Three-section permanent magnet synchronous linear motor and design and optimization method

Abstract

The invention discloses a three-section permanent magnet synchronous linear motor and a design and optimization method, and relates to the technical field of motors. An existing permanent magnet synchronous linear motor has remarkable problems in the aspects of thrust fluctuation, insufficient thrust output, low permanent magnet utilization rate, complex optimization process and the like, and the technical problems urgently need to be overcome through new design and optimization methods. According to the technical scheme, the three-section type permanent magnet synchronous linear motor is of a three-section type double-layer consequent pole permanent magnet structure and comprises a three-section type structure and a double-layer magnetic pole structure; the three-section type structure comprises at least two opposite convex iron poles, main magnetic poles are arranged at the top ends of the two opposite convex iron poles, and the auxiliary magnetic poles are arranged on the outer sides of the two main magnetic poles respectively and wrap the convex iron poles; the double-layer magnetic pole structure comprises an epoxy resin seal and a stator winding arranged in the epoxy resin seal; and the double-layer magnetic pole structure is arranged between the two opposite convex iron poles. And the method is suitable for being applied to the work of putting the PMSLM into various industrial fields.

Description

Technical Field

[0001] The invention relates to the field of motor technology, and in particular to a three-stage permanent magnet synchronous linear motor and its design and optimization. Background Art

[0002] In recent years, permanent magnet synchronous linear motors (PMSLM) have been widely used in many fields such as modern industrial automation, high-speed transportation systems and precision positioning due to their unique high efficiency, high precision and fast response characteristics. PMSLM achieves non-contact transmission and rapid start-up with high acceleration on automated production lines, giving it significant advantages over traditional rotary motors on high-speed, heavy-loaded assembly lines. For example, in electronic assembly and logistics transportation, the high-precision linear drive characteristics of PMSLM can achieve fast and accurate transmission of materials, greatly improving the production efficiency and stability of the system.

[0003] PMSLM also plays an important role in the fields of high-speed rail transportation and precision manufacturing equipment. For example, in the drive system of high-speed rail and in precision manufacturing equipment such as laser engraving and 3D printing, PMSLM can provide stable and precise motion control to meet the requirements of high-precision manufacturing and processing. These applications benefit from the high thrust output and non-contact characteristics of PMSLM without mechanical wear, which greatly improves the durability and economy of the equipment.

[0004] However, although PMSLM has shown great potential in many fields, there are still several key problems in the existing technology, which seriously restrict its wider application and performance improvement. First, the thrust output of traditional PMSLM often has obvious fluctuations. Due to the uneven distribution of the magnetic field, this fluctuation makes it difficult for the system to ensure the smoothness of movement in high-precision applications, which directly affects the processing accuracy and operation stability of the equipment. Secondly, the lack of thrust is also a bottleneck of current technology. Especially in the case of large load changes or high-speed operation, the traditional single-layer magnetic pole design is difficult to provide sufficient thrust, which limits the applicability of the motor in heavy-load situations.

[0005] In addition, the problem of low utilization of permanent magnets also needs to be solved. In existing designs, the magnetic field distribution of permanent magnets is difficult to be fully utilized, resulting in material waste and increased production costs. With the continuous rise in the cost of rare earth materials, improving the utilization efficiency of permanent magnets has become a key issue in reducing the cost of motor manufacturing. Finally, in order to accurately simulate and optimize the working state of the motor, it is usually necessary to establish complex mathematical models, such as finite element analysis (FEA), which makes the design and optimization process time-consuming and complicated. Summary of the invention

[0006] In summary, the existing permanent magnet synchronous linear motor has significant problems in thrust fluctuation, insufficient thrust output, low permanent magnet utilization and complex optimization process, and it is urgent to overcome these technical difficulties through new design and optimization methods. The present invention aims to solve the above technical problems through a three-stage double-layer Halbach structure design, improve the thrust output stability, thrust size and permanent magnet utilization efficiency of the motor. Specifically:

[0007] A three-stage permanent magnet synchronous linear motor, the linear motor is a three-stage double-layer alternating pole permanent magnet structure, comprising:

[0008] Three-section structure and double-layer magnetic pole structure;

[0009] The three-stage structure includes at least two opposite salient iron poles, the tops of the two opposite salient iron poles are both provided with main magnetic poles, and auxiliary magnetic poles are respectively arranged outside the two main magnetic poles and wrap the salient iron stage;

[0010] The double-layer magnetic pole structure includes an epoxy grease seal and a stator winding arranged inside the epoxy grease seal;

[0011] The double-layer magnetic pole structure is arranged between the two opposite convex iron stages.

