Simulation method, simulation device, and program

Abstract

To provide a simulation method capable of suppressing an increase in calculation load and analyzing a hysteresis loop.SOLUTION: The simulation method includes: generating a magnetic material model consisting of a collection of a plurality of super particles less than the original number of atoms by coarse-graining a plurality of atoms constituting a magnetic material to be simulated; giving a magnetic moment to each of the super particles; and changing an external magnetic field to execute, a plurality of times, a magnetization calculation procedure for calculating magnetization of the magnetic material model when a magnetic field is given to the magnetic material model. The magnetization calculation procedure executes, a plurality of times until the magnetization of the magnetic material model converges, a time step loop of calculating total magnetic fields acting on each of the plurality of super particles on the basis of the external magnetic field and the magnetic moment of each of the super particles, changing the magnetic moment of each of the super particles on the basis of the total magnetic fields acting on each of the super particles, and calculating the magnetization of the magnetic material model on the basis of the changed magnetic moment.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a simulation method, a simulation device, and a program. [Background technology]

[0002] To calculate losses in electromagnetic steel sheets, etc., a method is used that combines a method for solving the magnetic field distribution in a geometric model based on Maxwell's equations (magnetic field analysis method) with a magnetic material model that generates magnetization hysteresis.

[0003] Typical magnetic field analysis methods include the finite element method and finite volume method, which divide a geometric model into meshes, and a solution method using discrete domain division called the magnetic moment method.

[0004] A magnetic material model is assigned to a divided micro-region for use in magnetic field analysis methods, and is composed of the magnetic field generated in the micro-region, as well as multiple magnetizations, magnetic moments, spins, etc. Conventional magnetic material models generally return magnetization corresponding to an externally applied magnetic field, and known models include play models based on experimental data and micromagnetic models in which a magnetic material model is further divided into minute mesh regions and magnetization is calculated from first principles as a magnetic continuum.

[0005] When using a play model, experimental data is required, so a database must be created that matches the conditions under which the calculation target is placed. However, in reality, the conditions that can be prepared in the database do not necessarily match the conditions under which the calculation target is actually placed. This makes it difficult to improve calculation accuracy. For example, the hysteresis loss that occurs in the electromagnetic steel sheets used in motors depends on stress, and is strongly affected by the accuracy of the calculation or evaluation of the pressure distribution inside the motor.

[0006] In the micromagnetics method (Patent Document 1), a magnetic material is divided into very small meshes of about 10 nm, and analysis is performed using the finite element method. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Patent No. 5556882 Summary of the Invention [Problem to be solved by the invention]

[0008] To calculate the hysteresis loss of a magnetic material, the size of the magnetic material model must be set to at least several micrometers, which is the magnetic domain width, in order to clarify the effect of magnetocrystalline anisotropy interactions, which are the main factor in forming hysteresis loops. If the dimensions of the magnetic material model analyzed using micromagnetics are set to about several micrometers and this magnetic material model is divided into extremely small meshes of about 10 nm, the number of meshes becomes enormous, increasing the calculation load.

[0009] An object of the present invention is to provide a simulation method, a simulation device, and a program that are capable of suppressing an increase in the calculation load and analyzing a hysteresis loop. [Means for solving the problem]

[0010] According to one aspect of the present invention, a processing unit of the simulation device, By coarse-graining the atoms that make up the magnetic material to be simulated, a magnetic material model consisting of a collection of multiple superparticles, the number of which is less than the original number of atoms, is generated. The processing unit imparting a magnetic moment to each of the plurality of superparticles; The processing unit A simulation method for calculating magnetization of a magnetic material model when an external magnetic field is applied to the magnetic material model, the method comprising: executing a magnetization calculation procedure multiple times while changing the external magnetic field; The magnetization calculation procedure includes: a step of executing a time step loop multiple times until the magnetization of the magnetic material model converges, the time step loop including: calculating a total magnetic field acting on each of the plurality of superparticles based on the external magnetic field and the magnetic moments of the plurality of superparticles; changing the magnetic moment of each of the plurality of superparticles based on the total magnetic field acting on each of the plurality of superparticles; and calculating the magnetization of the magnetic material model based on the magnetic moments of the plurality of superparticles after the change; a step of determining the magnetization of the magnetic material model after the magnetization of the magnetic material model has converged; a step of storing the external magnetic field and the determined magnetization of the magnetic material model in association with each other; Including, In the magnetization calculation procedure, there is provided a simulation method in which the magnetic moments of the plurality of super-particles at the end of the immediately preceding magnetization calculation procedure are given as initial conditions.

