A spliced pole structure motor, its spliced pole design method and application

Abstract

The present invention relates to a spliced pole structure motor, its spliced pole design method and application, belonging to the technical field of motor manufacturing. The motor includes a stator core, a rotor core, spliced poles, a rotating shaft and an armature winding. Among them, the rotor core is fixedly arranged on the rotating shaft, the spliced poles are evenly arranged around the inside of the rotor core, the stator core is sleeved outside the rotor core, an air gap is left between the rotor core and the stator core, the motor slots are evenly arranged around the inside of the stator core, and the armature winding is arranged in the motor slots. The magnetic field mathematical model of the spliced pole structure of the present invention is approximately a step function, which can significantly reduce the high-order harmonics of the magnetic field, making the synthesized magnetic field of the segmented magnets closer to the standard sine wave, achieving the effect of significantly reducing the back electromotive force. It is beneficial to improve the safety performance of the motor and at the same time improve the economic benefits.

Description

Technical Field

[0001] The present invention relates to a spliced pole structure motor, a spliced pole design method and application thereof, belonging to the technical field of motor manufacturing. Background Art

[0002] At present, although permanent magnet motors have advantages such as high efficiency, high power density and high reliability, due to the scarcity of neodymium iron boron materials, coupled with the increasing market demand, the price is rising day by day. The extensive use will inevitably increase the production cost. In the current motor manufacturing industry, the poles are usually made of a single permanent magnet material with equal thickness. However, due to the magnetization problem, there will be problems such as incomplete utilization of materials and waste of materials and funds. The price of neodymium iron boron permanent magnet materials usually ranges from several hundred to several thousand yuan per kilogram, while the price of ferrite is less than one yuan per kilogram. Therefore, the splicing application of multiple permanent magnet materials is of great significance for saving rare earth materials and reducing production costs.

[0003] When the motor rotates, a back electromotive force will be generated. The value of this back electromotive force is related to the magnetic field generated by the permanent magnet (pole). When using a single type of permanent magnet alone, its magnetic field mathematical model is approximately a step function, and its magnetic field waveform contains many high-order harmonics after being expanded by series. This is likely to cause the back electromotive force of the motor to be too large during operation, which is not conducive to power output and is likely to cause voltage impact and damage the drive board. The pole made of spliced magnets can significantly reduce the magnetic field harmonic components. The traditional method of splicing magnets must perform finite element electromagnetic analysis based on the magnet data fixedly given by the researcher, and the workload of data preparation and electromagnetic analysis verification is extremely large. In the optimization process, a genetic algorithm combined with the Kriging method is used to optimize the magnetic field of the spliced magnets. Its algorithm process is difficult to implement, not easy to popularize, and can only be used to optimize sine pulse width modulation (SPWM) magnets and surface-mounted permanent magnet (SPM) motors, lacking research and application on interior permanent magnet motors.

[0004] In the prior art, the researcher selects the model of each level of magnet according to the remanence property of the permanent magnet. Since the soft magnetic material has no remanence, the present invention selects the materials of each level according to the magnetic permeability, and the spliced pole design method can be extended to the field of soft magnetic materials.

[0005] The implementation difficulty of the particle swarm algorithm is significantly lower than that of the genetic algorithm. The particle swarm optimization algorithm belongs to a kind of evolutionary algorithm. It starts from a random solution, evaluates the quality of the solution through fitness, and searches for the optimal solution through iteration, and follows the currently searched optimal value to find the global optimal value.

[0006] The fitness of the particle swarm algorithm is related to the function matching degree. Calculation techniques for the matching degree between two functions usually apply methods such as Euclidean distance (ED), dynamic time warping (DTW), and longest common substring (LCSS). However, these methods have complex logics, are vulnerable to noise interference, and cannot well fit the characteristic that the magnetic field equivalent model has the same period as the standard sine wave when analyzing the spliced pole magnetic field. Summary of the Invention

[0007] In view of the deficiencies of the prior art, the present invention provides a spliced pole structure motor, its spliced pole design method and application. The magnetic field mathematical model of the spliced pole structure is approximately a step function, which can significantly reduce the high-order harmonics of the magnetic field, making the segmented magnetic field of the magnet closer to the standard sine wave after synthesis, achieving the effect of significantly reducing the back electromotive force. It is beneficial to improve the safety performance of the motor and reduce the production cost.

[0008] The technical solution of the present invention is as follows:

[0009] A spliced pole structure motor includes a stator core, a rotor core, spliced poles, a rotating shaft, and an armature winding. Among them, the rotor core is fixedly arranged on the rotating shaft, the spliced poles are evenly arranged around the inside of the rotor core, the stator core is sleeved outside the rotor core, there is an air gap between the rotor core and the stator core, the motor slots are evenly arranged around the inside of the stator core, and the armature winding is arranged in the motor slots.

[0010] Preferably according to the present invention, the value range of the air gap is 0.75 - 1.10 mm. When the air gap is less than this range, the harmonic components in the magnetic field will be significantly increased, which is likely to produce a negative optimization effect. When the air gap is greater than this range, the motor efficiency and power factor will significantly decrease, which is not conducive to power output. Taking values within this range can ensure that the power factor and motor efficiency are high enough and do not affect the harmonic magnetic field.

[0011] Preferably according to the present invention, the spliced pole includes a first magnet, a second magnet, a third magnet, a fourth magnet, and a fifth magnet connected in sequence. The first magnet, the second magnet, the third magnet, the fourth magnet, and the fifth magnet are all cuboids.

