A low computational cost motor displacement detection method that eliminates zero-crossing distortion

Abstract

The present invention discloses a motor displacement detection method with low computational cost for eliminating zero-crossing distortion. An equation containing rotor displacement information is obtained by using a negative-sequence demodulation method based on different electrical angular frequencies. To avoid the tediousness of solving the equation and the existing zero-crossing distortion problem, an equation group symmetrical about the displacement vector angle is constructed, and useless signal cancellation is realized, and decoupling of the rotor displacement vector angle and the motor electrical frequency is realized, so that the denominator of the expression for identifying the displacement does not contain variables measured in real time by the motor, and the zero-crossing distortion problem is effectively solved. In addition, the calculation process is simple, the interruption period is reduced, and the control accuracy is improved.

Description

Technical Field

[0001] The invention relates to the technical field of bearingless motor control, and mainly to a motor displacement detection method with low calculation cost for eliminating zero-crossing distortion. Background Art

[0002] The bearingless thin-film motor has no rotating shaft, the stator and rotor are separated, the axial length is small, and it is easy to achieve five-degree-of-freedom suspension. It has unique applications in ultra-clean medical fields such as heart pumps. Heart pumps require small size and stable suspension performance, while traditional bearingless control requires displacement sensors to sample displacement signals to achieve closed loops. Commonly used displacement sensors such as eddy current sensors are large in size, expensive, and greatly affected by the environment. They do not meet the requirements of small size and stable suspension performance of heart pumps. Therefore, it is necessary to find a displacement-free algorithm to replace the displacement sensor.

[0003] Since the bearingless thin-film motor itself requires a Hall sensor to measure the angle, the patent "A non-displacement method for the rotor of a bearingless permanent magnet thin-film motor based on a Hall sensor" (application number: 202111409320.9) proposes to use the Hall sensor in one to measure the rotor displacement and angle at the same time. This solution can be used for displacement identification in the startup state, but it has the following two disadvantages:

[0004] (1) This method requires the installation of 12 Hall sensors on the machine, with two Hall sensors placed in each tooth slot, which limits the further reduction of the tooth slot space and thus limits the volume of the entire system.

[0005] (2) This method does not take into account the influence of the third harmonic of the permanent magnet on the displacement solution, and the solution accuracy is still insufficient.

[0006] The negative sequence demodulation method proposed in the document "Radial displacement detection method of bearingless permanent magnet thin-film motor based on Hall negative sequence demodulation method, Zhang Yi, Wang Yu, Zhang Chenggao, Proceedings of the CSEE" can separate the displacement information of the motor rotor from the Hall output signal. However, when the third harmonic appears, the algorithm fails.

[0007] The patent "Rotor displacement identification method for bearingless permanent magnet thin-film motor considering harmonic influence" (application number: 202311404004.1) provides a rotor displacement identification method for bearingless permanent magnet thin-film motor considering the influence of the third harmonic of the permanent magnet. Compared with the Hall method of identifying displacement, the amount of hardware is reduced and the system volume is reduced. The proposed dual negative-sequence demodulation method based on different electrical angles effectively solves the interference of the third harmonic, and the measured displacement accuracy is higher.

[0008] However, this method still has two problems:

[0009] (1) After obtaining the four linear equations, the rotor displacement is obtained by directly solving the equations. The solution process is cumbersome and the expression obtained is very complex. When the digital system is used for implementation, the amount of calculation is large, resulting in a long interruption cycle, which affects the control accuracy.

[0010] (2) What is more serious is that the displacement expressions obtained by solving are all in the form of numerator divided by denominator. The denominator contains not only the inherent structural parameters of the motor, but also the variables measured by the motor in real time. When the calculated value of the denominator is zero in a certain interruption cycle, the calculated displacement is infinite, and there is a zero-crossing distortion problem. Since the denominator contains the variables measured by the motor in real time, it is difficult to solve the zero-crossing distortion problem by changing the inherent structural parameters of the motor. Summary of the invention

[0011] Purpose of the invention: In view of the problems existing in the above-mentioned background technology, the present invention provides a low-computational cost motor displacement detection method that takes into account the influence of the third harmonic of the permanent magnet and eliminates zero-crossing distortion. Compared with the existing algorithm, a set of equations symmetrical about the displacement vector angle is constructed, and useless signal cancellation is realized, and the decoupling of the rotor displacement vector angle and the motor electrical frequency is realized, so that the denominator of the expression for identifying the displacement does not contain the variables measured in real time by the motor, which effectively solves the problem of zero-crossing distortion, and the calculation process is simple, the interruption period is reduced, and the control accuracy is improved.