[0011] According to another aspect of the present invention, an input unit to which simulation conditions including coarse-grained conditions are input; a processing unit that determines a relationship between an external magnetic field applied to a magnetic body to be simulated and the magnetization of the magnetic body based on the simulation conditions inputted to the input unit; Output section and Equipped with The processing unit The atoms constituting the magnetic material are coarse-grained based on the input coarse-graining conditions, thereby forming a magnetic material with a number of atoms smaller than the original number. Multiple A function to generate a magnetic model consisting of a collection of superparticles, a function of imparting a magnetic moment to each of the plurality of superparticles; a function of executing a magnetization calculation procedure for calculating the magnetization of the magnetic material model when the external magnetic field is applied to the magnetic material model multiple times while changing the external magnetic field; and The magnetization calculation procedure includes: a step of executing a time step loop multiple times until the magnetization of the magnetic material model converges, the time step loop including: calculating a total magnetic field acting on each of the plurality of superparticles based on the external magnetic field and the magnetic moments of the plurality of superparticles; changing the magnetic moment of each of the plurality of superparticles based on the total magnetic field acting on each of the plurality of superparticles; and calculating the magnetization of the magnetic material model based on the magnetic moments of the plurality of superparticles after the change; a step of determining the magnetization of the magnetic material model after the magnetization of the magnetic material model has converged; a step of storing the external magnetic field and the determined magnetization of the magnetic material model in association with each other; Including, The processing unit may be a simulation device that provides, as an initial condition in the magnetization calculation procedure, the magnetic moments of the plurality of super-particles at the end of the immediately preceding magnetization calculation procedure.

[0012] According to yet another aspect of the present invention, A function to generate a magnetic material model consisting of a collection of superparticles with a number smaller than the original number of atoms by coarse-graining the multiple atoms that make up the magnetic material to be simulated; a function of imparting a magnetic moment to each of the plurality of superparticles of the magnetic material model; a function of executing a magnetization calculation procedure for calculating the magnetization of the magnetic material model when an external magnetic field is applied to the magnetic material model multiple times while changing the external magnetic field; A program for causing a computer to realize the above, The magnetization calculation procedure includes: a step of executing a time step loop multiple times until the magnetization of the magnetic material model converges, the time step loop including: calculating a total magnetic field acting on each of the plurality of superparticles based on the external magnetic field and the magnetic moments of the plurality of superparticles; changing the magnetic moment of each of the plurality of superparticles based on the total magnetic field acting on each of the plurality of superparticles; and calculating the magnetization of the magnetic material model based on the magnetic moments of the plurality of superparticles after the change; a step of determining the magnetization of the magnetic material model after the magnetization of the magnetic material model has converged; a step of storing the external magnetic field and the determined magnetization of the magnetic material model in association with each other; Including, In the magnetization calculation procedure, a program is provided that provides the magnetic moments of the plurality of super-particles at the end of the immediately preceding magnetization calculation procedure as initial conditions. [Effects of the Invention]

[0013] By generating a magnetic model consisting of a collection of multiple superparticles by coarse-graining the atoms that make up the magnetic material, the computational load of the magnetic field can be reduced. By calculating the magnetization of the magnetic material model while changing the external magnetic field, the hysteresis loop can be reproduced. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1A is a diagram schematically showing a plurality of atoms constituting a magnetic body to be simulated, and FIG. 1B is a diagram schematically showing a magnetic body model generated by coarse-graining the plurality of atoms constituting the magnetic body shown in FIG. 1A. [Figure 2] Figure 2 is a schematic diagram of two superparticles to explain the parameters V, W, and S that define the Hamiltonian of the exchange interaction between the superparticles. [Figure 3] FIG. 3 is a block diagram of a simulation device according to an embodiment. [Figure 4] FIG. 4 is a flowchart showing the procedure of the simulation method according to the embodiment. [Figure 5] FIG. 5 is a graph showing an example of the magnetization calculation results. [Figure 6] FIG. 6 is a flowchart showing the procedure of magnetization calculation (step S3) in the simulation method according to the embodiment. [Figure 7] FIG. 7 is a graph showing an example of the change over time in normalized total magnetization M / Ms. [Figure 8]FIG. 8 is a graph showing the results of an actual simulation performed using the method according to this embodiment. [Figure 9] FIG. 9 is a graph showing other results of an actual simulation performed using the method according to this embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0015] [Coarse-graining of atoms that make up magnetic materials] In a simulation method according to an embodiment of the present invention, the atoms constituting a magnetic material are coarse-grained to generate a magnetic material model consisting of a collection of superparticles with a number smaller than the original number of atoms. This reduces the calculation load. First, the method for coarse-graining atoms will be described with reference to Figures 1A and 1B.

