Dual-source excitation compound rotor six-phase bearingless motor and decoupling control method thereof

Abstract

This invention discloses a dual-source excitation composite rotor six-phase bearingless motor and its decoupling control method. The motor includes a stator assembly, a rotor assembly, an inverter, and a control unit. The stator assembly has a double Y-shifted 30° six-phase winding, and the rotor assembly includes a composite excitation structure composed of built-in permanent magnets and squirrel-cage bars. The control unit performs spatial vector decoupling of the six-phase currents, forming an αβ fundamental subspace, a z1z2 harmonic subspace, and an o1o2 zero-sequence subspace. In the αβ subspace, d-q axis current control is performed, with the d-axis simultaneously adjusting the dual flux linkages and the q-axis controlling the dual torque output according to the distribution coefficient. In the z1z2 subspace, harmonic current control is performed to generate a harmonic magnetic field with a pole pair difference of 1 to produce radial levitation force. In the o1o2 subspace, the zero-sequence current is monitored, faults are identified, and feedback triggers control adjustments. This invention balances high efficiency and a wide speed range, achieving independent control of torque and levitation force, and fault tolerance, making it suitable for high-speed precision transmission applications.

Description

Technical Field

[0001] This invention belongs to the field of bearingless motor technology, specifically relating to a dual-source excitation composite rotor six-phase bearingless motor and its decoupling control method. Background Technology

[0002] As a new type of electromechanical equipment that integrates electromagnetic drive and magnetic levitation technology, bearingless motors eliminate the inherent defects of traditional mechanical bearings, such as friction loss and lubrication dependence, in principle. They have outstanding advantages such as high speed, high precision, long life and low noise. They have become a key core component for breaking through technical bottlenecks and achieving independent control in the field of high-end equipment manufacturing, and have important application value in national strategic fields such as aerospace, precision manufacturing and energy.

[0003] Currently, the bearingless motor technology system still has many structural shortcomings: traditional bearingless motors mostly adopt a single drive form (permanent magnet synchronous or asynchronous), which makes it difficult to balance high efficiency and wide speed range, and there is an inherent contradiction between low-speed starting performance and high-speed operation reliability; most adopt a three-phase or four-phase winding structure, which limits the degree of control freedom, making it difficult to achieve complete decoupled control of torque and levitation force, and resulting in insufficient disturbance resistance and fault tolerance; reliability improvement mostly relies on control strategy optimization, lacking fundamental innovation at the motor body topology level, making it difficult to meet the long-term reliable operation requirements of high-end equipment under extreme working conditions.

[0004] Meanwhile, the excitation methods of existing bearingless motors are mostly single excitation, and the air gap magnetic field adjustment is not flexible enough, which further limits the operating performance and adaptability of the motor. In the process of multi-physical quantity (synchronous torque, asynchronous torque, levitation force) control, the coupling between the physical quantities is serious, resulting in low control accuracy and slow dynamic response. Moreover, under abnormal conditions such as permanent magnet demagnetization and winding failure, it is difficult to achieve stable operation, which cannot meet the rigid requirements of high-end equipment for high reliability, wide speed adaptability and strong fault tolerance.

[0005] Furthermore, six-phase windings, with their multi-degree-of-freedom control advantages, possess significant advantages in fault tolerance and harmonic suppression. However, current technologies have not yet combined them with dual-source excitation composite rotors to achieve coordinated output and fully decoupled control of synchronous torque, asynchronous torque, and levitation force. This fails to fully leverage the control potential of six-phase windings and overcome the performance bottlenecks of traditional bearingless motors. Therefore, developing a six-phase bearingless motor with dual-source excitation composite rotor capable of decoupled control of multiple physical quantities, high reliability and fault tolerance, and wide speed adaptability, along with its control method, has become a pressing technical problem in this field. Summary of the Invention

[0006] The technical problem to be solved by this invention is to overcome the defects in the prior art and provide a dual-source excitation composite rotor six-phase bearingless motor and its decoupling control method. This invention solves the technical problems of existing bearingless motors, such as single drive form, severe coupling of multiple physical quantities, insufficient fault tolerance, and inflexible air gap magnetic field adjustment. It achieves independent decoupling control of synchronous torque, asynchronous torque and levitation force, thereby improving the reliability of motor operation, control accuracy and wide speed adaptability.

