A method for rotor polarity identification of a ferrite-assisted synchronous reluctance motor

By constructing an asymmetric magnetic circuit in the rotor and setting taps in the stator windings, the polarity of the +d and -d axes of the permanent magnet assisted synchronous reluctance motor is identified using the voltage difference signal under square wave voltage excitation. This solves the problem of difficult identification in the prior art and achieves high reliability and low risk polarity identification.

CN123122404APending Publication Date: 2026-11-17THE UNIV OF NOTTINGHAM NINGBO CHINA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610987778.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-03
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately identify the +d and -d axis polarities of the rotor in a permanent magnet assisted synchronous reluctance motor, leading to problems such as starting failure, reverse torque, or current surges. Furthermore, existing methods are highly dependent on ferrite permanent magnets, limiting their versatility and robustness.

Method used

By constructing an asymmetric magnetic circuit in the +d and -d axis directions of the rotor, and setting taps in the stator winding to divide the winding into two series-connected winding segments, the polarity is determined by the voltage difference signal under square wave voltage excitation, avoiding large current magnetic saturation injection, and a dual threshold decision mechanism is adopted for polarity identification.

Benefits of technology

It enables accurate identification of rotor polarity without the need for high current magnetic saturation, reduces the risk of current surges and torque disturbances, improves the reliability of identification and engineering applicability, and is suitable for three-phase ferrite permanent magnet assisted synchronous reluctance motors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123122404A_ABST
    Figure CN123122404A_ABST
Patent Text Reader

Abstract

This invention discloses a rotor polarity identification method for a ferrite-assisted synchronous reluctance motor, relating to the field of reluctance motor technology. It involves constructing an asymmetric magnetic circuit along the +d and -d axes of the rotor, and setting taps in the stator windings to form a first winding segment and a second winding segment with different spatial positions, which are connected in series. When the motor is stationary or at low speed, alternating positive and negative square wave voltages are applied to the series branch. During the stable plateau region, the voltages of the two winding segments are collected and the average voltage difference is calculated. This average voltage difference is compared with a preset first threshold and a second threshold. If the difference is less than the first threshold, it is identified as the +d axis; if it is greater than the second threshold, it is identified as the -d axis; and if it falls between the two, it is identified as a blind zone and no result is output. This invention eliminates the π electrical angle polarity ambiguity between the +d and -d axes in traditional methods without relying on large-current magnetic saturation pulses, while also reducing the risk of current surges and permanent magnet demagnetization.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of reluctance motor technology, and more specifically to a rotor polarity identification method for a ferrite-assisted synchronous reluctance motor. Background Technology

[0002] Permanent magnet assisted synchronous reluctance motors combine permanent magnet torque and reluctance torque, offering advantages such as simple structure, high efficiency, and low usage of rare-earth permanent magnet materials. They have been increasingly applied in industrial motor drive systems. For high-performance control of this type of motor, especially during standstill and low-speed start-up, accurate acquisition of the rotor's initial position and magnetic pole polarity is crucial. Incorrect polarity determination can lead to start-up failure, reverse torque, or current surges.

[0003] In existing technologies, high-frequency signal injection methods based on the salient polarity of the motor are commonly used for rotor position estimation. These methods utilize the difference in inductance between the d-axis and q-axis magnetic circuits of the motor, determining the rotor magnetic axis direction by detecting the current response under the injected signal. However, such methods typically only identify the linear direction of the rotor magnetic axis and struggle to distinguish between the +d-axis and -d-axis, as they exhibit similar or equivalent characteristics in their salient polarity responses, leading to phase differences. Ambiguity of the polarity of electrical angle.

[0004] To eliminate this ambiguity, existing methods typically employ DC current injection or voltage pulse injection to induce nonlinear magnetic saturation in the motor's magnetic circuit, then compare the differences in current response under different injection directions to determine polarity. However, these methods have several drawbacks: Firstly, they are highly dependent on the nonlinear response caused by magnetic saturation, resulting in a weak polarity discrimination signal. The identification effect is easily affected by the injection voltage amplitude, pulse width, sampling time, and non-ideal factors of the inverter, limiting their versatility and robustness. Secondly, for permanent magnet assisted synchronous reluctance motors using ferrite permanent magnets, the permanent magnet magnetic field bias is relatively weak, often requiring a large injection current to obtain a detectable polarity difference signal. This may lead to significant current surges, torque disturbances, and the risk of permanent magnet demagnetization. Furthermore, some existing methods require parameter tuning for different motor structures, making them difficult to directly apply. Summary of the Invention

