the electron density and the confinement constant of the initial magnetic system based on the first principles according to the system parameters and the target magnetic moment, so as to minimize the energy of the initial magnetic system.

[0146] A determination module 430, configured to determine the magnetic configuration of the confined magnetic system based on the optimized electron density and the optimized confinement constant.

[0147] The device proposed by the embodiment of the present application acquires the magnetic configuration and the target magnetic moment of the initial magnetic system through the acquisition module; the optimization module optimizes the electron density and the confinement constant of the initial magnetic system based on the first principles according to the system parameters and the target magnetic moment, so as to minimize the energy of the initial magnetic system; the determination module determines the magnetic configuration of the confined magnetic system based on the optimized electron density and the optimized confinement constant. Therefore, the confinement accuracy of the present application is high, and the confinement effect of the magnetic system can be significantly improved.

[0148] The present invention also proposes a method for generating the magnetic configuration of a magnetic material. The method is as Figure 5 shown and includes the following steps:

[0149] Step 1: Measure the magnetic material using a magneto-optical Kerr measurement system (MOKE) to obtain the initial magnetic configuration of the magnetic material, turn off the magneto-optical Kerr measurement system, and then perform diffraction on the magnetic material using X-rays to obtain the initial atomic configuration, and turn off the X-rays.

[0150] For a magnetic material, multiple initial magnetic configurations and initial atomic configurations can be obtained. The magnetic moment is obtained according to the magnetic configuration, and the position of each atom in the magnetic configuration is obtained according to the atomic configuration.

[0151] Step 2: Randomly generate A random magnetic configurations and B random atomic configurations for each initial magnetic configuration and initial atomic configuration, so as to generate A×B combinations of random configurations; where both A and B are positive integers.

[0152] The values of A and B can be the same or different; usually, the value ranges of A and B are [50, 200]; each magnetic material can contain multiple magnetic configurations, and each magnetic configuration can correspond to multiple atomic configurations. When randomly generating, it is preferably to use the Monte Carlo method.

[0153] Step 3: Select one of the randomly generated configuration combinations in Step 2 as the target magnetic configuration and target atomic configuration, and based on the first principles and according to the Hamiltonian of the magnetic configuration, use the self-consistent iteration method to obtain the constraint constant and wave function of the target magnetic configuration; it includes the following steps:

[0154] Step 331: Initialize the constraint constant v αa to 0, and obtain the target magnetic moment according to the target magnetic configuration.

[0155] Step 332: On the premise of knowing the constraint constant v αa , use the Hamiltonian of the initial magnetic configuration to solve the KS equation to obtain the corresponding wave function ψ as the electron density, and specifically determine the wave function ψ according to the following formula:

[0156] {H KS +H v}|ψ nσ > = ∈ n |ψ nσ >;

[0157] where H KS is the KS equation, H V is the Hamiltonian of the magnetic system, ψ nσ is the σ component of the spinor wave function, ∈ n is the eigenvalue, and n is the nth electron energy level.

[0158] Step 333: According to the wave function ψ and the constraint constant v αa , use the function M(△v αa ) to update the magnetic moment M of the atom; where

[0159] the function M(△v αa ) is specifically determined according to the following formula:

[0160]

[0161] {H KS +H v +ΔH v}|ψ nσ > = ∈ n |ψ nσ >

[0162]

[0163] M αa←∫d 3 rm α (r)Θ(r cut -|r - ra|);

[0164] where H V (r) is the Hamiltonian of the magnetic system, v αa is the constraint constant, is the Pauli matrix in the α direction, θ(·) is the cutoff function, r cut is the cutoff radius, r is the real - space coordinate, r a is the real - space coordinate of the a atom, M αa is the magnetic moment of the a atom in the α direction; m α (r) is the spin density in the α direction, f n is the occupancy number on the orbital n, ψ nσ is the σ component of the spinor wave function, ψ nσ’ is the σ′ component of the spinor wave function; d3r is the three - dimensional differential symbol.