[0007] The technical solution adopted by this invention to solve the technical problem is as follows: A dual-source excitation composite rotor six-phase bearingless motor includes a stator assembly, a rotor assembly, an inverter, and a control unit; The stator assembly includes a stator core and a single set of six-phase windings. The six-phase windings consist of a first three-phase winding and a second three-phase winding. The first three-phase winding and the second three-phase winding have a spatial electrical angle difference of 30°. The neutral points of the two sets of windings are isolated and they share the same stator core. The stator core has evenly distributed stator slots. The six-phase winding adopts a distributed winding structure and is embedded in the stator slots. The rotor assembly includes a rotor core, permanent magnets, squirrel cage bars, and end rings. The permanent magnets are embedded inside the rotor core in a built-in arrangement to form a permanent magnet excitation source with a pole pair number of p. The squirrel cage bars are uniformly embedded in the bar slots on the outer edge of the rotor core and short-circuited by the end rings to form a closed loop, constituting an induction excitation source. The permanent magnets and squirrel cage bars share the same air gap in space and are coupled in the magnetic circuit through the yoke of the rotor core, forming a dual-source excitation composite rotor structure with electromagnetic decoupling and spatial multiplexing. The inverter includes a first three-phase inverter and a second three-phase inverter, which are respectively connected to the first three-phase winding and the second three-phase winding. The control unit is used to perform spatial vector decoupling of the current of the six-phase winding, mapping it to three orthogonal subspaces: the αβ fundamental subspace, the z1z2 harmonic subspace, and the o1o2 zero-sequence subspace. In the αβ fundamental subspace, dq-axis current control is performed, where the d-axis current simultaneously adjusts the permanent magnet flux linkage of the permanent magnet and the induced flux linkage of the squirrel cage conductors, and the q-axis current controls the coordinated output of synchronous and asynchronous torque according to a speed-related distribution coefficient. In the z1z2 harmonic subspace, harmonic current control is performed to generate a harmonic rotating magnetic field with a pole pair difference of 1 from the fundamental magnetic field, which couples with the fundamental magnetic field to generate a controllable radial levitation force. In the o1o2 zero-sequence subspace, the zero-sequence current component is monitored to identify winding asymmetry faults. Simultaneously, a flux linkage observer is constructed in the αβ fundamental subspace to identify permanent magnet demagnetization faults, generating fault diagnosis signals that are fed back to the αβ fundamental subspace and the z1z2 harmonic subspace, triggering operating condition switching or compensating harmonic injection.

[0008] Furthermore, the stator core has 48 uniformly distributed stator slots; the αβ fundamental subspace corresponds to the fundamental magnetic field and the 12k±1 harmonic magnetic field, used to generate synchronous torque and asynchronous torque; the z1z2 harmonic subspace corresponds to the 6k±1 harmonic magnetic field, used to generate radial levitation force and fault magnetic field correction; the o1o2 zero-sequence subspace corresponds to the 3rd and multiples of 3 zero-sequence harmonic magnetic fields, used for fault detection and diagnosis.

[0009] Furthermore, the harmonic current control in the z1z2 harmonic subspace includes: controlling the 5th harmonic to generate a rotating magnetic field with a pole pair number of p-1, and controlling the 7th harmonic to generate a rotating magnetic field with a pole pair number of p+1. The 5th and 7th harmonics are coupled with the fundamental main magnetic field to generate a radial levitation force. When a winding fault occurs in the motor, the 11th or 13th compensating harmonics are injected into the z1z2 subspace to correct the distortion of the faulty magnetic field.

[0010] Furthermore, the control of the q-axis current according to the speed-related distribution coefficient includes: adjusting the synchronous / asynchronous torque distribution coefficient in real time according to the speed and slip rate, with asynchronous torque as the main force during the start-up phase, synchronous torque as the main force during the steady-state phase, and synchronous torque being shut off and completely replaced by asynchronous torque during the severe demagnetization phase.

[0011] Furthermore, the monitoring of the zero-sequence current component in the o1o2 zero-sequence subspace includes: calculating the amplitude and rate of change of the o1 and o2 components, and determining the winding asymmetry fault when the amplitude or rate of change exceeds the threshold; at the same time, constructing a flux linkage observer in the αβ fundamental subspace to observe the permanent magnet flux linkage amplitude in real time, and determining the permanent magnet demagnetization fault when the flux linkage amplitude is continuously lower than the preset flux linkage threshold.

[0012] Furthermore, the permanent magnet uses neodymium iron boron rare earth permanent magnet material with a pole pair number p=2; the squirrel cage conductors are made of copper, numbering 32, and are evenly distributed on the outer edge of the rotor core; the space vector decoupling employs extended Clark transform and Park transform, and the transformation matrix of the extended Clark transform is:

[0013] The first two rows of the matrix correspond to the αβ fundamental wave subspace, the middle two rows correspond to the z1z2 harmonic subspace, and the last two rows correspond to the o1o2 zero-order subspace.