[0012] Preferably according to the present invention, the third magnet uses the first-level permanent magnetic material, the second magnet and the fourth magnet use the same second-level permanent magnetic material. The second magnet and the fourth magnet have the same structure. The magnetic field intensity (magnetic permeability) of the second-level permanent magnetic material is lower than that of the first-level permanent magnetic material. The first magnet and the fifth magnet use the same third-level permanent magnetic material. The first magnet and the fifth magnet have the same structure. The magnetic field intensity (magnetic permeability) of the third-level permanent magnetic material is lower than that of the second-level permanent magnetic material.

[0013] The above-mentioned spliced pole design method of the spliced pole structure motor uses the particle swarm algorithm, and the algorithm process includes:

[0014] (1) Determination of magnetic property material parameters and iteration algebra: Initialize the magnetic permeability μ1 of the first-stage permanent magnet material, the magnetic permeability μ2 of the second-stage permanent magnet material, the magnetic permeability μ3 of the third-stage permanent magnet material, the magnetomotive force H1 of the first-stage permanent magnet material, the magnetomotive force H2 of the second-stage permanent magnet material, the magnetomotive force H3 of the third-stage permanent magnet material, and the air gap δ. Determine the iteration algebra N and the axial length h of the magnetic pole, and determine the amplitude A of the standard sine function according to the first-stage permanent magnet material and the third-stage permanent magnet material;

[0015] (2) Initialization of segmented material size parameters: Take the width of the first-stage permanent magnet material as (0.17, 0.25) unit lengths as x 11 , and the second-stage permanent magnet material and the third-stage permanent magnet material generate x 12 and x 13 respectively according to the random function. Save the generated above parameters x 11 , x 12 , and x 13 in a (1*3)-dimensional vector x1 (the 3-dimensional vector is represented by x1); Take the width of the first-stage permanent magnet material as (0.17, 0.25) unit lengths as x 21 , and the second-stage permanent magnet material and the third-stage permanent magnet material generate x 22 and x 23 respectively according to the random function. Save the generated above parameters in a (1*3)-dimensional vector x2; Take the width of the first-stage permanent magnet material as (0.17, 0.25) unit lengths as x 31 , and the second-stage permanent magnet material and the third-stage permanent magnet material generate x 32 and x 33 respectively according to the random function. Save the generated above parameters in a (1*3)-dimensional vector x3; And so on, a total of N vectors are generated. Save the N vectors in the vector group {x i}, where i = 1, 2,..., N. Randomly generate the velocity of each particle and also save it in a (1*3)-dimensional vector, v1 = (v 11 , v 12 , v 13 ); v2 = (v 21 , v 22 , v 23 ); v3 = (v 31 , v 32 , v 33 )... until v N ;

[0016] In the present invention, only a three-stage splicing structure is shown. If a four-stage splicing is adopted, the particle position and velocity vectors should be set to four dimensions; if a five-stage splicing is adopted, the particle position and velocity vectors should be set to five dimensions; if a k-stage splicing is adopted, the particle position and velocity vectors should be set to k dimensions.

[0017] (3) Generate the mathematical model of the spliced pole magnetic field. To make the generated mathematical model comparable to the standard sine function, the mathematical model of the synthesized magnetic field must have the same period (*) as the standard sine function y = Asinx. Multiply the vector x i by half of the period length π of the standard sine function to meet the condition required by (*);

[0018] (4) Obtain the Fourier expansion coefficients and distortion analysis. The closer the function shape is to the standard sine function, the fewer the harmonic components it contains. Calculate the Fourier expansion coefficients according to the Fourier coefficient formula. Based on the above obtained coefficients, use the distortion analysis to obtain the distortion degree of the mathematical model of the spliced pole magnetic field relative to the standard sine function;

[0019] (5) Obtain the matching degree. Use the distortion degree as the measurement standard for the matching degree between the mathematical model of the spliced pole magnetic field and the standard sine function. The smaller the distortion degree, the greater the matching degree, the higher the fitness of the corresponding position, and the more inclined the particle is to become the historical best position;

[0020] (6) Obtain the optimal value of the particle and update the particle.

[0021] Preferably according to the present invention, in step (3), for the vector x i *π represents the spliced pole parameters for abstracting the mathematical model, and the generation process is as follows:

[0022] a. For the single-group dimension parameter data of the vector x i , perform the following processing. As shown in Figure 9 , take the lower left corner at the edge of the first magnet as the origin O, and extend upward to the calibration point O1 at the 1 / 2 of the air gap. The magnetic field intensity at O1 is A 31 (31 represents the third-level permanent magnetic material, the first magnet), then the magnetic field intensity A 31 at the calibration point O1 is obtained by the following abstract Biot-Savart law formula:

[0023]

[0024]

[0025] In the formula, μ3 is the magnetic permeability of the third-level permanent magnetic material, S is the magnet area. In the present invention, when the magnet length is fixed, its surface area is only determined by the width. Therefore, use the width x i of the third-level permanent magnetic material saved in the above vector x i3 to replace it. h is the distance from this point to the origin O, and K is the distance proportionality constant;

[0026] b. Using the idea of equal magnetic lines, calculate the first-level permanent magnetic material and the second-level permanent magnetic material respectively to generate a magnetic field intensity of A31 The distance Y required for the magnetic field y

[0027]

[0028]

[0029] In the formula, y takes 1 or 2. When y = 1, the calculation target is the first-stage permanent magnetic material; when y = 2, the calculation target is the second-stage permanent magnetic material;

[0030] c. Fit the mathematical model of the spliced pole magnetic field according to the following formula:

[0031] X1 = x i1 *pi

[0032] X 21 = X 22 = x i2 *pi*0.5

[0033] X 31 = X 32 = x i3 *pi*0.5

[0034] y = Y3.*(x >= a & x < b)+Y2.*(x >= b & x < c)+Y1.*(x >= c & x < d)+Y2.*(x >= d & x < e)+Y3.*(x >= e & x < f);