[0012] Furthermore, a preferred embodiment is provided, in which in the double-layer magnetic pole structure, different layers of magnetic poles are arranged in a complementary manner.

[0013] Furthermore, a preferred embodiment is provided, wherein there are four auxiliary magnetic poles.

[0014] Furthermore, a preferred embodiment is provided, wherein the double-layer magnetic pole structure is a double-layer Halbach structure.

[0015] Furthermore, a preferred embodiment is provided, wherein the epoxy seal is used to provide insulation and mechanical support capabilities for the stator winding.

[0016] Based on the same inventive concept, the present invention also provides a three-stage permanent magnet synchronous linear motor design and optimization method for designing the linear motor, comprising:

[0017] The step of initializing a population using a chaotic mapping sequence according to a multi-objective raccoon optimization algorithm and enhancing the global search capability of the multi-objective raccoon optimization algorithm through a Lévy flight strategy;

[0018] The step of using a BP neural network proxy model to improve the accuracy of the multi-objective raccoon optimization algorithm;

[0019] The step of optimizing the linear motor according to the multi-objective raccoon optimization algorithm with improved accuracy.

[0020] Based on the same inventive concept, the present invention also provides a three-stage permanent magnet synchronous linear motor design and optimization device, which is used to design the linear motor, including:

[0021] A module for initializing a population using a chaotic mapping sequence according to a multi-objective raccoon optimization algorithm and enhancing the global search capability of the multi-objective raccoon optimization algorithm through a Lévy flight strategy;

[0022] A module for improving the accuracy of the multi-objective raccoon optimization algorithm using a BP neural network proxy model;

[0023] A module for optimizing the linear motor according to a multi-objective raccoon optimization algorithm with improved accuracy.

[0024] Based on the same inventive concept, the present invention also provides a computer storage medium for storing a computer program. When the computer reads the computer program, the computer executes the method described.

[0025] Based on the same inventive concept, the present invention also provides a computer, including a processor and a storage medium. When the processor reads a computer program stored in the storage medium, the computer executes the method described.

[0026] Based on the same inventive concept, the present invention also provides a computer program product, which is a computer program. When the computer program is executed, the method described above is implemented.

[0027] Compared with the prior art, the technical solution provided by the present invention is beneficial in that:

[0028] The three-stage double-layer Halbach structure significantly reduces thrust fluctuations by optimizing the magnetic field distribution. The design uses a three-stage structure and a double-layer Halbach magnetic array, which greatly improves the uniformity of the magnetic field and reduces the thrust fluctuation problem caused by uneven magnetic field in traditional designs. This structural improvement ensures that the motor can provide smooth thrust output in high-precision application scenarios, thereby significantly improving the operating stability and accuracy of the system.

[0029] The design of the double-layer Halbach structure enhances the thrust output capacity of the motor and improves its adaptability under heavy load and high-speed conditions. Compared with the traditional single-layer design, the double-layer structure greatly increases the magnetic flux density, enabling the motor to provide greater thrust under various load conditions. This improvement solves the limitation of traditional permanent magnet synchronous linear motors in terms of insufficient thrust, allowing it to maintain good performance in situations with large load changes and high-speed operation.

[0030] Through the rational use of the double-layer Halbach structure, this implementation significantly improves the utilization efficiency of permanent magnets and reduces material costs. The design of the double-layer structure enables the magnetic field to be distributed and utilized more efficiently, reduces magnetic leakage, and thus reduces the demand for permanent magnets. Compared with traditional designs, this optimization not only saves rare earth materials and reduces manufacturing costs, but also improves the economy and environmental protection of resource utilization.

[0031] The use of BP neural network proxy model significantly reduces the complexity of model calculation. Traditional finite element analysis and other methods are computationally intensive and time-consuming, while this implementation uses a neural network proxy model to reduce computing time and resource consumption while ensuring a certain degree of accuracy. This approach improves the efficiency of the optimization design process, allowing complex optimization problems to be solved at a lower cost and in a shorter time.

[0032] It is suitable for applying PMSLM to various industrial fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is the structural diagram of a three-stage double-layer Halbach alternating-pole permanent magnet linear synchronous motor;

[0034] Figure 2 for Figure 1 Side view of

[0035] Figure 3 for Figure 1 Parameter diagram of

[0036] Figure 4 Optimize the flow chart for multiple objectives;

[0037] Figure 5 This is the simulation verification result diagram of the optimized three-stage double-layer Halbach alternating pole permanent magnet linear synchronous motor;

[0038] Where a represents the distribution of magnetic lines of force, b represents the distribution of magnetic flux density, c represents the waveform of air gap magnetic flux density, and d represents the electromagnetic thrust. DETAILED DESCRIPTION

