A unified optimization method for the torque system and suspension system of a magnetic levitation motor

Abstract

The present invention provides a unified optimization method for the torque system and the suspension system of a magnetic levitation motor. The motor is divided into a torque system and a suspension system, and the torque system and the suspension system are optimized separately. A migration operation is performed on the optimized solution sets to determine the final optimized solution set of the system. Among them, performing the migration operation on the optimized solution sets is to execute a migration operator on the optimized solution sets of the torque system and the suspension system, and determine whether the optimized solutions of the migration operation are dominant, and then decide whether to replace, realizing information migration and feedback between multiple systems, and accelerating the optimization of global problems. The present invention divides the motor system into multiple systems for separate optimization, with fast optimization speed, flexible change, strong adaptability, and can effectively solve the problem of parameter coupling in the optimization process of complex systems.

Description

Technical Field

[0001] The present invention belongs to the field of motor optimization design, and in particular relates to a unified optimization method for a magnetic levitation motor torque system and a suspension system. Background Art

[0002] The novel structure of the new magnetic levitation motor allows the torque and suspension force flux paths to be offset, achieving functional decoupling of the torque and suspension forces. However, because the torque and suspension systems share a salient-pole rotor, complex coupling phenomena still exist during the motor optimization process.

[0003] Traditional motor optimization mostly uses multi-objective optimization methods. By selecting decision variables and target variables, an objective function and constraints are established. Algorithms are then used to optimize the decision variables, finding the optimal solution set and ultimately obtaining a combination of system variables that satisfies the requirements. Multi-objective optimization methods are effective in finding the optimal solution set when the system structure is simple and the number of decision and target variables is small. However, when the system structure is complex, the number of decision variables and targets is numerous, and coupled phenomena exist, multi-objective optimization methods can encounter problems such as numerous variables, high computational complexity, and slow performance, ultimately leading to unsatisfactory optimization results. Summary of the Invention

[0004] In view of the shortcomings in the prior art, the present invention provides a unified optimization method for the magnetic levitation motor torque system and the suspension system. The present invention can effectively solve the parameter coupling problem in the optimization process, with a fast optimization rate and low computational complexity.

[0005] The present invention achieves the above technical objectives through the following technical means.

[0006] A unified optimization method for the magnetic levitation motor torque system and the suspension system is as follows:

[0007] The motor system is divided into a torque system and a suspension system, and the torque system and the suspension system are optimized separately to obtain an optimized solution set, and the optimized solution set is migrated to determine the final optimized solution set of the system;

[0008] The optimization solution set is migrated, specifically: the optimization solution set of the torque system and the suspension system is migrated, when the optimization solution of the jth suspension / torque system is x jl After (s) is transferred to the i-th torque / suspension system, it is combined with the k-th optimal solution x of the i-th torque / suspension system. ik (s) performs non-dominated comparison, when x jl (s) dominates x ik (s), x jl (s) replace x ik (s), otherwise it is not replaced.

[0009] Furthermore, the division of the motor system into a torque system and a suspension system is based on motor performance, motor structure, or optimization requirements.

[0010] Furthermore, the separate optimization of the torque system and the suspension system is specifically as follows:

[0011] Determine the optimization objectives for the torque system and the suspension system respectively, and select preliminary optimization variables;

[0012] Conduct sensitivity analysis on the preliminary optimization variables of the torque system and the suspension system to determine the final optimization variables;

[0013] Based on the final optimization variables and optimization objectives, construct the optimization objective functions and constraint conditions for the torque system and the suspension system;

[0014] Use optimization algorithms to optimize the torque system and the suspension system respectively, and obtain the corresponding optimized solution sets.

[0015] Furthermore, the preliminary optimization variables are selected as the structural parameters of the motor body, and the optimization objectives are selected as the motor performance indicators.

[0016] Furthermore, there are identical optimization variables between the preliminary optimization variables of the torque system and the preliminary optimization variables of the suspension system.

[0017] Furthermore, the optimization objective function is as follows:

[0018] Set the final optimization variables as the input and the optimization objectives as the output, determine the variation range and step size, obtain the training samples between the final optimization variables and the optimization objectives through finite element simulation, and construct the optimization objective functions for the torque system and the suspension system.