[0012] Technical solution: To achieve the above purpose, the technical solution adopted by the present invention is:

[0013] A motor displacement detection method with low computational cost for eliminating zero-crossing distortion, wherein the bearingless permanent magnet thin-film motor adopts a motor structure with six teeth and one pole, including six L-shaped stators; each L-shaped stator includes an axial stator yoke and radial stator teeth, which surround the thin-film rotor, and the radial stator teeth are flush with the rotor, and each axial stator yoke is respectively wound with a suspension winding and a torque winding, the torque winding is a pair of poles, and the suspension winding is two pairs of poles, and suspension control and rotation control are realized at the same time; the bottom of the L-shaped stator is connected through an iron core magnetic ring; a pair of permanent magnets are attached to the outside of the thin-film rotor; six Hall sensors Hall1-Hall6 are installed in sequence in the radial stator slots, wherein Hall1 is located at the center of the stator slot facing the N-pole rotor, and Hall2-Hall6 are arranged in sequence in other stator slots counterclockwise, and the interval angles between adjacent Hall sensors are the same; a set of equations is constructed based on the output signals of Hall1-Hall6 to solve the displacement in the startup state;

[0014] Specifically, the sum of the output signals of three groups of two Hall sensors that are mechanically opposite to each other are derived by taking into account the third harmonic of the permanent magnet magnetic potential, namely S1, S2, and S3, and the derivation process is as follows:

[0015] The schematic diagram of rotor eccentricity is as follows Figure 1 As shown, the air gap length when the rotor eccentricity is taken into account is a function of the spatial angle θ, and its expression is

[0016] g(θ)=g0-(αcosθ+βsinθ)

[0017] Among them, g0 is the air gap length when the rotor is not eccentric, α is the horizontal displacement in the x direction, β is the vertical displacement in the y direction, l is the eccentric displacement length, θ l is the eccentricity angle (rotor displacement vector angle).

[0018] Assume (αcosθ+βsinθ) 2 <<g0 2 , then the permeance formula per unit angle can be simplified to

[0019]

[0020] Where μ0 is the air magnetic permeability, R is the outer radius of the rotor, l a is the axial length of the motor.

[0021] The sum of the output signals of two mechanically opposite Hall sensors can be analogized to a winding with two pairs of poles distributed in space. Taking Hall1+Hall4 as an example, the spatial distribution function is N s cos2θ and higher harmonics, such as N s cos4θ、N s cos6θ, etc., only the fundamental wave N is considered here s cos2θ, higher harmonics can be ignored and are not considered for now.

[0022] The magnetic potential formed by the rotor permanent magnet is in the form of a traveling wave when the third harmonic is considered:

[0023] N t I f [cos(θ-ωt)+Bcos(3θ-3ωt)]

[0024] Among them, N s 、N t ,I f , B are the equivalent winding turns of the Hall sensor respectively, and the permanent magnet takes into account the equivalent winding turns of the third harmonic and the amplitude of its equivalent excitation current.

[0025] According to the continuity of magnetic flux:

[0026]

[0027] in

[0028]

[0029] Then finally

[0030]

[0031] In the above formula, E c is the compensation value of the rotor magnetic potential, E r For the final rotor magnetic potential the third harmonic is considered.

[0032] The flux of the winding equivalent to the Hall sensor is the integral of the product of the spatial distribution function and the magnetic flux on the circumference. The higher harmonics of the spatial distribution function can be ignored, so the formula for the flux is:

[0033]

[0034] Among them, k1 and k2 are the horizontal displacement coefficient and vertical displacement coefficient of the permanent magnet flux, respectively.

[0035] The self-inductance L of the suspension winding s Independent of the eccentric displacement of the rotor, the total flux Ψ 14 The magnetic flux component of the suspension winding Ψ 14_s It can be written as follows:

[0036]

[0037] Among them, i s is the suspension current, I s is the amplitude of the suspension current, θ s is the current electrical angle of the suspension current.

[0038] Magnetic Link 14 The torque flux component Ψ 14_t It can be written as follows:

[0039]

[0040] Among them, k1 and k2 are the horizontal displacement coefficient and vertical displacement coefficient of the torque flux respectively.