[0039] In order to make the advantages and benefits of the technical solution provided by the present invention more clearly reflected, the technical solution provided by the present invention is now further described in detail with reference to the accompanying drawings, specifically:

[0040] Embodiment 1: This embodiment provides a three-stage permanent magnet synchronous linear motor, which is a three-stage double-layer alternating pole permanent magnet structure, including:

[0041] Three-section structure and double-layer magnetic pole structure;

[0042] The three-stage structure includes at least two opposite salient iron poles, the tops of the two opposite salient iron poles are both provided with main magnetic poles, and auxiliary magnetic poles are respectively arranged outside the two main magnetic poles and wrap the salient iron stage;

[0043] The double-layer magnetic pole structure includes an epoxy grease seal and a stator winding arranged inside the epoxy grease seal;

[0044] The double-layer magnetic pole structure is arranged between the two opposite convex iron stages.

[0045] Specifically, the design method of the above linear motor is as follows:

[0046] Step 1: Three-stage double-layer Halbach permanent magnet structure design

[0047] Brief description: The three-stage double-layer Halbach structure design makes the magnetic field distribution of the motor more uniform, effectively improving the stability and strength of thrust output.

[0048] The three-section double-layer Halbach structure consists of two main parts. The first part is the primary three-section structure, and the second part is the secondary double-layer structure. The primary structure adopts a three-section Halbach magnetic pole arrangement. Each pair of magnetic poles consists of two convex iron poles and four auxiliary magnetic poles to ensure the concentration effect of the magnetic field and improve the magnetic flux density. The secondary structure adopts a double-layer arrangement. By reasonably arranging the polarity of the magnetic poles, the magnetic field is complementary between different layers, thereby reducing magnetic leakage and achieving uniform distribution of the magnetic field. This design greatly reduces the thrust fluctuation caused by the uneven magnetic field in the traditional single-layer structure, thereby significantly improving the stability and accuracy of the system.

[0049] Step 2: Enhancement of thrust output

[0050] Brief description: The double-layer Halbach structure design enhances the thrust output capability, enabling the motor to provide higher thrust under various load conditions.

[0051] In the double-layer Halbach structure, the reasonable arrangement and arrangement of the magnetic poles can significantly improve the strength of the magnetic field, so that the motor can still maintain stable thrust output when the load changes greatly or when high-speed operation is required. The magnetic flux density between the first layer and the second layer in the double-layer structure reinforces each other, maximizing the utilization of the magnetic field, thereby enhancing the thrust output capacity of the motor. Compared with the traditional single-layer structure design, this solution significantly improves the thrust output capacity, solves the problem of insufficient thrust in the prior art, and enables the motor to maintain good performance under heavy load and high-speed conditions.

[0052] Step 3: Improvement of permanent magnet utilization

[0053] Brief description: The double-layer Halbach structure improves the utilization rate of permanent magnets and reduces material costs.

[0054] The double-layer Halbach structure design in this embodiment fully optimizes the magnetic field direction and strength of the permanent magnet in a limited space, reduces magnetic leakage, and thus improves the utilization efficiency of the permanent magnet. In the traditional single-layer design, the utilization efficiency of the permanent magnet is low due to the uneven distribution of the magnetic field. In the double-layer Halbach structure, by optimizing the arrangement of the magnetic poles, more magnetic fields can be effectively involved in the generation of thrust, thereby reducing the demand for permanent magnets, saving the use of rare earth materials, and reducing production costs. This improvement achieves the dual advantages of economy and environmental protection while ensuring thrust output.

[0055] Step 4: BP neural network agent model in the optimization design process

[0056] Brief description: The BP neural network agent model is used to optimize the design process and reduce the computational complexity.

[0057] In order to simplify the amount of calculation in the optimization design process, the present embodiment introduces a BP neural network proxy model. Although traditional finite element analysis (FEA) can provide high-precision calculation results, its calculation process is complex and time-consuming. The present embodiment adopts a BP neural network proxy model, which is trained through a small number of high-precision calculation points to establish an approximate model that can accurately predict the magnetic field distribution and thrust output. During the optimization process, the proxy model is used to quickly evaluate the performance of different design schemes, which can significantly reduce the calculation time and resource consumption while ensuring a certain accuracy. The BP neural network has strong nonlinear mapping capabilities and good generalization performance, which enables it to provide reliable support in complex optimization problems and greatly improves the design efficiency.

[0058] Step 5: Improvement of optimization algorithm

[0059] Brief description: Improve the optimization algorithm to further improve the overall performance of the motor.

