Magnetic thrust assembly design method, device and magnetic thrust prediction method

Abstract

This application relates to a design method, apparatus, and thrust prediction method for a magnetic thrust assembly. The method includes: obtaining the structural type and constraints of the magnetic thrust assembly, wherein the structural type corresponds to at least one basic structure; constructing structural parameter conditions based on the structural type; determining at least one set of target parameters for the magnetic thrust assembly based on the structural parameter conditions and constraints, using an optimization algorithm and a thrust prediction model; and automating the design of the magnetic thrust assembly through the optimization algorithm and a pre-trained thrust prediction model, thereby reducing the design time and further reducing the design cost of the magnetic thrust assembly.

Description

Technical Field

[0001] This application relates to the field of camera technology, and in particular to a design method, device and method for predicting magnetic thrust components. Background Technology

[0002] With the development of camera modules, from the initial fixed-focus modules, optical image stabilization camera modules, periscope telephoto camera modules, and variable aperture camera modules have evolved. The voice coil motor (VCM) is a crucial component of the camera module. The VCM is a drive device based on the principle of a current-carrying coil operating within a permanent magnetic field created by a magnet, subject to the Lorentz force. The VCM consists of a magnet, a coil assembly, and a guiding mechanism. The magnet comprises a permanent magnet and a magnetic yoke, generating a constant magnetic field; the coil assembly contains precision-wound copper wire windings that generate an alternating magnetic field when energized; the guiding mechanism limits the axis of motion via spring plates or slider rails. The magnet and coil assembly in the VCM constitute the thrust assembly, providing driving force to the lens. Driven by the VCM, the camera module achieves core functions such as autofocus and optical image stabilization. The design of the thrust assembly plays a key role in the image sharpness, optical zoom, and optical image stabilization performance of the camera module.

[0003] Currently, the design of thrust components typically requires experienced engineers to perform design, simulation verification, and optimization. This current approach to thrust component design consumes a significant amount of development time and incurs excessively high labor costs. Summary of the Invention

[0004] Therefore, it is necessary to provide a magnetic thrust component design method, device, and magnetic thrust prediction method to address the above-mentioned technical problems.

[0005] In a first aspect, this application provides a design method for a magnetic thrust assembly, the method comprising: obtaining the structural type and constraints of the magnetic thrust assembly; the structural type corresponding to at least one basic structure; constructing structural parameter conditions based on the structural type; determining at least one set of target parameters of the magnetic thrust assembly based on the structural parameter conditions and the constraints, using an optimization algorithm and a thrust prediction model; the thrust prediction model being used to predict the thrust for each of the basic structures respectively.

[0006] In one embodiment, any of the foundation structures includes a magnet and a coil; any of the foundation structures is of one of a first foundation structure type, a second foundation structure type, a third foundation structure type, or a fourth foundation structure type; in the first foundation structure type, the coil is parallel to the magnet and the coil is a lapped winding; in the second foundation structure type, the coil is parallel to the magnet and the coil is a staggered winding; in the third foundation structure type, the coil is perpendicular to the magnet and the coil is a lapped winding; in the fourth foundation structure type, the coil is perpendicular to the magnet and the coil is a staggered winding.

[0007] In one embodiment, the structural parameter conditions include parameter conditions for at least one of the basic structures.

[0008] In one embodiment, the parameter conditions of each of the basic structures include at least one of the following: magnet length condition, magnet width condition, magnet height condition, coil center length condition, coil center width condition, coil and magnet spacing condition, coil and magnet relative displacement condition, coil winding wire diameter condition, coil winding number of turns in the height direction condition, and coil winding number of turns in the width direction condition.

[0009] In one embodiment, the constraint conditions include at least one of the following: a thrust constraint condition and a spatial constraint condition; wherein the thrust constraint condition is used to constrain the thrust magnitude of the magnetic thrust assembly, and the spatial constraint condition is used to constrain the spatial dimensions of the magnetic thrust assembly.

[0010] In one embodiment, determining at least one set of target parameters for the magnetic thrust assembly based on the structural parameter conditions and the constraint conditions, using an optimization algorithm and a thrust prediction model, includes: generating an initial population based on the structural parameter conditions and an optimization algorithm; the initial population includes multiple sets of structural parameters for the magnetic thrust assembly; determining thrust data corresponding to each set of structural parameters based on the initial population and the thrust prediction model; and iterating the population based on the initial population, the thrust data corresponding to each set of structural parameters, and the constraint conditions, using the optimization algorithm and the thrust prediction model, until a preset termination condition is met, thereby generating at least one set of target parameters for the magnetic thrust assembly.

[0011] In one embodiment, generating an initial population based on the structural parameter conditions and an optimization algorithm includes: generating multiple sets of original parameters based on the structural parameter conditions and an optimization algorithm; each set of original parameters includes basic parameters for each basic structure corresponding to the structural type; determining the positive and negative relative displacements of each basic structure to obtain positive and negative basic data for each basic structure; the positive basic data includes the basic parameters and the positive relative displacement, and the negative basic data includes the basic parameters and the negative relative displacement; and determining multiple sets of structural parameters based on the positive and negative basic data of each basic structure corresponding to each set of original parameters.

[0012] In one embodiment, determining the thrust data corresponding to each set of structural parameters based on the initial population and the thrust prediction model includes: determining target structural parameters; the target structural parameters are any set of structural parameters from multiple sets of structural parameters; determining the positive stroke thrust and negative stroke thrust of each basic structure corresponding to the target structural parameters based on the positive stroke basic data and negative stroke basic data of each basic structure in the target structural parameters, and the thrust prediction model corresponding to each basic structure; and determining the total positive stroke thrust and total negative stroke thrust corresponding to the target structural parameters based on the positive stroke thrust and negative stroke thrust of each basic structure corresponding to the target structural parameters.

[0013] In one embodiment, the thrust prediction model includes: a first thrust prediction model, a second thrust prediction model, a third thrust prediction model, and a fourth thrust prediction model; the first thrust prediction model is used to predict thrust for the first infrastructure type; the second thrust prediction model is used to predict thrust for the second infrastructure type; the third thrust prediction model is used to predict thrust for the third infrastructure type; and the fourth thrust prediction model is used to predict thrust for the fourth infrastructure type.

[0014] In one embodiment, the method further includes: constructing a first training dataset corresponding to the first infrastructure type, the first training dataset including sample structure parameters and corresponding sample thrust data; and training a neural network model using the first training dataset to obtain the first thrust prediction model.

[0015] In one embodiment, before constructing the first training dataset corresponding to the first infrastructure type, the method further includes: constructing a simulation model of the first infrastructure type and parameter conditions corresponding to the first infrastructure type; performing parameter sampling based on the parameter conditions corresponding to the first infrastructure type to determine multiple sets of sample structure parameters; and performing simulation based on the simulation model of the first infrastructure type and the multiple sets of sample structure parameters to determine sample thrust data for each set of sample structure parameters.

[0016] Secondly, this application also provides a magnetic thrust prediction method, the method comprising: obtaining the structural type and structural parameters to be predicted of a magnetic thrust assembly, wherein the structural type corresponds to at least one basic structure, and the structural parameters to be predicted include the basic parameters to be predicted for each of the at least one basic structure corresponding to the structural type; determining the thrust magnitude of the magnetic thrust assembly based on a thrust prediction model according to the structural type and the structural parameters to be predicted; wherein the thrust prediction model is used to predict the thrust for each of the basic structures respectively.

[0017] In one embodiment, any of the foundation structures includes a magnet and a coil; any of the foundation structures is of one of a first foundation structure type, a second foundation structure type, a third foundation structure type, or a fourth foundation structure type; in the first foundation structure type, the coil is parallel to the magnet and the coil is a lapped winding; in the second foundation structure type, the coil is parallel to the magnet and the coil is a staggered winding; in the third foundation structure type, the coil is perpendicular to the magnet and the coil is a lapped winding; in the fourth foundation structure type, the coil is perpendicular to the magnet and the coil is a staggered winding.

[0018] In one embodiment, the thrust prediction model includes: a first thrust prediction model, a second thrust prediction model, a third thrust prediction model, and a fourth thrust prediction model; the first thrust prediction model is used to predict thrust for the first infrastructure type; the second thrust prediction model is used to predict thrust for the second infrastructure type; the third thrust prediction model is used to predict thrust for the third infrastructure type; and the fourth thrust prediction model is used to predict thrust for the fourth infrastructure type.

[0019] In one embodiment, the predicted basic parameters corresponding to the basic structure include at least one of the following: magnet length, magnet width, magnet height, coil center length, coil center width, distance between the coil and the magnet, relative displacement between the coil and the magnet, diameter of the coil winding, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding.

[0020] Thirdly, this application also provides a magnetic thrust assembly design device, the device comprising: an acquisition module for acquiring the structural type and constraints of the magnetic thrust assembly; the structural type corresponding to at least one basic structure; a construction module for constructing structural parameter conditions according to the structural type; and a design module for determining at least one set of target parameters of the magnetic thrust assembly based on the structural parameter conditions and the constraints, using an optimization algorithm and a thrust prediction model; the thrust prediction model is used to predict the thrust for each of the basic structures respectively.