[0041] Combining the above flux components, the total flux Ψ under all excitations considering the third harmonic of the permanent magnet potential can be obtained: 14 for:

[0042]

[0043] The above formula can also be written as follows:

[0044]

[0045] Furthermore, a pair of poles θ f With θt Synthesized into θ1, 3θ f Rewritten as θ3, the above formula can be organized into Formula S1:

[0046] S1=ψ 14 =L s I s cosθ s +k1l cosθ l [I1 cosθ1+I3 cosθ3]-k2l sinθ l [I1 sinθ1-I3 sinθ3]

[0047] Among them, I1 is the composite current amplitude of one pair of poles, I3 is the current amplitude of three pairs of poles, Ψ 14 is the total flux linkage of Hall1 and Hall4 under all excitations considering the third harmonic of the permanent magnet potential.

[0048] From the symmetry of the mechanical structure, it can be seen that the sum of the output signals S2 and S3 of the remaining two sets of two Hall sensors that are opposite to each other in the mechanical structure are respectively:

[0049]

[0050] Among them, 36 is the total flux linkage of Hall 3 and Hall 6 under all excitations considering the third harmonic of the permanent magnet potential, Ψ 52 is the total flux linkage of Hall5 and Hall2 under all excitations considering the third harmonic of the permanent magnet potential.

[0051] Then, it is necessary to identify the winding self-inductance L of the suspension winding. s and the suspension current amplitude I s , the specific steps are as follows:

[0052] Only the suspension current excitation is retained, while the torque current excitation and permanent magnet excitation are all removed. The amplitude of the suspension current input is the current amplitude I in the formula s ;

[0053] The magnetic flux size Ψ is measured when there is only suspension current excitation 14_s , then the winding self-inductance of the suspension winding

[0054]

[0055] Next, using the identified self-inductance L of the suspension winding S and the suspension current I S, removing the influence of the suspension current in the sum of the output signals S1, S2, and S3 of the three groups of two Hall sensors that are mechanically opposite to each other in consideration of the third harmonic of the permanent magnet magnetic potential, and obtaining three groups of new output signals S4, S5, and S6, the specific steps of which include:

[0056] The flux linkage expressions S1, S2, and S3 obtained previously are removed by removing the self-inductance L of the suspension winding identified previously. S and the suspension current I S The influence on the total flux linkage is

[0057]

[0058] Finally, the negative sequence demodulation method is used to obtain the expression of the radial displacement of the rotor in the starting state. The specific steps include:

[0059]

[0060]

[0061] Construct y1, y2, y3, y4 as the angle θ about the rotor displacement vector l The symmetric function of is as follows:

[0062]

[0063] By canceling E1, E2, E3, and E4, we can achieve θ l Decoupling from the motor electrical frequency, the specific process is as follows:

[0064]

[0065] The expression of displacement with only motor intrinsic parameters in the denominator is obtained by using E1+E2 and E3+E4. The specific process is as follows:

[0066]

[0067] It can be seen that lcos(θ l ) and lsin(θ l The denominator in the expression of ) does not contain the detection parameters of the motor, but only contains the inherent parameters of the motor. It only needs to be designed by the motor itself. The value of is designed to be greater than -2 to solve the zero-crossing distortion problem. In addition, the above transformation process does not involve the complex process of solving linear equations, and is simple to implement using a digital controller, occupying less resources.

[0068] The present invention also provides a bearingless thin-film motor control system, which uses the above-mentioned motor displacement detection method to obtain the radial displacement of the rotor in the starting state.

[0069] The present invention provides a motor displacement detection method with low computational cost for eliminating zero-crossing distortion, which has the following beneficial effects compared with the prior art:

[0070] (1) A set of equations symmetrical about the displacement vector angle is constructed, and useless signal cancellation is achieved, thereby decoupling the rotor displacement vector angle from the motor electrical frequency. This ensures that the denominator of the expression for identifying the displacement does not contain variables measured in real time by the motor, effectively solving the zero-crossing distortion problem.

