Permanent magnet synchronous motor dead-time effect compensation method and permanent magnet synchronous motor

Abstract

A permanent magnet synchronous motor dead-time effect compensation method and a permanent magnet synchronous motor are disclosed. The method comprises: obtaining the current sector where a synthesized current vector is located and the flow direction of ABC three-phase currents; compensating the three-phase opening time of ABC three-phase currents respectively according to the flow direction of ABC three-phase currents respectively in the current sector, so that the effect in the direction of the compensated voltage vector corresponding to the current sector is increased by a predetermined time; and when the synthesized current vector enters a near-zero sector where the zero-crossing point of ABC three-phase currents is located, reducing the compensation of the three-phase opening time by a compensation gain k. The disclosure realizes compensation by inserting a relatively fixed dead-time at a suitable position to reduce the output voltage error and other effects caused by the dead-time effect, and reduces the above compensation near the zero-crossing point of ABC three-phase currents to avoid compensation errors caused by repeated zero-crossing of the current near the zero-crossing point. Further, the loss of the above fixed-time compensation can also be compensated by hardware feedforward compensation.

Description

Technical Field

[0001] This disclosure relates to the field of electric motors, and more particularly to a method for compensating for the dead zone effect of a permanent magnet synchronous motor and the permanent magnet synchronous motor itself. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in new energy vehicles, robot servos, and home appliances due to their advantages such as low noise and high energy efficiency. In order to make the PMSM rotor rotate continuously and smoothly, FOC (Field-Oriented Control) is required to control the motor.

[0003] In FOC-based PMSM speed control systems, the application of pulse width modulation (PWM) technology enables real-time control of the dynamic torque of the servo motor, significantly improving the dynamic performance of the servo system. To apply PWM, an inverter is needed to convert DC power to AC power. In the three-phase inverters used in FOC systems, the switching devices need to operate at a high switching frequency to output an ideal current waveform. While ideally, the two switching devices (usually power transistors) on the upper and lower arms of the same phase in a three-phase inverter system operate in a complementary state, in practical applications, due to the charge storage effect, there is a certain delay in the switching of the power transistors, especially the turn-off time t. off Comparison of conduction time t on Therefore, during the commutation process of the two switching devices in the same bridge arm, in order to prevent the two power transistors from shooting through and burning out, a certain dead time (t) needs to be added between the turn-on and turn-off times of the two power transistors. d ).

[0004] Dead time and the nonlinear characteristics of power devices are collectively referred to as the dead-time effect. While the output voltage error caused by a single cycle is small, the cumulative effect due to the high switching frequency cannot be ignored. The dead-time effect causes stator current distortion and torque ripple, and the higher the switching frequency, the more pronounced the impact of the dead-time effect, potentially leading to severe zero-point clamping.

[0005] Therefore, a solution is needed that can effectively compensate for the dead zone effect. Summary of the Invention

[0006] One technical problem this disclosure aims to solve is to provide a dead-time effect compensation scheme for permanent magnet synchronous motors. This scheme achieves compensation by inserting a relatively fixed dead time at an appropriate position (determined by the direction of the A, B, and C phase currents determined by the sector where the synthesized current vector is located) to reduce output voltage errors and other effects caused by the dead-time effect. Furthermore, the compensation is reduced near the zero-crossing points of the A, B, and C phase currents to avoid compensation errors caused by repeated zero-crossings of the current near these points. Additionally, hardware feedforward compensation can be used to compensate for the losses from the fixed-time compensation mentioned above.

[0007] According to a first aspect of this disclosure, a method for compensating for dead-zone effect in a permanent magnet synchronous motor is provided, comprising: obtaining the current flow direction of the three phases A, B, and C in the current sector where the composite current vector is located; compensating the three-phase turn-on time of each of the three phases A, B, and C according to the current flow direction of each of the three phases A, B, and C in the current sector so that the effect of the compensation voltage vector in the direction corresponding to the current sector is increased by a predetermined time; and reducing the compensation of the three-phase turn-on time of each of the three phases A, B, and C by a compensation gain k when the composite current vector enters the near-zero sector where the zero-crossing point of the three phases A, B, and C current is located.

[0008] Optionally, compensating for the three-phase turn-on time of each of the three phases A, B, and C in the current sector according to their respective current flow directions includes: extending the conduction time of the upper bridge power transistor of the phase with positive current flow by t. d The conduction time t of the lower arm bridge power transistor with the current flowing in the positive direction is shortened. d ; and shorten the conduction time t of the upper bridge power transistor that directs current to the negative phase. d Extend the conduction time t of the lower arm bridge power transistor that directs current to the negative phase. d .