[0014] This invention also provides a decoupling control method for a dual-source excitation composite rotor six-phase bearingless motor, implemented based on the aforementioned motor, comprising the following steps: A. Parameter acquisition steps: Acquire rotor speed, rotor position signal, phase current signal of six-phase winding and inverter output voltage signal, and filter and condition the acquired signals; B. Subspace decoupling steps: Perform space vector decoupling on the phase current signal and map it to the αβ fundamental subspace, z1z2 harmonic subspace and o1o2 zero-sequence subspace to obtain the current components of each subspace; C. Multi-physical quantity command generation steps: Based on the real-time operating conditions of the motor, synchronous torque command, asynchronous torque command, and radial suspension force command are adaptively generated; the weights of the synchronous torque command and asynchronous torque command are adjusted according to the operating conditions, and the radial suspension force command is generated independently of the torque command. D. Subspace Independent Control Steps: (d1) In the αβ fundamental subspace, based on synchronous torque command and asynchronous torque command, the proportional-integral regulation algorithm is used to generate d-axis and q-axis current control commands to control the amplitude and phase of the fundamental magnetic field, thereby achieving the coordinated output of synchronous torque and asynchronous torque; under normal operating conditions, the permanent magnet synchronous motor control mode is adopted, and under starting conditions or permanent magnet demagnetization fault conditions, the induction motor control mode is switched. (d2) In the z1z2 harmonic subspace, based on the radial levitation force command, a robust control and feedforward compensation algorithm is used to generate a harmonic current control command, which generates a harmonic rotating magnetic field with a pole pair difference of 1 from the fundamental main magnetic field, and couples with the fundamental main magnetic field to generate a radial levitation force; when the motor fails, a compensation harmonic is injected to correct the distortion of the fault magnetic field. (d3) In the o1o2 zero-sequence subspace, the fault type and location are identified based on the zero-sequence current component, and a fault diagnosis signal is generated. E. Control signal output steps: Convert the current control command of each subspace into a pulse width modulation control signal, and drive the six-phase winding through the inverter; F. Closed-loop control steps: Compare the actual operating parameters of the motor with the command parameters in real time, and dynamically adjust the current control commands of each subspace according to the deviation signal.

[0015] Furthermore, in step (d2), the 5th harmonic is controlled to generate a rotating magnetic field with a pole pair number of p-1, and the 7th harmonic is controlled to generate a rotating magnetic field with a pole pair number of p+1. The 5th and 7th harmonics are coupled with the fundamental main magnetic field to generate a radial levitation force; the compensation harmonic is the 11th or 13th harmonic.

[0016] Furthermore, in step C, under normal steady-state operating conditions, the synchronous torque command dominates and the asynchronous torque command is used to compensate for torque ripple; under low-speed start-up operating conditions, the asynchronous torque command dominates; and under permanent magnet demagnetization fault operating conditions, the asynchronous torque command completely replaces the synchronous torque command.

[0017] Furthermore, in step F, when the deviation between the actual rotational speed and the commanded rotational speed exceeds a preset rotational speed threshold, the current command of the αβ subspace is adjusted; when the deviation between the actual levitation force and the commanded levitation force exceeds a preset levitation force threshold, the current command of the z1z2 subspace is adjusted.

[0018] This invention is based on three core mechanisms: electromagnetic decoupling-spatial multiplexing, three-subspace orthogonal decoupling, and operating condition adaptive cooperative control. Details are as follows: 1. Electromagnetic decoupling-spatial multiplexing mechanism of dual-source excitation: (1) The permanent magnet and the squirrel cage conductor share the same air gap and are coupled to the same rotor core magnetic circuit, forming a special spatial-electromagnetic relationship. (2) Spatial reuse: The magnetic field generated by a single set of stator windings acts on both the permanent magnet and the squirrel cage conductors simultaneously, realizing "one winding, dual drive". (3) Electromagnetic decoupling: Although the direct-axis flux generated by the d-axis current links with both excitation sources, the permanent magnet flux linkage and the induction flux linkage are made independent in control through magnetic circuit design. The strength of the two magnetic fields can be adjusted separately by the same d-axis current, without interfering with each other's working state. (4) The torque generated by the two excitation sources is flexibly allocated through the distribution coefficient of the q-axis current: synchronous torque has fast response and high efficiency, while asynchronous torque has strong starting capability and good fault tolerance.

[0019] 2. Three-subspace orthogonal decoupling mechanism for six-phase windings: (1) The six-phase winding with double Y shift of 30° is decoupled by space vector and naturally mapped into three functionally independent orthogonal subspaces. (2) αβ fundamental subspace: carries the fundamental wave and specific subharmonics, and is responsible for electromechanical energy conversion. After rotation transformation, the d-axis controls the flux linkage and the q-axis controls the torque, realizing the decoupling of the magnetic field and the torque. (3) z1z2 harmonic subspace: carries specific subharmonics and is responsible for the generation of levitation force; among them, the harmonic components with a difference of 1 in the number of pole pairs are coupled with the fundamental main magnetic field to generate controllable radial levitation force, the amplitude and direction of which are independently adjusted by the harmonic current and have no cross interference with the torque control. (4) o1o2 zero-sequence subspace: carries zero-sequence harmonics and is responsible for fault diagnosis. Due to the isolation of the winding neutral point, the zero-sequence current only reflects the system asymmetry or excitation source abnormality, and the fault type is identified through feature analysis. (5) The three subspaces are mathematically completely orthogonal and have clearly separated physical functions, realizing "multi-functional reuse of a single winding".