[0060] In the optimization design, this implementation proposes an improved multi-objective optimization algorithm to further improve the performance of the three-stage double-layer Halbach permanent magnet synchronous linear motor. The use of chaotic mapping and Lévy flight strategy enhances the uniformity of the population and the global search capability, thereby avoiding the local optimal problem that may be encountered in traditional optimization algorithms. By reasonably setting the population initialization and search mechanism, the improved algorithm can converge to the optimal solution in a shorter time, effectively improving the thrust density, thrust fluctuation stability and utilization rate of the permanent magnet of the motor, and improving the performance of the overall system.

[0061] Embodiment 2: This embodiment further limits the three-segment permanent magnet linear synchronous motor provided in embodiment 1. In the double-layer magnetic pole structure, magnetic poles in different layers are arranged in a complementary manner.

[0062] Embodiment 3: This embodiment further limits the three-segment permanent magnet linear synchronous motor provided in embodiment 1, and the number of the auxiliary magnetic poles is four.

[0063] Embodiment 4: This embodiment further limits the three-stage permanent magnet linear synchronous motor provided in Embodiment 1, and the double-layer magnetic pole structure is a double-layer Halbach structure.

[0064] Embodiment 5: This embodiment further limits the three-stage permanent magnet synchronous linear motor provided in embodiment 1, and the epoxy seal is used to provide insulation and mechanical support capabilities for the stator winding.

[0065] Embodiment 6: This embodiment provides a three-stage permanent magnet synchronous linear motor design and optimization method, which is used to design the linear motor provided in embodiment 1, including:

[0066] The step of initializing a population using a chaotic mapping sequence according to a multi-objective raccoon optimization algorithm and enhancing the global search capability of the multi-objective raccoon optimization algorithm through a Lévy flight strategy;

[0067] The step of using a BP neural network proxy model to improve the accuracy of the multi-objective raccoon optimization algorithm;

[0068] The step of optimizing the linear motor according to the multi-objective raccoon optimization algorithm with improved accuracy.

[0069] Specific:

[0070] Step 1: Design of three-stage structure

[0071] First, a three-stage structure is designed, which includes at least two relatively arranged salient iron poles. The tops of the two salient iron poles are respectively provided with main magnetic poles, and the function of the main magnetic poles is to enhance the concentration effect of the magnetic field and increase the magnetic flux density. On the outside of each salient iron pole, a plurality of auxiliary magnetic poles are also provided, which are wrapped around the salient iron pole to supplement and guide the uniform distribution of the magnetic field. Through this three-stage structural design, the concentration of the magnetic field is significantly improved, thereby achieving higher thrust output and system stability.

[0072] Step 2: Construction of double-layer magnetic pole structure

[0073] Next, a double-layer magnetic pole structure is designed, which includes an epoxy seal and a stator winding set inside the epoxy seal. The epoxy seal is mainly used to provide insulation protection and mechanical support for the stator winding to ensure that the winding will not be damaged by external factors during long-term operation. The stator winding is combined with a double-layer magnetic pole structure through an optimized winding design to achieve a more uniform magnetic field distribution and improve the output efficiency of the motor. This design ensures the reliability and stability of the motor under high load and high precision conditions.

[0074] Step 3: Optimization based on the improved multi-objective raccoon algorithm

[0075] The optimization design is based on the improved multi-objective raccoon algorithm (MOCOA). Aiming at the problem that the initial population distribution of the MOCOA algorithm is uneven and it is extremely easy to fall into the local optimal solution, an improvement strategy is proposed. First, the population is initialized by the tent chaotic mapping sequence to enhance the uniformity and ergodicity of the population, thereby improving the convergence speed in the early stage. Then, the levy flight search mechanism is used to improve the global search ability of the raccoon in the exploration stage and the local search ability in the development stage. Finally, the nonlinear control factor is used to enhance the balance between the exploration stage and the mining stage. This improvement method effectively improves the thrust density of the motor, reduces the thrust pulsation, and enhances the operation stability of the motor and the utilization rate of the permanent magnet.

[0076] Step 4: Combination of double-layer magnetic pole structure and three-stage structure

[0077] The double-layer magnetic pole structure is set between two opposite convex iron poles to form a complete magnetic circuit structure, so that the magnetic poles of different layers complement and interact with each other, thereby reducing magnetic leakage. Through this reasonable arrangement, the interaction between the magnetic field of different magnetic pole layers is optimized, making the magnetic flux density more uniform and the magnetic field distribution more reasonable. This not only improves the thrust density of the motor, but also improves the adaptability and stability of the system under various working conditions.