[0165] Step 334, determine whether the difference between the magnetic moment of the atom and the target magnetic moment is greater than the set threshold; if the difference between the magnetic moment of the atom and the target magnetic moment is not greater than the set threshold, then execute Step 335, if it is greater, then update the constraint constant vαa according to the following formula, and return to Step 332, using the new constraint constant:

[0166]

[0167] v αa ←v αa +Δv αa ;

[0168] where is the target magnetic moment; vαa is the constraint constant, is the Lagrangian coefficient optimization for M and , M is the magnetic moment of the atom. In a preferred embodiment, the value of the set threshold can be 10 -4 Bohr magnetons.

[0169] In the above calculation, a linear energy contribution is added to the Hamiltonian. The linear energy contribution includes the constraint constant and the difference between the target magnetic moment and the magnetic moment of the atom. The dimension of the constraint constant in the linear energy contribution is three times the number of atoms, and the constraint constants of any atom in different directions are different. The change amount introduced in the local situation is proportional to the constraint constant.

[0170] Step 335, output the constraint constant v αa and the wave function ψ at this time; the constraint constant and the wave function at this time are the constraint constant and the wave function of the target magnetic configuration.

[0171] Step 4: Determine the magnetic system energy, temperature, and pressure of the target magnetic configuration based on the target magnetic configuration, target atomic configuration, constraint constant of the target magnetic configuration, and wave function, and store the target atomic configuration, target magnetic configuration, constraint constant of the target magnetic configuration, wave function, magnetic system energy, temperature, and pressure in the magnetic configuration library. It should be noted that the temperature and pressure here refer to the temperature and pressure at which the target magnetic configuration can exist.

[0172] Obtain the magnetic moment force based on the constraint constant, and the magnetic moment force is also known as the magnetic interaction force exerted on the atom; obtain the magnetic system energy based on the wave function and constraint constant, or obtain the magnetic system energy based on the magnetic moment force; obtain the temperature and pressure based on the magnetic moment force, magnetic moment, and atomic configuration.

[0173] Step 5: Return to Step 3, select a new random configuration combination in Step 2 as the target magnetic configuration and target atomic configuration until all random configuration combinations have been selected as the target magnetic configuration and target atomic configuration, then execute Step 6, where the new random configuration combination means a random configuration combination that has not been used in Step 3 before.

[0174] Step 6: After updating the magnetic material, restart the magneto-optical Kerr measurement system and the X-ray instrument, and execute Steps 1 to 5; when the number of magnetic configurations and atomic configurations in the magnetic configuration library reaches the preset value, execute Step 7. In a preferred embodiment, the preset value can be 2000.

[0175] Step 7: Use a superconducting quantum interference device such as SQUID to measure the phase transition temperature of a known, i.e., existing magnetic material, and use the phase transition temperature, magnetic configuration, atomic configuration, magnetic system energy, temperature, and pressure of the known magnetic material to train a deep learning model. Use the trained deep learning model to obtain magnetic materials in a specified phase transition temperature range through the magnetic configuration library. Subsequently, the synthesis process can be determined based on the magnetic configuration and phase transition temperature of the magnetic material. In a preferred embodiment, the specified phase transition temperature range can be, for example: 600 - 700 0 C.

[0176] When training the deep learning model, use the magnetic configuration, atomic configuration, magnetic system energy, temperature, and pressure of the known magnetic material as the input, and the phase transition temperature as the output to train the deep learning model.

[0177] In another aspect of the present invention, a magnetic configuration simulation method for generating magnetic materials is disclosed. The simulation method includes the following steps:

[0178] Step 310: Obtain the initial magnetic configuration and the initial atomic configuration;

[0179] The initial magnetic configuration is in an ideal state environment of absolute zero temperature, zero pressure, and zero magnetic field. The magnetic moment is obtained based on the magnetic configuration, and the positions of each atom in the magnetic configuration are obtained based on the atomic configuration.

[0180] Step 320: Set the target magnetic configuration and the target atomic configuration; obtain the target magnetic moment according to the target magnetic configuration.

[0181] Step 330: Optimize the electron density and the constraint constant in a self-consistent iterative manner according to the Hamiltonian of the magnetic configuration, specifically including the following steps:

[0182] Step 3301: Initialize the constraint constant v αa to be 0, and obtain the target magnetic moment according to the target magnetic configuration.