[0020] 3. Fault-control closed-loop feedback mechanism: The zero-order subspace monitors the system status in real time, and the diagnostic results are fed back to the torque control subspace and the suspension force control subspace, triggering control strategy adjustments or the injection of compensation measures, forming a closed-loop fault-tolerant system of "detection-diagnosis-decision-execution".

[0021] The working process of this invention is divided into three modes according to the working conditions, and the function allocation of the control unit is dynamically adjusted in each mode: 1. Start-up process: (1) When the motor is stationary or at low speed, it is difficult for the permanent magnet to synchronize with the rotating magnetic field, and the torque output is limited. At this time, the control unit tilts the torque weight to the induction excitation source, and generates the dominant torque by relying on the asynchronous characteristics of the squirrel cage conductor, so as to achieve smooth self-start. The synchronous excitation source is only used as an auxiliary to provide initial excitation. (2) During the speed increase process, the control unit smoothly switches the torque weight and gradually increases the proportion of synchronous torque until the rated speed is reached.

[0022] 2. Steady-state operation process: (1) After the motor reaches the rated speed, the control unit tilts the torque weight to the permanent magnet excitation source, with synchronous torque as the main factor to ensure high-efficiency operation; the induction excitation source becomes auxiliary, and its torque component is used to compensate for the torque pulsation caused by air gap harmonics and suppress vibration noise. (2) The suspension force control operates independently, and the rotor is kept in contactless suspension by harmonic current adjustment, and is not affected by torque conditions.

[0023] 3. Fault tolerance process: (1) If an abnormality of the excitation source is detected during operation, the control unit identifies the fault type: if the permanent magnet excitation source fails, the torque weight is immediately transferred to the induction excitation source to maintain continuous torque output; if the winding is asymmetrical, compensation harmonics are injected into the levitation force control subspace to correct the magnetic field distortion and maintain levitation stability. (2) During the fault period, the motor continues to run without dereasing, and maintenance is carried out when maintenance conditions permit.

[0024] The positive and beneficial effects of this invention are as follows: 1. This invention adopts a dual-source excitation composite rotor structure of "electromagnetic decoupling-spatial reuse". The permanent magnet and the squirrel cage conductor share the same air gap and are coupled in the same magnetic circuit, realizing the coordinated control of dual excitation sources by a single winding. Under normal operating conditions, synchronous torque is the main force to ensure high-efficiency operation; under starting conditions, asynchronous torque is the main force to achieve smooth self-starting. It effectively solves the problems of the single drive form and poor wide speed adaptability of traditional bearingless motors, and takes into account both high efficiency and strong starting capability.

[0025] 2. This invention uses a six-phase single-set stator winding, which is decoupled by space vector to form three orthogonal subspaces. The αβ subspace is responsible for dual torque coordinated output, the z1z2 subspace is responsible for levitation force generation and fault magnetic field correction, and the o1o2 subspace is responsible for fault diagnosis and feedback control. Each subspace has a clear function and no coupling interference, realizing the full decoupling control of synchronous torque, asynchronous torque and levitation force, with high control accuracy and fast dynamic response.

[0026] 3. This invention achieves a collaborative control mechanism of "single winding driving dual sources" by simultaneously adjusting the permanent magnet flux linkage and the induction flux linkage through the d-axis current and controlling the dual torque weights according to the distribution coefficients through the q-axis current. This invention also forms a closed-loop fault-tolerant mechanism of "detection-diagnosis-decision-execution" by triggering the adjustment of control parameters in the αβ and z1z2 subspaces through the feedback of fault diagnosis signals in the o1o2 subspace. When the permanent magnet loses its magnetism, the asynchronous torque can be completely replaced, and when there is a winding fault, the harmonic injection can correct the magnetic field, which significantly improves the fault tolerance capability and operational reliability of the motor.

[0027] 4. This invention does not require additional suspension windings. It achieves multi-physical quantity control through a single set of six-phase windings, resulting in a compact structure and low cost. It eliminates the independent suspension windings found in traditional solutions, simplifies the motor topology, reduces manufacturing costs, and is suitable for the operational needs of high-end equipment such as aerospace and precision manufacturing.

[0028] 5. The air gap magnetic field strength of this invention can be continuously adjusted within a wide range to adapt to various operating requirements such as weakening the field to increase speed, increasing the field to increase torque, etc., thereby enhancing the motor's adaptability to complex loads.