[0016] 1A is a diagram schematically showing a plurality of atoms 11 constituting a magnetic body 10 to be simulated. In reality, the plurality of atoms 11 are distributed three-dimensionally within the magnetic body 10 according to the crystal structure (simple cubic lattice, face-centered cubic lattice, body-centered cubic lattice, etc.), but FIG. 1A shows an example in which the plurality of atoms 11 are distributed two-dimensionally. FIG. 1A may be considered as a plurality of atoms 11 located on a single imaginary plane within the magnetic body 10.

[0017] FIG. 1B is a schematic diagram showing a magnetic body model 20 generated by coarse-graining the multiple atoms 11 that make up the magnetic body 10 shown in FIG. 1A. The magnetic body model 20 is made up of a collection of coarse-grained superparticles 21, the number of which is less than the number of atoms in the original magnetic body 10. A magnetic moment μ is assigned to each of the multiple superparticles 21 based on the atomic spin s of the atoms 11 of the magnetic body 10. In the calculation, the magnetic moment μ of the superparticle 21 is assumed to be, for example, a unit vector with a length of 1.

[0018] The total magnetic field h acting on the i-th superparticle 21 i can be calculated by the following formula:

number

[0019] external magnetic field h i ext occurs throughout the entire region of interest and is given as a simulation condition. The magnetic field due to the exchange interaction, h i exch can be calculated using the following formula, as described in, for example, JP 2021-110997 A.

number

number

[0020] Figure 2 is a schematic diagram of two superparticles 21 to explain the parameters V, W, and S. The i-th superparticle 21i and the j-th superparticle 21j are adjacent to each other. V on the right side of equation (3) represents the volume of the superparticle 21. S represents the surface area of ​​the i-th superparticle 21i within the solid angle Ω seen from the center O of the i-th superparticle 21i to the j-th superparticle 21j. W is a parameter with a length dimension. For example, the value of W can be the thickness of one atomic layer located on the surface of the i-th superparticle 21i. In this case, the value of W is equal to the diameter of the atom 11 in the magnetic material 10 (Figure 1A). In Figure 2, the area corresponding to the volume of W·S is hatched.

[0021] Magnetic field h due to magnetic dipole interaction i dip can be calculated by the following formula:

number

[0022] Magnetic field h due to triaxial magnetocrystalline anisotropy interaction i anis can be calculated by the following formula:

number

[0023] demagnetizing field h i demag can be calculated by the following formula:

number

[0024]

number

[0025] Thermal fluctuation magnetic field h i th The calculation method is described, for example, in Japanese Patent Application Laid-Open No. 2021-110997, and can be calculated using the following formula.

number

number

[0026] In equation (8), λ is the particle expansion ratio, which is defined as the ratio of the radius of the superparticle 21 to the radius of the atom 11. The function f(λ) in equation (8) is defined by the following equation:

number

[0027] The total magnetic field h acting on the i-th superparticle 21 i Once (Equation (1)) is found, the magnetic moment μ of the i-th superparticle 21 i The time variation of can be expressed by the Landau-Lifshitz-Gilbert equation (LLG equation) below.

number

[0028] Magnetic moment μ at time t+Δt i (t+Δt) is the magnetic moment μ at time t i It is expressed by the following formula using (t):

number

[0029] [Simulation device] 3 is a block diagram of a simulation device according to an embodiment. The simulation device according to the embodiment includes an input unit 50, a processing unit 51, an output unit 52, and a storage unit 53. Simulation conditions and the like are input from the input unit 50 to the processing unit 51. Furthermore, various instructions (commands) and the like are input from an operator to the input unit 50. The input unit 50 is composed of, for example, a communication device, a removable media reader, a keyboard, and the like.