[0078] Step 5: Application of Neural Network Agent Model

[0079] A proxy model based on BP neural network is proposed to further improve the accuracy of multi-objective optimization. The model performs multi-objective optimization with average output thrust density, thrust fluctuation, no-load back EMF fundamental amplitude and thrust density as optimization targets. Due to the assumptions in the establishment of the linear motor analytical model, errors in model calculation may occur, which is not suitable for high-precision optimization design. Therefore, the proxy model based on BP neural network model provides a high-precision foundation for subsequent optimization work, making the optimization process more reliable and efficient.

[0080] Step 6: Optimize the motor's thrust output

[0081] The thrust output of the motor has been significantly improved through the combined design of the three-stage structure, the double-layer magnetic pole structure and the optimization algorithm. In this process, the design of the auxiliary magnetic pole improves the concentration and uniformity of the magnetic field, while the double-layer magnetic pole structure further optimizes the overall magnetic circuit configuration, effectively reducing magnetic leakage, so that the motor achieves the best effect in thrust output and energy utilization. The overall design ensures that the motor can provide efficient and stable thrust output under various load conditions.

[0082] Supplementary description: Applicable to industrial automation equipment

[0083] Finally, through the above design steps, the three-stage double-layer Halbach permanent magnet synchronous linear motor has been significantly improved in terms of thrust output stability, thrust density, and permanent magnet utilization. This design is particularly suitable for industrial automation equipment with high-precision positioning and high-speed motion requirements. It can work stably under various complex working conditions and achieve efficient energy conversion and motion control.

[0084] Embodiment 7: This embodiment provides a three-stage permanent magnet synchronous linear motor design and optimization device, which is used to design the linear motor provided in embodiment 1, including:

[0085] A module for initializing a population using a chaotic mapping sequence according to a multi-objective raccoon optimization algorithm and enhancing the global search capability of the multi-objective raccoon optimization algorithm through a Lévy flight strategy;

[0086] A module for improving the accuracy of the multi-objective raccoon optimization algorithm using a BP neural network proxy model;

[0087] A module for optimizing the linear motor according to a multi-objective raccoon optimization algorithm with improved accuracy.

[0088] Embodiment 8: This embodiment provides a computer storage medium for storing a computer program. When the computer reads the computer program, the computer executes the method provided in embodiment 6.

[0089] Embodiment 9: This embodiment provides a computer, including a processor and a storage medium. When the processor reads a computer program stored in the storage medium, the computer executes the method provided in embodiment 6.

[0090] Embodiment 10: This embodiment provides a computer program product, which is a computer program. When the computer program is executed, the method provided in embodiment 6 is implemented.

[0091] Implementation Method XI: Combination Figure 1-5This embodiment is described in detail by using specific examples to further describe the technical solution provided above. Specifically:

[0092] The purpose of this embodiment is to design a three-stage double-layer Halbach permanent magnet synchronous linear motor (PMSLM) to effectively improve the thrust output, thrust stability and utilization rate of the permanent magnet of the motor, and solve the key technical limitations of the traditional design. The specific goals are as follows:

[0093] 1) Improve thrust stability and reduce thrust fluctuations: The three-stage structure and double-layer Halbach magnetic array design are used to optimize the magnetic field distribution and permanent magnet arrangement to significantly reduce thrust fluctuations during motor operation. This structure can provide more stable thrust output in high-precision and high-stability applications, meeting the needs of scenarios with high requirements for stability.

[0094] 2) Enhanced thrust output and improved load adaptability: The double-layer Halbach structure design not only enhances the magnetic flux density of the motor, but also effectively improves the thrust output capacity, making up for the lack of thrust of traditional permanent magnet synchronous linear motors under heavy load or high speed. The design goal is to enable the motor to provide higher thrust output under various load conditions, suitable for applications with large load changes.

[0095] 3) Improve the utilization efficiency of permanent magnets and reduce material costs: This implementation method uses a double-layer Halbach structure to achieve efficient utilization of the magnetic field, reducing the demand for permanent magnets while ensuring thrust output. This can not only save the use of rare earth permanent magnet materials and reduce production costs, but also reduce resource consumption and improve the economy and environmental protection of motor design.

[0096] 4) Reduce model computational complexity: A proxy model is an approximate model that replaces the original high-cost computational model. By running a high-precision model on a small number of key points to obtain data, and then using this data to train the proxy model, the computational time and resource consumption can be greatly reduced while ensuring a certain level of accuracy. This approach is particularly suitable for fields such as optimization design and uncertainty quantification, where each evaluation of the original model may require a lot of computing resources. Commonly used proxy models include polynomial regression, radial basis function (RBF), Kriging (Gaussian process regression), support vector machine (SVM), etc. By rationally selecting and constructing a proxy model, the relationship between computational efficiency and model accuracy can be effectively balanced, thereby improving the efficiency of overall research and development work.