[0029] 6. This invention utilizes asynchronous torque components to compensate for torque pulsation caused by air gap harmonics, and combines this with the harmonic suppression characteristics of the six-phase winding to effectively reduce vibration and noise during motor operation and improve operational stability. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the structure of the dual-source excitation composite rotor six-phase bearingless motor of the present invention (wherein, 1-stator assembly, 2-rotor assembly, 3-stator core, 4-six-phase winding, 5-rotor core, 6-permanent magnet, 7-squirrel cage conductor). Figure 2 This is an overall block diagram of the multi-subspace decoupling control of the present invention; Figure 3 This is the main circuit diagram of the present invention; Figure 4 This is a schematic diagram of the vector control system of the present invention; Figure 5 This is a structural diagram of the double Y-shifted 30° winding in this invention; Figure 6 This is a structural diagram of the 48-slot double three-phase winding in this invention. Detailed Implementation

[0031] The present invention will be further described in detail below with reference to specific embodiments, but these are not intended to limit the present invention: Example 1: Structure of a six-phase bearingless motor with dual-source excitation composite rotor: like Figure 1As shown, the dual-source excitation composite rotor six-phase bearingless motor of this embodiment includes a stator assembly 1, a rotor assembly 2, an inverter, and a control unit; the stator assembly 1 includes a stator core 3 and a single set of six-phase windings 4. The stator core 3 is made of laminated silicon steel sheets, and 48 evenly distributed stator slots are formed on the stator core 3; the six-phase windings 4 are composed of a first three-phase winding (A1, B1, C1) and a second three-phase winding (A2, B2, C2). The spatial electrical angles of the two sets of three-phase windings differ by 30°, the neutral points of the two sets of windings are isolated, a distributed winding structure is adopted, the pitch is 1-7, and they are embedded in the stator slots; as shown Figure 6 As shown, the 48-slot double three-phase winding structure provides sufficient slots to achieve distributed winding arrangement and optimize magnetic field waveform.

[0032] The rotor assembly 2 includes a rotor core 5, permanent magnets 6, squirrel cage bars 7, and end rings. The rotor core 5 is made of laminated silicon steel sheets. The permanent magnets 6 are made of neodymium iron boron rare earth permanent magnet material (NdFeB) with a pole pair number p=2. They are evenly arranged along the circumference of the rotor core 5 and embedded in the permanent magnet slots inside the rotor core 5 to form a permanent magnet excitation source. The squirrel cage bars 7 are made of copper and consist of 32 bars. They are evenly embedded in the bar slots on the outer edge of the rotor core 5 and fixed at both ends by welding with copper end rings to form a closed loop, which constitutes an induction excitation source.

[0033] In this embodiment, the permanent magnet 6 and the squirrel cage conductor 7 form a special structural relationship of "electromagnetic decoupling-spatial reuse": spatially, the two share the same air gap, and the permanent magnet magnetic field generated by the permanent magnet 6 and the magnetic field induced by the squirrel cage conductor 7 are superimposed in the same air gap; in the magnetic circuit, the two are coupled through the yoke of the rotor core 5, and the permanent magnet flux and the induced flux share part of the magnetic circuit, but through the magnetic circuit design, the two are electromagnetically decoupled, that is, the excitation state of the permanent magnet 6 is not directly affected by the induced current of the squirrel cage conductor 7, and the induced current of the squirrel cage conductor 7 is not significantly affected by the saturation of the magnetic field of the permanent magnet 6. This structure realizes "dual-source excitation", that is, the magnetic field generated by the same set of stator windings can act on the permanent magnet 6 and the squirrel cage conductor 7 at the same time, and the torque generated by the two excitation sources can be flexibly adjusted through the control strategy.

[0034] The inverter is a six-phase voltage-source inverter, including a first three-phase inverter and a second three-phase inverter, which are respectively connected to the first three-phase winding and the second three-phase winding. The inverter uses IGBTs as switching devices with a switching frequency of 10kHz, which are used to convert the control signals output by the control unit into frequency- and voltage-adjustable AC power.

[0035] The control unit adopts a DSP controller + FPGA architecture and is electrically connected to the inverter, photoelectric encoder (position sensor), and Hall current sensor, respectively. The photoelectric encoder is used to acquire rotor speed and rotor position signals, and the Hall current sensor is used to acquire phase current signals of the six-phase windings.

[0036] Example 2: Spatial Vector Decoupling and Three-Subspace Control. This example details the spatial vector decoupling process and the independent control mechanism for the three subspaces. The control unit collects the six-phase current signals (i A1 i B1 i C1 i A2 i B2 i C2 Perform the extended Clark transformation, and the transformation matrix is:

[0037] This matrix maps the six-phase current into six components: α, β, z1, z2, o1, and o2. The first two rows correspond to the αβ fundamental subspace, the middle two rows correspond to the z1z2 harmonic subspace, and the last two rows correspond to the o1o2 zero-sequence subspace. The α and β components are transformed into d and q-axis rotating coordinate system components by Park transformation; the z1 and z2 components are transformed into z1d and z2q-axis components by Park transformation; the o1 and o2 components remain in the stationary coordinate system and are directly used for fault detection.