[0005] To improve the reliability and engineering applicability of rotor polarity identification in permanent magnet assisted synchronous reluctance motors during the stationary or low-speed start-up phase, this invention proposes a rotor polarity identification method for ferrite assisted synchronous reluctance motors. The motor's rotor has an asymmetric magnetic circuit structure in the +d-axis and -d-axis directions, and at least one phase of the stator winding is tapped to divide that phase winding into a first winding and a second winding with different spatial positions. The first winding segment and the second winding segment are configured to be connected in series. The rotor polarity identification method includes the following steps: S1: When the motor is stationary or at low speed, the first winding segment and the second winding segment connected in series are treated as a whole branch, and alternating positive and negative square wave voltage excitation is applied to both ends of the whole branch. S2: In each positive half-cycle of the square wave voltage excitation, the first voltage at both ends of the first winding segment and the second voltage at both ends of the second winding segment are collected in the stable plateau interval, and the average voltage difference in the stable plateau interval is calculated. S3: Compare the average voltage difference with a preset first threshold and a second threshold, wherein the first threshold is less than the second threshold; S4: If the average voltage difference is less than the first threshold, the current d-axis polarity of the rotor is determined to be +d-axis; if the average voltage difference is greater than the second threshold, the current d-axis polarity of the rotor is determined to be -d-axis; if the average voltage difference is between the first threshold and the second threshold, it is determined that the rotor is currently in the polarity recognition blind zone and no polarity judgment result is output.

[0006] This invention constructs an asymmetric magnetic circuit in the +d and -d axis directions of the rotor and sets taps in the stator winding to form two series-connected winding segments. Under square wave excitation, the voltage difference between the two winding segments is collected and compared with a preset double threshold. The rotor polarity is directly determined by the stable voltage difference signal generated by the asymmetry of the magnetic circuit. This eliminates the π electrical angle polarity ambiguity between the +d and -d axes without relying on large current magnetic saturation injection, while avoiding the risks of current impact and permanent magnet demagnetization.

[0007] Furthermore, the asymmetric magnetic circuit structure is achieved by removing or thinning the radial and tangential magnetic bridges on the +d axis magnetic circuit of the rotor, while retaining the radial and tangential magnetic bridges on the -d axis magnetic circuit.

[0008] Furthermore, the regions after the radial and tangential magnetic bridges are removed or thinned are filled with a non-magnetic reinforcing material.

[0009] Furthermore, the motor has P poles, and the single-phase winding is divided into P sequentially distributed equivalent coil groups along the circumferential direction according to the magnetic pole period. Adjacent equivalent coil groups are separated by a mechanical angle of 360° / P. With the number of turns and structure of each equivalent coil group being consistent, the tap is located at the 1 / P node and the 2 / P node of the total number of series turns of the phase winding, such that the first winding segment corresponds to the winding portion from the starting end of the winding to the 1 / P node, and the second winding segment corresponds to the winding portion from the 1 / P node to the 2 / P node. The first winding segment and the second winding segment respectively correspond to the equivalent coil groups of two adjacent magnetic pole regions on the rotor.

[0010] Furthermore, the motor is a three-phase motor, including phase A, phase B and phase C. The rotor polarity identification method performs steps S1 to S4 on the windings of phase A, phase B and phase C respectively to obtain the average voltage difference of each phase. When a phase is in the polarity identification blind zone, the current d-axis polarity of the rotor is determined according to the comparison results of the average voltage difference of the other two phases with the first threshold and the second threshold, based on the preset decision logic.

[0011] Furthermore, the preset decision logic is as follows: when phase A is in the polarity identification blind zone and the average voltage difference of phase B is less than the first threshold and the average voltage difference of phase C is greater than the second threshold, the current d-axis polarity of the rotor is determined to be -d-axis; when phase A is in the polarity identification blind zone and the average voltage difference of phase B is greater than the second threshold and the average voltage difference of phase C is less than the first threshold, the current d-axis polarity of the rotor is determined to be +d-axis.

[0012] Furthermore, the first threshold and the second threshold are obtained in advance through finite element simulation or prototype calibration. The first threshold corresponds to the average voltage difference when the rotor is located at the first theoretical boundary position between adjacent +d and -d axes, and the second threshold corresponds to the average voltage difference when the rotor is located at the second theoretical boundary position between adjacent +d and -d axes. The first theoretical boundary position and the second theoretical boundary position are located in the +q and -q axis directions of the rotor, respectively.

[0013] Furthermore, the motor has P poles, the mechanical angle interval between adjacent +d and -d axes is 360° / P, the first theoretical dividing position is located at a mechanical angle of 180° / P relative to the +d axis, and the second theoretical dividing position is located at a mechanical angle of 540° / P relative to the +d axis.

[0014] Furthermore, the stable platform interval is the quasi-steady-state time period after the end of the switching transient in each positive half-cycle of the square wave voltage excitation. The first voltage and the second voltage are sampled multiple times and the arithmetic mean is calculated during the quasi-steady-state time period.

[0015] Compared with the prior art, the present invention has at least the following beneficial effects: (1) The present invention proposes a rotor polarity identification method for a ferrite-assisted synchronous reluctance motor. By constructing an asymmetric magnetic circuit in the +d and -d axis directions of the rotor, the two winding segments exhibit different equivalent inductances under the same injected current. The inductance difference is converted into a measurable voltage difference signal by using winding taps. Since it is not necessary to apply a large DC current or voltage pulse to saturate the motor magnetic circuit, it can effectively reduce current impact and torque disturbance, and at the same time reduce the risk of demagnetization of ferrite permanent magnets under high current excitation. (2) By detecting the average voltage difference between two series winding segments under square wave excitation and determining the polarity according to the preset threshold, the detection signal is the average DC voltage. There is no need for complex current demodulation or high-frequency signal processing. The detection process is direct and the amount of calculation is small, which makes it easy to implement in conventional motor controllers. (3) By drawing out taps from specific nodes of the electric phase winding to form a detection section, the tap structure is only used for polarity identification during the start-up or low-speed phase, and does not affect the winding connection and working mode of the motor during normal three-phase operation, thus having good engineering compatibility. (4) Taking advantage of the 120° electrical angle difference between the three-phase windings in space, each phase is detected in sequence. When a phase is in the identification blind zone near the zero point of the voltage difference, the detection results of the other two phases can be used for compensation and judgment. This can effectively eliminate the single-phase detection blind zone and realize polarity identification within the entire rotor position range. Attached Figure Description