[0035] In the above formula, X 21 is the width of the second magnet, and X 22 is the width of the fourth magnet; X 31 and X 32 are the widths of the first magnet and the fifth magnet respectively; X1, X 21 , X 22 , X 31 , X 32 sum up to pi; x represents the unit length of the abscissa, and the ordinate y represents the magnetic field strength value; a, b, c, d, e, f are all the values of the abscissa x, where a = 0, b = X 31 , c = b + X 21 , d = c + X1, e = d + X 22 , f = e + X 32 ;

[0036] The amplitude A of the standard sine function is approximately obtained by the following formula:

[0037] A = 0.5*(Y1 - Y3)

[0038] Since the process of obtaining the matching degree is to perform a difference analysis between the equivalent model of the spliced ​​pole 3 magnetic field and the standard sine function, although these two signals are not electrical signals, the standard sine function has the same composition as the typical electrical signal in the Fourier transform, and the starting point and period of the abstract mathematical model of the generated synthetic magnetic field are the same as the standard sine function, which is convenient for distortion analysis. Therefore, this design adopts an electrical signal analysis method, that is, using FFT (Fast Fourier Transform) and distortion analysis to obtain the matching degree.

[0039] Preferably, according to the present invention, in step (4), the specific steps are as follows:

[0040] Note: The x in this step is a symbolic variable defined using the smys statement. Different from the previous vector xi, x here represents a symbol or an independent variable in a function f(x);

[0041] Calculate the distortion degree fai for the i-th particle i , i=1,2,3..N;

[0042] f i (x) = y i *x

[0043] Where y i The mathematical model generated by the spliced ​​magnetic pole parameters represented by the i-th particle, x is the symbolic function independent variable;

[0044] Find the mathematical function f i (x) Fourier nth harmonic coefficient B in (n=2, 3........) The formula is as follows:

[0045] B in =0 (n is an even number)

[0046]

[0047] Since the DC component B i0 =0, distortion fai i The calculation formula is simplified to:

[0048]

[0049] In the formula, B i1 is the fundamental wave amplitude, which is numerically the same as the amplitude A of the required standard sine function;

[0050] Therefore, we can solve for the particle x i The corresponding distortion fai i ; Since the step function is similar to the standard sine function, there is no n=even number among its harmonic components, that is, the cosine function cos component;

[0051] Preferably according to the present invention, in step (5), the distortion degree is used as a measure of the matching degree between the spliced pole magnetic field mathematical model and the standard sine function, and the fitness γ is defined by the following formula:

[0052]

[0053] The smaller the distortion degree, the greater the matching degree, the stronger the fitness, and the more inclined the particle is to become the optimal position.

[0054] Preferably according to the present invention, in step (6), the iteration formula is:

[0055] v id = w * v id + c1r1(p id - x id ) + c2r1(p gd - x id );

[0056] x id = x id + v id

[0057] The iteration formula is used to update the particle position and velocity: In the formula, x id represents the particle position (the vector xi obtained by initializing at the beginning is used as the coordinate, which represents the particle position); v id represents the particle velocity (the particle moves during iteration, similar to a bird flying when foraging for food. The velocity represents the ability to move in each coordinate direction (in the example algorithm, the particle position has three dimensions, so its velocity should also be three-dimensional). The velocity vector is also randomly generated and iterated); c1 and c2 are learning factors, also known as acceleration constants, r1 and r2 are uniformly distributed random numbers in the range of [0, 1], and the iteration effect is the best when the values of r1 and r2 are in the range of (0.25, 0.75). When higher than this range, the particle quickly moves towards the optimal position and converges prematurely. When lower than this range, the particle will not move within the optimal region and cannot produce the optimal solution, and w is the inertia constant;

[0058] The formula x id = x id + v id means that when the particle is updated, the new position = the old position + the moving ability;

[0059] The right side of the formula v id = w * v id + c1r1(p id - x id ) + c2r1(p gd - x id ) consists of three parts:

[0060] The first part is the inertial part, which reflects the movement habit of the particle and represents that the particle has the tendency to maintain its previous speed.

[0061] The second part is the cognitive part, which reflects the memory or recollection of the particle's own historical experience and represents that the particle has the tendency to approach its own historical best position.

[0062] The third part is the social part, which reflects the group historical experience of cooperation and knowledge sharing among particles and represents that the particle has the tendency to approach the historical best position of the group or neighborhood.

[0063] The optimal solution generated by iteration is a particle. The result of the iteration is to output the three-dimensional vector of this particle. The content stored in the output three-dimensional vector is the proportion of the first-level permanent magnet material, the second-level permanent magnet material (the sum of the second magnet and the fourth magnet), and the third-level permanent magnet material (the first magnet and the fifth magnet) in the unit length 1. Assuming that the required single-pole width is D, multiplying the output three-dimensional vector by D can obtain the determined width of each level of material in the spliced magnetic pole. The process from the obtained optimal solution to obtaining the width of each level in the spliced pole is essentially a process of equal-proportion amplification.