[0019] Furthermore, the constraint conditions are the size requirements of the torque system or the suspension system.

[0020] Furthermore, it is used for the unified optimization of the torque system and the suspension system of a 12 / 14-pole magnetic levitation switched reluctance motor.

[0021] Furthermore, the final optimization variables of the torque system are: torque pole arc, rotor pole arc, stator axial length, and air gap length, and the final optimization variables of the suspension system are: air gap length, stator radius, permanent magnet ring thickness, and rotor pole arc.

[0022] Furthermore, the optimization objectives of the torque system are the average torque and torque ripple of the motor, and the optimization objectives of the suspension system are the average suspension force and suspension force ripple of the motor.

[0023] The beneficial effects of the present invention are:

[0024] (1) The present invention decomposes the motor system and adopts the idea of divide and conquer to optimize each system separately, which can effectively solve the problem of parameter coupling in the optimization process of complex coupled systems;

[0025] (2) The present invention divides the motor system into multiple systems for separate optimization, thereby reducing the optimization variables and the amount of calculation; performs a migration operation on the optimization solution set to determine the final optimization solution set of the system, realizes the information migration and feedback between multiple systems, and accelerates the optimization of global problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a schematic structural diagram of the optimized motor according to the present invention;

[0027] Figure 2 is a unified optimization flowchart of the torque system and the suspension system of the magnetic levitation motor according to the present invention;

[0028] In the figure: 101 - motor stator yoke, 102 - motor suspension pole teeth, 103 - suspension winding, 104 - magnetic isolation ring, 105 - motor torque stator core, 106 - torque winding, 107 - motor rotor. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but the protection scope of the present invention is not limited thereto.

[0030] The optimization method of the present invention will be described in detail below with a 12 / 14 pole magnetic levitation switched reluctance motor. Figure 1 is a schematic structural diagram of the 12 / 14 pole magnetic levitation switched reluctance motor. The 12 / 14 magnetic levitation switched reluctance motor is composed of a motor stator yoke 101, four motor suspension poles 102, four suspension windings 103, four magnetic isolation rings 104, four motor torque stator cores 105, eight torque windings 106 and a motor rotor 107; the four motor suspension poles 102 are evenly distributed on the motor stator yoke 101, and the motor stator yoke 101 and the motor suspension poles 102 form a motor suspension stator. The motor suspension stator is located outside the motor rotor 107, and the two are coaxially arranged; the suspension windings 103 are respectively wound on the motor suspension poles 102; a magnetic isolation ring 104 is fixed between the two motor suspension poles 102, and the motor torque stator core 105 is embedded on the magnetic isolation ring 104, and the torque poles of the motor torque stator core 105 are respectively wound with torque windings 106.

[0031] As Figure 2 shown, a unified optimization method for the torque system and the suspension system of a magnetic levitation motor according to the present invention specifically includes the following steps:

[0032] Step (1), divide the motor system into a torque system and a suspension system according to the motor performance (or motor structure or optimization requirements). In this embodiment, it is preferably divided according to the motor performance. The motor body structure size parameters affecting torque (such as torque pole arc, rotor radius, rotor pole arc, torque system stator axial length, suspension system stator axial length, stator inner diameter, rotor pole height, rotor yoke height, torque stator yoke height, suspension stator yoke height, etc.) are divided into the torque system, and the motor body structure size parameters affecting suspension force (such as air gap length, rotor pole height, stator inner diameter, torque system stator axial length, suspension system stator axial length, suspension pole arc, suspension stator pole height, suspension stator yoke height, etc.) are divided into the suspension system.

[0033] Step (2), determine the optimization objectives for the torque system and the suspension system respectively, and select the preliminary optimization variables.

[0034] For the torque system, take the average motor torque T avg and torque ripple T rip as the optimization objectives, and select the torque pole yoke height, torque pole arc, rotor pole arc, torque system stator axial length, air gap length, rotor outer diameter and rotor inner diameter as the preliminary optimization variables of the torque system.

[0035] For the suspension system, take the average motor suspension force F avg and suspension force ripple F rip as the optimization objectives, and select the permanent magnet ring thickness, suspension pole arc, stator outer diameter, stator radius, stator axial length, air gap length, torque pole yoke height, torque pole arc, rotor pole arc, rotor outer diameter and rotor inner diameter as the preliminary optimization variables of the suspension system.