[0097] In summary, the present embodiment aims to improve the comprehensive performance of the permanent magnet synchronous linear motor in terms of thrust stability, output thrust, and material utilization through an innovatively designed structure, thereby solving the limitations of the existing design and making it adaptable to a wider range of industrial application needs.

[0098] 2. Technical solution:

[0099] 1) Three-stage double-layer Halbach alternating pole permanent magnet structure design:

[0100] The magnetic pole adopts a double-layer three-section Halbach structure. The first layer adopts a three-section Halbach structure. Each pair of magnetic poles consists of two convex iron poles and four auxiliary magnetic poles. The second layer adopts a three-section Halbach magnetic pole structure. In order to simplify the plastic packaging process, the primary winding of TDHCP-PMSLM adopts a single-layer rectangular concentrated winding design (see Appendix Figure 1 , 2, 3).

[0101] 2) Multi-objective optimization based on improved multi-objective raccoon algorithm (attached Figure 4 ):

[0102] Aiming at the problem that the initial population of MOCOA algorithm is unevenly distributed and easy to fall into the local optimal solution, an improvement strategy is proposed. The tent chaotic mapping sequence is used to initialize the population, enhance the uniformity and ergodicity of the population, and improve the early convergence speed. Then, the levy flight search mechanism is used to improve the global search ability of coatis in the exploration stage and the local search ability in the development stage; finally, the nonlinear control factor is used to enhance the balance between the exploration stage and the mining stage. In order to further improve the output thrust density of the double-layer permanent magnet linear synchronous motor, reduce thrust pulsation, improve the stability of motor operation and the utilization rate of permanent magnets, the average output thrust density F is used. av , thrust fluctuation F r , no-load back EMF fundamental amplitude E m To optimize the goal.

[0103] 3) Neural Network Agent Model:

[0104] A BP neural network proxy model is proposed to average the output thrust density F av , thrust fluctuation F r , no-load back EMF fundamental amplitude E m and thrust density ρ as optimization targets for multi-objective optimization. Due to the assumptions in the process of establishing the linear motor analytical model, errors in model calculation will occur, which is not suitable for high-precision calculation of optimization design. Therefore, a proxy model based on the BP neural network model is established to provide a high-precision model foundation for subsequent optimization work.

[0105] Among them, the design and optimization method flow chart of the three-stage Halbach permanent magnet synchronous linear motor is shown in the attached Figure 1 As shown, the specific steps include:

[0106] 1) Three-stage double-layer Halbach permanent magnet structure design:

[0107] An ironless permanent magnet synchronous linear motor is used, which consists of a mover (primary) and a stator (secondary). Permanent magnets are installed on the mover, and the back iron is used to support and enhance the magnetic circuit. The stator consists of a coreless multi-phase winding, which generates an electromagnetic field to interact with the mover. Epoxy resin is used to wrap the stator winding to provide insulation and mechanical protection. The TDHCP-PMSLM pole adopts a double-layer three-section halbach structure. The first layer adopts a three-section halbach structure. Each pair of poles consists of two convex iron poles and four auxiliary poles. The second layer adopts a three-section halbach pole structure. In order to simplify the plastic encapsulation process, the primary winding of TDHCP-PMSLM adopts a single-layer rectangular concentrated winding design.

[0108] Table 1 Parameter design table

[0109]

[0110] 2) Selection of parameters to be optimized:

[0111] Spearman correlation analysis was used for correlation calculation, and the formula of Spearman correlation coefficient was applied:

[0112]

[0113] Where: d i is the rank difference of each pair of observations, and n is the number of observations.

[0114] If there are tied data in variables X or Y (that is, multiple observations have the same rank), adjustments need to be made. In this case, the formula for the Spearman correlation coefficient becomes:

[0115]

[0116] Among them, R(X) and R(Y) are the ranks of variables X and Y, cov represents the covariance, σ R(X) and σ R(Y) is the standard deviation of the ranks.

[0117] Five optimization parameters with the highest correlation were obtained: pole pitch, polarization coefficient of the first layer of permanent magnets, polarization coefficient of the second layer of permanent magnets, winding thickness and air gap spacing as parameters to be optimized.

[0118] 3) Improved multi-objective raccoon algorithm

[0119] Aiming at the problem that the initial population of MOCOA algorithm is unevenly distributed and easy to fall into the local optimal solution, an improvement strategy is proposed. The tent chaotic mapping sequence is used to initialize the population, enhance the uniformity and ergodicity of the population, and improve the early convergence speed. Then, the levy flight search mechanism is used to improve the global search ability of coatis in the exploration stage and the local search ability in the development stage; finally, nonlinear control factors are used to enhance the balance between the exploration stage and the mining stage.