[0038] αβ fundamental subspace control: dq-axis current control is performed in the αβ fundamental subspace, where the d-axis current i d Simultaneously adjust the permanent magnet flux linkage ψpm of the permanent magnet and the induced flux linkage ψcage of the squirrel cage bars. This is because the direct-axis magnetic flux generated by the d-axis current links with both the permanent magnet flux linkage of the permanent magnet and the induced flux linkage of the squirrel cage bars. By adjusting i d It can simultaneously affect the magnetic field strength of two excitation sources, achieving a synergistic control effect of "single current dual adjustment".

[0039] q-axis current i q According to the speed-related allocation coefficient k ω The coordinated output of synchronous and asynchronous torque is controlled by a distribution coefficient k(ω), which is a function of the rotational speed and is defined as follows:

[0040] Where ω is the actual rotational speed, ω1 is the start-transition switching threshold (e.g., 500 r / min), ω2 is the transition-normal switching threshold (e.g., 1500 r / min), and k max This is the synchronous torque ratio coefficient under normal operating conditions, with a value ranging from 0.8 to 0.95.

[0041] Synchronous torque current component i qs = k ω ·i q asynchronous torque current component i qa = [1-k ω ]·i qUnder normal steady-state conditions, k ω >0.8, synchronous torque dominates, asynchronous torque is used to compensate for torque ripple; under low-speed starting conditions, k ω <0.2, asynchronous torque dominates, relying on the inductive characteristics of the squirrel cage conductor to achieve smooth self-starting; under permanent magnet demagnetization fault conditions, k ω =0, asynchronous torque completely replaces synchronous torque, ensuring continuous operation of the motor.

[0042] z1z2 Harmonic Subspace Control: Harmonic current control is performed in the z1z2 harmonic subspace. The z1z2 subspace corresponds to 6k±1 harmonic magnetic fields, including the 5th, 7th, 11th, and 13th harmonics. Among them, the 5th harmonic generates a reverse rotating magnetic field with a pole pair number of p-1=1, and the 7th harmonic generates a forward rotating magnetic field with a pole pair number of p+1=3. According to the levitation force generation mechanism of bearingless motors, when the difference in the pole pair number between the two sets of magnetic fields is 1 and they rotate synchronously, a constant and controllable radial levitation force can be generated. In this embodiment, the fundamental main magnetic field has a pole pair number of p=2, and the 5th and 7th harmonic magnetic fields have pole pairs of 1 and 3 respectively, satisfying the pole pair number matching conditions of |2-1|=1 and |3-2|=1. Radial levitation force is generated through coupling with the fundamental magnetic field.

[0043] The amplitude and phase of the 5th and 7th harmonic currents are independently controllable, completely decoupling the magnitude and direction of the radial levitation force from the torque output. When the motor experiences an open-circuit fault in its windings, 11th or 13th compensating harmonics (also belonging to 6k±1, when k=2) are injected into the z1z2 subspace to correct the magnetic field distortion caused by the fault and maintain the stability of the levitation force.

[0044] o1o2 zero-sequence subspace control: Monitor the zero-sequence currents of the o1 and o2 components in the o1o2 zero-sequence subspace. The o1o2 subspace corresponds to the 3rd and multiples of 3 zero-sequence harmonic magnetic fields. Calculate the amplitudes |o1|, |o2| and the rates of change d|o1| / dt, d|o2| / dt of the zero-sequence currents.

[0045] When the amplitude exceeds the first threshold or the rate of change exceeds the second threshold, it is determined to be a winding asymmetry fault; at the same time, a flux linkage observer is constructed in the αβ fundamental wave subspace to observe the permanent magnet flux linkage amplitude in real time. When the flux linkage amplitude is continuously lower than the preset flux linkage threshold, it is determined to be a permanent magnet demagnetization fault.

[0046] The fault diagnosis signal is fed back to the αβ fundamental subspace and the z1z2 harmonic subspace: For permanent magnet demagnetization faults, the αβ subspace is triggered to force k(ω) to 0, and switch to asynchronous torque drive completely; for winding asymmetry faults, the z1z2 subspace is triggered to inject the corresponding number of compensation harmonics to correct the magnetic field distortion.