[0071] (2) The transformation process does not involve the complex process of solving linear equations. It is simple to implement using a digital controller and occupies less resources. The calculation process is simple, which reduces the interruption cycle and improves the control accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 This is a schematic diagram of the eccentric rotor of the bearingless permanent magnet thin-film motor provided by the present invention;

[0073] Figure 2 An axial cross-sectional view of a six-tooth one-pole bearingless permanent magnet thin-film motor provided by the present invention;

[0074] Figure 3 It is a schematic diagram of the mechanical position of the Hall sensor of the bearingless permanent magnet thin-film motor adopted by the present invention;

[0075] Figure 4 This is a comparison chart between simulation and theoretical curves when the third harmonic of the permanent magnet magnetic potential is considered;

[0076] Figure 5 This is a comparison chart between simulation and theoretical curves when the third harmonic of the permanent magnet magnetic potential is not considered;

[0077] Figure 6 It is a bearingless thin-film motor control system;

[0078] Figure 7 It is the zero-crossing distortion problem existing in the solution of file (202311404004.1);

[0079] Figure 8 is the displacement identified by the solution of the present invention (there is no zero-crossing distortion problem);

[0080] Fig. 9 It is a comparison of the time required for digital control implementation of the solution in document (202311404004.1) and the solution of the present invention;

[0081] Fig.10 It is a flow chart of the motor displacement detection method of the present invention. DETAILED DESCRIPTION

[0082] The present invention is further described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0083] The present invention adopts Figure 2 The bearingless permanent magnet thin-film motor shown adopts a six-teeth-one-pole motor structure, including six L-shaped stators 2. Each L-shaped stator 2 includes an axial stator yoke and radial stator teeth, which surround the thin-film rotor 1. The radial stator teeth are flush with the rotor. Each axial stator yoke is respectively wound with a suspension winding 4 and a torque winding 5. The torque winding is a pair of poles, and the suspension winding is two pairs of poles, so that suspension control and rotation control are realized at the same time. The bottom of the L-shaped stator is connected through an iron core magnetic ring 3. A pair of permanent magnets are attached to the outside of the thin-film rotor 1.

[0084] Six Hall sensors Hall1-Hall6 are installed in the radial stator slots in sequence, where Hall1 is located at the center of the stator slot facing the N-pole rotor, and Hall2-Hall6 are arranged in the other stator slots in a counterclockwise direction. The interval angles between adjacent Hall sensors are the same. The specific positions are as follows: Figure 3 shown.

[0085] First, the sum of the output signals of three groups of two Hall sensors that are mechanically opposite to each other are derived by considering the third harmonic of the permanent magnet's magnetic potential, which are S1, S2, and S3, respectively. The derivation process is as follows:

[0086] The schematic diagram of rotor eccentricity is as follows Figure 1 As shown, the air gap length when the rotor eccentricity is taken into account is a function of the spatial angle θ, and its expression is

[0087] g(θ)=g0-(αcosθ+βsinθ)

[0088] Among them, g0 is the air gap length when the rotor is not eccentric, α is the horizontal displacement in the x direction, β is the vertical displacement in the y direction, l is the eccentric displacement length, θ l is the eccentric angle.

[0089] Assume (αcosθ+βsinθ) 2 <<g0 2 , then the permeance formula per unit angle can be simplified to

[0090]

[0091] Where μ0 is the air magnetic permeability, R is the outer radius of the rotor, l a is the axial length of the motor.

[0092] The sum of the output signals of two mechanically opposite Hall sensors can be analogized to a winding with two pairs of poles distributed in space. Taking Hall1+Hall4 as an example, the spatial distribution function is N s cos2θ and higher harmonics, such as N s cos4θ、N s cos6θ, etc., only the fundamental wave N is considered here s cos2θ, higher harmonics can be ignored and are not considered for now.

[0093] The magnetic potential formed by the rotor permanent magnet is in the form of a traveling wave when the third harmonic is considered:

[0094] N t I f [cos(θ-ωt)+Bcos(3θ-3ωt)]

[0095] Among them, N s 、N t ,I f , B are the equivalent winding turns of the Hall sensor respectively, and the permanent magnet takes into account the equivalent winding turns of the third harmonic and the amplitude of its equivalent excitation current.

[0096] According to the continuity of magnetic flux:

[0097]

[0098] in

[0099]

[0100] Then finally

[0101]

[0102] In the above formula, E c is the compensation value of the rotor magnetic potential, E r For the final rotor magnetic potential the third harmonic is considered.