[0064] The application of the spliced pole of the above-mentioned spliced pole structure motor in the surface-mounted pole motor is as follows:

[0065] ① The surface-mounted pole motor includes a stator core, a rotor core, a rotating shaft, and a magnetic ring sleeve. The rotor core is fixedly arranged on the rotating shaft, and the magnetic ring sleeve is sleeved on the rotor core. The magnetic ring sleeve is a hollow cylindrical sleeve, and grooves are arranged at intervals on the outer side of the magnetic ring sleeve. Magnets are arranged in the grooves. The stator core is sleeved on the outer side of the rotor core, and an armature winding is arranged in the stator core;

[0066] ② Calculate for the surface-mounted pole motor described in step ①:

[0067]

[0068] In the formula, n is the motor speed, f is the power supply frequency, and p is the number of pole pairs. It can be seen from this formula that when the power supply frequency of the surface-mounted pole motor is constant, the smaller the speed, the more the number of pole pairs;

[0069]

[0070] In the formula, α is the central angle occupied by each pole, and p is the number of pole pairs;

[0071] As can be seen from the above formula, the more pole pairs there are, the smaller the central angle occupied by each magnetic pole. When the central angle α occupied by each magnetic pole is less than one degree, it can be considered that the arc carried by the surface-mounted magnet is small, and the magnet can be approximately considered as a rectangular magnet. For a surface-mounted pole motor operating at low speeds, a spliced pole is used to replace the magnet, and the spliced pole design method is applicable.

[0072] The beneficial effects of the present invention are as follows:

[0073] 1. The magnetic field mathematical model of the spliced pole structure of the present invention is approximately a step function, which can significantly reduce the high-order harmonics of the magnetic field, making the segmented magnetic field of the magnet closer to a standard sine wave after synthesis, achieving the effect of significantly reducing the back electromotive force. It is beneficial to improve the safety performance of the motor.

[0074] 2. In order to generate a stepped equivalent magnetic field in the present invention, there are certain differences in the magnetic field intensities (measured by magnetic permeability) of the three-stage magnetic pole materials of the spliced pole. The magnetic permeability of the permanent magnetic material should decrease in sequence, effectively saving the material cost and significantly reducing the use of neodymium iron boron materials.

[0075] 3. Since the magnetic field equivalent model has the same period as the standard sine wave when analyzing the spliced pole magnetic field, it is similar to the distortion analysis of electrical signals. Therefore, the present invention also opens up a method for measuring the matching degree based on the fast Fourier transform (FFT) and distortion analysis of signals in the mathematical analysis of the spliced pole magnetic field. This analysis method is a method in the field of electrical signal analysis. Compared with the algorithms in the machine learning field of traditional matching degree analysis, this method has simple logical thinking, is easy to implement and is more suitable for the magnetic field analysis problem of spliced magnetic poles. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 is the algorithm flow chart of the present invention

[0077] Figure 2 is the schematic diagram of the radial structure of the motor body of the present invention

[0078] Figure 3 is the schematic diagram of the axial structure of the motor body of the present invention

[0079] Figure 4 is the schematic diagram of the one-slot one-pole model of the motor body of the present invention

[0080] Figure 5 is the back electromotive force curve diagram of a conventional pole motor

[0081] Figure 6 is the back electromotive force curve diagram of the motor of the present invention

[0082] Figure 7 is the magnetic cloud diagram of a conventional pole motor

[0083] Figure 8 is the magnetic cloud diagram of the motor of the present invention

[0084] Figure 9 is the vector x in step (3) of the present invention i processing diagram of single-group dimensional parameter data

[0085] The reference numerals in the figure are: 1 - rotating shaft; 2 - rotor core; 3 - split pole; 4 - air gap; 5 - stator core; 6 - motor slot. Specific embodiments

[0086] The present invention will be further described below by way of examples in conjunction with the accompanying drawings, but not limited thereto.

[0087] Example 1:

[0088] As Figure 2-4 shown, this embodiment provides a split-pole structure motor, including a stator core 5, a rotor core 2, a split pole 3, a rotating shaft 1 and an armature winding. Among them, the rotor core 2 is fixedly arranged on the rotating shaft 1, the split pole 3 is uniformly arranged around the inside of the rotor core 2, the stator core 5 is sleeved outside the rotor core 2, an air gap 4 is left between the rotor core 2 and the stator core 5, the motor slots 6 are uniformly arranged around the inside of the stator core 5, and the armature winding is arranged in the motor slots.

[0089] The value range of the air gap is 0.75 - 1.10 mm. When the air gap is less than this range, the harmonic components in the magnetic field will be significantly increased, which is likely to produce a negative optimization effect. When the air gap is greater than this range, the motor efficiency and power factor will decrease significantly, which is not conducive to power output. Taking values within this range can ensure that the power factor and motor efficiency are high enough and do not affect the harmonic magnetic field.

[0090] The split pole includes a first magnet, a second magnet, a third magnet, a fourth magnet and a fifth magnet connected in sequence. The first magnet, the second magnet, the third magnet, the fourth magnet and the fifth magnet are all cuboids.

[0091] The third magnet uses the first-level permanent magnetic material, the second magnet and the fourth magnet use the same second-level permanent magnetic material. The second magnet and the fourth magnet have the same structure. The magnetic field intensity (magnetic permeability) of the second-level permanent magnetic material is lower than that of the first-level permanent magnetic material. The first magnet and the fifth magnet use the same third-level permanent magnetic material. The first magnet and the fifth magnet have the same structure. The magnetic field intensity (magnetic permeability) of the third-level permanent magnetic material is lower than that of the second-level permanent magnetic material.