[0009] Optionally, the power transistor's on-time can be extended by t. d This includes: delaying the turn-off time of the power transistor by t. d / 2, and advance the conduction time by t. d / 2; shorten the power transistor's on-time by t d This includes: advancing the turn-off time of the power transistor by t. d / 2, and postpone the conduction time by t d / 2.

[0010] Optionally, the near-zero sector includes: a predetermined angle range of -Δd to Δd above and below each phase current zero-crossing point, where Δd is a predetermined positive angle value.

[0011] Optionally, the compensation gain k is a function that varies with the angle of the synthesized current vector between [predetermined angle -Δd to predetermined angle +Δd] and increases between [-1, 1].

[0012] Optionally, the compensation gain k is 0 at each zero-crossing point of the angle of the synthesized current vector.

[0013] Optionally, when the synthesized current vector enters the near-zero sector where the zero-crossing point of the three-phase currents ABC are located, the compensation gain k is used to reduce the three-phase turn-on time of each of the three phases ABC, which includes: when entering the near-zero sector where the zero-crossing point of the A-phase current is located, multiplying the compensation gain k by the compensation amounts of each of the A-phase, B-phase, and C-phase; when entering the near-zero sector where the zero-crossing point of the B-phase current is located, multiplying the compensation gain k by the compensation amounts of each of the A-phase, B-phase, and C-phase; and when entering the near-zero sector where the zero-crossing point of the C-phase current is located, multiplying the compensation gain k by the compensation amounts of each of the A-phase, B-phase, and C-phase.

[0014] Optionally, obtaining the current sector where the composite current vector is located includes: calculating the d-axis current and q-axis current in a two-phase rotating coordinate system; calculating the composite current vector based on the d-axis current and the q-axis current; calculating the spatial angle between the composite current vector and the α-axis in a two-phase stationary coordinate system; and determining the current sector in the current vector circle based on the spatial angle.

[0015] Optionally, the method further includes: calculating the d-axis compensation voltage and q-axis compensation voltage for the current cycle based on the d-axis current, q-axis voltage, d-axis voltage, q-axis speed and rotor speed of the previous cycle; and performing hardware compensation on the d-axis voltage and q-axis voltage for the current cycle using the d-axis compensation voltage and q-axis compensation voltage for the current cycle.

[0016] Optionally, calculating the d-axis compensation voltage and q-axis compensation voltage for the current cycle based on the collected d-axis current, q-axis current, d-axis voltage, q-axis voltage, and rotor speed of the previous cycle includes: determining the d-axis voltage error value between the collected d-axis voltage of the previous cycle and the theoretical d-axis voltage calculated based on the d-axis current and q-axis current of the previous cycle, and using this value as the d-axis compensation voltage for the current cycle; and determining the q-axis voltage error value between the collected q-axis voltage of the previous cycle and the theoretical q-axis voltage calculated based on the d-axis current and q-axis current of the previous cycle, and using this value as the q-axis compensation voltage for the current cycle.

[0017] According to a second aspect of this disclosure, a permanent magnet synchronous motor is provided, comprising: a motor; and a vector control circuit, the vector control circuit comprising: a software time compensation module located between SVPWM and a three-phase inverter bridge, for performing the method according to the first aspect of the present invention to compensate for the three-phase turn-on time of each of the three phases ABC.

[0018] Optionally, the vector control circuit further includes a hardware feedforward compensation module, used to calculate the d-axis compensation voltage and q-axis compensation voltage for the current cycle based on the d-axis current, q-axis current, d-axis voltage, q-axis voltage and rotor speed of the previous cycle, so as to perform hardware compensation on the d-axis voltage and q-axis voltage for the current cycle.

[0019] Therefore, the software-inserted fixed dead time scheme of the present invention can provide fixed compensation for the dead time, and avoid the problems caused by the current zero crossing by introducing a compensation gain k to reduce the fixed compensation near the zero crossing point. Furthermore, the software fixed dead time compensation of the present invention can be supplemented by a novel hardware feedforward dead time compensation scheme. This scheme can dynamically compensate the dq-axis voltage by real-time reading of the dq-axis voltage, current, and speed. The above-mentioned hardware-assisted compensation scheme does not require additional hardware, has a small computational load, and when combined with the software fixed dead time compensation method, can greatly reduce current harmonics and improve motor control performance. Attached Figure Description

[0020] The above and other objects, features and advantages of this disclosure will become more apparent from the more detailed description of exemplary embodiments thereof taken in conjunction with the accompanying drawings, wherein like reference numerals generally denote like parts.