[0036] Among them, when selecting optimization variables for the torque system and the suspension system, it is necessary to ensure that the two systems have a certain number of the same optimization variables for later migration operations between systems. For example, in this embodiment, the torque system and the suspension system both select the torque pole yoke height, torque pole arc, rotor pole arc, rotor outer diameter and rotor inner diameter as the preliminary optimization variables.

[0037] Step (3), perform sensitivity analysis on the preliminary optimization variables of the torque system and the suspension system to determine the final optimization variables.

[0038] Perform sensitivity analysis on the preliminary optimization variables of the torque system respectively, and screen out the sensitivity parameters with higher influence (larger absolute value of the parameter) on the average motor torque T avg and torque ripple T rip ; perform sensitivity analysis on the preliminary optimization variables of the suspension system respectively, and screen out the sensitivity parameters with higher influence on the average suspension force F avg and suspension force ripple F ripHigher-influence sensitivity parameters; based on finite element simulation, a one-variable-at-a-time method is used for sensitivity analysis to further generate sensitivity indices, which represent the degree of influence of parameters on performance. The positive or negative sign of the sensitivity index indicates that the parameter can correspondingly improve or suppress performance; among them, the sensitivity index H(x i ) can be expressed by the formula:

[0039]

[0040] In the formula, represents the average value of the optimization objective y when the preliminary optimization variable x i is a constant, and and V(y) are respectively and the variance of y.

[0041] According to the above method, the final optimization variables of the torque system of the 12 / 14-pole magnetic levitation switched reluctance motor are: torque pole arc, rotor pole arc, stator axial length, and air gap length; the final optimization variables of the suspension system of the 12 / 14-pole magnetic levitation switched reluctance motor are: air gap length, stator radius, permanent magnet ring thickness, and rotor pole arc.

[0042] Step (4), construct the optimization objective functions and constraint conditions for the torque system and the suspension system

[0043] Set the final optimization variables of the torque system and the suspension system as inputs and the optimization objectives as outputs, determine the appropriate range of variation and step size, obtain the training samples between the final optimization variables and the optimization objectives through finite element simulation, and finally use the extreme learning machine or apply the response surface method to construct the optimization objective functions for the torque system and the suspension system;

[0044] Take the conventional dimension requirements of the motor torque system as the system constraint conditions, and on this basis, in this embodiment, use the extreme learning machine to construct the objective functions of the final optimization variable x of the torque system respectively with the maximum average torque T avg and the minimum torque ripple T rip ;

[0045] The optimization objective function of the torque system is:

[0046]

[0047]

[0048] In the formula, is the expression of the average torque with respect to the final optimization variable, and is the expression of the torque ripple with respect to the final optimization variable;

[0049] The constraint conditions are:

[0050]

[0051] In the formula, β st is the torque pole arc, β r is the rotor pole arc, d1 is the outer diameter of the rotor, and d2 is the inner diameter of the rotor;

[0052] Taking the conventional dimension requirements of the motor suspension system as system constraints, and on this basis, in this embodiment, the extreme learning machine is used to construct the objective functions of the final optimized variable x of the suspension system with respect to the maximum average suspension force F avg and the minimum suspension force ripple F rip respectively;

[0053] The optimization objective function of the suspension system is:

[0054]

[0055]

[0056] In the formula, is the expression of the average suspension force with respect to the final optimized variable, is the expression of the suspension force ripple with respect to the final optimized variable;

[0057] The constraint conditions are:

[0058] d3 > d1

[0059] In the formula, d3 is the outer diameter of the stator.

[0060] Step (5), according to the number of optimization objectives of the torque system and the suspension system, respectively use optimization algorithms for single-objective or multi-objective optimization, and obtain the corresponding optimization solution sets

[0061] Optimizing the torque system and the suspension system is a relatively independent process. Different optimization algorithms (such as non-dominated genetic algorithm, particle swarm algorithm, simulated annealing algorithm, etc.) can be used for optimization, and different optimization effects can be obtained.