[0092] The split-pole design method of the above split-pole structure motor adopts the particle swarm algorithm, and the algorithm flow is as Figure 1 shown, and the algorithm flow includes:

[0093] (1) Determination of magnetic property material parameters and iteration algebra, initialize the magnetic permeability μ1 of the first-stage permanent magnet material, the magnetic permeability μ2 of the second-stage permanent magnet material, the magnetic permeability μ3 of the third-stage permanent magnet material, the magnetomotive force H1 of the first-stage permanent magnet material, the magnetomotive force H2 of the second-stage permanent magnet material, the magnetomotive force H3 of the third-stage permanent magnet material and the air gap δ, determine the iteration algebra N and the axial length h of the magnetic pole, and determine the amplitude A of the standard sine function according to the first-stage permanent magnet material and the third-stage permanent magnet material;

[0094] (2) Initialization of segmented material size parameters, take the width of the first-stage permanent magnet material as (0.17, 0.25) unit lengths as x 11 , the second-stage permanent magnet material and the third-stage permanent magnet material respectively generate x 12 , x 13 according to the random function, and save the generated above parameters x 11 , x 12 , x 13 in the (1*3)-dimensional vector x1 (the 3-dimensional vector is represented by x1); take the width of the first-stage permanent magnet material as (0.17, 0.25) unit lengths as x 21 , the second-stage permanent magnet material and the third-stage permanent magnet material respectively generate x 22 , x 23 according to the random function, and save the generated above parameters in the (1*3)-dimensional vector x2; take the width of the first-stage permanent magnet material as (0.17, 0.25) unit lengths as x 31 , the second-stage permanent magnet material and the third-stage permanent magnet material respectively generate x 32 , x 33 according to the random function, and save the generated above parameters in the (1*3)-dimensional vector x3; and so on, a total of N vectors are generated, and the N vectors are saved in the vector group {x i}, where i = 1, 2,......N, randomly generate the velocity of each particle, and also save it in the (1*3)-dimensional vector, v1 = (v 11 , v 12 , v 13 ); v2 = (v 21 , v 22 , v 23 ); v3 = (v 31 , v 32 , v 33 )...... until v N ;

[0095] In the present invention, only a three-stage splicing structure is shown. If a four-stage splicing is adopted, the particle position and velocity vectors should be set to four-dimensional; if a five-stage splicing is adopted, the particle position and velocity vectors should be set to five-dimensional; if a k-stage splicing is adopted, the particle position and velocity vectors should be set to k-dimensional.

[0096] (3) Generate the mathematical model of the spliced pole magnetic field. To make the generated mathematical model comparable to the standard sine function, the mathematical model of the synthesized magnetic field must have the same period (*) as the standard sine function y = Asinx. Multiply the vector x i by half of the period length π of the standard sine function to meet the conditions required by (*);

[0097] (4) Calculate the Fourier expansion coefficients and distortion analysis. The closer the function shape is to the standard sine function, the fewer harmonic components it contains. Calculate the Fourier expansion coefficients according to the Fourier coefficient formula. Based on the above calculated coefficients, use distortion analysis to calculate the distortion of the mathematical model of the spliced pole magnetic field relative to the standard sine function;

[0098] (5) Calculate the matching degree. Use the distortion as the measurement standard for the matching degree between the mathematical model of the spliced pole magnetic field and the standard sine function. The smaller the distortion, the greater the matching degree, the higher the fitness at the corresponding position, and the more likely the particle is to become the historical best position;

[0099] (6) Obtain the optimal value of the particle and update the particle.

[0100] In step (3), for the vector x i *π representing the spliced pole parameters, an abstract mathematical model is generated as follows:

[0101] a. For the single-group dimension parameter data of the vector x i , perform the following processing. As shown in Figure 9 , take the lower left corner at the edge of the first magnet as the origin O, and extend upward to the calibration point O1 at the 1 / 2 of the air gap. The magnetic field intensity at O1 is A 31 (31 represents the third-level permanent magnetic material, the first magnet), then the magnetic field intensity A 31 at the calibration point O1 is obtained by the following abstract Biot-Savart law formula:

[0102]

[0103]

[0104] In the formula, μ3 is the magnetic permeability of the third-level permanent magnetic material, S is the area of the magnet. In the present invention, when the length of the magnet is fixed, its surface area is only determined by the width. Therefore, use the width x i of the third-level permanent magnetic material saved in the above vector x i3 to replace it, h is the distance from this point to the origin O, and K is the distance proportionality constant;

[0105] b. Using the idea of equal magnetic lines, calculate the intensities of the first-level permanent magnetic material and the second-level permanent magnetic material to generate A 31The distance Y required for the magnetic field y

[0106]

[0107]

[0108] Where y takes 1 or 2. When y = 1, the calculation target is the first-stage permanent magnetic material; when y = 2, the calculation target is the second-stage permanent magnetic material;

[0109] c. Fit the mathematical model of the spliced pole magnetic field according to the following formula:

[0110] X1 = x i1 *pi

[0111] X 21 = X 22 = x i2 *pi*0.5

[0112] X 31 = X 32 = x i3 *pi*0.5

[0113] y = Y3.*(x >= a & x < b)+Y2.*(x >= b & x < c)+Y1.*(x >= c & x < d)+Y2.*(x >= d & x < e)+Y3.*(x >= e & x < f);

[0114] In the above formula, X 21 is the width of the second magnet, and X 22 is the width of the fourth magnet; X 31 , X 32 are the widths of the first magnet and the fifth magnet respectively; X1, X 21 , X 22 , X 31 , X 32 sum up to pi; x represents the unit length of the abscissa, and the ordinate y represents the magnetic field strength value; a, b, c, d, e, f are all the values of the abscissa x, where a = 0, b = X 31 , c = b + X 21 , d = c + X1, e = d + X 22 , f = e + X 32 ;

[0115] The amplitude A of the standard sine function is approximately obtained by the following formula:

[0116] A = 0.5*(Y1 - Y3)

[0117] Since the process of obtaining the matching degree involves analyzing the difference between the equivalent model of the magnetic field of the spliced pole 3 and the standard sine function, although these two signals are not electrical signals, because the composition of the standard sine function is the same as that of the typical electrical signals in Fourier transform, and the starting point and period of the abstract mathematical model of the generated synthetic magnetic field are the same as those of the standard sine function, which is convenient for distortion analysis, this design adopts the electrical signal analysis method, that is, uses FFT (Fast Fourier Transform) and distortion analysis to obtain the matching degree.