[0030] The processing unit 51 performs simulation calculations based on input simulation conditions and commands. The processing unit 51 is realized by a computer including a central processing unit (CPU), a main memory, etc. A simulation program executed by the computer is stored in the storage unit 53. The storage unit 53 may be, for example, a hard disk drive (HDD), a solid state drive (SSD), etc. The processing unit 51 reads the program stored in the storage unit 53 into the main memory and executes it.

[0031] The processing unit 51 outputs the simulation results to the output unit 52. The output unit 52 includes, for example, a communication device, a removable media writing device, a display, a printer, and the like.

[0032] [Simulation method] Next, a simulation method according to an embodiment will be described with reference to FIGS. 4 is a flowchart showing the procedure of the simulation method according to the embodiment. First, the processing unit 51 acquires the simulation conditions input to the input unit 50 (step S1). The simulation conditions include coarse-graining conditions for generating superparticles 21 by coarse-graining atoms 11, such as a coarse-graining magnification λ.

[0033] Next, the superparticle group is initialized based on the acquired simulation conditions (step S2). For example, for each of the multiple superparticles 21, initial values ​​are given to the position, direction of magnetic moment, total magnetic field acting on the superparticle 21, and vector information of the crystal orientation. Furthermore, information specifying the crystal structure and coarse-graining magnification λ are given. The information specifying the crystal structure determines the crystal structure, such as body-centered cubic lattice (BCC), face-centered cubic lattice (FCC), etc. The number of nearest superparticles, the number of superparticles per unit lattice, and the lattice constant are determined depending on the crystal structure. The magnetic field h due to exchange interaction is determined depending on the number of nearest superparticles. i exch When calculating the number of superparticles 21 to be summed up is determined by the Σ symbol on the right side of equation (3).

[0034] Next, the magnetic field step loop L1 is executed. In the magnetic field step loop L1, the external magnetic field h ext The procedure of performing the magnetization calculation (step S3) and saving the magnetization calculation results (step S4) with the value of the external magnetic field h ext In the magnetization calculation (step S3), the total magnetic field h acting on each of the superparticles 21 is calculated using equations (1), (2), (4), (5), (6), and (8). i Furthermore, using equations (11) and (12), the magnetic moment μ of each of the superparticles 21 is calculated. iThe updated magnetic moment μ i Based on this, the total magnetization M is calculated using equation (7). In the storage of the magnetization calculation results in step S4, the external magnetic field h given as a fixed value is used. ext and the calculated total magnetization M are stored in association with each other.

[0035] The magnetization calculation result obtained by executing the magnetic field step loop L1 will be described with reference to Fig. 5. Fig. 5 is a graph showing an example of the magnetization calculation result. The horizontal axis represents the external magnetic field h ext The vertical axis represents the total magnetization M. For example, the external magnetic field h ext H k and run the magnetic field step loop L1 once to obtain the total magnetization M k is obtained.

[0036] external magnetic field h ext H k+1 Then, by executing the next magnetic field step loop L1 once, the total magnetization M k+1 At this time, the magnetic moment μ of the superparticle 21 i is set to the value obtained by the previous magnetic field step loop L1 calculation. ext By varying and running the magnetic field step loop L1 multiple times, a hysteresis loop is obtained.

[0037] Next, the procedure of magnetization calculation (step S3) will be described in detail with reference to Figures 6 and 7. Figure 6 is a flowchart showing the procedure of magnetization calculation (step S3) in the simulation method according to this embodiment.

[0038] The magnetization calculation procedure includes a time step loop LA1, which includes a superparticle loop LA2. The superparticle loop LA2 numerically solves the LLG equation (11) to calculate the magnetic moment μ of each superparticle 21. i The time step loop LA1 updates the magnetic moment μ of each superparticle 21 until the total magnetization M (equation (7)) converges. iThis is a procedure to evolve the time and calculate the value of the total magnetization M after convergence.