[0183] Step 3302: On the premise of knowing the constraint constant v αa , solve the KS equation using the Hamiltonian of the initial magnetic configuration to obtain the corresponding wave function ψ as the electron density, and specifically determine the wave function ψ according to the following formula:

[0184] {H KS +H v}|ψ nσ > = ∈ n |ψ nσ >;

[0185] where H KS is the KS equation, H V is the Hamiltonian of the magnetic system, ψ nσ is the σ component of the spinor wave function, ∈ n is the eigenvalue, and n is the nth electron energy level.

[0186] Step 3303: Update the magnetic moment M of the atom using the function M(△v αa ) according to the wave function ψ and the constraint constant v αa ; where

[0187] the function M(△v αa ) is specifically determined according to the following formula:

[0188]

[0189] {H KS +H v +ΔH v}|ψ nσ > = ∈ n |ψ nσ >

[0190]

[0191] M αa←∫d 3 rm α (r)Θ(r cut -|r - r a |);

[0192] Among them, H V (r) is the Hamiltonian of the magnetic system, v αa is the constraint constant, is the Pauli matrix in the α direction, θ(·) is the truncation function, r cut is the truncation radius, r is the real - space coordinate, r a is the real - space coordinate of the a atom, M αa is the magnetic moment of the a atom in the α direction; m α (r) is the spin density in the α direction, f n is the occupancy number on the orbital n, ψ nσ is the σ component of the spinor wave function, ψ nσ’ is the σ' component of the spinor wave function; d 3 r is the three - dimensional differential symbol.

[0193] Step 3304, determine whether the difference between the magnetic moment of the atom and the target magnetic moment is greater than the set threshold; if not, execute Step 335, if so, update the constraint constant v according to the following formula αa , and return to Step 332, using the new constraint constant:

[0194]

[0195] v αa ←v αa +Δv αa ;

[0196] Among them is the target magnetic moment; v αa is the constraint constant, is the Lagrangian coefficient optimization for M and , M is the magnetic moment of the atom.

[0197] Preferably, a linear energy contribution is added to the Hamiltonian. The linear energy contribution includes the constraint constant and the difference between the target magnetic moment and the magnetic moment of the atom, and the dimension of the constraint constant is three times the number of atoms. The spatial position of at least one atom in the initial magnetic configuration is fixed, or the magnetic moment of at least one atom in the initial magnetic configuration is fixed, so as to obtain the relaxation process of the magnetic moments of the remaining atoms, where the remaining atoms refer to the atoms other than the fixed atoms in the initial magnetic configuration.

[0198] Step 3305, output the constraint constant v at this time αaand the wave function ψ; the constraint constant and the wave function at this time are the constraint constant and the wave function of the target magnetic configuration.

[0199] Step 340, perform a simulation of the magnetic configuration of the magnetic material.

[0200] Based on the constraint constant and the wave function, determine the spatial positions of all the constrained atoms corresponding to the magnetic configuration, the final magnetic moment of each atom, the magnetic system energy of the constrained magnetic system, the atomic force received by each atom, the magnetic interaction force received by each atom, and the full-degree-of-freedom constraint field, so as to realize the simulation of the target magnetic configuration.

[0201] The self-consistent iteration method adopted by the present invention optimizes the electron density and the constraint constant. Taking the magnitude and direction of the atomic magnetic moment as free variables, the constraint parameters are adaptively adjusted to realize efficient and accurate calculation of the system energy and the wave function. The self-consistent iteration method can provide information about the spin-lattice interaction, which helps to guide the preparation of the magnetic excitation state of the material. In addition, it can also be used to calculate the electronic structure of specific magnetic excitation states and even spin fluctuation configurations, and its effects include the orbital characteristics and topological characteristics of the energy band structure and the Fermi surface, which can be used to study basic physical problems such as spin and electron interactions, and guide rich functional designs in terms of dynamic or optical properties.

[0202] Using the method proposed by the present invention, the energy path and dynamic behavior of the AFM-FM phase transition of antiferromagnetic NiO bulk under a uniform external field are studied. It is found that the Ni atomic magnetic moment has both the power of tangential rotation in the direction of the external field and the power of radial stretching along the external field. And under a strong external field, the latter trend is much stronger than the former. In addition, it is found that the potential barrier for the phase transition of the periodically restricted single-crystal Nio is significantly higher, indicating the magnetic domain mechanism in its actual transformation process. In the spin-lattice dynamics potential function model based on deep learning, using the magnetic configuration generated by this method as the generation end of the high-throughput self-consistent data set can greatly improve the training speed. The energy prediction of the NiO magnetic excitation state using the spin-lattice dynamics potential function model based on deep learning shows that the model has high prediction accuracy, can simulate the spin-lattice dynamics of large-scale systems, and has good extrapolation ability.