[0118] In step (4), the specific steps are as follows:

[0119] Note: The x in this step is a symbolic variable defined using the smys statement. Different from the previous vector xi, x here represents a symbol or the independent variable in a function f(x); and the uppercase X appears in the following narrative, where uppercase X represents the value of the independent variable lowercase x in the function f(x).

[0120] Obtain the distortion degree fai of the i-th particle i , i = 1, 2, 3..N;

[0121] f i (x) = y i *x

[0122] In the formula, y i is the mathematical model generated by the spliced magnetic pole parameters represented by the i-th particle, and x is the independent variable of the symbolic function;

[0123] Obtain the Fourier n-th harmonic coefficient B i of the mathematical function f in (n = 2, 3........) The formula is as follows:

[0124] B in = 0 (n is even)

[0125]

[0126] Since the DC component B i0 = 0, the calculation formula for the distortion degree fai i is simplified to:

[0127]

[0128] In the formula, B i1 is the fundamental wave amplitude, which is numerically the same as the amplitude A of the required standard sine function;

[0129] Thus, the distortion degree fai corresponding to the particle x i is solved. i; Since the step function is similar to the standard sine sin function, there is no quantity with n = even in its harmonic components;

[0130] In step (5), taking the distortion as the measurement standard for the matching degree between the spliced extreme magnetic field mathematical model and the standard sine function, the fitness γ is defined by the following formula:

[0131]

[0132] The smaller the distortion, the greater the matching degree, the stronger the fitness, and the more inclined the particle is to become the optimal position.

[0133] In step (6), the iteration formula is:

[0134] v id = w * v id + c1r1(p id - x id ) + c2r1(p gd - x id );

[0135] x id = x id + v id

[0136] The iteration formula is for updating the particle position and velocity: In the formula, x id represents the particle position (the vector xi obtained by initializing at the beginning is used as the coordinate, which represents the particle position); v id represents the particle velocity (the particle moves during iteration, similar to a bird flying when foraging. The velocity represents the ability to move in each coordinate direction (in the example algorithm, the particle position has three dimensions, so its velocity should also be three-dimensional). The velocity vector is also randomly generated and iterated); c1 and c2 are learning factors, also known as acceleration constants, r1 and r2 are uniformly distributed random numbers in the range of [0,1], and the iteration effect is the best when the values of r1 and r2 are in the range of (0.25, 0.75). When higher than this range, the particle moves quickly towards the optimal position and converges prematurely. When lower than this range, the particle will not move within the optimal region and no optimal solution can be generated. w is the inertia constant;

[0137] The formula x id = x id + v id means that when the particle is updated, the new position = the old position + the moving ability;

[0138] The formula v id = w * v id + c1r1(p id - x id ) + c2r1(p gd - xid ) The right side consists of three parts:

[0139] The first part is the inertia part, which reflects the movement habit of the particle and represents the tendency of the particle to maintain its previous speed.

[0140] The second part is the cognitive part, which reflects the memory or recollection of the particle's own historical experience and represents the tendency of the particle to approach its own historical best position.

[0141] The third part is the social part, which reflects the group historical experience of collaborative cooperation and knowledge sharing among particles and represents the tendency of the particle to approach the historical best position of the group or neighborhood.

[0142] The optimal solution generated by iteration is a particle. The result of the iteration is to output the three-dimensional vector of this particle. The content saved in the output three-dimensional vector is the proportion of the first-level permanent magnet material, the second-level permanent magnet material (the sum of the second magnet and the fourth magnet), and the third-level permanent magnet material (the first magnet and the fifth magnet) in the unit length 1. Assuming that the required single-pole width is D, multiplying the output three-dimensional vector by D can obtain the determined width of each level of material in the spliced magnetic pole. The process from the obtained optimal solution to obtaining the width of each level in the spliced pole is essentially a process of equal-proportion amplification.

[0143] The first-level permanent magnet material, the second-level permanent magnet material, and the third-level permanent magnet material are selected as N42UH, N35UH, and FB3X respectively. Using the spliced pole design method of this embodiment for calculation, the width ratio of the first-level permanent magnet material, the second-level permanent magnet material, and the third-level permanent magnet material is 20:17:23. Compared with the motor with pure N35UH magnetic poles, the consumption of neodymium iron boron material can be reduced by 46%. The cost is only

[0144] See Figure 5 , when the ordinary magnetic pole motor designed with N35UH neodymium iron boron material runs, the peak value of the back electromotive force generated is 600v. See Figure 6 , when the spliced pole motor of the same specification runs, the peak value of the back electromotive force generated is 400v. By comparison, it can be known that the back electromotive force of the spliced pole motor described in the present invention during operation is only 66.7% of that of the conventional motor.

[0145] See Figure 7 , it can be known the magnetic flux density distribution of the ordinary magnetic pole motor designed with N35UH material; see Figure 8 , it can be known the magnetic flux density distribution of the spliced pole motor of the same specification. By comparison, it can be known that the magnetic performance of the spliced pole motor described in the present invention is similar to that of the conventional motor.