[0021] Figure 1 The FOC control principle diagram for PMSM is shown.

[0022] Figure 2 The main structure diagram of a three-phase PWM inverter PMSM load is shown.

[0023] Figure 3A -B shows a schematic diagram of the current freewheeling direction before and after entering the dead time.

[0024] Figure 4 The timing diagram of the drive waveform of phase A bridge arm and an example of the output voltage are shown.

[0025] Figure 5 A schematic flowchart of a method for compensating for the dead-zone effect of a permanent magnet synchronous motor according to an embodiment of the present invention is shown.

[0026] Figure 6 It shows i d and iq The relationship between the currents is used to calculate the current synthesis vector i. s One example.

[0027] Figure 7 A diagram for determining current polarity based on a vector circle is shown.

[0028] Figure 8 The dead time compensation waveform is shown within the 60° to 90° range of sector II.

[0029] Figure 9 The dead time compensation waveform is shown within the 90° to 120° range of sector II.

[0030] Figure 10 An example of a compensation gain function is shown.

[0031] Figure 11 An example of the voltage output error waveform diagram of the dead zone effect in phase A is shown.

[0032] Figure 12 A schematic diagram of the composition of a permanent magnet synchronous motor according to an embodiment of the present invention is shown. Detailed Implementation

[0033] Preferred embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0034] To ensure that the PMSM rotor rotates continuously and smoothly, FOC (Field-Oriented Control) is required to control the motor.

[0035] Figure 1 The FOC control principle diagram for PMSM is shown.

[0036] As shown in the figure, a position sensor, such as a magnetic encoder, acquires the rotor speed n and rotor position θ of motor M. In practical applications, for example, a speed reference n obtained from user input... ref The difference between the measured rotor speed n and the actual rotor speed n is input to PID1 (i.e., the speed loop PID). The output of the speed loop PID is the q-axis reference current i. qref For simplicity, without additional control, the d-axis reference current i can be set. dref =0. At this point, the reference currents i on the q-axis and d-axis can be set to 0. qref and i drefThe actual feedback from the motor in the two-phase rotating coordinate system shows the q-axis and d-axis currents i. q and i d (This can also be referred to as the q-axis feedback current i) q and d-axis feedback current i d The difference is calculated, and after being regulated by PID2 (i.e., q-axis current loop) and PID3 (i.e., d-axis current loop) respectively, the output q-axis and d-axis voltages U are obtained. d and U q Then, through the inverse Park transformation, it is converted into the α-axis and β-axis voltages U in a two-phase stationary coordinate system. α and U β Then, the three-phase voltage U is obtained through SVPWM (Space Vector Modulation). a U b U c The motor M is driven to rotate via a three-phase inverter bridge. In FOC control, the three-phase current i is collected. a i b i c After Clark transformation, the α-axis and β-axis currents i in the two-phase stationary coordinate system are obtained. α and i β Then, after Park transformation, the q-axis and d-axis currents i in the two-phase rotating coordinate system are obtained. d and i q The inputs are used in the current loop PID2 and PID3 control, respectively. The rotor position θ obtained by the position sensor is used in the Park transform and inverse Park transform.

[0037] In a FOC-based PMSM speed control system, an inverter is needed to convert DC power into AC power in order to apply PWM. Figure 2 The main structure diagram of a three-phase PWM inverter (PMSM) load is shown. Power transistors VT1 and VT4 are the upper and lower arm switches of phase A, respectively. D1 and D4 are used to control the voltage i during the dead time. a The freewheeling diode; power transistors VT2 and VT5 are the upper and lower arm switching transistors of phase B, respectively (here, "switching transistor" can be a "switching device" implemented by a "power transistor", and can also be referred to as "power transistor", "power device" or "switching device"). D2 and D5 are used to provide i during the dead time. b The freewheeling diode; power transistors VT3 and VT6 are the upper and lower arm switching transistors of phase C, respectively; D3 and D6 are used to provide i during the dead time. c A freewheeling diode.