[0039] First, the time average count number is set to 0 (step SA01). The time average count number is a parameter that specifies the number of iterations of calculation to calculate the time average value of the total magnetization M after the total magnetization M has converged. Next, the magnetization average flag is set to OFF (step SA02). The magnetization average flag is a flag that is set to ON when the amount of change in the total magnetization M for one time step becomes equal to or less than a specified value.

[0040] Next, the superparticle loop LA2 will be explained. First, the total magnetic field h acting on the i-th superparticle 21 is i (Step SA03). For this calculation, equations (1), (2), (4), (5), (6), and (8) are used. Next, the magnetic moment μ of the i-th superparticle 21 is calculated. i The change in the magnetic moment μ of the i-th superparticle 21 is calculated using equation (11) (step SA04). i is updated using equation (12) (step SA05). In the superparticle loop LA2, the procedure from step SA03 to step SA05 is executed for all the superparticles 21.

[0041] Next, the time step loop LA1 will be explained. When the superparticle loop LA2 is completed, the normalized total magnetization is calculated. The normalized total magnetization is obtained by dividing the total magnetization M obtained by equation (7) by the saturation magnetization Ms. The saturation magnetization Ms is the magnetic moment μ of all the superparticles 21. i In this example, the magnetic moment μ i Since the magnitude of is set to 1, the saturation magnetization Ms is equal to the number N of the superparticles 21. Therefore, the normalized total magnetization M / M s is expressed by the following formula:

number

[0042] Next, the time average flag is judged (step SA07). That is, the normalized total magnetization M / M s Determine whether the value of has converged. If the time average flag is OFF, that is, the normalized total magnetization M / M s If the value of has not yet converged, the normalized total magnetization M / M s The amount of change in one time step is determined (step SA08). If the amount of change is less than the reference value ε, the normalized total magnetization M / M s has converged, the time average flag is set to ON (step SA09), and the procedure for the next time step in the time step loop LA1 is executed. s If the change in is equal to or greater than the reference value ε, the time average flag remains OFF and the procedure for the next time step is executed.

[0043] Figure 7 shows the normalized total magnetization M / M s , where the horizontal axis represents the time step and the vertical axis represents the normalized total magnetization M / M s As the time step progresses, the normalized total magnetization M / M s For example, at the time step number of 34, the normalized total magnetization M / M s The change in becomes less than the reference value ε.

[0044] Standardized total magnetization M / M s After the magnetization average flag is set to ON in step SA09, the normalized total magnetization M / M s After calculating the normalized total magnetization M / Ms, the normalized total magnetization M / Ms is accumulated (step SA10), and 1 is added to the time-averaged count number (step SA11). In the example shown in FIG. 7, the normalized total magnetization M / Ms calculated in the procedure after the time step number 35 is s The time-average count number is then determined (step SA12). If the time-average count number is less than 10, the procedure for the next time step is executed.

[0045] When the time average count reaches 10, the normalized total magnetization M / Ms In the example shown in FIG. 7, the time average of the normalized total magnetization M / M for 10 time steps from 35 to 44 (range Tm in FIG. 7) is calculated (step SA13). s Calculate the time average of the normalized total magnetization M / M s After calculating the time average of , the magnetization calculation (step S3) is terminated and the process returns to the magnetic field step loop L1 in Figure 4. The normalized total magnetization M / M after convergence s The time average value of the total magnetization M k , M k+1 The external magnetic field h i ext In the next magnetic field step loop L1, the magnetic moment μ of the superparticle 21 at the end of the time step loop LA1 is i is set as the initial condition.

[0046] Next, the results of calculating magnetization using the method according to this embodiment will be described with reference to Figs. 8 and 9. Figs. 8 and 9 are graphs showing the results of actual simulations performed using the method according to this embodiment. The horizontal axis represents the external magnetic field h ext is expressed in units of [A / m], and the vertical axis is the normalized total magnetization M / M s The circle symbols in the graph represent the external magnetic field h given in the magnetization calculation (step S3). i ext The arrows in the graph indicate the order in which the calculation results are moved.