[0203] It should be understood that each part of the present application can be implemented by hardware, software, firmware, or a combination thereof. In particular, the data of the present application can be obtained by a variety of devices, and the magnetic configuration obtained by the method of the present application can not only be used with a variety of instruments to study existing magnetic materials, but also guide the synthesis of new magnetic materials with better phase transition temperatures in fields such as magnetic materials for electric vehicle engines, magnetic materials for robot motors, and magnetic materials for computer hard disks.

[0204] Moreover, in the above-described embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

[0205] Those of ordinary skill in the art can understand that all or part of the steps carried by the method of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0206] In addition, in each embodiment of the present application, each functional unit can be integrated in a processing module, or each unit can exist physically alone, or two or more units can be integrated in a module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0207] The above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disk, or the like.

[0208] Any process or method description shown in a flowchart or described in other ways herein can be understood to represent a module, segment, or part of code including one or more executable instructions for implementing a specific logical function or process. The scope of the preferred embodiments of the present application includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed. This should be understood by those skilled in the art to which the embodiments of the present application belong.

[0209] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0210] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for generating a magnetic configuration of a magnetic material, characterized in that: It includes the following steps: Step 1: Use a magneto-optical Kerr measurement system to measure a known magnetic material and obtain a plurality of initial magnetic configurations of the magnetic material, turn off the magneto-optical Kerr measurement system, then use an X-ray instrument to diffract the magnetic material to obtain a plurality of initial atomic configurations, and turn off the X-ray instrument; Step 2, randomly generating A random magnetic configurations and B random atomic configurations for each initial magnetic configuration and initial atomic configuration, thereby generating A×B random configuration combinations; wherein A and B are both positive integers; Step 3, selecting a random configuration combination in step 2 as the target magnetic configuration and target atomic configuration, and obtaining the constraint constant and wave function of the target magnetic configuration by a self-consistent iteration method according to the Hamiltonian of the magnetic configuration; Step 4, determining the magnetic system energy, temperature and pressure of the target magnetic configuration according to the target magnetic configuration, the target atomic configuration, the constraint constant and the wave function of the target magnetic configuration, and storing the target atomic configuration, the target magnetic configuration, the constraint constant, the wave function, the magnetic system energy, the temperature and the pressure of the target magnetic configuration in a magnetic configuration library; Step 5, return to step 3, select a new random configuration combination in step 2 as the target magnetic configuration and target atomic configuration and repeat steps 4 to 5 until all random configuration combinations have been selected as the target magnetic configuration and target atomic configuration, then execute step 6; Step 6, after the magnetic material is updated, the magneto-optical Kerr measurement system and the X-ray instrument are restarted, and steps 1 to 5 are executed; when the number of magnetic configurations and atomic configurations in the magnetic configuration library reaches a preset value, step 7 is executed; Step 7: Use a superconducting quantum interference device to measure the phase transition temperature of a known magnetic material, and use the phase transition temperature, magnetic configuration, atomic configuration, and magnetic system energy, temperature, and pressure of the known magnetic material to train a deep learning model. Use the trained deep learning model to obtain magnetic materials in a specified phase transition temperature range through a magnetic configuration library.