[0062] Step (6), perform a migration operation on the optimization solution sets of the torque system and the suspension system to determine the final optimization solution set of the system

[0063] Execute the migration operator on the optimization solution sets of the torque system and the suspension system according to the threshold probability to obtain the final optimization solution set of the system; where the migration operator is:

[0064] x ik (s) ← x jl (s)

[0065] In the formula, x ik (s) is the kth optimization solution of the ith torque / suspension system, x jl(s) is the l-th optimal solution of the j-th suspension / torque system.

[0066] When x jl (s) is migrated to the i-th torque / suspension system, it needs to be non-dominated compared with x ik (s). When x jl (s) dominates x ik (s), the threshold probability is 1, and x jl (s) replaces x ik (s); when x jl (s) does not dominate x ik (s), the threshold probability is 0, and x jl (s) does not replace x ik (s).

[0067] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0068] The described embodiments are the preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Without departing from the essence of the present invention, any obvious improvements, replacements, or variations that those skilled in the art can make all belong to the protection scope of the present invention.

Claims

1. A unified optimization method for the torque system and suspension system of a magnetic levitation motor, characterized in that: The motor system is divided into a torque system and a suspension system, the torque system and the suspension system are respectively optimized to obtain an optimization solution set, and a migration operation is performed on the optimization solution set to determine the final optimization solution set of the system; Perform a migration operation on the optimized solution set, specifically: execute the migration operator on the optimized solution sets of the torque system and the suspension system. When the l-th optimized solution x jl (s) of the j-th suspension / torque system is migrated to the i-th torque / suspension system, it is compared with the k-th optimized solution x ik (s) of the i-th torque / suspension system. When x jl (s) dominates x ik (s), x jl (s) replaces x ik (s); otherwise, it does not replace. The specific steps of respectively optimizing the torque system and the suspension system are as follows: Determine the optimization objectives for the torque system and the suspension system respectively, and select the preliminary optimization variables; Perform sensitivity analysis on the preliminary optimization variables of the torque system and the suspension system to determine the final optimization variables; Based on the final optimization variables and optimization objectives, construct the optimization objective functions and constraint conditions for the torque system and the suspension system; Use optimization algorithms to optimize the torque system and the suspension system respectively, and obtain the corresponding optimization solution sets.

2. The unified optimization method for the torque system and suspension system of a maglev motor according to claim 1, characterized in that, The division of the motor system into a torque system and a suspension system is based on the motor performance, motor structure, or optimization requirements.

3. The unified optimization method for the torque system and suspension system of a magnetic levitation motor according to claim 1, characterized in that, The preliminary optimization variables are selected as the structural parameters of the motor body, and the optimization objectives are selected as the motor performance indicators.

4. The unified optimization method for the torque system and suspension system of a magnetic levitation motor according to claim 3, characterized in that There are identical optimization variables between the preliminary optimization variables of the torque system and the preliminary optimization variables of the suspension system.

5. The unified optimization method for the torque system and suspension system of a magnetic levitation motor according to claim 1, wherein The optimization objective function is: Set the final optimization variables as the input and the optimization objectives as the output, determine the variation range and step size, obtain the training samples between the final optimization variables and the optimization objectives through finite element simulation, and construct the optimization objective functions for the torque system and the suspension system.

6. The unified optimization method for the torque system and suspension system of a maglev motor according to claim 1, characterized in that, The constraint conditions are the size requirements of the torque system or the suspension system.

7. The unified optimization method for the torque system and suspension system of a maglev motor according to claim 1, characterized in that It is used for the unified optimization of the torque system and suspension system of a 12 / 14 pole magnetic levitation switched reluctance motor.

8. The unified optimization method for the torque system and suspension system of a magnetic levitation motor according to claim 7, characterized in that The final optimization variables of the torque system are: torque pole arc, rotor pole arc, stator axial length, and air gap length, and the final optimization variables of the suspension system are: air gap length, stator radius, permanent magnet ring thickness, and rotor pole arc.

9. The unified optimization method for the torque system and suspension system of a maglev motor according to claim 1, characterized in that, The optimization objectives of the torque system are the average torque and torque ripple of the motor, and the optimization objectives of the suspension system are the average suspension force and suspension force ripple of the motor.

Images