[0120] Tent Chaotic Map Sequence

[0121] In order to maintain the richness of the coati population, this paper proposes to use the Tent chaotic mapping sequence to initialize the raccoon population so that the coati individuals are distributed as evenly as possible within the patrol range.

[0122] Chaos is a nonlinear phenomenon that is ubiquitous in nature. Chaotic variables are often widely used in the generation of initial populations for multi-objective optimization algorithms because of their randomness, ergodicity, and regularity. The sequences generated by Tent mapping have good randomness and uniform distribution characteristics, which is particularly beneficial for applications that require random sampling or initialization. Since Tent mapping exhibits chaotic behavior within a specific parameter range and is highly sensitive to initial conditions, this means that even a small initial change can lead to a huge difference in the sequence, which helps explore a wide area of ​​the search space and avoid local optimal traps. In this paper, the Tent chaotic mapping sequence is selected to initialize the population to improve the MOCOA algorithm.

[0123] Its mathematical expression is as follows:

[0124]

[0125] Where: i = 1, 2, ..., N corresponds to the individual numbers of the coati population, and d = 1, 2, ..., D is the variable dimension. In the process of generating the Tent mapping sequence in the feasible domain, there may be small cycles {0.2, 0.4, 0.6, 0.8}, and there are also unstable points {0, 0.25, 0.5, 0.75}. In order to avoid falling into small cycles and unstable cycle points during iteration, random variables are introduced in the above expression to perturb the sequence. Within the controllable random range, the randomness, ergodicity and orderliness of the Tent mapping sequence are also guaranteed.

[0126]

[0127] Lévy flight strategy

[0128] In multi-objective optimization, maintaining the diversity of the solution set is the key to ensuring that the algorithm can cover the Pareto front. The randomness and wide-area exploration capabilities of Lévy flight can promote the dispersion of solutions, prevent the algorithm from converging to a local area of ​​the solution space too early, and help maintain the diversity of the solution set and the uniformity of distribution. The long-range jumping characteristics of Lévy flight allow the raccoon individuals (i.e., solutions) in the algorithm to quickly jump from their current positions to other distant areas of the solution space. This helps the algorithm escape from the local optimal solution and explore a wider solution space, which may lead to the discovery of new Pareto front solutions. This paper improves the MOCOA algorithm by adding the Lévy flight strategy during the coati exploration phase.

[0129] Lévy flight is a random walk model derived from the Lévy process in mathematics, especially related to the α-stable distribution. It has a wide range of applications in nature and engineering optimization. Unlike traditional Brownian motion, Lévy flight is characterized by a power-law tail in its step size distribution, which means that it contains a large number of small steps and fewer long-distance jumps. This feature allows Lévy flight to perform both local searches and occasional large-scale jumps when exploring the solution space of a problem, which helps avoid local optimal traps, thereby improving search efficiency and global optimization capabilities.

[0130] The probability density function of the Lévy flight follows the Lévy distribution, and its function is as follows:

[0131]

[0132] Among them: 0<α≤2 controls the sharpness of the peak in the Lévy distribution graph, and r>0 controls the span of the distribution graph. When α=2, the Lévy distribution is a Gaussian distribution, and when α=1, the Lévy distribution is a Cauchy distribution. For non-integer α values, it is difficult to analyze the probability density function of the Lévy distribution, and a simulation method is required to generate a random step size of the Lévy distribution. The position update formula is as follows:

[0133]

[0134] Where: α is the random step size; ⊕ is the dot product; Lévy is a random search path that conforms to the Lévy distribution and meets the following constraints:

[0135]

[0136] Among them: u and v obey the standard normal distribution; λ=1.5.

[0137]

[0138] The nonlinear dynamic factor enables the search direction of individual raccoons to be adaptively adjusted according to the current search status and historical information, thereby better coping with the nonlinear characteristics and dynamic changes of the problem, and improving the robustness and effectiveness of the algorithm. In the basic raccoon optimization algorithm, the global exploration and local development capabilities of the coati population mainly depend on the convergence factor r In the early stage of COA, there are fewer optimal individuals in the coati population, and the global exploration ability of the population needs to be strengthened; in the later stage of the algorithm, the population is more likely to have local optimal problems, and it is necessary to change the amount of convergence factor change and strengthen the local development ability to jump out of the local optimal value. Therefore, this paper proposes a nonlinear dynamic factor, which is expressed as follows:

[0139]

[0140] Among them, t is the current iteration number, T MAX is the maximum number of iterations,