[0047] Example 3: Complete Control Flow. This example specifically illustrates the complete decoupling control method flow: S1. Parameter Acquisition: The rotor speed (0-3000 r / min) and rotor position signal (1024 lines resolution) are acquired by photoelectric encoder; the phase current of the six-phase winding (0-50A) is acquired by Hall current sensor; the inverter output voltage (0-380V) is acquired by voltage sensor; the signals are transmitted to the control unit after low-pass filtering and A / D conversion. S2, Subspace Decoupling: Perform extended Clark transformation and Park transformation as described in Example 2 to obtain the components d, q, z1d, z2q, o1, and o2; S3. Multi-physical quantity command generation: Generate torque command and levitation force command according to the working condition; Normal working condition: synchronous torque command 100 N·m, asynchronous torque command 5 N·m, levitation force command 500 N; Start-up working condition: asynchronous torque command 80 N·m, synchronous torque command 20 N·m; Permanent magnet demagnetization fault: asynchronous torque command 100 N·m, replacing synchronous torque. S4, Independent Subspace Control: (1) αβ fundamental subspace: d-axis and q-axis current commands are generated using PI adjustment algorithm (Kp=0.5, Ki=0.01); permanent magnet synchronous motor control mode is used under normal operating conditions, and induction motor control mode is switched to during startup or fault conditions. (2) z1z2 harmonic subspace: Robust control and feedforward compensation algorithm are used to generate z1d and z2q axis current commands to control the 5th and 7th harmonics to generate radial levitation force; 11th or 13th harmonic compensation is injected in case of fault; (3) o1o2 zero-sequence subspace: monitor zero-sequence current, threshold 5A, identify fault type and generate diagnostic signal; S5. Control signal output: After inverse Park transform and inverse Clark transform, a PWM signal is generated through SVPWM modulation to drive the inverter; S6. Closed-loop regulation: Real-time comparison of actual parameters and command parameters; when the speed deviation exceeds 10 r / min, adjust the αβ subspace current command; when the levitation force deviation exceeds 20 N, adjust the z1z2 subspace current command.

[0048] In this embodiment, the motor achieves smooth speed regulation from 0 to 3000 r / min, torque pulsation is less than 5%, and suspension accuracy is 0.01 mm; in the event of permanent magnet demagnetization fault, the operating condition is switched within 0.1 s, and the motor continues to run stably; in the event of winding open circuit fault, the suspension force fluctuation is less than 10% by injecting a corrective magnetic field through harmonics.

Claims

1. A dual-source field excited compound rotor six-phase bearingless motor, characterized in that: Includes stator assembly, rotor assembly, inverter and control unit; The stator assembly includes a stator core and a single set of six-phase windings. The six-phase windings consist of a first three-phase winding and a second three-phase winding. The first three-phase winding and the second three-phase winding have a spatial electrical angle difference of 30°. The neutral points of the two sets of windings are isolated and they share the same stator core. The stator core has evenly distributed stator slots. The six-phase winding adopts a distributed winding structure and is embedded in the stator slots. The rotor assembly includes a rotor core, permanent magnets, squirrel cage bars, and end rings. The permanent magnets are embedded inside the rotor core in a built-in arrangement to form a permanent magnet excitation source with a pole pair number of p. The squirrel cage bars are uniformly embedded in the bar slots on the outer edge of the rotor core and short-circuited by the end rings to form a closed loop, constituting an induction excitation source. The permanent magnets and squirrel cage bars share the same air gap in space and are coupled in the magnetic circuit through the yoke of the rotor core, forming a dual-source excitation composite rotor structure with electromagnetic decoupling and spatial multiplexing. The inverter includes a first three-phase inverter and a second three-phase inverter, which are respectively connected to the first three-phase winding and the second three-phase winding. The control unit is used to perform spatial vector decoupling of the current of the six-phase winding, mapping it to three orthogonal subspaces: the αβ fundamental subspace, the z1z2 harmonic subspace, and the o1o2 zero-sequence subspace. In the αβ fundamental subspace, dq-axis current control is performed, where the d-axis current simultaneously adjusts the permanent magnet flux linkage of the permanent magnet and the induced flux linkage of the squirrel cage conductors, and the q-axis current controls the coordinated output of synchronous and asynchronous torque according to a speed-related distribution coefficient. In the z1z2 harmonic subspace, harmonic current control is performed to generate a harmonic rotating magnetic field with a pole pair difference of 1 from the fundamental magnetic field, which couples with the fundamental magnetic field to generate a controllable radial levitation force. In the o1o2 zero-sequence subspace, the zero-sequence current component is monitored to identify winding asymmetry faults. Simultaneously, a flux linkage observer is constructed in the αβ fundamental subspace to identify permanent magnet demagnetization faults, generating fault diagnosis signals that are fed back to the αβ fundamental subspace and the z1z2 harmonic subspace, triggering operating condition switching or compensating harmonic injection.

2. The dual-source field compounded rotor six-phase bearingless motor of claim 1, wherein: The stator core has 48 uniformly distributed stator slots; the αβ fundamental subspace corresponds to the fundamental magnetic field and the 12k±1 harmonic magnetic field, which is used to generate synchronous torque and asynchronous torque; the z1z2 harmonic subspace corresponds to the 6k±1 harmonic magnetic field, which is used to generate radial levitation force and fault magnetic field correction; the o1o2 zero-sequence subspace corresponds to the 3rd and multiples of 3 zero-sequence harmonic magnetic fields, which are used for fault detection and diagnosis.