[0146] Embodiment 2:

[0147] Application of the spliced pole of the spliced pole structure motor as described in Embodiment 1 in a surface-mounted permanent magnet motor, the steps are as follows:

[0148] ① The surface-mounted permanent magnet motor includes a stator core, a rotor core, a rotating shaft, and a magnetic ring sleeve. The rotor core is fixedly arranged on the rotating shaft, and the magnetic ring sleeve is sleeved on the rotor core. The magnetic ring sleeve is a hollow cylindrical sleeve, and grooves are arranged at intervals on the outer side of the magnetic ring sleeve. Magnets are arranged in the grooves. The stator core is sleeved on the outer side of the rotor core, and an armature winding is arranged in the stator core;

[0149] ② Calculate the surface-mounted permanent magnet motor described in step ①:

[0150]

[0151] In the formula, n is the motor speed, f is the power supply frequency, and p is the number of pole pairs. It can be seen from this formula that when the power supply frequency of the surface-mounted permanent magnet motor is constant, the smaller the speed, the more the number of pole pairs;

[0152]

[0153] In the formula, α is the central angle occupied by each pole, and p is the number of pole pairs;

[0154] It can be seen from the above formula that the more the number of pole pairs, the smaller the central angle occupied by each pole. When the central angle α occupied by each pole is less than one degree, it can be considered that the arc carried by the surface-mounted magnet is small, and the magnet can be approximately considered as a rectangular magnet. For a surface-mounted permanent magnet motor operating at low speed, use a spliced pole to replace the original pole, and apply the spliced pole design method described in Embodiment 1.

Claims

1. A spliced pole structure motor, characterized in that, It includes a stator core, a rotor core, a spliced pole, a rotating shaft, and an armature winding. Among them, the rotor core is fixedly arranged on the rotating shaft, the spliced poles are evenly arranged around the inside of the rotor core, the stator core is sleeved outside the rotor core, there is an air gap between the rotor core and the stator core, the motor slots are evenly arranged around the inside of the stator core, and the armature winding is arranged in the motor slots; The spliced pole includes a first magnet, a second magnet, a third magnet, a fourth magnet, and a fifth magnet connected in sequence. The first magnet, the second magnet, the third magnet, the fourth magnet, and the fifth magnet are all cuboids; The third magnet uses the first-level permanent magnetic material, the second magnet and the fourth magnet use the same second-level permanent magnetic material, the second magnet and the fourth magnet have the same structure, the magnetic field strength of the second-level permanent magnetic material is lower than that of the first-level permanent magnetic material, the first magnet and the fifth magnet use the same third-level permanent magnetic material, the first magnet and the fifth magnet have the same structure, and the magnetic field strength of the third-level permanent magnetic material is lower than that of the second-level permanent magnetic material.

2. The spliced pole structure motor according to claim 1, characterized in that, The value range of the air gap is 0.75 - 1.10 mm.

3. A splicing pole design method for a motor with a splicing pole structure as described in claim 1, characterized in that, The particle swarm algorithm is adopted, and the algorithm process includes: (1) Determination of magnetic performance material parameters and the number of iteration generations. Initialize the magnetic permeability μ1 of the first-level permanent magnetic material, the magnetic permeability μ2 of the second-level permanent magnetic material, the magnetic permeability μ3 of the third-level permanent magnetic material, the magnetomotive force H1 of the first-level permanent magnetic material, the magnetomotive force H2 of the second-level permanent magnetic material, the magnetomotive force H3 of the third-level permanent magnetic material, and the air gap δ. Determine the number of iteration generations N and the axial length h of the magnetic pole, and determine the amplitude A of the standard sine function according to the first-level permanent magnetic material and the third-level permanent magnetic material; (2) Initialize the segmented material size parameters. Take the width of the first - stage permanent magnet material as \(x=(0.17, 0.25)\) unit lengths 11 , and generate \(x\) for the second - stage and third - stage permanent magnet materials respectively according to a random function 12 、\(x\) 13 . Save the generated above - mentioned parameters \(x\) 11 、\(x\) 12 、\(x\) 13 in a \((1\times3)\) - dimensional vector \(x1\); Take the width of the first - stage permanent magnet material as \(x\) 21 , and generate \(x\) for the second - stage and third - stage permanent magnet materials respectively according to a random function 22 、\(x\) 23 . Save the generated above - mentioned parameters in a \((1\times3)\) - dimensional vector \(x2\); Take the width of the first - stage permanent magnet material as \(x\) 31 , and generate \(x\) for the second - stage and third - stage permanent magnet materials respectively according to a random function 32 、\(x\) 33 . Save the generated above - mentioned parameters in a \((1\times3)\) - dimensional vector \(x3\); And so on, a total of \(N\) vectors are generated. Save the \(N\) vectors in the vector group \(\{x\) i \}, where \(i = 1,2,\cdots,N\). Randomly generate the velocity of each particle and also save it in a \((1\times3)\) - dimensional vector, \(v1=(v\) 11 , \(v\) 12 , \(v\) 13 ); \(v2=(v\) 21 , \(v\) 22 , \(v\) 23 ); \(v3=(v\) 31 , \(v\) 32 , \(v\) 33 ) \(\cdots\) until \(v\) N ; (3) Generate a mathematical model of the spliced pole magnetic field. To make the generated mathematical model comparable to the standard sine function, and make the mathematical model of the synthesized magnetic field have the same period (*) as the standard sine function y = Asinx, multiply the vector x i by half of the period length π of the standard sine function to meet the conditions required by (*); (4) Calculate the Fourier expansion coefficients and distortion analysis. The closer the function shape is to the standard sine function, the fewer the harmonic components it contains. Calculate the Fourier expansion coefficients according to the Fourier coefficient formula. Based on the above calculated coefficients, use distortion analysis to calculate the distortion of the magnetic field mathematical model of the spliced pole relative to the standard sine function; (5) Calculate the matching degree. Use the distortion as the measurement standard for the matching degree between the magnetic field mathematical model of the spliced pole and the standard sine function; (6) Obtain the optimal value of the particle and update the particle.