2. The method for generating a magnetic configuration of a magnetic material according to claim 1, characterized in that: In step 3, the constraint constants and wave functions of the target magnetic configuration are obtained by a self-consistent iteration method, which specifically includes the following steps: Step 331, initialize constraint constant v αa is 0, and the target magnetic moment is obtained according to the target magnetic configuration; Step 332, given the known constraint constant v αa Under the premise of using the Hamiltonian of the initial magnetic configuration to solve the KS equation, the corresponding wave function ψ as the electron density is obtained. The wave function ψ is determined according to the following formula: {H KS +H v }|ψ nσ >=∈ n |ψ nσ >; Among them, H KS is the KS equation, H V is the Hamiltonian of the magnetic system, ψ nσ is the σ component of the spin sub-wave function, ∈ n is the eigenvalue, n is the nth electronic energy level; Step 333, according to the wave function ψ and the constraint constant v αa , using the function M(△v αa ) updates the magnetic moment M of the atom; where, Function M(△v αa ) is determined according to the following formula: {H KS +H v +ΔH v }|ψ nσ >=∈ n |ψ nσ > M αa ←∫d 3 rm α (r)Θ(r cut -|r-r a |); Among them, H V (r) is the Hamiltonian of the magnetic system, v αa is the constraint constant, is the Pauli matrix in the α direction, θ(·) is the cutoff function, r cut is the cutoff radius, r is the real space coordinate, r a is the real space coordinate of atom a, M αa is the magnetic moment of atom in the α direction; m α (r) is the spin density in the α direction, f n is the occupation number of orbital n, ψ nσ is the σ component of the spin sub-wave function, ψ nσ ' is the σ' component of the spin sub-wave function; d 3 r is the three-dimensional differential symbol; Step 334, determine whether the difference between the magnetic moment of the atom and the target magnetic moment is greater than a set threshold; if it is less than or equal to, execute step 335; if it is greater, update the constraint constant v according to the following formula αa , and return to step 332, using the new constraint constants: v αa ←v αa +Δv αa in is the target magnetic moment; M is the magnetic moment of the atom, v αa is the constraint constant, For M and Optimize the Lagrangian coefficients; Step 335, output the constraint constant v at this time αa and wave function ψ serve as the constraint constant and wave function of the target magnetic configuration.

3. The method for generating a magnetic configuration of a magnetic material according to claim 1, characterized in that: The specific method for obtaining the energy, temperature and pressure of the magnetic system is: The force on the magnetic moment is obtained based on the constraint constant, and the energy of the magnetic system is obtained based on the wave function and the constraint constant or based on the force on the magnetic moment; the temperature and pressure of the magnetic configuration are obtained based on the force on the magnetic moment, the magnetic moment and the atomic configuration.

4. The method for generating a magnetic configuration of a magnetic material according to claim 1, characterized in that: In step 7, the phase transition temperature, magnetic configuration, atomic configuration, magnetic system energy, temperature and pressure of the known magnetic material are used to train the deep learning model, specifically: The deep learning model is trained by taking the magnetic configuration, atomic configuration, magnetic system energy, temperature and pressure of the known magnetic material as input and the phase transition temperature of the known magnetic material as output.

5. The method for generating a magnetic configuration of a magnetic material according to claim 1, characterized in that: Step 7 also includes: determining the synthesis process of the obtained magnetic material according to the magnetic configuration and phase transition temperature range of the obtained magnetic material.