[0021] Fourthly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement any one of the magnetic thrust component design methods in the first aspect or any one of the magnetic thrust prediction methods in the second aspect.

[0022] Fifthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements any one of the magnetic thrust component design methods in the first aspect or any one of the magnetic thrust prediction methods in the second aspect.

[0023] The aforementioned magnetic thrust assembly design method, device, and magnetic thrust prediction method obtain the structural type and constraints of the magnetic thrust assembly, wherein the structural type corresponds to at least one basic structure. Based on the structural type, structural parameter conditions are constructed. Based on the structural parameter conditions and constraints, at least one set of target parameters for the magnetic thrust assembly is determined using an optimization algorithm and a thrust prediction model. The automated design of the magnetic thrust assembly is achieved through the optimization algorithm and the pre-trained thrust prediction model, thereby reducing the design time and further lowering the design cost of the magnetic thrust assembly. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the design method of a magnetic thrust component in one embodiment;

[0025] Figure 2 This is a structural schematic diagram of a magnet design type in one embodiment;

[0026] Figure 3 A schematic diagram illustrating the design of the infrastructure type in one embodiment;

[0027] Figure 4 This is a flowchart illustrating a target parameter generation method in one embodiment;

[0028] Figure 5 This is a schematic diagram of the thrust data direction in one embodiment;

[0029] Figure 6 This is a flowchart illustrating a method for generating an initial population in one embodiment;

[0030] Figure 7 This is a flowchart illustrating a thrust data prediction method in one embodiment;

[0031] Figure 8 This is a flowchart illustrating the training method for the first thrust prediction model in one embodiment;

[0032] Figure 9 This is a flowchart illustrating a magnetic thrust prediction method in one embodiment;

[0033] Figure 10 This is a structural block diagram of a magnetic thrust assembly design device in one embodiment;

[0034] Figure 11 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0036] With the development of camera modules, from the initial fixed-focus modules, optical image stabilization camera modules, periscope telephoto camera modules, and variable aperture camera modules have evolved. The voice coil motor (VCM) is a crucial component of the camera module. A VCM is a drive device that operates based on the principle of a energized coil operating within a permanent magnetic field created by a magnet, subject to the Lorentz force. The VCM consists of a magnet, a coil assembly, and a guiding mechanism. The magnet comprises a permanent magnet and a magnetic yoke, generating a constant magnetic field; the coil assembly contains precisely wound copper wire windings that generate an alternating magnetic field when energized; the guiding mechanism limits the axial movement via spring plates or slider rails. The magnet and coil assembly in the VCM constitute the thrust assembly, providing driving force to the lens. Driven by the VCM, the camera module achieves core functions such as autofocus and optical image stabilization. The quality of the thrust assembly design plays a decisive role in the image sharpness, optical zoom, and optical image stabilization performance of the camera module.

[0037] Currently, thrust assembly design is a complex, specialized, and meticulous task in the industry. It typically requires experienced engineers to design, simulate, verify, and optimize the thrust assembly. Current thrust assembly design methods consume significant development time and are extremely costly in terms of manpower. This long design cycle conflicts with the market demand for low-cost camera modules, making efficient thrust assembly design simulation a pressing market requirement.

[0038] In one embodiment, such as Figure 1 As shown, a design method for a magnetic thrust assembly is provided, including the following steps:

[0039] Step 101: Obtain the structural type and constraints of the magnetic thrust assembly.

[0040] The structural type is the design structure of the magnetic thrust assembly selected according to actual needs. This structural type consists of at least one basic structure, each of which includes a magnet and a coil.

[0041] The copper wire winding method of the coil in the basic structure can be either lapped winding or staggered winding. The positional relationship between the magnet and the coil in the basic structure can be either the coil parallel to the magnet or the coil perpendicular to the magnet. Lapped winding refers to adjacent layers of coil being aligned, while staggered winding refers to adjacent layers of coil being staggered. Therefore, in this application, based on the positional relationship between the magnet and the coil and the copper wire winding method of the coil, the basic structure can be divided into the following basic structure types:

[0042] The first basic structural type: the coil is parallel to the magnet, and the coil is a multi-layered winding;

[0043] The second basic structure type: the coil is parallel to the magnet, and the coil is wound with interlaced wires;

[0044] The third basic structural type: the coil is perpendicular to the magnet, and the coil is a multi-layered winding;

[0045] The fourth basic structural type: the coil is perpendicular to the magnet, and the coil is wound with interlaced wires.

[0046] The structure type can include the magnet design type and the copper wire winding type of the coil. Based on the magnet design type and the copper wire winding type of the coil, at least one basic structure corresponding to the structure type can be determined. For example... Figure 2 As shown, the magnet design types include: single-pole magnet and single coil, double single-pole magnet and single coil, Halebeck magnet and single coil, shared double single-pole magnet and double coil, and shared Halebeck magnet and double coil. The copper wire winding type of the coil can be lapped winding or staggered winding. The structure type needs to be entered according to the actual design requirements.

[0047] Specifically, such as Figure 3 As shown, based on single-pole magnet and single coil, double single-pole magnet and single coil, Halebeck magnet and single coil, shared double single-pole magnet and double coil, and shared Halebeck magnet and double coil, five design structures are obtained. Figure 3 The five design structures, from left to right, are structures 1 to 5. Structure 1 is a single-pole magnet with the magnetic poles pointing left or right. The coil is parallel to the magnet's direction, and the main thrust direction is up-down. Structure 2 is a double-monopolar magnet group with the upper and lower magnets pointing left and right, respectively. The coil is parallel to the magnet group, and the main thrust direction is up-down. Structure 3 is a Helbeck magnet group with the upper, middle, and lower magnets pointing left, down, and right, respectively. The coil is parallel to the magnet group, and the main thrust direction is up-down. This structure's magnet group directionally enhances the magnetic field at the coil, and its thrust is greater than that of Structure 2. Structure 4 is a double-monopolar magnet group with the upper and lower magnets pointing left and right, respectively. The coil is perpendicular to the magnet group, and its main thrust direction is left-right. Structure 5 is a Helbeck magnet group with the upper, middle, and lower magnets pointing left, down, and right, respectively. The coil is perpendicular to the magnet group, and its main thrust direction is left-right, with a thrust greater than that of Structure 4.

[0048] Based on the above, and according to the magnetic charge theory of magnetic physics, the magnetic field of a multi-magnet group can be composed of multiple magnets superimposed. Therefore, the above five design structures can be broken down into two types of relative positions between the single magnet and the coil: either the coil is parallel to the magnet or the coil is perpendicular to the magnet. Combining this with the coil's copper wire winding method—either layered or staggered—four basic structural types can be obtained.

[0049] For example, such as Figure 2 As shown, a single-stage magnet includes a magnet and a coil. The coil is parallel to the magnet. The coil can be wound in a stacked or interleaved manner. The single-stage magnet can be composed of a first basic structure type or a second basic structure type.

[0050] In a double-monopolar magnet, the positions of the single magnet and the coil include: the coil of magnet 1 and the coil combination is parallel to the magnet; the coil of magnet 2 and the coil combination is parallel to the magnet. In other words, a double-monopolar magnet includes two positional relationships corresponding to two magnets and one coil. If the coil winding method is lap winding, the double-monopolar magnet consists of two first basic structural types; if the coil winding method is staggered winding, the double-monopolar magnet consists of two second basic structural types.

[0051] The positions of a single magnet and coil in a Helbeck magnet include: the coil of magnet 1 is parallel to the magnet; the coil of magnet 2 is perpendicular to the magnet; and the coil of magnet 3 is parallel to the magnet. In other words, a Helbeck magnet includes three positional relationships corresponding to three magnets and one coil. If the coil is wound in a lapped manner, the Helbeck magnet consists of two first basic structural types and one third basic structural type; if the coil is wound in an interleaved manner, the Helbeck magnet consists of two second basic structural types and one fourth basic structural type.

[0052] The positions of the single magnet and coil in the shared double single-pole magnet include: the coil of magnet 1 and coil 1 is parallel to the magnet; the coil of magnet 2 and coil 1 is parallel to the magnet; the coil of magnet 1 and coil 2 is perpendicular to the magnet; and the coil of magnet 2 and coil 2 is perpendicular to the magnet. That is, the shared double single-pole magnet includes four positional relationships corresponding to two magnets and two coils. If both coil 1 and coil 2 are wound in a lapped winding pattern, the shared double-pole magnet consists of two first basic structural types and two third basic structural types. If both coil 1 and coil 2 are wound in an alternating winding pattern, the shared double-pole magnet consists of two first basic structural types and two fourth basic structural types. If both coil 1 and coil 2 are wound in an alternating winding pattern, the shared double-pole magnet consists of two second basic structural types and two fourth basic structural types. If both coil 1 and coil 2 are wound in an alternating winding pattern and coil 2 are wound in a lapped winding pattern, the shared double-pole magnet consists of two second basic structural types and two third basic structural types.