[0103] The flux of the winding equivalent to the Hall sensor is the integral of the product of the spatial distribution function and the magnetic flux on the circumference. The higher harmonics of the spatial distribution function can be ignored, so the formula for the flux is:

[0104]

[0105] Among them, k1 and k2 are the horizontal displacement coefficient and vertical displacement coefficient of the permanent magnetic flux, respectively, and the self-inductance L of the suspension winding is s Independent of the eccentric displacement of the rotor, the total flux Ψ 14 The magnetic flux component of the suspension winding Ψ 14_s It can be written as follows:

[0106]

[0107] Among them, i s is the suspension current, I s is the amplitude of the suspension current, θ s is the current electrical angle of the suspension current.

[0108] Magnetic Link 14 The torque flux component Ψ 14_t It can be written as follows:

[0109]

[0110] Among them, k1 and k2 are the horizontal displacement coefficient and vertical displacement coefficient of the torque flux respectively.

[0111] Combining the above flux components, the total flux Ψ under all excitations considering the third harmonic of the permanent magnet potential can be obtained: 14 for:

[0112]

[0113] The above formula can also be written as follows:

[0114]

[0115] Furthermore, a pair of poles θ f With θ t Synthesized into θ1, 3θ f Rewritten as θ3, the above formula can be organized into Formula S1:

[0116] S1=ψ 14 =L s I s cosθ s +k1l cosθ l [I1 cosθ1+I3 cosθ3]-k2l sinθ l [I1 sinθ1-I3 sinθ3]

[0117] Among them, I1 is the composite current amplitude of one pair of electrodes, and I3 is the current amplitude of three pairs of electrodes.

[0118] From the symmetry of the mechanical structure, it can be seen that the sum of the output signals S2 and S3 of the remaining two sets of two Hall sensors that are opposite to each other in the mechanical structure are respectively:

[0119]

[0120] Then, it is necessary to identify the winding self-inductance L of the suspension winding.s and the suspension current amplitude I s , the specific steps are as follows:

[0121] Only the suspension current excitation is retained, while the torque current excitation and permanent magnet excitation are all removed. The amplitude of the suspension current input is the current amplitude I in the formula s ;

[0122] The magnetic flux size Ψ is measured when there is only suspension current excitation 14_s , then the winding self-inductance of the suspension winding

[0123]

[0124] Next, using the identified self-inductance L of the suspension winding S and the suspension current I S , removing the influence of the suspension current in the sum of the output signals S1, S2, and S3 of the three groups of two Hall sensors that are mechanically opposite to each other in consideration of the third harmonic of the permanent magnet magnetic potential, and obtaining three groups of new output signals S4, S5, and S6, the specific steps of which include:

[0125] The flux linkage expressions S1, S2, and S3 obtained previously are removed by removing the self-inductance L of the suspension winding identified previously. S and the suspension current I S The influence on the total flux linkage is

[0126]

[0127] Finally, the negative sequence demodulation method is used to obtain the expression of the radial displacement of the rotor in the starting state. The specific steps include:

[0128]

[0129] After obtaining y1, y2, y3, and y4, the displacement solution process provided by the patent "Rotor displacement identification method of bearingless permanent magnet thin-film motor considering harmonic influence" (application number: 202311404004.1) is as follows:

[0130]

[0131] Then the above formula can be simplified into the following form:

[0132]

[0133] The above coefficients can be obtained through simulation and experimental measurement, that is, they can be regarded as known numbers. Then the above formula is a set of four-variable linear equations about x. x1 and x3 are the required horizontal and vertical displacements x and y. After calculation, x1 and x3 are respectively:

[0134]

[0135] According to the above formula, the rotor horizontal displacement x and vertical displacement y of the bearingless permanent magnet thin-film motor can be obtained when the third harmonic of the permanent magnet magnetic potential is taken into account.

[0136] To verify whether the derivation is correct, in the MAXWELL simulation, given a rotor horizontal displacement x and vertical horizontal displacement y, check whether the final simulated magnetic flux is consistent with the theoretically derived magnetic flux, thereby verifying whether the theoretical derivation of the rotor horizontal displacement x and vertical horizontal displacement y is correct.

[0137] Figure 4 It is a comparison diagram of simulation and theoretical curves when the third harmonic of the permanent magnet magnetic potential is considered, that is, the displacement is obtained by using the double negative sequence demodulation method based on different electrical angles proposed in this paper.

[0138] Figure 5 It is a comparison diagram of the simulation and theoretical curves when the third harmonic of the permanent magnet's magnetic potential is not considered, that is, the displacement is obtained by using the negative-sequence demodulation method of the document "Radial displacement detection method of bearingless permanent magnet thin-film motor based on Hall negative-sequence demodulation method, Zhang Yi, Wang Yu, Zhang Chenggao, Proceedings of the CSEE".