6. A method for simulating the magnetic configuration of a magnetic material, characterized in that: It includes the following steps: Step 310, obtaining an initial magnetic configuration and an initial atomic configuration; Step 320, setting a target magnetic configuration and a target atomic configuration; obtaining a target magnetic moment according to the target magnetic configuration; Step 330, according to the Hamiltonian of the magnetic configuration, the wave function and the constraint constant are optimized by a self-consistent iteration method, which specifically includes the following steps: Step 3301, initialize constraint constant v αa is 0, and the target magnetic moment is obtained according to the target magnetic configuration; Step 3302, given the known constraint constant v αa Under the premise of using the Hamiltonian of the initial magnetic configuration to solve the KS equation, the corresponding wave function ψ as the electron density is obtained. The wave function ψ is determined according to the following formula: {H KS +H v }|ψ nσ >=∈ n |ψ nσ >; Among them, H KS is the KS equation, H V is the Hamiltonian of the magnetic system, ψ nσ is the σ component of the spin sub-wave function, ∈ n is the eigenvalue, n is the nth electronic energy level; Step 3303, according to the wave function ψ and the constraint constant v αa , using the function M(△v αa ) updates the magnetic moment M of the atom; where, Function M(△v αa ) is determined according to the following formula: {H KS +H v +ΔH v }|ψ nσ >=∈ n |ψ nσ > M αa ←∫d 3 rm α (r)Θ(r cut -|r-r a |); Among them, H V (r) is the Hamiltonian of the magnetic system, v αa is the constraint constant, is the Pauli matrix in the α direction, θ(·) is the cutoff function, r cut is the cutoff radius, r is the real space coordinate, r a is the real space coordinate of atom a, M αa is the magnetic moment of atom in the α direction; m α (r) is the spin density in the α direction, f n is the occupation number of orbital n, ψ nσ is the σ component of the spin sub-wave function, ψ nσ ' is the σ' component of the spin sub-wave function; d 3 r is the three-dimensional differential symbol; Step 3304, determine whether the difference between the atomic magnetic moment and the target magnetic moment is greater than a set threshold; if less than or equal to, execute step 3305; if greater than, update the constraint constant v according to the following formula αa , and return to step 3302, using the new constraint constants: v αa ←v αa +Δv αa ; in is the target magnetic moment; v αa is the constraint constant, For M and Optimize the Lagrangian coefficient, where M is the magnetic moment of the atom; Step 3305, output the constraint constant v at this time αa and wave function ψ as the constraint constant and wave function of the target magnetic configuration; Step 340, realizing magnetic configuration simulation of magnetic materials: Based on the constraint constants and wave functions, the spatial positions of all constrained atoms corresponding to the magnetic configuration, the final magnetic moment of each atom, the magnetic system energy of the constrained magnetic system, the atomic force exerted on each atom, the magnetic interaction force exerted on each atom and the full-degree-of-freedom constraint field are determined to achieve the simulation of the target magnetic configuration.

7. The method for simulating the magnetic configuration of a magnetic material according to claim 6, characterized in that: The step 330 also includes: Adding a linear energy contribution to the Hamiltonian, wherein the linear energy contribution includes the constraint constant and the difference between the target magnetic moment and the magnetic moment of the atom, and the dimension of the constraint constant is three times the number of atoms, and the constraint constant of any atom in different directions is different; The spatial position of at least one atom in the initial magnetic configuration is fixed, or the magnetic moment of at least one atom in the initial magnetic configuration is fixed, so as to obtain a relaxation process of the magnetic moments of the remaining atoms.

8. A first-principles magnetic confinement method, characterized in that: It includes: Obtaining the magnetic configuration and target magnetic moment of the initial magnetic system; The method comprises obtaining the spatial positions of all atoms corresponding to the magnetic configuration of the initial magnetic system and the magnetic moment of each atom; Based on the first principles, the electron density and constraint constant of the initial magnetic system are optimized according to the system parameters and the target magnetic moment, so as to minimize the energy of the initial magnetic system; including: optimizing the electron density and constraint constant based on self-consistent iteration; in the process of self-consistent iteration, the change introduced in the situation domain is proportional to the constraint constant, and in the process of optimizing the electron density, the constraint constant remains unchanged; in the process of optimizing the constraint constant, the electron density remains unchanged; The magnetic configuration of the constrained magnetic system is determined based on the optimized electron density and the optimized constraint constants. The spatial positions of all constrained atoms, the final magnetic moment of each atom, the system energy of the constrained magnetic system, the atomic force exerted on each atom, the magnetic interaction force exerted on each atom and the full-degree-of-freedom constraint field corresponding to the magnetic configuration of the constrained magnetic system are determined based on the optimized electron density and the optimized constraint constants.

9. The first principle magnetic confinement method according to claim 8, characterized in that: In the process of optimizing the electron density and constraint constant of the initial magnetic system, a linear energy contribution is added to the Hamiltonian. The linear energy contribution includes the constraint constant and the difference between the target magnetic moment and the magnetic moment of the atom. In the linear energy contribution, the dimension of the constraint constant is three times the number of atoms, and the constraint constant of any atom in different directions is different.

10. A magnetic confinement device used in the magnetic confinement method according to the first principle of claim 8, characterized in that: It includes: An acquisition module, used to acquire the magnetic configuration and target magnetic moment of the initial magnetic system; An optimization module, for optimizing the electron density and constraint constant of the initial magnetic system based on the first principles according to the system parameters and the target magnetic moment, so as to minimize the energy of the initial magnetic system; A determination module is used to determine the magnetic configuration of the constrained magnetic system based on the optimized electron density and the optimized constraint constant.

Images