[0053] The positions of a single magnet and coil in a shared Helbeck magnet include: the coil of magnet 1 and coil 1 is parallel to the magnet; the coil of magnet 2 and coil 1 is perpendicular to the magnet; the coil of magnet 3 and coil 1 is parallel to the magnet; the coil of magnet 1 and coil 2 is perpendicular to the magnet; the coil of magnet 2 and coil 2 is parallel to the magnet; and the coil of magnet 3 and coil 2 is perpendicular to the magnet. In other words, the shared Helbeck magnet includes six possible positional relationships corresponding to 3 magnets and 2 coils. If both coil 1 and coil 2 are wound in a lapped winding pattern, the shared Hellbeck magnet consists of three first basic structure types and three third basic structure types. If both coil 1 and coil 2 are wound in a staggered winding pattern, the shared Hellbeck magnet consists of two first basic structure types, one second basic structure type, one third basic structure type, and two fourth basic structure types. If both coil 1 and coil 2 are wound in a staggered winding pattern, the shared Hellbeck magnet consists of three second basic structure types and three fourth basic structure types. If both coil 1 and coil 2 are wound in a staggered winding pattern, the shared Hellbeck magnet consists of one first basic structure type, two second basic structure types, two third basic structure types, and one fourth basic structure type.

[0054] The structural type includes any magnet design type and the winding design type of each coil within that magnet design type. Therefore, based on the positional relationships of the magnet design types within the structural type and the winding design type of each coil, the basic structural type of at least one basic structure corresponding to the structural type can be determined. For example, in the structural type, the magnet design type is a shared double-single-pole magnet, where the winding design type of coil 1 is a stacked winding, and the winding design type of coil 2 is an interleaved winding. Then, the basic structural types corresponding to the structural type include: the first basic structural type combining magnet 1 and coil 1, the first basic structural type combining magnet 2 and coil 1, the fourth basic structural type combining magnet 1 and coil 2, and the fourth basic structural type combining magnet 2 and coil 2. That is, in this case, the structural type includes four basic structures, two of which are the first basic structural type, and two of which are the fourth basic structural type.

[0055] Constraints are mathematical expressions of boundary limitations or performance thresholds that the design of a magnetic thrust assembly must meet. These constraints ensure that the output results conform to actual engineering application conditions. Constraints include at least one of thrust constraints and spatial constraints. The thrust constraint represents the total thrust condition of the magnetic thrust assembly formed by the final designed target parameters; for example, the total thrust of the magnetic thrust assembly needs to be greater than a preset target thrust threshold. The spatial constraint represents the space occupied by the magnetic thrust assembly formed by the final designed target parameters; for example, based on actual needs, size limitations are set for the magnets and coils in the magnetic thrust assembly, and the final magnetic thrust assembly formed by the target parameters should meet the corresponding size limitations. Constraints can be set according to actual needs; this embodiment does not impose specific limitations.

[0056] Step 102: Construct structural parameter conditions based on the structure type.

[0057] The structural parameter conditions are a set of adjustable variables representing the geometric and physical configuration of the magnetic thrust component. These variables are used to limit the adjustable range of each variable during the optimization process, forming the basic dimension of the search space. The structural parameter conditions include conditions for various parameter variables. Specifically, they include at least one of the following: magnet length condition, magnet width condition, magnet height condition, coil center length condition, coil center width condition, coil-magnet spacing condition, coil-magnet relative displacement condition, coil winding wire diameter condition, coil winding number of turns in the height direction condition, and coil winding number of turns in the width direction condition. The coil center length represents the central spatial length of the coil. The coil center width represents the central spatial width of the coil. The coil-magnet spacing represents the distance between the coil and the magnet. The relative displacement between the coil and the magnet represents the relative offset between the center of the magnet and the center of the coil. The coil winding wire diameter represents the diameter of the copper wire used to wind the coil. The conditions for each of the above parameter variables, i.e., the value range of each parameter, can be set according to actual needs; this embodiment does not impose specific limitations.

[0058] Specifically, based on the structure type, at least one basic structure corresponding to the structure type can be determined. Parameter conditions are then constructed for each basic structure type, thereby obtaining the structural parameter conditions for the structure type. For example, taking a structure type where the magnet design type is a double-single-pole magnet and the coil winding design type is a lapped winding, the structure type includes: a first basic structure of the first basic structure type combining magnet 1 and the coil, and a second basic structure of the first basic structure type combining magnet 2 and the coil. That is, the structure type includes a first basic structure and a second basic structure. Parameter conditions for the first basic structure and the second basic structure are constructed separately for the first basic structure and the second basic structure, respectively, thereby obtaining the structural parameter conditions for the structure type.

[0059] Step 103: Based on the structural parameter conditions and the constraint conditions, determine at least one set of target parameters for the magnetic thrust assembly using an optimization algorithm and a thrust prediction model.

[0060] The optimization algorithm is a global optimization algorithm that simulates the mechanism of biological evolution. It can find the optimal solution set among multiple conflicting objectives and coordinate the trade-offs between multiple design objectives while satisfying constraints. The optimization algorithm can be a multi-objective optimization genetic algorithm, specifically the NSGA-II algorithm. The multi-objective optimization genetic algorithm searches under structural parameter conditions through iterative operations such as population initialization, selection, crossover, mutation, and fitness evaluation. During the search, it receives structural parameter conditions and constraints as the definition of the search space, calls the thrust prediction model to evaluate the thrust data corresponding to each set of structural parameters in the population, and drives the population iteration. Population iteration is a repetitive computation process in the multi-objective optimization genetic algorithm that gradually approaches the optimal solution set through intergenerational evolution. It can be used to achieve asymptotic convergence from an initial random solution to a high-quality candidate solution. Population iteration generates a new population through selection, recombination, and mutation of individuals in each generation, and its performance is evaluated by the thrust prediction model.

[0061] The thrust prediction model is a computational model used to estimate the output thrust of a magnetic thrust assembly. It can replace traditional finite element simulation for performance evaluation, significantly improving the efficiency of a single evaluation. The thrust prediction model is used to predict thrust for each of the aforementioned foundation structures. That is, a thrust prediction model predicts thrust data for structural parameters corresponding to a specific foundation structure type. The thrust prediction models include: a first thrust prediction model for a first foundation structure type, a second thrust prediction model for a second foundation structure type, a third thrust prediction model for a third foundation structure type, and a fourth thrust prediction model for a fourth foundation structure type. Thrust prediction is performed using multiple thrust prediction models, and the thrust data is superimposed to obtain a set of thrust data corresponding to the structural parameters. The thrust prediction model can be a neural network model, trained using historical simulation data or experimental samples. The thrust data corresponding to the structural parameters output by the thrust prediction model can be used as a fitness evaluation basis for a multi-objective optimization genetic algorithm.

[0062] Specifically, within the structural parameter conditions of the structural type, an initial population is generated based on an optimization algorithm. Each population includes multiple sets of structural parameters for the magnetic thrust component. For example, these multiple sets of structural parameters can include 100 sets or 200 sets; this embodiment does not impose a specific limitation and can be set according to actual usage requirements. The parameter types in each set of structural parameters are the same as the parameter types in the structural parameter conditions. For example, taking a structure type with a double-single-stage magnet design and a coil winding design of layered winding, the structural parameter conditions of the structural type include parameter conditions for the first basic structure and condition parameters for the second basic structure. The first basic structure is the first basic structure type, and the second basic structure is the first basic structure type. Therefore, each set of structural parameters includes the structural parameters of the first basic structure and the structural parameters of the second basic structure. The structural parameters of each basic structure include: magnet length, magnet width, magnet height, coil center length, coil center width, spacing between the coil and magnet, relative displacement between the coil and magnet, coil winding wire diameter, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding. For a set of structural parameters in the population, the structural parameters of each basic structure type are input into the thrust prediction model corresponding to that basic structure type to obtain thrust data for each basic structure. The thrust data of each basic structure are then superimposed to obtain the total thrust of the set of structural parameters in the population. The total thrust of each set of structural parameters in the population is fed back to the optimization algorithm, allowing the algorithm to evaluate individual performance based on the total thrust of the set of structural parameters in the population. Genetic operations are performed to generate offspring, and this process is iterated until a preset termination condition is reached. From the final population, a solution vector that satisfies all constraints and has balanced performance is extracted, and the formatted output is a combination of structural parameters, i.e., at least one set of target parameters. The target parameters can be a numerical set of combinations of structural parameters of the magnetic thrust component that satisfy all design requirements. The target parameters include the structural parameters of each basic structure corresponding to the structural type of the magnetic thrust component. For example, the target parameters consist of 100 sets of predicted data output by the overall optimization algorithm.

[0063] This embodiment obtains the structural type and constraints of the magnetic thrust assembly, where each structural type corresponds to at least one basic structure. Based on the structural type, structural parameter conditions are constructed. Based on the structural parameter conditions and constraints, at least one set of target parameters for the magnetic thrust assembly is determined using an optimization algorithm and a thrust prediction model. The magnetic thrust assembly is designed automatically using the optimization algorithm and a pre-trained thrust prediction model, thereby reducing the design time and cost. By identifying the structural type, the parameter system construction is initiated, allowing the structural parameter conditions and constraints to jointly define the optimization direction. The optimization algorithm iterates along this direction, and the thrust prediction model replaces traditional time-consuming simulations to complete high-frequency performance evaluations, achieving multi-objective collaborative optimization of constraints. When preset termination conditions are met, one or more sets of target parameters meeting engineering requirements are output. This transforms the original design mode, which relied on repeated trial and error based on human experience, into an algorithm-driven automated search process, reducing the frequency of highly experienced human intervention, compressing the single design cycle, and lowering the design time and cost of the magnetic thrust assembly.