[0139] By comparison, it can be found that when the third harmonic of the permanent magnet's magnetic potential is not considered, the curve overlap is relatively low, proving that the accuracy of the measured rotor horizontal displacement x and vertical horizontal displacement is not high; when the third harmonic of the permanent magnet's magnetic potential is considered, the curve overlap is very high, proving that the accuracy of the rotor horizontal displacement x and vertical horizontal displacement y measured by the double negative sequence demodulation method based on different electrical angles is greatly improved.

[0140] However, after obtaining the four linear equations, the rotor displacement is obtained by directly solving the equations. The solution process is cumbersome and the expression obtained is very complex. When the digital system is used for implementation, the amount of calculation is large, resulting in a long interruption cycle and affecting the control accuracy.

[0141] Figure 6 is a bearingless thin-film motor control system, where ω * is the given mechanical angular frequency, ω is the actual mechanical angular frequency, Given the torque current of the torque winding, Given the excitation current of the torque winding, i td is the actual torque current of the torque winding, i tq is the actual excitation current of the torque winding, u td is the torque winding d-axis voltage, u tq is the q-axis voltage of the torque winding, u tα is the torque winding α-axis voltage, u tβis the torque winding β-axis voltage, SVPWM is the space voltage vector, i ta 、i tb 、i tc is the actual three-phase current of the torque winding, H 1-4 The difference between Hall1 and Hall4, H 3-6 Make a difference between Hall3 and Hall6, H 5-2 The difference between Hall5 and Hall2, H 1+4 Make the sum for Hall1 and Hall4, H 3+6 Make sum for Hall3 and Hall6, H 5+2 is the sum of Hall5 and Hall2, θ is the permanent magnet flux vector angle, abc / αβ is the transformation from three-phase stationary coordinates to two-phase stationary coordinates, αβ / dq is the transformation from two-phase stationary coordinates to two-phase rotating coordinates, LPF is a low-pass filter, Given the torque current of the suspension winding, Given the excitation current of the suspension winding, i sd is the actual torque current of the suspension winding, isq is the actual excitation current of the suspension winding, u sd is the d-axis current of the suspension winding, u sq is the q-axis voltage of the suspension winding, u sα is the α-axis voltage of the suspension winding, u sβ is the β-axis current of the suspension winding, i sa 、i sb 、i sc is the actual three-phase current of the suspension winding, α * For the rotor given α axis displacement, β * For the given β-axis displacement of the rotor, Figure 6 If the displacement detection in the method adopts the identification result of "Rotor displacement identification method of bearingless permanent magnet thin-film motor considering harmonic influence" (application number: 202311404004.1), the serious problem is that the displacement expressions obtained by solution are all in the form of numerator divided by denominator. The denominators x1 and x3 contain not only the inherent structural parameters of the motor, but also the variables measured in real time by the motor. When the calculated value of the denominator is zero in a certain interruption cycle, the calculated displacement is infinite, and there is a zero-crossing distortion problem, such as Figure 7 shown.

[0142] Since the denominator contains variables measured in real time by the motor, it is difficult to solve the zero-crossing distortion problem by changing the inherent structural parameters of the motor.

[0143] The solution provided by the present invention is as follows:

[0144] Construct y1, y2, y3, y4 as the angle θ about the rotor displacement vector l The symmetric function of is as follows:

[0145]

[0146] By canceling E1, E2, E3, and E4, we can achieve θ l Decoupling from the motor electrical frequency, the specific process is as follows:

[0147]

[0148] The expression of displacement with only motor intrinsic parameters in the denominator is obtained by using E1+E2 and E3+E4. The specific process is as follows:

[0149]

[0150] It can be seen that l cos(θ l ) and l sin(θ l The denominator in the expression of ) does not contain the detection parameters of the motor, but only contains the inherent parameters of the motor. It only needs to be designed by the motor itself. The value of is designed to be greater than -2 to solve the zero-crossing distortion problem. Figure 6 If the displacement detection in the present invention adopts the identification result, there will be no displacement distortion, such as Figure 8 As shown. In addition, the above transformation process does not involve a complicated process of solving linear equations, and is simple to implement using a digital controller, occupying less resources.

[0151] Fig. 9 The time required for digital control implementation of the solution in file (202311404004.1) and the solution in the present invention is compared. The solution in file (202311404004.1) requires 9 microseconds, while the present invention only requires 6 microseconds.