3. The dual-excitation compound rotor six-phase bearingless motor of claim 2, wherein: The harmonic current control in the z1z2 harmonic subspace includes: controlling the 5th harmonic to generate a rotating magnetic field with a pole pair number of p-1, and controlling the 7th harmonic to generate a rotating magnetic field with a pole pair number of p+1. The 5th and 7th harmonics are coupled with the fundamental main magnetic field to generate a radial levitation force. When a winding fault occurs in the motor, the 11th or 13th compensating harmonics are injected into the z1z2 subspace to correct the distortion of the faulty magnetic field.

4. The dual-source excitation composite rotor six-phase bearingless motor according to claim 1, characterized in that: The synchronous / asynchronous torque distribution coefficient is adjusted in real time according to the speed and slip rate. During the start-up phase, asynchronous torque is the main force, during the steady-state phase, synchronous torque is the main force, and during the severe demagnetization phase, synchronous torque is turned off and completely replaced by asynchronous torque.

5. The dual-source excitation composite rotor six-phase bearingless motor according to claim 1, characterized in that: The monitoring of the zero-sequence current component in the o1o2 zero-sequence subspace includes: calculating the amplitude and rate of change of the o1 and o2 components; when the amplitude or rate of change exceeds a threshold, it is determined to be a winding asymmetry fault; at the same time, a flux linkage observer is constructed in the αβ fundamental subspace to observe the permanent magnet flux linkage amplitude in real time; when the flux linkage amplitude is continuously lower than the preset flux linkage threshold, it is determined to be a permanent magnet demagnetization fault.

6. The dual-source excitation composite rotor six-phase bearingless motor according to claim 1, characterized in that: The permanent magnet is made of neodymium iron boron rare earth permanent magnet material with a pole pair number p=2; the squirrel cage conductor is made of copper material, and there are 32 of them, which are evenly distributed on the outer edge of the rotor core.

7. A decoupling control method for a dual-source excitation composite rotor six-phase bearingless motor, implemented based on the motor described in any one of claims 1 to 6, characterized in that, Includes the following steps: A. Parameter acquisition steps: Acquire rotor speed, rotor position signal, phase current signal of six-phase winding and inverter output voltage signal, and filter and condition the acquired signals; B. Subspace decoupling steps: Perform space vector decoupling on the phase current signal and map it to the αβ fundamental subspace, z1z2 harmonic subspace and o1o2 zero-sequence subspace to obtain the current components of each subspace; C. Multi-physical quantity command generation steps: Based on the real-time operating conditions of the motor, synchronous torque command, asynchronous torque command, and radial suspension force command are adaptively generated; the weights of the synchronous torque command and asynchronous torque command are adjusted according to the operating conditions, and the radial suspension force command is generated independently of the torque command. D. Subspace Independent Control Steps: (d1) In the αβ fundamental subspace, based on synchronous torque command and asynchronous torque command, the proportional-integral regulation algorithm is used to generate d-axis and q-axis current control commands to control the amplitude and phase of the fundamental magnetic field, thereby achieving the coordinated output of synchronous torque and asynchronous torque; under normal operating conditions, the permanent magnet synchronous motor control mode is adopted, and under starting conditions or permanent magnet demagnetization fault conditions, the induction motor control mode is switched. (d2) In the z1z2 harmonic subspace, based on the radial levitation force command, a robust control and feedforward compensation algorithm is used to generate a harmonic current control command, which generates a harmonic rotating magnetic field with a pole pair difference of 1 from the fundamental main magnetic field, and couples with the fundamental main magnetic field to generate a radial levitation force; when the motor fails, a compensation harmonic is injected to correct the distortion of the fault magnetic field. (d3) In the o1o2 zero-sequence subspace, the fault type and location are identified based on the zero-sequence current component, and a fault diagnosis signal is generated. E. Control signal output steps: Convert the current control command of each subspace into a pulse width modulation control signal, and drive the six-phase winding through the inverter; F. Closed-loop control steps: Compare the actual operating parameters of the motor with the command parameters in real time, and dynamically adjust the current control commands of each subspace according to the deviation signal.

8. The decoupling control method according to claim 7, characterized in that: In step (d2), the 5th harmonic is controlled to generate a rotating magnetic field with a pole pair number of p-1, and the 7th harmonic is controlled to generate a rotating magnetic field with a pole pair number of p+1. The 5th and 7th harmonics are coupled with the fundamental main magnetic field to generate a radial levitation force. The compensation harmonic is the 11th or 13th harmonic.

9. The decoupling control method according to claim 7, characterized in that: In step C, under normal steady-state conditions, synchronous torque command dominates and asynchronous torque command is used to compensate for torque pulsation; under low-speed start-up conditions, asynchronous torque command dominates; and under permanent magnet demagnetization fault conditions, asynchronous torque command completely replaces synchronous torque command.

10. The decoupling control method according to claim 7, characterized in that: In step F, when the deviation between the actual rotational speed and the commanded rotational speed exceeds a preset rotational speed threshold, the current command of the αβ subspace is adjusted; when the deviation between the actual levitation force and the commanded levitation force exceeds a preset levitation force threshold, the current command of the z1z2 subspace is adjusted.

Images