[0064] In one embodiment, any of the said basic structures includes a magnet and a coil. The type of any said basic structure is one of a first basic structure type, a second basic structure type, a third basic structure type, or a fourth basic structure type. Each structure type corresponds to at least one basic structure. Each basic structure includes a magnet and a coil. The basic structures include four basic structure types, with different positions of the magnet and coil and different winding methods for the copper wire of the coil among each basic structure type.

[0065] In one embodiment, the structural parameter conditions include parameter conditions for at least one of the basic structures.

[0066] Specifically, after obtaining the user-specified structure type, its corresponding basic structure needs to be determined based on the structure type. Structure types include magnet design type and winding design type. Magnet design types include: single-pole magnet, double single-pole magnet, Halebeck magnet, shared double single-pole magnet, and shared Halebeck magnet. Winding design types include stacked winding or interleaved winding. The basic structure corresponding to the structure type can be determined based on the magnet design type and winding design type. Taking a structure type with a double single-pole magnet design and a stacked winding design as an example, the positions of the single magnet and coil in the double single-pole magnet include: the coil of magnet 1 and coil combination is parallel to the magnet, and the coil of magnet 2 and coil combination is parallel to the magnet. Therefore, the structure types include: the first basic structure of the first basic structure type of magnet 1 and coil combination, and the second basic structure of the first basic structure type of magnet 2 and coil combination.

[0067] Based on the basic structures corresponding to the structure type, structural parameter conditions are constructed. These structural parameter conditions include the parameter conditions for each basic structure corresponding to the structure type. Specifically, for each basic structure corresponding to the structure type, corresponding parameter conditions need to be constructed to obtain the structural parameter conditions for the structure type. Taking a structure type with a double-single-pole magnet design and a coil winding design of lap winding as an example, the structure type includes: a first basic structure of the first basic structure type combining magnet 1 and coil, and a second basic structure of the first basic structure type combining magnet 2 and coil. For the combination of magnet 1 and coil in the first basic structure, parameter conditions for the first basic structure are constructed; for the combination of magnet 2 and coil in the second basic structure, parameter conditions for the second basic structure are constructed. Integrating the parameter conditions of the first and second basic structures yields the structural parameter conditions for the structure type.

[0068] Each of the aforementioned basic structures includes at least one of the following parameter conditions: magnet length condition, magnet width condition, magnet height condition, coil center length condition, coil center width condition, spacing condition between the coil and the magnet, relative displacement condition between the coil and the magnet, coil winding wire diameter condition, coil winding number of turns in the height direction condition, and coil winding number of turns in the width direction condition. Taking a structure type where the magnet design type is a double single-stage magnet and the coil winding design type is a cascaded winding as an example, the parameter conditions of the first basic structure include: magnet 1 length condition, magnet 1 width condition, magnet 1 height condition, coil center length condition, coil center width condition, spacing condition between the coil and magnet 1, relative displacement condition between the coil and magnet 1, coil winding wire diameter condition, coil winding number of turns in the height direction condition, and coil winding number of turns in the width direction condition. The parameters of the second basic structure include: the length of magnet 2 corresponding to the coil, the width of magnet 2, the height of magnet 2, the center length of the coil, the center width of the coil, the distance between the coil and magnet 2, the relative displacement between the coil and magnet 2, the wire diameter of the coil winding, the number of turns in the height direction of the coil winding, and the number of turns in the width direction of the coil winding.

[0069] The magnet length condition is a pre-set range, for example, 0.35mm-18mm. The magnet width condition is a pre-set range, for example, 0.35mm-6mm. The magnet height condition is a pre-set range, for example, 0.35mm-18mm. The coil center length condition is a pre-set range of the central space length of the coil, for example, 0.3mm-18mm. The coil center width condition is a pre-set range of the central space width of the coil, for example, 0.3mm-12mm. The coil and magnet spacing condition is a pre-set range of the distance between the coil and the magnet, for example, 0.08mm-17mm. The relative displacement condition between the coil and the magnet can be a pre-set range of relative offset between the center of the magnet and the center of the coil, for example, -15mm to +15mm. The wire diameter of the coil winding can be a preset copper wire diameter, for example, ranging from 0.039mm, 0.044mm, 0.05mm, 0.055mm, 0.062mm, 0.067mm, 0.073mm, 0.078mm, 0.083mm, 0.088mm, and 0.094mm. The number of turns in the height direction of the coil winding can be a preset number of turns in the height direction of the copper wire, for example, ranging from 2 to 36. The number of turns in the width direction of the coil winding can be a preset number of turns in the width direction of the copper wire, for example, ranging from 3 to 24.

[0070] In one embodiment, the constraints include at least one of the following: a thrust constraint and a spatial constraint; wherein the thrust constraint is used to constrain the thrust magnitude of the magnetic thrust assembly, for example, the total thrust of the magnetic thrust assembly corresponding to the target parameter needs to be greater than a preset target thrust threshold. The spatial constraint is used to constrain the spatial dimensions of the magnetic thrust assembly. For example, the total length of the multiple magnets in the target parameter needs to be less than or equal to a preset magnet length limit, the total length of the multiple magnets in the target parameter needs to be greater than or equal to 0.7 times the preset magnet length limit, the total width of the multiple magnets in the target parameter needs to be less than or equal to a preset magnet width limit, the total height of the multiple magnets in the target parameter needs to be less than or equal to a preset magnet height limit, the corresponding coil height in the target parameter needs to be less than or equal to a preset coil height limit, the corresponding coil length in the target parameter needs to be less than or equal to a preset coil length limit, the corresponding coil resistance in the target parameter needs to be less than or equal to a preset coil resistance limit, and the difference between the corresponding coil length and the magnet length in the target parameter needs to be less than or equal to a preset difference limit.

[0071] This embodiment first maps design requirements to a basic structure with engineering rationality, and then generates structural parameter conditions based on this basic structure. This ensures that the selection of optimization variables and boundary settings have structural consistency and physical feasibility. By solidifying empirical knowledge into a basic structural template, it reduces the differences and redundant attempts caused by human judgment, and improves the automation and reliability of condition construction.

[0072] In one embodiment, such as Figure 4 As shown, a method for generating target parameters is provided, which specifically includes the following steps:

[0073] Step 401: Based on the structural parameter conditions, generate an initial population using an optimization algorithm.

[0074] The initial population comprises multiple sets of structural parameters for the magnetic thrust component. It is the first set of candidate solutions generated during the iteration of the optimization algorithm. The initial population can be generated based on the value range of each parameter variable in the structural parameter conditions, using uniform random sampling or Latin hypercube sampling methods. The initial population includes multiple sets of structural parameters. For example, it may include 100 sets of structural parameters or 200 sets; this embodiment does not impose a specific limitation, and the settings can be adjusted according to actual usage requirements.

[0075] The parameter types in the structural parameters are the same as those in the structural parameter conditions. Each set of structural parameters includes the structural parameters of each basic structure in at least one basic structure. For example, taking a structure type with a double-single-pole magnet design and a coil winding design of lap winding, the structural parameter conditions of the structure type include the parameter conditions of the first basic structure and the parameter conditions of the second basic structure. Therefore, each set of structural parameters in the initial population includes the structural parameters of the first basic structure and the structural parameters of the second basic structure. The structural parameters of the first basic structure include: magnet length, magnet width, magnet height, coil center length, coil center width, spacing between the coil and the magnet, relative displacement between the coil and the magnet, coil winding wire diameter, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding. The structural parameters of the second basic structure include: magnet length, magnet width, magnet height, coil center length, coil center width, spacing between the coil and the magnet, relative displacement between the coil and the magnet, coil winding wire diameter, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding.

[0076] Step 402: Determine the thrust data corresponding to each set of structural parameters based on the initial population and the thrust prediction model.

[0077] After generating the initial population, for a set of structural parameters in the initial population, the structural parameters corresponding to each basic structure are input into the thrust prediction model corresponding to the basic structure type to obtain thrust data for each basic structure. The thrust data for each basic structure are then superimposed to obtain the thrust data for a set of structural parameters in the population. The thrust data includes thrust data in the Y direction and thrust data in the Z direction. The Y and Z directions are perpendicular to each other, as shown in the diagram. Figure 5 As shown, the thrust prediction model can output thrust data in the Y and Z directions for each foundation structure type. By superimposing the thrust data in the Y direction for each foundation structure, a set of thrust data in the Y direction for structural parameters is obtained; by superimposing the thrust data in the Z direction for each foundation structure, a set of thrust data in the Z direction for structural parameters is obtained.