[0016] Figure 1 A step diagram illustrating a rotor polarity identification method for a ferrite-assisted synchronous reluctance motor; Figure 2 This is a schematic diagram of a symmetrical rotor structure; Figure 3 This is a schematic diagram of an asymmetric rotor structure; Figure 4 This is a schematic diagram of the structure of the tap terminals of the electrical phase winding; Figure 5 This is a schematic diagram of square wave voltage excitation; Figure 6 A schematic diagram showing the selection of the quasi-steady-state time period. Detailed Implementation

[0017] The following are specific embodiments of the present invention, which are described in conjunction with the accompanying drawings. However, the present invention is not limited to these embodiments.

[0018] Permanent magnet assisted synchronous reluctance motors combine permanent magnet torque and reluctance torque, offering advantages such as simple structure, high efficiency, and low rare earth element usage, and have been increasingly applied in industrial motor drive systems. For high-performance control of this type of motor, especially during stationary and low-speed start-up phases, accurate acquisition of the rotor's initial position and magnetic pole polarity information is crucial. Existing technologies often employ high-frequency signal injection methods based on the motor's salient polarity for rotor position estimation. However, these methods typically only identify the rotor's magnetic axis direction, making it difficult to directly distinguish between the +d and -d axes, resulting in π-electric angle polarity ambiguity. To address this issue, existing methods usually inject a large current to induce nonlinear magnetic saturation in the motor's magnetic circuit, comparing the current response differences under different injection directions to determine polarity. However, these methods rely on the nonlinear response caused by magnetic saturation, making their identification effectiveness susceptible to injection parameters and inverter non-ideal factors, limiting their versatility and robustness. Furthermore, for motors with weak ferrite permanent magnet magnetic field bias, a large injection current is required to obtain a detectable polarity difference signal, posing risks of current surges, torque disturbances, and permanent magnet demagnetization. To address the above problems, this invention proposes a rotor polarity identification method for a ferrite-assisted synchronous reluctance motor, such as... Figure 1 As shown, the method mainly includes the following steps: S1: When the motor is stationary or at low speed, the first winding segment and the second winding segment connected in series are treated as a whole branch, and alternating positive and negative square wave voltage excitation is applied to both ends of the whole branch. S2: In each positive half-cycle of the square wave voltage excitation, the first voltage at both ends of the first winding segment and the second voltage at both ends of the second winding segment are collected in the stable plateau interval, and the average voltage difference in the stable plateau interval is calculated. S3: Compare the average voltage difference with a preset first threshold and a second threshold, wherein the first threshold is less than the second threshold; S4: If the average voltage difference is less than the first threshold, the current d-axis polarity of the rotor is determined to be +d-axis; if the average voltage difference is greater than the second threshold, the current d-axis polarity of the rotor is determined to be -d-axis; if the average voltage difference is between the first threshold and the second threshold, it is determined that the rotor is currently in the polarity recognition blind zone and no polarity judgment result is output.

[0019] The realization of the above-mentioned functions relies on the specific rotor magnetic circuit design proposed in this invention. Specifically, this invention constructs an asymmetric magnetic circuit structure for the rotor in the +d-axis and -d-axis directions, and sets taps in the stator windings to divide the same phase winding into two winding segments with different spatial positions. The rotor d-axis polarity is determined by the voltage difference generated by the difference in the magnetic circuits of the two winding segments under the same excitation. This method is particularly suitable for three-phase ferrite permanent magnet assisted synchronous reluctance motors, which are widely used in industrial drive fields. Their rotor poles are usually 4, with a few cases having 6 poles, but generally not exceeding 6 poles, and the motor windings are three-phase windings, not involving five-phase, six-phase, or other multi-phase motor structures.

[0020] First, the rotor structure of the ferrite-assisted synchronous reluctance motor in this invention will be described. The rotor consists of a rotor core, a magnetic barrier structure, and ferrite permanent magnets embedded within the magnetic barriers. For example... Figure 2 As shown (the dark area represents ferrite permanent magnets), the "main magnetic barrier" of a ferrite-assisted reluctance motor is filled with ferrite permanent magnets, while the "side magnetic barriers" on the left and right sides are composed of air. The magnetization direction of the ferrite is defined as the electrical -q axis. Taking clockwise as the positive direction in the diagram, the axes are, in order, +q, +d, -q, and -d. Each axis differs from the other by 90 electrical degrees, which is equal to the mechanical angle multiplied by P / 2, where P is the number of rotor poles. For a 4-pole motor, each axis differs from the other by 45 mechanical degrees. A 4-pole motor has two complete qd coordinate systems, while a P-pole motor has P / 2 complete qd coordinate systems.