[0047] In this simulation, multiple superparticles 21 are assumed to be magnetic nanoparticles, and the magnetic field h i exch was set to 0. The number N of the superparticles 21 was set to 1000, and the easy axes e1, e2, and e3 of the equation (5) were randomly arranged. As an initial condition, the external magnetic field h ext is set to 400A / m, and the magnetic moment μ of all superparticles 21 i The direction of the external magnetic field h ext Therefore, in the initial state, the normalized total magnetization M / M s becomes 1. The external magnetic field is reversed from 400 A / m and changed back to the original state.

[0048] Figure 8 shows the external magnetic field h i ext The calculation results obtained when the external magnetic field h is inverted and then restored are shown. It can be seen that the major loop of the hysteresis loop is reproduced. i ext The calculation results obtained by folding back once during the inversion are shown below. It can be seen that the minor loop of the hysteresis loop is reproduced.

[0049] Next, the excellent effects of the above embodiment will be described. To calculate the hysteresis loss of a magnetic material, the hysteresis loop must be reproduced in the simulation. To express the effect of magnetocrystalline anisotropy interactions, which are the main factor in forming the hysteresis loop, the dimensions of the magnetic material model must be at least several micrometers, which is the width of the magnetic domain.

[0050] In the above example, the magnetic material model is formed from a plurality of superparticles 21 (FIG. 1B) that are coarse-grained versions of the atoms 11 (FIG. 1A) that make up the magnetic material, thereby reducing the number of particles to be calculated. Therefore, even if the dimensions of the magnetic material model are set to a magnetic domain width of several micrometers or more, it is possible to perform calculations without increasing the calculation load. By employing the method according to the above example, it is possible to reproduce the major and minor loops of the hysteresis loop, as shown in FIGS. 8 and 9.

[0051] In the above example, it is possible to adjust the constituent elements of the magnetic material and arrange the superparticles taking into account the crystal structure.Furthermore, it is possible to analyze the hysteresis loop, iron loss, and magnetic properties of bulk materials containing scattered magnetic materials such as magnetic nanoparticles.

[0052] Next, various modifications of the above embodiment will be described. In the above embodiment, when the time-average count number becomes 10 or more (step SA10), the normalized total magnetization M / M sThe time average of the normalized total magnetization M / M is calculated (step SA13). s The time average count number as a condition for calculating the time average may be other than 10.

[0053] In the above example, the magnetic field step loop L1 calculates the external magnetic field h per time step. i ext There is no particular limitation on the range of change of the external magnetic field h i ext A fixed value may be given as a simulation condition as the variation range of .

[0054] In the above embodiment, as shown in equation (1), the total magnetic field h acting on the superparticle 21 i , the external magnetic field h i ext , magnetic field due to exchange interaction h i exch , magnetic field h due to magnetic dipole interaction i dip , the magnetic field h due to the triaxial magnetocrystalline anisotropy interaction i anis , demagnetizing field h i demag , and the thermal fluctuation magnetic field h i th However, magnetic fields that have little effect on the formation of the hysteresis loop can be excluded from the calculation. i ext and the magnetic field h due to the triaxial magnetocrystalline anisotropy interaction i anis is the total magnetic field h acting on the superparticle 21. i If the magnetic material under analysis has uniaxial magnetocrystalline anisotropy, the magnetic field h due to the triaxial magnetocrystalline anisotropy interaction can be included in i anis Instead, a magnetic field due to uniaxial magnetocrystalline anisotropy interaction may be used.

[0055] The above-described embodiments are merely examples, and the present invention is not limited to the above-described embodiments. For example, it will be obvious to those skilled in the art that various modifications, improvements, combinations, etc. are possible. [Explanation of symbols]

[0056] 10 Magnetic material 11 atoms 20 Magnetic Material Model 21, 21i, 21j superparticles 50 Input section 51 Processing section 52 Output section 53 Storage section