[0078] Taking a structure type with a double-single-pole magnet design and a coil winding design of lap winding as an example, the structural parameter conditions of the structure type include the parameter conditions of the first basic structure and the parameter conditions of the second basic structure. Therefore, each set of structural parameters in the initial population includes the structural parameters of the first basic structure and the structural parameters of the second basic structure. Here, the first basic structure is the first basic structure type, and the second basic structure is the first basic structure type. For any set of structural parameters, the structural parameters of the first basic structure are input into the first thrust prediction model corresponding to the first basic structure type to obtain the Y-direction thrust data and Z-direction thrust data of the first basic structure; the structural parameters of the second basic structure are input into the first thrust prediction model corresponding to the first basic structure type to obtain the Y-direction thrust data and Z-direction thrust data of the second basic structure. The Y-direction thrust data of the first basic structure and the Y-direction thrust data of the second basic structure are superimposed to obtain the Y-direction thrust data of this set of structural parameters in the initial population; the Z-direction thrust data of the first basic structure and the Z-direction thrust data of the second basic structure are superimposed to obtain the Z-direction thrust data of this set of structural parameters in the initial population.

[0079] Step 403: Based on the initial population, the thrust data corresponding to each set of structural parameters, and the constraints, the population is iterated using an optimization algorithm and the thrust prediction model until the preset termination condition is met, generating at least one set of target parameters for the magnetic thrust component.

[0080] The preset termination condition is used to determine whether the population iteration is complete, ensuring that the algorithm stops and outputs the current optimal result within a reasonable time. For example, the termination condition could be stopping iteration when the change in population thrust data over the last 20 generations is less than 10 to the power of -6, or stopping iteration after reaching a preset number of iterations. This embodiment does not impose specific limitations on the termination condition; it can be set according to actual usage requirements.

[0081] The thrust data corresponding to each set of structural parameters in the initial population is fed back to the optimization algorithm. This allows the algorithm to evaluate individual performance using the thrust data and constraints of the initial population's structural parameters, perform genetic operations to generate offspring, and then use a thrust prediction model to determine the thrust data corresponding to each set of structural parameters in the offspring population. This process iterates until a preset termination condition is reached. Finally, a solution vector satisfying all constraints and exhibiting balanced performance is extracted from the final population and formatted as a combination of structural parameters, i.e., at least one set of target parameters. This set of target parameters includes the structural parameters corresponding to each basic structure.

[0082] This embodiment defines the range of parameter variables by structural parameter conditions and generates a diverse initial population as the starting point for optimization. It uses a thrust prediction model to efficiently evaluate the performance of structural parameters to obtain thrust data. Under the optimization algorithm framework, it continuously performs population evolution operations by combining thrust data and constraints to gradually screen out high-performance and compliant candidate solutions. Finally, when the preset termination conditions are met, the target parameters are output. By introducing an automated search mechanism to overcome the limitations of manual setting, and by accelerating the performance evaluation process through the thrust prediction model, it reduces the reliance on repeated manual intervention and high-cost physical simulation, thereby reducing the design cycle and R&D cost of the magnetic thrust component.

[0083] In one embodiment, such as Figure 6 As shown, a method for generating an initial population is provided, which specifically includes the following steps:

[0084] Step 601: Based on the structural parameter conditions, generate multiple sets of original parameters using an optimization algorithm.

[0085] Multiple sets of initial parameters constitute the first-generation candidate solution set generated during the iteration of the optimization algorithm. These initial parameters are generated using uniform random sampling or Latin hypercube sampling methods, based on the value ranges of each parameter variable in the structural parameter conditions. Each set of initial parameters includes the fundamental parameters for each basic structure corresponding to the structural type.

[0086] Taking a structure type with a double-single-pole magnet design and a coil winding design of lapped winding as an example, the structural parameter conditions of the structure type include the parameter conditions of the first basic structure and the parameter conditions of the second basic structure. Therefore, each set of original parameters includes the basic parameters of the first and second basic structures. More specifically, each set of original parameters includes: the basic parameters of the first basic structure and the basic parameters of the second basic structure. The basic parameters of the first basic structure include: magnet length, magnet width, magnet height, coil center length, coil center width, spacing between the coil and the magnet, coil winding wire diameter, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding. The basic parameters of the second basic structure include: magnet length, magnet width, magnet height, coil center length, coil center width, spacing between the coil and the magnet, coil winding wire diameter, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding.

[0087] Step 602: Determine the positive stroke relative displacement and negative stroke relative displacement of each of the foundation structures to obtain the positive stroke basic data and negative stroke basic data of each foundation structure.

[0088] Given a set of initial parameters, the relative displacement between each magnet and the coil can be calculated based on the total height of all magnets and the height of each individual magnet. Taking a structure with a double-single-pole magnet design and a coil winding design of lap winding as an example, each set of initial parameters includes: the basic parameters of the first base structure and the basic parameters of the second base structure. Based on the magnet heights corresponding to the first and second base structures, the total magnet height is calculated. In the first base structure, the magnet is the upper magnet, so the relative displacement between this magnet and the coil = (total magnet height / 2) - (upper magnet height / 2); in the second base structure, the magnet is the lower magnet, so the relative displacement between this magnet and the coil = (total magnet height / 2) - (lower magnet height / 2). Thus, the relative displacements between the magnet and the coil corresponding to the first and second base structures are obtained. After obtaining the relative displacements of the magnets and coils corresponding to the first and second base structures, the positive stroke displacement is added to the relative displacements of the magnets and coils to obtain the positive stroke relative displacement, and the negative stroke displacement is added to the relative displacements of the magnets and coils to obtain the negative stroke relative displacement. This determines the positive and negative stroke relative displacements for each base structure. The positive and negative stroke displacements are the positive and negative stroke displacements required by the design.

[0089] For example, adding a positive stroke displacement to the relative displacement of the magnet and coil in the first base structure yields the positive stroke relative displacement of the first base structure; adding a negative stroke displacement to the relative displacement of the magnet and coil in the first base structure yields the negative stroke relative displacement of the first base structure. Similarly, adding a positive stroke displacement to the relative displacement of the magnet and coil in the second base structure yields the positive stroke relative displacement of the second base structure; adding a negative stroke displacement to the relative displacement of the magnet and coil in the second base structure yields the negative stroke relative displacement of the second base structure. After obtaining the positive and negative stroke relative displacements, they are combined with the basic parameters of the corresponding base structure to obtain the positive stroke basic data and negative stroke basic data for each base structure. The positive stroke basic data includes the basic parameters and the positive stroke relative displacement, and the negative stroke basic data includes the basic parameters and the negative stroke relative displacement.

[0090] For example, the basic data for positive stroke includes: magnet length, magnet width, magnet height, coil center length, coil center width, distance between coil and magnet, relative displacement during positive stroke, wire diameter of coil winding, number of turns in the height direction of coil winding, and number of turns in the width direction of coil winding; the basic data for negative stroke includes: magnet length, magnet width, magnet height, coil center length, coil center width, distance between coil and magnet, relative displacement during negative stroke, wire diameter of coil winding, number of turns in the height direction of coil winding, and number of turns in the width direction of coil winding.

[0091] Step 603: Based on the positive stroke basic data and the negative stroke basic data of each basic structure corresponding to each set of original parameters, determine multiple sets of structural parameters.

[0092] A set of structural parameters includes: the basic data for the positive stroke and the basic data for the negative stroke of each basic structure. Taking a structure type with a double single-pole magnet design and a coil winding design of lap winding as an example, each set of structural parameters includes: the basic data for the positive stroke and the basic data for the negative stroke of the first basic structure, and the basic data for the positive stroke and the basic data for the negative stroke of the second basic structure.

[0093] This embodiment can further improve the reliability of structural parameters by adding positive and negative relative displacements to the basic parameters.

[0094] In one embodiment, such as Figure 7 As shown, a thrust data prediction method is provided, which specifically includes the following steps:

[0095] Step 701: Determine the target structural parameters.

[0096] The target structural parameters are any one of multiple sets of structural parameters.

[0097] Step 702: Based on the positive stroke basic data and negative stroke basic data of each basic structure in the target structural parameters, and the thrust prediction model corresponding to each basic structure, determine the positive stroke thrust and negative stroke thrust of each basic structure corresponding to the target structural parameters.

[0098] The positive and negative stroke basic data of each basic structure in the target structural parameters are input into the thrust prediction model of the corresponding basic structure type to determine the positive and negative stroke thrust of each basic structure corresponding to the target structural parameters.

[0099] Taking a structure with a double-single-pole magnet design and a coil winding design of lapped winding as an example, each set of structural parameters includes: basic positive and negative stroke data for the first basic structure, and basic positive and negative stroke data for the second basic structure. The first basic structure is a first basic structure type, and the second basic structure is a first basic structure type. The basic positive stroke data of the first basic structure is input into the first thrust prediction model corresponding to the first basic structure type to obtain the positive thrust in the Y and Z directions of the first basic structure; the basic negative stroke data of the first basic structure is input into the first thrust prediction model corresponding to the first basic structure type to obtain the negative thrust in the Y and Z directions of the first basic structure; the basic positive stroke data of the second basic structure is input into the first thrust prediction model corresponding to the first basic structure type to obtain the positive thrust in the Y and Z directions of the second basic structure; the basic negative stroke data of the second basic structure is input into the first thrust prediction model corresponding to the first basic structure type to obtain the negative thrust in the Y and Z directions of the second basic structure.

[0100] Step 703: Determine the total positive thrust and total negative thrust corresponding to the target structural parameters based on the positive and negative thrust of each basic structure corresponding to the target structural parameters.