[0021] In conventional designs, rotors typically have radial and tangential magnetic bridges along the d-axis to connect the core regions between adjacent magnetic barriers and provide mechanical strength. However, this symmetrical magnetic bridge structure results in symmetrical magnetic circuits along the +d and -d axes. Before reaching magnetic saturation, the equivalent inductance in both directions is essentially the same, which is the fundamental reason why traditional high-frequency signal injection methods cannot distinguish the polarity of the +d and -d axes. To address this problem, this invention modifies the rotor structure asymmetrically. Specifically, as follows... Figure 3 As shown, radial and tangential magnetic bridges are removed or thinned on one side of the magnetic circuit in the +d-axis direction of the rotor, while the corresponding radial and tangential magnetic bridges are retained on the opposite side, i.e., the -d-axis direction. This treatment increases the magnetic reluctance and decreases the equivalent inductance of the magnetic circuit in the +d-axis direction, while maintaining the original low magnetic reluctance and high inductance characteristics of the magnetic circuit in the -d-axis direction. This results in a stable, asymmetric magnetic circuit between the +d and -d axes that is independent of magnetic saturation.

[0022] In a preferred embodiment of the present invention, the magnetic bridge is removed from the +d-axis magnetic circuit and retained on the -d-axis magnetic circuit. However, those skilled in the art should understand that the opposite approach can also be used, i.e., the magnetic bridge is removed from the -d-axis magnetic circuit and retained on the +d-axis magnetic circuit, as long as there is a clear asymmetry in the magnetic circuits in both directions. Both approaches can achieve the technical effects of the present invention. In actual manufacturing, considering that the rotor experiences significant centrifugal force during high-speed rotation, removing or thinning the area after the magnetic bridge is removed may weaken the rotor's mechanical strength. Therefore, the present invention fills this area with a non-magnetic reinforcing material, such as epoxy resin or other polymer materials with good mechanical strength that do not participate in magnetic conduction. The main function of this non-magnetic reinforcing material is to support the magnetic barrier area, connect adjacent iron core segments, and prevent structural damage to the rotor during high-speed operation. Furthermore, since this material does not participate in magnetic conduction, its filling will not substantially affect the characteristics of the constructed asymmetric magnetic circuit.

[0023] The tap arrangement in the stator windings will be explained next. In the ferrite-assisted synchronous reluctance motor of this invention, the stator is wound with three phase windings: phase A, phase B, and phase C. Taking phase A winding as an example, phase B and phase C windings use the same tap structure. Let the number of poles of the motor be P. For a three-phase ferrite-assisted synchronous reluctance motor, P is usually 4 or 6. The phase A winding is naturally divided into P sequentially distributed equivalent coil groups along the stator circumference according to the magnetic pole period. Each equivalent coil group corresponds in physical position to a sequentially distributed magnetic pole region on the rotor. Adjacent equivalent coil groups are separated by a mechanical angle of 360° / P, which corresponds to an electrical angle of 180°. Since adjacent magnetic pole regions are +d-axis and -d-axis polarities respectively, the spatial positions of two adjacent equivalent coil groups correspond exactly to the +d-axis magnetic pole region and the -d-axis magnetic pole region, respectively. Based on this distribution pattern, in order to compare the electrical quantity differences caused by magnetic circuit differences between two winding segments under adjacent +d-axis and -d-axis magnetic pole regions, this invention selects two adjacent equivalent coil groups in the same phase winding as the detection segment. Specifically, regarding the setting of the tap position, starting from the beginning of the winding of that phase, a tap terminal is led out at the 1 / P node, which represents the total number of series turns of the winding, and another tap terminal is led out at the 2 / P node. In this way, the winding portion from the beginning of the winding to the 1 / P node constitutes the first winding segment, and the winding portion from the 1 / P node to the 2 / P node constitutes the second winding segment. The first winding segment and the second winding segment correspond exactly to the equivalent coil groups of two adjacent magnetic pole regions on the rotor.

[0024] Taking the most common 4-pole motor as an example, the A-phase winding is divided into four sequentially distributed equivalent coil groups along the circumferential direction according to the magnetic pole period, named coil 1, coil 2, coil 3, and coil 4, respectively. Their spatial positions correspond to adjacent +d, -d, +d, and -d magnetic pole regions, and adjacent coil groups are separated by a 90° mechanical angle. At this time, as... Figure 4 As shown, tap terminals are led out at the 1 / 4 node and 1 / 2 node of the total number of turns of phase A winding, respectively. The portion from the start of phase A winding to the 1 / 4 node is considered as the first winding segment. This corresponds to coil 1; the portion between the 1 / 4 node and the 1 / 2 node serves as the second winding segment. This corresponds to coil 2. Thus, and These are two adjacent equivalent coil groups, located near the +d-axis and -d-axis magnetic pole regions, respectively. For a 6-pole motor, taps are similarly placed at the 1 / 6 and 1 / 3 nodes of the total winding turns, forming the first and second winding segments, which also correspond to two adjacent magnetic pole regions on the rotor. It should be noted that this tapping method assumes that the number of turns and structure of each equivalent coil group are basically consistent. If the motor windings have parallel branches, special pitches, or non-uniform winding structures, the tap positions should not be mechanically determined solely by the wire length or turn ratio, but rather adjusted according to the actual correspondence between the equivalent coil groups and the spatial positions of the magnetic poles, ensuring that the first and second winding segments each correspond to a complete equivalent coil group and are spatially adjacent.