4. The splicing pole design method of the spliced pole structure motor according to claim 3, characterized in that, In step (3), for the vector x i The process of generating an abstract mathematical model for the splicing pole parameter represented by *pi is as follows: a. For the vector x i The single-group dimensional parameter data is processed as follows. Taking the lower left corner at the edge of the first magnet as the origin O, extending upward to the calibration point O1 at the 1 / 2 of the air gap, the magnetic field strength at O1 is A 31 , then the magnetic field strength A at the calibration point O1 31 is obtained by the following abstract Biot-Savart law formula: where μ3 is the magnetic permeability of the third - level permanent magnetic material, S is the area of the magnet. Given a fixed magnet length, its surface area is only determined by the width. Therefore, the width x of the third - level permanent magnetic material saved in the above - mentioned vector x i is used instead, h is the distance from this point to the origin O, and K is the distance proportionality constant; i3 ​ b. Using the idea of isomagnetic lines, calculate the distances Y required for the first-stage permanent magnet material and the second-stage permanent magnet material to generate magnetic fields with intensities of A respectively 31 y ​ In the formula, y takes 1 or 2. When y = 1, the calculation target is the first-level permanent magnetic material; when y = 2, the calculation target is the second-level permanent magnetic material; c. Fit the mathematical model of the magnetic field of the spliced pole according to the following formula: X1 = x i1 *pi X 21 = X 22 = x i2 * pi * 0.5 X 31 = X 32 = x i3 * pi * 0.5 y = Y3.*(x >= a & x < b)+Y2.*(x >= b & x < c)+Y1.*(x >= c & x < d)+Y2.*(x >= d & x < e)+Y3.*(x >= e & x < f); In the above formula, X 21 is the width of the second magnet, and X 22 is the width of the fourth magnet; X 31 and X 32 are the widths of the first magnet and the fifth magnet respectively; the sum of X1, X 21 , X 22 , X 31 , X 32 is π; x represents the unit length of the abscissa, and the ordinate y represents the magnetic field strength value; a, b, c, d, e, f are all the values of the abscissa x, where a = 0, b = X 31 , c = b + X 21 , d = c + X1, e = d + X 22 , f = e + X 32 ; The amplitude A of the standard sine function is obtained by the following formula: A = 0.5*(Y1 - Y3).

5. The splicing pole design method of the spliced pole structure motor according to claim 4, characterized in that, In step (4), the specific steps are as follows: Calculate the distortion degree φi of the i-th particle i , where i = 1, 2, 3..N; f i (x) = y i *x where y i is the mathematical model generated by the splicing magnetic pole parameters represented by the i-th particle, and x is the independent variable of the sign function; Obtain the mathematical function f i (x) Fourier n-th harmonic coefficient B in (n = 2, 3........) The formula is as follows: B in = 0 (n is even) Due to the DC component B i0 = 0, the distortion degree φ i The calculation formula is simplified to: where B i1 is the fundamental wave amplitude, which is numerically the same as the amplitude A of the required standard sine function; Thus, the particle x is solved i corresponding distortion degree φ i .

6. The splicing pole design method of the motor with a splicing pole structure as described in claim 5, characterized in that, In step (5), use the distortion as the measurement standard for the matching degree between the magnetic field mathematical model of the spliced pole and the standard sine function, and define the fitness γ by the following formula: The smaller the distortion, the greater the matching degree, the stronger the fitness, and the more inclined the particle is to become the optimal position.

7. The splicing pole design method of the spliced pole structure motor according to claim 6, characterized in that In step (6), the iteration formula is: v id = w * v id + c1r1(p id - x id ) + c2r1(p gd - x id ); x id = x id + v id The iterative formula is used to update the particle position and velocity: where, x id represents the particle position; v id represents the particle velocity; c1 and c2 are learning factors, also known as acceleration constants, r1 and r2 are uniformly distributed random numbers in the range of [0, 1], and the iterative effect is best when the values of r1 and r2 are in the range of (0.25, 0.75). When it is higher than this range, the particles move rapidly towards the optimal position and premature convergence occurs. When it is lower than this range, the particles will not move within the optimal region and no optimal solution can be generated. w is the inertia constant; The optimal solution generated by iteration is a particle. The result of iteration is to output the three-dimensional vector of this particle. The content stored in the output three-dimensional vector is the proportion of the first-level permanent magnet material, the second-level permanent magnet material, and the third-level permanent magnet material in the unit length of 1. Assuming that the required width of a single magnetic pole is D, multiplying the output three-dimensional vector by D can obtain the determined width of each level of material in the spliced magnetic pole. The process from the obtained optimal solution to obtaining the width of each level in the spliced pole is essentially a process of equal-proportion magnification.

8. Application of the spliced pole of the spliced pole structure motor as described in Claim 1 in a surface-mounted pole motor, the steps are as follows: ① The surface-mounted pole motor includes a stator core, a rotor core, a rotating shaft, and a magnetic ring sleeve. The rotor core is fixedly arranged on the rotating shaft, and the magnetic ring sleeve is sleeved on the rotor core. The magnetic ring sleeve is a hollow cylindrical sleeve, and grooves are arranged at intervals on the outer side of the magnetic ring sleeve. Magnets are arranged in the grooves. The stator core is sleeved on the outer side of the rotor core, and an armature winding is arranged in the stator core; ② Calculate the surface-mounted pole motor described in step ①: In the formula, n is the motor speed, f is the power supply frequency, and p is the number of pole pairs. It can be seen from this formula that when the power supply frequency of the surface-mounted pole motor is constant, the smaller the speed, the more the number of pole pairs; In the formula, α is the central angle occupied by each magnetic pole, and p is the number of pole pairs; It can be seen from the above formula that the more the number of pole pairs, the smaller the central angle occupied by each magnetic pole. When the central angle α occupied by each magnetic pole is less than one degree, it can be considered that the radian carried by the surface-mounted magnet is small, and the magnet can be approximately considered as a rectangular parallelepiped magnet. For a surface-mounted pole motor operating at low speed, a spliced pole is used to replace the magnet, and the spliced pole design method is applicable.

Images