[0101] The total thrust in the positive Y-direction stroke of each foundation structure is obtained by superimposing the thrust in the positive Y-direction stroke corresponding to the target structural parameters; the total thrust in the negative Y-direction stroke of each foundation structure is obtained by superimposing the thrust in the negative Y-direction stroke corresponding to the target structural parameters; the total thrust in the positive Z-direction stroke of each foundation structure is obtained by superimposing the thrust in the positive Z-direction stroke corresponding to the target structural parameters; and the total thrust in the negative Z-direction stroke of each foundation structure is obtained by superimposing the thrust in the negative Z-direction stroke corresponding to the target structural parameters. Thrust superposition simply involves adding the thrusts with direction. Thrust superposition can also be achieved by adding the absolute values ​​of all thrusts.

[0102] This embodiment uses the positive and negative stroke base data of the foundation structure, along with the corresponding thrust prediction model, to predict thrust data, avoiding the time-consuming problem of traditional finite element simulation. Furthermore, using the thrust prediction model improves prediction accuracy.

[0103] In one embodiment, the thrust prediction model includes: a first thrust prediction model, a second thrust prediction model, a third thrust prediction model, and a fourth thrust prediction model; the first thrust prediction model is used to predict thrust for the first infrastructure type; the second thrust prediction model is used to predict thrust for the second infrastructure type; the third thrust prediction model is used to predict thrust for the third infrastructure type; and the fourth thrust prediction model is used to predict thrust for the fourth infrastructure type.

[0104] The following explanation uses the training of the first thrust prediction model as an example. Figure 8 As shown, a method for training a first thrust prediction model is provided, which specifically includes the following steps:

[0105] Step 801: Construct the first training dataset corresponding to the first basic structure type.

[0106] In the first basic structure type, the coil is parallel to the magnet, and the coil is a multi-layered winding. A first training dataset is constructed for the first basic structure type. This first training dataset includes sample structural parameters and corresponding sample thrust data.

[0107] Specifically, a simulation model of the first basic structure type and corresponding parameter conditions are constructed. For the first basic structure type where the coil is parallel to the magnet and the coil is a multi-layered winding, a simulation model of the first basic structure type is constructed. Corresponding parameter conditions for the first basic structure type are set, including: magnet length, magnet width, magnet height, coil center length, coil center width, spacing between the coil and the magnet, relative displacement between the coil and the magnet, coil winding diameter, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding. The numerical ranges of each parameter condition are described in the above embodiments and will not be repeated here.

[0108] Parameter sampling is performed based on the parameter conditions corresponding to the first basic structure type to determine multiple sets of sample structural parameters. Parameter sampling is the process of selecting discretized parameter combinations under given parameter conditions according to a specific strategy. Parameter sampling can be one or more of Latin hypercube sampling, random sampling, and full factorial experimental sampling. Latin hypercube sampling is the preferred parameter sampling method. The multiple sets of sample structural parameters are a series of specific numerical combinations generated during the parameter sampling process, each set representing a possible magnetic thrust component structural configuration. Specifically, a set of sample structural parameters includes: magnet length, magnet width, magnet height, coil center length, coil center width, spacing between the coil and magnet, relative displacement between the coil and magnet, coil winding diameter, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding.

[0109] Simulations are performed based on a simulation model of the first basic structure type and multiple sets of sample structural parameters to determine the sample thrust data for each set of sample structural parameters. The sample thrust data refers to the thrust data corresponding to each set of sample structural parameters obtained through simulation using the simulation model of the first basic structure type. The thrust data includes thrust data in the Y direction and thrust data in the Z direction. Each set of sample structural parameters is input into the simulation model of the first basic structure type to obtain the sample thrust data corresponding to each set of sample structural parameters. Specifically, a set of sample structural parameters is input into the simulation model of the first basic structure type and simulation is performed to obtain the sample thrust data corresponding to the first basic structure type. Specifically, each set of sample structural parameters is substituted into the simulation model of the first basic structure type to run simulation calculations, and the thrust output results are extracted. Simulation calculations can be automatically executed by batch calling the Maxwell simulation software API.

[0110] After obtaining the sample thrust data corresponding to the structural parameters of each group of samples, the first training dataset corresponding to the first basic structure type is constructed based on the structural parameters of each group of samples and their corresponding sample thrust data.

[0111] Step 802: Train the neural network model using the first training dataset to obtain the first thrust prediction model.

[0112] The neural network model can be any type of neural network model; this embodiment does not impose any specific limitations. The original neural network model in the example can be an FNN network model, which is configured with 5 fully connected layers and uses LeakyReLU as the activation function. The original neural network model can also be a CNN convolutional neural network.

[0113] This embodiment determines multiple sets of sample structural parameters by sampling based on the parameter conditions corresponding to the first basic structure type. Discrete parameter combinations are then uniformly extracted within the parameter conditions to obtain a high-quality dataset. A nonlinear mapping from input parameters to output performance is established through iterative optimization of network weights, generating a specialized, high-precision thrust prediction model.

[0114] Understandably, a second thrust prediction model can also be trained. Specifically, a simulation model for a second basic structure type and corresponding parameter conditions are constructed. Parameter sampling is performed based on the parameter conditions corresponding to the second basic structure type to determine multiple sets of sample structural parameters. Simulations are conducted based on the simulation model for the second basic structure type and the multiple sets of sample structural parameters to determine the sample thrust data for each set of sample structural parameters. A second training dataset corresponding to the second basic structure type is constructed, including sample structural parameters and corresponding sample thrust data. The neural network model is trained using the second training dataset to obtain the second thrust prediction model. The specific training method for the second thrust prediction model is the same as that for the first thrust prediction model described above, and will not be repeated here.

[0115] Understandably, a third thrust prediction model can also be trained. Specifically, a simulation model for a third infrastructure type and corresponding parameter conditions are constructed. Parameter sampling is performed based on the parameter conditions corresponding to the third infrastructure type to determine multiple sets of sample structural parameters. Simulations are conducted based on the simulation model for the third infrastructure type and the multiple sets of sample structural parameters to determine the sample thrust data for each set of sample structural parameters. A third training dataset corresponding to the third infrastructure type is constructed, including sample structural parameters and corresponding sample thrust data. The neural network model is trained using the third training dataset to obtain the third thrust prediction model. The specific training method for the third thrust prediction model is the same as that for the first thrust prediction model described above, and will not be repeated here.

[0116] Understandably, a fourth thrust prediction model can also be trained. Specifically, a simulation model for a fourth basic structure type and the corresponding parameter conditions for that type are constructed. Parameter sampling is performed based on the parameter conditions corresponding to the fourth basic structure type to determine multiple sets of sample structural parameters. Simulations are then performed based on the simulation model for the fourth basic structure type and the multiple sets of sample structural parameters to determine the sample thrust data for each set of sample structural parameters. A fourth training dataset corresponding to the fourth basic structure type is constructed, including sample structural parameters and corresponding sample thrust data. The neural network model is trained using the fourth training dataset to obtain the fourth thrust prediction model. The specific training method for the fourth thrust prediction model is the same as that for the first thrust prediction model described above, and will not be repeated here.

[0117] In one embodiment, an AI-based automatic design method for a camera coil motor magnetic thrust structure is provided, specifically including the following steps:

[0118] Step 1: Sampling of Basic Structure Design. Investigate different motor thrust design structures, analyze thrust design and simulation methods, and sample common basic structures to achieve a parametric design of the basic structure that encompasses the overall design of the motor module's magnetic thrust. The final basic structure types include: First basic structure type: coil parallel to the magnet, with the coil being a stacked winding. Second basic structure type: coil parallel to the magnet, with the coil being a staggered winding. Third basic structure type: coil perpendicular to the magnet, with the coil being a stacked winding. Fourth basic structure type: coil perpendicular to the magnet, with the coil being a staggered winding.

[0119] Step 2: Perform parametric modeling and parameter design for the basic structures. Perform parametric structural design for the four basic structures mentioned above, removing critical structural parameters and retaining important ones. Use Maxwell simulation software for secondary development parametric modeling and design the parameter range.

[0120] Specifically, the main parameters corresponding to the basic structural design include: the length, width, and height of the magnet; the length and width of the central space of the coil; the diameter of the copper wire in the coil; the number of winding layers (i.e., the number of turns in the height direction); the number of winding coils (i.e., the number of turns in the thickness direction); the boundary distance between the magnet and the coil; and the relative displacement (vertical direction) between the magnet and the coil. Please refer to the description of the above embodiment for the range of each main parameter; it will not be repeated here. Non-critical parameters include: a 0.1mm radius for the magnet's edge; and a 0.15mm radius for the inner edge of the coil's center. Maxwell software supports user-defined parameter models (UDP), implemented through Python scripting. Scripting mainly involves: defining structural parameters, creating the magnet model, creating the coil model, and creating radius rounding, among other software-based operations.