[0025] After the first and second winding segments are constructed, they are configured in the circuit to be connected in series. Specifically, one end of the first winding segment is the starting end of the phase winding, and the other end is a tap terminal at the 1 / P node; one end of the second winding segment is a tap terminal at the 1 / P node, and the other end is a tap terminal at the 2 / P node. Therefore, by using the starting end of the phase winding and the tap terminal at the 2 / P node as the two endpoints of the series branch, the first and second winding segments are connected in series through the common tap at the 1 / P node, allowing the same current to flow through them. This series connection can be achieved externally to the motor through circuit switching, or the corresponding tap terminals can be pre-connected to the detection circuit, and the relay or semiconductor switch inside the detection circuit can establish a series circuit in detection mode. Regardless of the specific connection method used, it is a direct application of the technical solution of this invention.

[0026] The specific process of rotor polarity identification using the above structure is described in detail below. During polarity identification, the motor is in a stationary state or rotating at extremely low speed. At this time, the motor's main driver has not yet applied driving voltage, or has only applied a low-amplitude voltage signal for initial position detection. For example... Figure 5As shown, the first and second winding segments connected in series are treated as a single branch, and alternating positive and negative square wave voltage excitations are applied to both ends of this single branch. This square wave voltage excitation can be generated by an independent external AC power supply or directly by the inverter bridge circuit of the motor driver. When using an inverter bridge to generate the square wave excitation, no additional excitation power supply is required, which simplifies the system structure and reduces costs. The amplitude and frequency of the square wave voltage excitation need to be selected and adjusted according to the specific motor winding parameters, magnetic circuit structure, and bus voltage. In one specific embodiment, the amplitude of the square wave voltage excitation is ±40V, and the frequency is 1.5kHz. The amplitude selection should ensure that a sufficiently large detection current is generated in the winding to obtain a measurable voltage difference signal, but it should not be too large to cause significant torque disturbances or the risk of demagnetization of the ferrite permanent magnet. The frequency selection should consider avoiding the mechanical resonant frequency of the motor and the bandwidth limitations of the current control loop, while ensuring that multiple voltage samples can be completed within one square wave cycle. For motors with different power ratings and parameters, the above amplitude and frequency can be adjusted appropriately according to the actual situation. For example, the amplitude can be selected in the range of ±20V to ±60V, and the frequency can be selected in the range of 500Hz to 5kHz. These adjustments do not depart from the protection scope of this invention.

[0027] After applying a square wave voltage excitation, the first and second winding segments, connected in series, flow with the same injected current. According to the basic principle of series circuits, under the same current conditions, the voltage distribution on the two winding segments is proportional to their respective equivalent impedances. Since the two winding segments have essentially the same resistance value, their impedance difference is mainly determined by the difference in equivalent inductance. However, as mentioned earlier, due to the asymmetry of the rotor's magnetic circuit structure in the +d and -d axis directions, when the +d and -d axes are close to the spatial positions of the first and second winding segments respectively, the equivalent magnetic circuits corresponding to the two winding segments exhibit different reluctances, resulting in a difference in their equivalent inductances. This inductance difference directly manifests as a difference in voltage amplitude on the two winding segments under series AC excitation. This physical relationship forms the basis for the polarity determination of this invention. Specifically, after the square wave voltage excitation is applied to both ends of the series branch, the voltage across the first winding segment is collected and recorded as the first voltage. The voltage across the second winding section is recorded as the second voltage. Voltage acquisition can be achieved using a resistive voltage divider sampling circuit, an isolated voltage sensor, or other conventional voltage detection methods. The acquired analog voltage signal is converted into a digital signal by an analog-to-digital converter and then sent to a microcontroller for processing.

[0028] Regarding the timing of voltage signal sampling, since the square wave voltage excitation is generated by power electronic switching devices (such as IGBTs or MOSFETs in an inverter bridge), large voltage spikes and oscillations are generated in the circuit at the instants of switching devices turning on and off. This transient process typically lasts from several microseconds to tens of microseconds. If voltage sampling is performed within this transient interval, the obtained voltage value will deviate significantly from the quasi-steady-state value, thus introducing a large measurement error. Therefore, this invention avoids the switching transient interval within each square wave voltage excitation cycle, sampling during the quasi-steady-state period after the switching transient ends (e.g., ...). Figure 6 Voltage sampling is performed on the portion indicated by the blue box shown. This quasi-steady-state time period is referred to as the stable plateau interval in this invention. Within this interval, the oscillations in the circuit have essentially decayed, and the voltage waveform enters a relatively stable phase. Within one square wave half-cycle, multiple samplings can typically be performed within the stable plateau interval, and the arithmetic mean of the sampled voltage values ​​is calculated to eliminate the influence of random noise and residual small-amplitude oscillations on the measurement results. Specifically, for the first winding segment, multiple first voltage samples are collected within the stable plateau interval, and their arithmetic mean is calculated as the first representative voltage value within this half-cycle; for the second winding segment, the second representative voltage value is obtained in the same way. Then, the difference between the first and second representative voltage values ​​is calculated to obtain the voltage difference within this half-cycle. More preferably, the voltage difference within multiple consecutive square wave half-cycles can be further averaged to further improve the measurement stability and anti-interference capability.