[0141] 4) Neural Network Agent Model

[0142] To optimize the parameters τ, α p ,β p ,h w , g is the input layer parameter of the BP neural network proxy model, E m ,F avg ,F r ,ρ is the output layer parameter, the number of training is set to 2000 times, the learning rate is set to 0.001, and the minimum error between the target value obtained by training and the true value is set to 10-6. The transfer functions of the hidden layer and the output layer are Tansig and Purelin functions respectively. The final TDHCP-PMSLM agent model is:

[0143] [F avg , F r ,ρ,E m ]=BPNN[τ,α p , β p ,h w , g]

[0144] In the formula, F avg ,F r ,ρ,E m They are respectively expressed as the output layer parameters average thrust, thrust fluctuation, thrust density and no-load back EMF fundamental amplitude τ, α p ,β p ,h w , g represent the input layer parameters pole pitch, first layer pole arc coefficient, second layer pole arc coefficient, coil thickness, and air gap spacing respectively.

[0145] 80% of the data set of sample points is randomly selected as the training set of the neural network, and the remaining 20% ​​of the data set is used as the test set.

[0146] According to the calculation of the coefficient of determination R 2 Verify the accuracy of the proxy model built through neural network training. 2 The closer it is to 1, the closer the model fitting effect is to the true value. Through analysis, F avg ,F r ,ρ,E m Calculated R 2 They are 0.98, 0.979, 0.997 and 0.998 respectively, all of which are greater than 0.95, indicating that the model construction is reasonable and the model has high reliability.

[0147] The IMOCOA program was written based on the Matlab platform, with the population size set to 200, the number of iterations set to 200, and the input variable parameters set to τ, α, β, and h. w ,g, respectively for F avg ,F r ,ρ,E m Four objectives are used for multi-objective optimization.

[0148] The technical solution provided by the present invention is further described in detail above through several specific implementation modes in order to highlight the advantages and benefits of the technical solution provided by the present invention. However, the several specific implementation modes described above are not intended to be used as limitations on the present invention. Any reasonable modification and improvement of the present invention, combination of implementation modes and equivalent substitution within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A three-stage permanent magnet synchronous linear motor, characterized in that: The linear motor is a three-section double-layer alternating pole permanent magnet structure, comprising: Three-section structure and double-layer magnetic pole structure; The three-stage structure includes at least two opposite salient iron poles, the tops of the two opposite salient iron poles are both provided with main magnetic poles, and auxiliary magnetic poles are respectively arranged outside the two main magnetic poles and wrap the salient iron stage; The double-layer magnetic pole structure includes an epoxy grease seal and a stator winding arranged inside the epoxy grease seal; The double-layer magnetic pole structure is arranged between the two opposite convex iron stages.

2. A three-stage permanent magnet synchronous linear motor according to claim 1, characterized in that: In the double-layer magnetic pole structure, magnetic poles in different layers are arranged in a complementary manner.

3. A three-stage permanent magnet synchronous linear motor according to claim 1, characterized in that: There are four auxiliary magnetic poles.

4. A three-stage permanent magnet synchronous linear motor according to claim 1, characterized in that: The double-layer magnetic pole structure is a double-layer Halbach structure.

5. The three-stage permanent magnet synchronous linear motor according to claim 1, characterized in that: The epoxy seal is used to provide insulation and mechanical support capabilities for the stator winding.

6. A three-stage permanent magnet synchronous linear motor design and optimization method, used to design the linear motor according to claim 1, characterized in that: include: The step of initializing a population using a chaotic mapping sequence according to a multi-objective raccoon optimization algorithm and enhancing the global search capability of the multi-objective raccoon optimization algorithm through a Lévy flight strategy; The step of using a BP neural network proxy model to improve the accuracy of the multi-objective raccoon optimization algorithm; The step of optimizing the linear motor according to the multi-objective raccoon optimization algorithm with improved accuracy.

7. A three-stage permanent magnet synchronous linear motor design and optimization device, used to design the linear motor according to claim 1, characterized in that: include: A module for initializing a population using a chaotic mapping sequence according to a multi-objective raccoon optimization algorithm and enhancing the global search capability of the multi-objective raccoon optimization algorithm through a Lévy flight strategy; A module for improving the accuracy of the multi-objective raccoon optimization algorithm using a BP neural network proxy model; A module for optimizing the linear motor according to a multi-objective raccoon optimization algorithm with improved accuracy.

8. A computer storage medium for storing a computer program, characterized in that: When the computer reads the computer program, the computer executes the method of claim 6.

9. A computer, comprising a processor and a storage medium, characterized in that: When the processor reads the computer program stored in the storage medium, the computer executes the method of claim 6 .

10. A computer program product, being a computer program, characterized in that When the computer program is executed, the method according to claim 6 is implemented.

Images