[0152] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A motor displacement detection method with low computational cost for eliminating zero-crossing distortion, wherein the motor is a bearingless permanent magnet thin-film motor, adopts a motor structure with six teeth and one pole, and includes six L-shaped stators; each L-shaped stator includes an axial stator yoke and radial stator teeth, which surround the thin-film rotor, and the radial stator teeth are flush with the rotor. Each axial stator yoke is respectively wound with a suspension winding and a torque winding, wherein the torque winding is a pair of poles, and the suspension winding is two pairs of poles, and suspension control and rotation control are realized at the same time; the bottom of the L-shaped stator is connected by an iron core magnetic ring; a pair of permanent magnets are attached to the outside of the thin-film rotor; six Hall sensors Hall1-Hall6 are sequentially installed in the radial stator slot, wherein Hall1 is located at the center of the stator slot facing the N-pole rotor, and the Hall2-Hall6 are sequentially arranged in other stator slots counterclockwise, and the interval angles between adjacent Hall sensors are the same; the specific steps of the method include: Step 1: Considering the third harmonic of the permanent magnet magnetic potential, construct the expressions of the sum S1, S2, and S3 of the output signals of three groups of two Hall sensors that are mechanically opposite to each other. The expressions of S1, S2, and S3 are: S1=ψ 14 =L s I s cosθ s + k1lcosθ l [I1cosθ1+I3cosθ3]- k2lsinθ l [I1sinθ1-I3sinθ3] Among them, θ s is the current electrical angle of the suspension current, k1 and k2 are the horizontal displacement coefficient and vertical displacement coefficient of the permanent magnetic flux, l is the eccentric displacement length, θ l is the eccentricity angle, θ1 is a pair of poles θ f With θ t The sum of θ3 is 3θ f ,Ψ 14 is the total flux linkage of Hall 1 and Hall 4 under all excitations considering the third harmonic of the permanent magnet potential, Ψ 36 is the total flux linkage of Hall 3 and Hall 6 under all excitations considering the third harmonic of the permanent magnet potential, Ψ 52 is the total flux linkage of Hall5 and Hall2 under all excitations considering the third harmonic of the permanent magnet potential, I1 is the composite current amplitude of 1 pair of poles, and I3 is the current amplitude of 3 pairs of poles; Step 2: Identify the winding self-inductance L of the suspension winding s and the suspension current amplitude I s , the specific steps are as follows: Only the suspension current excitation is retained, while the torque current excitation and permanent magnet excitation are all removed. The amplitude of the suspension current input is the current amplitude I in the formula s ; The magnetic flux size Ψ is measured when there is only suspension current excitation 14_s , then the winding self-inductance of the suspension winding Step 3: Use the identified self-inductance L of the suspended winding s and the suspension current I s , removing the influence of the suspension current in the sum of the output signals S1, S2, and S3 of the three groups of two Hall sensors that are mechanically opposite to each other in consideration of the third harmonic of the permanent magnet magnetic potential, and obtaining three groups of new output signals S4, S5, and S6, the specific steps include: The expressions of S1, S2, and S3 are obtained by removing the self-inductance L of the identified suspension winding. s and the suspension current I s The influence on the total flux linkage is Among them, k1 and k2 are the horizontal displacement coefficient and vertical displacement coefficient of the permanent magnetic flux, respectively; Step 4: Define four linear equations y1, y2, y3, y4, the negative sequence projection angle of y1 and y3 is θ1, and the negative sequence projection angle of y2 and y3 is θ3, specifically: Step 5: Solve the four linear equations to obtain the radial displacement of the rotor in the starting state; It is characterized in that the specific steps of step 5 include: Step 5.1: Construct about the rotor eccentricity angle θ l The symmetric function of is as follows: Step 5.2: Eliminate E1, E2, E3, and E4 to achieve θ l Decoupling from the motor electrical frequency, the specific process is as follows: Step 5.3: Use E1+E2 and E3+E4 to find the expression of displacement with only the motor intrinsic parameters in the denominator. The specific process is as follows: lcos(θ l ) and lsin(θ l ) are the expressions of the radial displacement of the rotor in the starting state.

2. A bearingless thin-film motor control system, characterized in that: The motor displacement detection method as claimed in claim 1 is used to obtain the radial displacement of the rotor in the starting state.

Images