[0029] After obtaining the average voltage difference, it needs to be compared with a preset threshold to determine polarity. This invention employs a dual-threshold decision mechanism, namely a preset first threshold and a second threshold, where the first threshold is less than the second threshold. These two thresholds cannot be arbitrarily set but should be obtained in advance based on the electromagnetic characteristics of the motor itself through finite element simulation or actual prototype calibration. The finite element simulation method involves establishing a motor model in electromagnetic simulation software that includes the aforementioned asymmetric rotor structure and tapped winding structure. The inductance values ​​of the two winding segments are calculated separately when the rotor is in different positions, thereby simulating a complete curve of voltage difference versus rotor position under square wave excitation. The prototype calibration method involves manually rotating the rotor to different positions on an actual manufactured prototype, applying the same square wave excitation, and measuring the voltage difference to obtain the measured voltage difference-position curve. Regardless of the method used, the obtained voltage difference-position curve exhibits a periodic variation with the rotor position. On this curve, there are some special positions where the voltage difference is exactly zero or close to zero. These positions correspond to the transition region between adjacent +d and -d axes, which is the position of the rotor's q-axis direction.

[0030] In the 4-pole motor embodiment of the present invention, the +d and -d axes are alternately distributed on the rotor circumference, with adjacent +d and -d axes separated by a 90° mechanical angle. The theoretical boundary region is located in the q-axis direction, i.e., positions with mechanical angles of 45° and 135° relative to the +d axis. At these two positions, the distances of the first and second winding segments from the +d and -d axes are equal, and the equivalent inductances of the two winding segments are essentially the same, therefore the voltage difference is zero or minimal. Based on this physical relationship, the present invention sets the voltage difference values ​​corresponding to the two zero-crossing points on the voltage difference-position curve as the first threshold and the second threshold, respectively. Specifically, in the 4-pole motor, the first threshold corresponds to the voltage difference value when the rotor is located at a mechanical angle of 45° relative to the +d axis, and the second threshold corresponds to the voltage difference value when the rotor is located at a mechanical angle of 135° relative to the +d axis. For the general case of P poles, the mechanical angular interval between adjacent +d and -d axes is 360° / P. The first theoretical boundary position is located at a mechanical angle of 180° / P relative to the +d axis, and the second theoretical boundary position is located at a mechanical angle of 540° / P relative to the +d axis. It should be noted that due to factors such as core saturation, uneven magnetic field distribution in permanent magnets, and manufacturing tolerances in actual motors, the actual zero-crossing position may deviate from the theoretically calculated position. Therefore, in practical engineering applications, the theoretically calculated position should be used as an initial reference, and then corrected using measured data from finite element simulation or prototype calibration to obtain a more accurate and reliable actual threshold.

[0031] In the specific execution of the dual-threshold decision mechanism, the average voltage difference calculated in step S2 is compared with the first threshold and the second threshold. When the average voltage difference is less than the first threshold, it indicates that the rotor position is on one side of the boundary position corresponding to the first threshold, which corresponds to the +d-axis polarity region. Therefore, the current d-axis polarity of the rotor is determined to be +d-axis. When the average voltage difference is greater than the second threshold, it indicates that the rotor position is on the other side of the boundary position corresponding to the second threshold, which corresponds to the -d-axis polarity region. Therefore, the current d-axis polarity of the rotor is determined to be -d-axis. When the average voltage difference is between the first and second thresholds, it indicates that the rotor position is exactly in the transition region between the two boundary positions, i.e., near the q-axis. In this region, the inductance difference between the two winding segments is too small, and the amplitude of the voltage difference signal is insufficient to reliably distinguish the polarity. Therefore, it is determined that the current polarity identification blind zone is in which no polarity judgment result is output. The determination of the polarity identification blind zone is of great significance for the safety of the motor drive system, because incorrect polarity judgment may cause the motor to reverse or generate abnormal torque during startup, thereby causing equipment failure or even safety accidents.