[0121] Step 3: Parameter Sampling and Batch Simulation. Sampling is performed on the main parameters corresponding to the four basic structures. Latin hypercube sampling can be used. The sampled data should number at least 5 million sets. After sampling, Maxwell batch simulations are performed using a high-performance computer to process the simulation data and construct training datasets for different basic structures. The simulation results are the Y-direction thrust data and Z-direction thrust data corresponding to each set of main parameters. During Latin hypercube sampling, the value range of each parameter variable is divided into equally probable intervals, and a value is randomly selected within each interval. These values ​​are then randomly combined with the values ​​of other variables to ensure uniform spatial distribution and coverage of the sample points.

[0122] Step 4, Model Training. Using Maxwell batch simulation results, construct training datasets for different infrastructure types. Provide this million-level dataset to the AI ​​model (FNN network model) for training. The trained model will possess parameterized structural thrust prediction capabilities. After training, we obtain a first thrust prediction model for the first infrastructure type, a second thrust prediction model for the second infrastructure type, a third thrust prediction model for the third infrastructure type, and a fourth thrust model for the fourth infrastructure type.

[0123] Step 5, Thrust and Structural Prediction: Applying a trained AI model, thrust prediction is performed on the magnet coil design structure using thrust combinations from multiple basic structures. A genetic algorithm combined with the AI ​​model predicts optimal structural design parameters based on thrust requirements and constraints. Client software development integrates the AI ​​prediction model, providing convenient and efficient interaction.

[0124] For thrust prediction, by inputting a set of key parameters corresponding to the basic structure, the corresponding thrust data in the Y direction and thrust data in the Z direction can be predicted in real time.

[0125] For structure prediction, based on the structural parameters of each basic structure corresponding to the structure type, the NSGA2 multi-objective optimization genetic algorithm is used for design optimization. The parameter optimization range for each basic structure is the same as the parameter range during AI model training. The population size of the multi-objective optimization genetic algorithm is set to 100 groups. The termination condition of the multi-objective optimization genetic algorithm is that the change in thrust data in the last 20 generations is less than 10 to the power of -6. The overall structural parameters of the structure type can be decomposed into parameter combinations of multiple basic structures. The trained AI model predicts the thrust data of each basic structure, and the thrust data of each basic structure are superimposed to form the thrust data of the overall structure.

[0126] For example, this example focuses on predicting the thrust of a shared dual-single-stage magnet structure. First, the prediction parameters are designed as follows: upper magnet (length, width, height), lower magnet (length, width, height), center length, diameter, and number of turns of the parallel coil, and center length, diameter, and number of turns of the vertical coil. The minimum target thrust values ​​for the parallel coil and the vertical coil are then set. Each generation of 100 sets of structural parameters generated by the NSGA2 algorithm is then split into four basic structures: upper magnet + parallel coil, upper magnet + vertical coil, lower magnet + parallel coil, and lower magnet + vertical coil. These four basic structures correspond to two structural types and each has a trained AI model. After the AI ​​model predicts the thrust for each of the four basic structures, the thrust of the parallel coil and the vertical coil are summed to obtain the corresponding thrust magnitude. The NSGA2 algorithm iteratively optimizes the structural design based on the predicted thrust magnitude until the thrust requirements and termination conditions are met.

[0127] This embodiment applies AI models to the development of VCM motors, integrating multiple intelligent prediction functions such as AI prediction of magnet coil thrust and structural prediction optimization. This solution boasts rapid prediction capabilities and high accuracy. It significantly enhances the efficient design capability of motor magnetic thrust, enabling automatic parametric generation of designs from models, thus reducing the workload and increasing efficiency for designers. By automatically iterating to optimize the structure, it reduces trial-and-error costs and significantly improves design efficiency.

[0128] In one embodiment, such as Figure 9 As shown, a magnetic thrust prediction method is provided, including the following steps:

[0129] Step 901: Obtain the structural type and structural parameters to be predicted for the magnetic thrust component.

[0130] This embodiment can be used to predict the thrust magnitude corresponding to the structural parameters of a magnetic thrust assembly after the structural parameter design has been completed, based on the structural parameters and using a thrust prediction model. The structural parameters to be predicted are the structural parameters of the magnetic thrust assembly for which thrust prediction is required. Each structural type corresponds to at least one basic structure, and the structural parameters to be predicted include the basic parameters to be predicted for each of the at least one basic structure corresponding to the structural type. The structural type and the structural parameters to be predicted for the magnetic thrust assembly are obtained, and the basic parameters to be predicted for each basic structure corresponding to the structural type are determined.

[0131] Taking a structure with a double single-pole magnet design and a coil winding design of lap winding as an example, the structural parameters to be predicted include: the basic parameters to be predicted for the first basic structure and the basic parameters to be predicted for the second basic structure.

[0132] Step 902: Determine the thrust magnitude of the magnetic thrust assembly based on the thrust prediction model according to the structure type and the parameters of the structure to be predicted.

[0133] The thrust prediction model is used to predict the thrust of each of the aforementioned foundation structures. The predicted foundation parameters for each foundation structure corresponding to the structure type are input into the corresponding thrust prediction model to obtain the thrust magnitude of each foundation structure. These thrust magnitudes are then superimposed to obtain the thrust magnitude of the magnetic thrust assembly. Taking a structure type with a double-single-pole magnet design and a coil winding design of lapped winding as an example, the predicted foundation parameters for the first foundation structure are input into the first thrust prediction model corresponding to the first foundation structure type to obtain the Y-direction and Z-direction thrusts of the first foundation structure. The predicted foundation parameters for the second foundation structure are input into the first thrust prediction model corresponding to the first foundation structure type to obtain the Y-direction and Z-direction thrusts of the second foundation structure. The Y-direction thrusts of the first and second foundation structures are superimposed to obtain the Y-direction thrust magnitude of the magnetic thrust assembly. The Z-direction thrusts of the first and second foundation structures are superimposed to obtain the Z-direction thrust magnitude of the magnetic thrust assembly.

[0134] This embodiment improves prediction accuracy by using a thrust prediction model to predict thrust data. The structure of the magnetic thrust assembly is divided into multiple basic structural types, and thrust prediction is performed for each basic structural type, enabling adaptation to different structural types and improving versatility. It is understood that detailed descriptions of structural types, structural parameters, and thrust prediction models can be found in the above embodiments and will not be repeated here.

[0135] In one embodiment, any of the foundation structures includes a magnet and a coil. The type of any of the foundation structures is one of a first foundation structure type, a second foundation structure type, a third foundation structure type, or a fourth foundation structure type. Each structure type corresponds to at least one foundation structure. Each foundation structure includes a magnet and a coil. The foundation structures include four types, with different positions of the magnet and coil and different copper wire winding methods for each type. Specifically, in the first foundation structure type, the coil is parallel to the magnet and the coil is lapped. In the second foundation structure type, the coil is parallel to the magnet and the coil is staggered. In the third foundation structure type, the coil is perpendicular to the magnet and the coil is lapped. In the fourth foundation structure type, the coil is perpendicular to the magnet and the coil is staggered.

[0136] In one embodiment, the thrust prediction model includes: a first thrust prediction model, a second thrust prediction model, a third thrust prediction model, and a fourth thrust prediction model; the first thrust prediction model is used to predict thrust for the first infrastructure type; the second thrust prediction model is used to predict thrust for the second infrastructure type; the third thrust prediction model is used to predict thrust for the third infrastructure type; and the fourth thrust prediction model is used to predict thrust for the fourth infrastructure type.

[0137] In one embodiment, the predicted basic parameters corresponding to the basic structure include at least one of the following: magnet length, magnet width, magnet height, coil center length, coil center width, distance between the coil and the magnet, relative displacement between the coil and the magnet, diameter of the coil winding, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding.

[0138] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0139] Based on the same inventive concept, this application also provides a magnetic thrust component design apparatus for implementing the magnetic thrust component design method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more embodiments of the magnetic thrust component design apparatus provided below can be found in the limitations of the magnetic thrust component design method described above, and will not be repeated here.

[0140] In one embodiment, such as Figure 10 As shown, a magnetic thrust component design device is provided, comprising: an acquisition module 100, a construction module 200, and a design module 300, wherein:

[0141] The acquisition module 100 is used to acquire the structural type and constraints of the magnetic thrust assembly; the structural type corresponds to at least one basic structure.

[0142] The construction module 200 is used to construct structural parameter conditions according to the structure type.

[0143] Design module 300 is used to determine at least one set of target parameters for the magnetic thrust component based on the structural parameter conditions and the constraint conditions, using an optimization algorithm and a thrust prediction model; the thrust prediction model is used to predict the thrust for each of the foundation structures respectively.

[0144] The design module 300 is further configured to generate an initial population based on the structural parameter conditions and an optimization algorithm; the initial population includes multiple sets of structural parameters of the magnetic thrust component; determine the thrust data corresponding to each set of structural parameters based on the initial population and the thrust prediction model; and perform population iteration based on the initial population, the thrust data corresponding to each set of structural parameters, and the constraints, according to the optimization algorithm and the thrust prediction model, until a preset termination condition is met, to generate at least one set of target parameters for the magnetic thrust component.