Claims

1. A processing unit of a simulation device generates a magnetic material model consisting of a collection of multiple superparticles, the number of which is less than the number of original atoms, by coarse-graining the multiple atoms that constitute the magnetic material to be simulated; the processing unit imparts a magnetic moment to each of the plurality of superparticles; a simulation method in which the processing unit executes a magnetization calculation procedure for calculating magnetization of the magnetic material model when an external magnetic field is applied to the magnetic material model multiple times while changing the external magnetic field, The magnetization calculation procedure includes: a step of executing a time step loop multiple times until the magnetization of the magnetic material model converges, the time step loop including: calculating a total magnetic field acting on each of the plurality of superparticles based on the external magnetic field and the magnetic moments of the plurality of superparticles; changing the magnetic moment of each of the plurality of superparticles based on the total magnetic field acting on each of the plurality of superparticles; and calculating the magnetization of the magnetic material model based on the magnetic moments of the plurality of superparticles after the change; a step of determining the magnetization of the magnetic material model after the magnetization of the magnetic material model has converged; a step of storing the external magnetic field and the determined magnetization of the magnetic material model in association with each other; Including, A simulation method in which, in the magnetization calculation procedure, the magnetic moments of the plurality of super-particles at the end of the immediately preceding magnetization calculation procedure are given as initial conditions.

2. The simulation method according to claim 1 , wherein in the step of calculating the total magnetic field acting on each of the plurality of super-particles, at least a magnetic field based on magnetocrystalline anisotropy interaction is included in the total magnetic field.

3. 3. The simulation method according to claim 1, wherein, in the magnetization calculation procedure, when the amount of change in magnetization of the magnetic material model becomes equal to or less than a specified value, the time step loop is then executed a plurality of times, and the average value of the magnetization of the magnetic material model calculated in the plurality of time step loops executed after the amount of change in magnetization of the magnetic material model becomes equal to or less than the specified value is set as the magnetization of the magnetic material model determined in the magnetization calculation procedure.

4. an input unit to which simulation conditions including coarse-grained conditions are input; a processing unit that determines a relationship between an external magnetic field applied to a magnetic body to be simulated and the magnetization of the magnetic body based on the simulation conditions inputted to the input unit; Output section and Equipped with The processing unit a function of generating a magnetic body model consisting of a collection of a plurality of superparticles, the number of which is less than the number of original atoms, by coarse-graining the plurality of atoms constituting the magnetic body based on input coarse-graining conditions; a function of imparting a magnetic moment to each of the plurality of superparticles; a function of executing a magnetization calculation procedure for calculating the magnetization of the magnetic material model when the external magnetic field is applied to the magnetic material model multiple times while changing the external magnetic field; and The magnetization calculation procedure includes: a step of executing a time step loop multiple times until the magnetization of the magnetic material model converges, the time step loop including: calculating a total magnetic field acting on each of the plurality of superparticles based on the external magnetic field and the magnetic moments of the plurality of superparticles; changing the magnetic moment of each of the plurality of superparticles based on the total magnetic field acting on each of the plurality of superparticles; and calculating the magnetization of the magnetic material model based on the magnetic moments of the plurality of superparticles after the change; a step of determining the magnetization of the magnetic material model after the magnetization of the magnetic material model has converged; a step of storing the external magnetic field and the determined magnetization of the magnetic material model in association with each other; Including, The processing unit is a simulation device that provides, as initial conditions in the magnetization calculation procedure, magnetic moments of the plurality of superparticles at the end of the immediately preceding magnetization calculation procedure.

5. A function to generate a magnetic material model consisting of a collection of superparticles with a number smaller than the original number of atoms by coarse-graining the multiple atoms that make up the magnetic material to be simulated; a function of imparting a magnetic moment to each of the plurality of superparticles of the magnetic material model; a function of executing a magnetization calculation procedure for calculating the magnetization of the magnetic material model when an external magnetic field is applied to the magnetic material model multiple times while changing the external magnetic field; A program for causing a computer to realize the above, The magnetization calculation procedure includes: a step of executing a time step loop multiple times until the magnetization of the magnetic material model converges, the time step loop including: calculating a total magnetic field acting on each of the plurality of superparticles based on the external magnetic field and the magnetic moments of the plurality of superparticles; changing the magnetic moment of each of the plurality of superparticles based on the total magnetic field acting on each of the plurality of superparticles; and calculating the magnetization of the magnetic material model based on the magnetic moments of the plurality of superparticles after the change; a step of determining the magnetization of the magnetic material model after the magnetization of the magnetic material model has converged; a step of storing the external magnetic field and the determined magnetization of the magnetic material model in association with each other; Including, a program for providing, as initial conditions in the magnetization calculation procedure, the magnetic moments of the plurality of super-particles at the end of the immediately preceding magnetization calculation procedure;

Images