[0032] In the aforementioned single-phase detection scheme, although polarity identification can be achieved at most rotor positions, reliable judgment results cannot be output in the blind zone near the q-axis. To solve this problem, this invention further proposes a three-phase extended detection scheme. Since the ferrite-assisted synchronous reluctance motor is a three-phase motor, it includes A-phase, B-phase, and C-phase windings, which are spatially spaced at 120° electrical angles on the stator circumference. Therefore, the same tap structure can be used in the A-phase, B-phase, and C-phase windings; that is, each phase winding has tap terminals at the 1 / P and 2 / P nodes of its total series turns, forming its own first winding segment and second winding segment respectively. During polarity identification, steps S1 to S4 are sequentially executed on the A-phase, B-phase, and C-phase windings, i.e., square wave voltage excitation is applied sequentially to the first and second winding segments connected in series for each phase, and the voltages of the two winding segments of each phase are collected and the average voltage difference is calculated. Three average voltage difference values ​​are thus obtained, denoted as... , and Since the three-phase windings are spatially 120° electrically out of phase, corresponding to a 60° mechanical angle in a 4-pole motor, the voltage difference-rotor position curves of phases B and C are basically the same as the curve of phase A, but lag behind and lead phase A by 60° mechanical angles, respectively. Utilizing this spatial phase relationship, when one phase (e.g., phase A) is in its own dead zone, the other two phases (phases B and C) are usually not simultaneously in their dead zones, but rather fall within a clearly identifiable region. Therefore, the detection results of the other two phases can be used to compensate for the dead zone.

[0033] Specifically, let's take phase A as an example to explain the compensation decision logic. First, based on the average voltage difference of phase A... Determine whether it is in the blind spot, i.e. Whether it falls between the first and second thresholds. If phase A is not in the dead zone, the rotor polarity is output directly based on the judgment result of phase A. If phase A is in the dead zone, the detection results of phases B and C are checked: when the average voltage difference of phase B is detected... Less than the first threshold, and at the same time, the average voltage difference of phase C When the value exceeds the second threshold, the rotor polarity can be uniquely determined to correspond to a certain side of the A-phase dead zone based on the spatial relationship of the three phases, specifically the -d axis; when the average voltage difference of phase B is detected... Greater than the second threshold, and simultaneously the average voltage difference of phase C If the value is less than the first threshold, the rotor polarity is determined to be the +d axis. Through the aforementioned three-phase compensation mechanism, the impact of the single-phase detection blind zone on polarity identification can be significantly reduced, achieving reliable polarity identification over a wider range, even across the entire rotor position.

[0034] In the specific implementation of sampling and calculation, voltage acquisition and average difference calculation can be accomplished by the timer and analog-to-digital converter inside the microcontroller, or by a field-programmable gate array (FPGA) or other digital signal processing chips. Alternatively, analog circuits can be used to directly perform differential amplification and average value extraction of the two voltages before sending them to the FPGA. All of these implementation methods are conventional choices for those skilled in the art. When calculating the average voltage difference, the averaging operation used can be a simple arithmetic average within a stable plateau range of a square wave half-cycle, or a moving average or exponentially weighted average over multiple consecutive square wave half-cycles. The choice of different averaging methods mainly depends on the signal-to-noise ratio of the actual signal and the controller's response speed requirements; these do not affect the basic implementation of this invention.

[0035] In practical engineering applications, users can perform a one-time threshold calibration operation during the initial installation of the motor. During the calibration process, the motor rotor is slowly rotated one revolution via an external device or manually. The detection system automatically records the voltage difference at each rotor position and automatically calculates the threshold, then writes the threshold parameters into the controller's non-volatile memory. During subsequent normal operation, a polarity identification program is automatically executed before each motor start-up, obtaining accurate rotor polarity information without the need for rotor position sensors. This provides a reliable initial magnetic pole direction for subsequent sensorless control algorithms or initial position positioning algorithms. The polarity identification result can be further used to optimize the motor's start-up control strategy, ensuring that the motor starts with the correct torque direction when stationary or at low speed, avoiding start-up failures or reverse rotation caused by incorrect polarity judgment, thereby improving the start-up reliability and operational safety of the entire motor drive system.

[0036] In summary, this invention constructs an asymmetric magnetic circuit along the +d and -d axes of the rotor, causing the two winding segments to exhibit different equivalent inductances under the same injected current. The inductance difference is then converted into a measurable voltage difference signal using winding taps. Compared to existing methods that rely on magnetic saturation to determine polarity, this invention does not require applying a large DC current or voltage pulse to saturate the motor's magnetic circuit. Therefore, it effectively reduces current surges and torque disturbances, while also minimizing the risk of demagnetization of ferrite permanent magnets under high-current excitation. This invention detects the average voltage difference between the two series-connected winding segments under square wave excitation and determines polarity based on a preset threshold. The detection signal is the average DC voltage, eliminating the need for complex current demodulation or high-frequency signal processing. The detection process is direct, computationally efficient, and easily implemented in conventional motor controllers.

[0037] By establishing a detection section through taps at specific nodes of the phase windings, this invention uses the tap structure only for polarity identification during startup or low-speed operation. It does not affect the winding connections and operating mode during normal three-phase motor operation, exhibiting good engineering compatibility. Furthermore, this invention utilizes the 120° electrical angle difference between the three-phase windings in space to sequentially detect each phase. When a phase is in the identification blind zone near the zero-crossing point of the voltage difference, compensation can be made using the detection results of the other two phases, effectively eliminating single-phase detection blind zones and achieving polarity identification across the entire rotor position range. This invention also provides reliable rotor polarity information for sensorless control or initial position identification, helping to avoid startup failures, reverse rotation, or current surges caused by incorrect polarity determination.