[0145] The design module 300 is further configured to generate multiple sets of original parameters based on the structural parameter conditions and an optimization algorithm; one set of original parameters includes the basic parameters of each basic structure corresponding to the structural type; determine the positive stroke relative displacement and negative stroke relative displacement of each basic structure to obtain the positive stroke basic data and negative stroke basic data of each basic structure; the positive stroke basic data includes the basic parameters and the positive stroke relative displacement, and the negative stroke basic data includes the basic parameters and the negative stroke relative displacement; and determine multiple sets of structural parameters based on the positive stroke basic data and negative stroke basic data of each basic structure corresponding to each set of original parameters.

[0146] The design module 300 is further configured to determine target structural parameters; the target structural parameters are any one set of structural parameters from a plurality of sets of structural parameters; based on the positive stroke basic data and the negative stroke basic data of each basic structure in the target structural parameters, and the thrust prediction model corresponding to each basic structure, the positive stroke thrust and negative stroke thrust of each basic structure corresponding to the target structural parameters are determined; based on the positive stroke thrust and negative stroke thrust of each basic structure corresponding to the target structural parameters, the total positive stroke thrust and total negative stroke thrust of the target structural parameters are determined.

[0147] The magnetic thrust component design device also includes a training module.

[0148] The training module is used to construct a first training dataset corresponding to the first infrastructure type. The first training dataset includes sample structure parameters and corresponding sample thrust data. The neural network model is trained using the first training dataset to obtain the first thrust prediction model.

[0149] The training module is also used to construct a simulation model of the first basic structure type and the parameter conditions corresponding to the first basic structure type; to perform parameter sampling based on the parameter conditions corresponding to the first basic structure type to determine multiple sets of sample structure parameters; and to perform simulation based on the simulation model of the first basic structure type and the multiple sets of sample structure parameters to determine the sample thrust data of each set of sample structure parameters.

[0150] Each module in the aforementioned magnetic thrust component design device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0151] In one embodiment, a magnetic thrust prediction device is provided, comprising:

[0152] A receiving module is used to acquire the structure type and the structural parameters to be predicted of the magnetic thrust assembly, wherein the structure type corresponds to at least one basic structure, and the structural parameters to be predicted include the basic parameters to be predicted for each of the at least one basic structure corresponding to the structure type.

[0153] The prediction module is used to determine the thrust magnitude of the magnetic thrust assembly based on the structure type and the parameters of the structure to be predicted, using a thrust prediction model; wherein the thrust prediction model is used to predict the thrust for each of the foundation structures.

[0154] Each module in the aforementioned magnetic thrust prediction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0155] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 11As shown, the computer device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a magnetic thrust component design method or a magnetic thrust prediction method.

[0156] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0157] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement any of the magnetic thrust component design methods or a magnetic thrust prediction method described in the above embodiments.

[0158] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements any of the magnetic thrust component design methods or a magnetic thrust prediction method described in the above embodiments.

[0159] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0160] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0161] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A design method for a magnetic thrust assembly, characterized in that, The method includes: Obtain the structural type and constraints of the magnetic thrust assembly; the structural type corresponds to at least one basic structure. Based on the structure type, construct the structural parameter conditions; Based on the structural parameter conditions and the constraint conditions, at least one set of target parameters for the magnetic thrust assembly is determined using an optimization algorithm and a thrust prediction model; the thrust prediction model is used to predict the thrust for each of the foundation structures respectively. Each of the aforementioned basic structures includes a magnet and a coil; each of the aforementioned basic structures is of one of a first basic structure type, a second basic structure type, a third basic structure type, or a fourth basic structure type; in the first basic structure type, the coil is parallel to the magnet and the coil is a lapped winding; in the second basic structure type, the coil is parallel to the magnet and the coil is a staggered winding; in the third basic structure type, the coil is perpendicular to the magnet and the coil is a lapped winding; in the fourth basic structure type, the coil is perpendicular to the magnet and the coil is a staggered winding.

2. The method according to claim 1, characterized in that, The structural parameter conditions include parameter conditions for at least one of the basic structures.

3. The method according to claim 2, characterized in that, Each of the aforementioned basic structures includes at least one of the following parameter conditions: magnet length condition, magnet width condition, magnet height condition, coil center length condition, coil center width condition, coil and magnet spacing condition, coil and magnet relative displacement condition, coil winding wire diameter condition, coil winding number of turns in the height direction condition, and coil winding number of turns in the width direction condition.

4. The method according to claim 1, characterized in that, The constraints include at least one of the following: thrust constraints and spatial constraints; wherein the thrust constraints are used to constrain the thrust magnitude of the magnetic thrust assembly, and the spatial constraints are used to constrain the spatial dimensions of the magnetic thrust assembly.

5. The method according to claim 1, characterized in that, The step of determining at least one set of target parameters for the magnetic thrust assembly based on the structural parameter conditions and the constraint conditions, using an optimization algorithm and a thrust prediction model, includes: Based on the structural parameter conditions, an initial population is generated using an optimization algorithm; the initial population includes multiple sets of structural parameters of the magnetic thrust component. Based on the initial population and the thrust prediction model, determine the thrust data corresponding to each set of structural parameters; Based on the initial population, the thrust data corresponding to each set of structural parameters, and the constraints, the population is iterated using the optimization algorithm and the thrust prediction model until a preset termination condition is met, generating at least one set of target parameters for the magnetic thrust component.

6. The method according to claim 1, characterized in that, The thrust prediction model includes: a first thrust prediction model, a second thrust prediction model, a third thrust prediction model, and a fourth thrust prediction model; The first thrust prediction model is used to predict thrust for the first type of infrastructure. The second thrust prediction model is used to predict thrust for the second type of infrastructure. The third thrust prediction model is used to predict thrust for the third type of infrastructure. The fourth thrust prediction model is used to predict thrust for the fourth type of infrastructure.

7. The method according to claim 6, characterized in that, The method further includes: Construct a first training dataset corresponding to the first basic structure type. The first training dataset includes sample structure parameters and corresponding sample thrust data. The neural network model is trained using the first training dataset to obtain the first thrust prediction model.

8. The method according to claim 7, characterized in that, Before constructing the first training dataset corresponding to the first infrastructure type, the method further includes: Construct a simulation model of the first basic structure type and the corresponding parameter conditions for the first basic structure type; Based on the parameter conditions corresponding to the first basic structure type, parameter sampling is performed to determine multiple sets of sample structure parameters; Simulations are performed based on the simulation model of the first basic structure type and multiple sets of sample structure parameters to determine the sample thrust data for each set of sample structure parameters.

9. A method for predicting magnetic thrust, characterized in that, The method includes: Obtain the structural type and predicted structural parameters of the magnetic thrust assembly, wherein the structural type corresponds to at least one basic structure, and the predicted structural parameters include the predicted basic parameters corresponding to each of the at least one basic structure corresponding to the structural type. Based on the structure type and the parameters of the structure to be predicted, the thrust magnitude of the magnetic thrust assembly is determined using a thrust prediction model; wherein, the thrust prediction model is used to predict the thrust for each of the foundation structures respectively; Each of the aforementioned basic structures includes a magnet and a coil; each of the aforementioned basic structures is of one of a first basic structure type, a second basic structure type, a third basic structure type, or a fourth basic structure type; in the first basic structure type, the coil is parallel to the magnet and the coil is a lapped winding; in the second basic structure type, the coil is parallel to the magnet and the coil is a staggered winding; in the third basic structure type, the coil is perpendicular to the magnet and the coil is a lapped winding; in the fourth basic structure type, the coil is perpendicular to the magnet and the coil is a staggered winding.

10. The method according to claim 9, characterized in that, The thrust prediction model includes: a first thrust prediction model, a second thrust prediction model, a third thrust prediction model, and a fourth thrust prediction model; The first thrust prediction model is used to predict thrust for the first type of infrastructure. The second thrust prediction model is used to predict thrust for the second type of infrastructure. The third thrust prediction model is used to predict thrust for the third type of infrastructure. The fourth thrust prediction model is used to predict thrust for the fourth type of infrastructure.

11. The method according to claim 9, characterized in that, The predicted basic parameters corresponding to the basic structure include at least one of the following: magnet length, magnet width, magnet height, coil center length, coil center width, distance between the coil and the magnet, relative displacement between the coil and the magnet, diameter of the coil winding, number of turns in the height direction of the coil winding, and number of turns in the width direction of the coil winding.

12. A magnetic thrust assembly design device, characterized in that, The device includes: The acquisition module is used to acquire the structural type and constraints of the magnetic thrust assembly; the structural type corresponds to at least one basic structure. A construction module is used to construct structural parameter conditions based on the structure type; The design module is used to determine at least one set of target parameters for the magnetic thrust assembly based on the structural parameter conditions and the constraint conditions, using an optimization algorithm and a thrust prediction model; the thrust prediction model is used to predict the thrust for each of the foundation structures respectively. Each of the aforementioned basic structures includes a magnet and a coil; each of the aforementioned basic structures is of one of a first basic structure type, a second basic structure type, a third basic structure type, or a fourth basic structure type; in the first basic structure type, the coil is parallel to the magnet and the coil is a lapped winding; in the second basic structure type, the coil is parallel to the magnet and the coil is a staggered winding; in the third basic structure type, the coil is perpendicular to the magnet and the coil is a lapped winding; in the fourth basic structure type, the coil is perpendicular to the magnet and the coil is a staggered winding.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8 or claims 9 to 11.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8 or claims 9 to 11.

Images