[0038] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0039] Furthermore, in this invention, descriptions involving terms such as "first," "second," and "a" are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0040] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0041] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

Claims

1. A method for rotor polarity identification of a ferrite-assisted synchronous reluctance motor, characterized in that, The rotor of the motor has an asymmetric magnetic circuit structure in the +d-axis and -d-axis directions, and at least one phase of the stator winding of the motor is provided with a tap to divide the phase winding into a first winding and a second winding with different spatial positions. The first winding segment and the second winding segment are configured to be connected in series. The rotor polarity identification method includes the following steps: S1: When the motor is stationary or at low speed, the first winding segment and the second winding segment connected in series are treated as a whole branch, and alternating positive and negative square wave voltage excitation is applied to both ends of the whole branch. S2: In each positive half-cycle of the square wave voltage excitation, the first voltage at both ends of the first winding segment and the second voltage at both ends of the second winding segment are collected in the stable plateau interval, and the average voltage difference in the stable plateau interval is calculated. S3: Compare the average voltage difference with a preset first threshold and a second threshold, wherein the first threshold is less than the second threshold; S4: If the average voltage difference is less than the first threshold, the current d-axis polarity of the rotor is determined to be +d-axis; if the average voltage difference is greater than the second threshold, the current d-axis polarity of the rotor is determined to be -d-axis; if the average voltage difference is between the first threshold and the second threshold, it is determined that the rotor is currently in the polarity recognition blind zone and no polarity judgment result is output.

2. The rotor polarity identification method for a ferrite-assisted synchronous reluctance motor as described in claim 1, characterized in that, The asymmetric magnetic circuit structure is achieved by removing or thinning the radial and tangential magnetic bridges on the +d axis magnetic circuit of the rotor, while retaining the radial and tangential magnetic bridges on the -d axis magnetic circuit.

3. The rotor polarity identification method for a ferrite-assisted synchronous reluctance motor as described in claim 2, characterized in that, The areas after the radial and tangential magnetic bridges are removed or thinned are filled with non-magnetic reinforcing material.

4. The rotor polarity identification method for a ferrite-assisted synchronous reluctance motor as described in claim 1, characterized in that, The motor has P poles. The single-phase winding is divided into P sequentially distributed equivalent coil groups along the circumferential direction according to the magnetic pole period. Adjacent equivalent coil groups are separated by a mechanical angle of 360° / P. With the number of turns and structure of each equivalent coil group being consistent, the tap is located at the 1 / P node and the 2 / P node of the total number of series turns of the phase winding, so that the first winding segment corresponds to the winding portion from the starting end of the winding to the 1 / P node, and the second winding segment corresponds to the winding portion from the 1 / P node to the 2 / P node. The first winding segment and the second winding segment respectively correspond to the equivalent coil groups of two adjacent magnetic pole regions on the rotor.

5. The rotor polarity identification method for a ferrite-assisted synchronous reluctance motor as described in claim 1, characterized in that, The motor is a three-phase motor, including phase A, phase B and phase C. The rotor polarity identification method performs steps S1 to S4 on the windings of phase A, phase B and phase C respectively to obtain the average voltage difference of each phase. When a phase is in the polarity identification blind zone, the current d-axis polarity of the rotor is determined according to the comparison results of the average voltage difference of the other two phases with the first threshold and the second threshold, based on the preset decision logic.

6. The rotor polarity identification method for a ferrite-assisted synchronous reluctance motor as described in claim 5, characterized in that, The preset decision logic is as follows: when phase A is in the polarity identification blind zone and the average voltage difference of phase B is less than the first threshold and the average voltage difference of phase C is greater than the second threshold, the current d-axis polarity of the rotor is determined to be -d-axis; when phase A is in the polarity identification blind zone and the average voltage difference of phase B is greater than the second threshold and the average voltage difference of phase C is less than the first threshold, the current d-axis polarity of the rotor is determined to be +d-axis.

7. The rotor polarity identification method for a ferrite-assisted synchronous reluctance motor as described in claim 1, characterized in that, The first threshold and the second threshold are obtained in advance through finite element simulation or prototype calibration. The first threshold corresponds to the average voltage difference when the rotor is located at the first theoretical boundary position between adjacent +d and -d axes, and the second threshold corresponds to the average voltage difference when the rotor is located at the second theoretical boundary position between adjacent +d and -d axes. The first theoretical boundary position and the second theoretical boundary position are located in the +q and -q axis directions of the rotor, respectively.

8. The rotor polarity identification method for a ferrite-assisted synchronous reluctance motor as described in claim 7, characterized in that, The motor has P poles, and the mechanical angle interval between adjacent +d and -d axes is 360° / P. The first theoretical dividing position is located at a mechanical angle of 180° / P relative to the +d axis, and the second theoretical dividing position is located at a mechanical angle of 540° / P relative to the +d axis.

9. The rotor polarity identification method for a ferrite-assisted synchronous reluctance motor as described in claim 1, characterized in that, The stable platform interval is the quasi-steady-state time period after the end of the switching transient in each positive half-cycle of the square wave voltage excitation. The first voltage and the second voltage are sampled multiple times and the arithmetic mean is calculated during the quasi-steady-state time period.