[0038] Figure 3A-B illustrates the current freewheeling direction before and after entering the dead time. As shown in the figure, taking phase A as an example, the flow into the motor is considered positive, and the flow out is considered negative. When the power transistor switches states, the polarity of the current changes, but because the motor is an inductive load, the current does not change abruptly and freewheels through the anti-parallel diode of the power device. Therefore, when i a When >0, such as Figure 3A As shown, VT1 changes from the on state to the off state, and the current changes from flowing through VT1 on the left side of the diagram to freewheeling through D4. And when i a When <0, such as Figure 3B As shown, VT4 changes from the on state to the off state, and the current changes from flowing through VT4 on the left side of the diagram to freewheeling through D1. During the dead time, the output voltage of phase A will not be controlled by the power devices, but will be determined by the polarity of the phase current.

[0039] For ease of understanding, Figure 4 The timing diagram of the drive waveform of phase A bridge arm and an example of the output voltage are shown. As shown in the figure, the output terminal voltage waveforms under ideal and actual conditions are given respectively. Specifically, (a) shows the ideal switching transistor control signal, where S... AU and S AL These represent the switching control signals for the upper and lower arms of phase A, respectively. As shown in the figure, under ideal conditions, the two power transistors in the upper and lower arms of the same phase are in a completely complementary state. (b) shows the addition of dead time t. d The switching control signal of the transistor. During the transition process where the upper bridge arm power transistor turns on and the lower bridge arm power transistor turns off, to prevent short circuits, the turn-on of the upper bridge arm power transistor is delayed by a time t compared to the turn-off of the lower bridge arm power transistor. d Similarly, during the transition from the upper arm power transistor to the lower arm power transistor to the upper arm power transistor, to prevent short circuits, the turn-on of the lower arm power transistor is delayed by a time t compared to the turn-off of the upper arm power transistor. d (b) The switching control signal shown has deviated from the ideal state.

[0040] Furthermore, in practical applications, factors such as turn-on and turn-off delays also need to be considered. As shown in (c), since the switching transistor requires time t to turn on... on Closing takes time t off This causes the actual switching control signal to deviate further from the ideal signal shown in (a). (d) and (e) respectively show i a >0 and i a The output voltage of phase A when <0.

[0041] As shown in (d), i a When the value is greater than 0, the output voltage of phase A begins to be delayed by t. on +td (This can be considered as part of the voltage pulse width adjustment loss), the output of phase A's output voltage is delayed by t. off (This can be seen as an increase in voltage pulse width adjustment). Accordingly, as shown in (e), i a When < 0, the output voltage of phase A starts to be delayed by t. off (This can be considered as part of the voltage pulse width adjustment loss), the output of phase A's output voltage is delayed by t. on +t d (This can be seen as the increase in voltage pulse width adjustment).

[0042] As can be seen from the above, in practical applications, power devices are difficult to operate in the ideal region. In addition to dead time, a certain amount of turn-on and turn-off time is also required, following the principle of turn-off before turn-on. After adding dead time, the output of phase A will also change accordingly. When i a When >0, such as Figure 4 As shown in (d) in the figure, the conduction time error is:

[0043] t err =(-t) d -t on +t off (1-1)

[0044] when i a When <0, as shown in Figure 3(e), the conduction time error is:

[0045] t err =(t d +t on -t off (1-2)

[0046] From the above two equations, we can deduce that the conduction time error is:

[0047] t err =sign(i x )(t off -t on -t d ) = sign(i x M (1-3)

[0048] in,

[0049] M=(t off -t on -t d (1-2)

[0050]

[0051] In summary, the actual conduction time T of phase j (j = a, b, c) j Equal to expected conduction time With conduction time error t err The sum is:

[0052]

[0053] Therefore, compensation is needed to address the time error described above to ensure real-time control of motor torque based on PWM. To this end, this invention proposes a dead-time effect compensation scheme for permanent magnet synchronous motors. This scheme first compensates for the dead-time effect by inserting a relatively fixed dead-time at an appropriate location (determined by the direction of the A, B, and C phase currents determined by the sector where the synthesized current vector is located), thereby reducing the output voltage error and other effects caused by the dead-time effect. This fixed-time compensation method is a software dead-time compensation method, which can be implemented by adding fixed-time dead-time compensation to the control program. By adjusting the duration of the actual PWM pulse, the deviation between the given voltage and the actual output voltage is corrected. Before pulse distortion occurs, a dead-time compensation time is added to that phase, making the compensated PWM pulse closer to the ideal pulse in terms of phase and width.

[0054] Taking phase A as an example, when the phase A current i a When the voltage is greater than 0, due to the existence of the dead zone, the voltage output is reduced by a dead zone duration t. d The compensation method is to extend the ideal commissioning time of the upper arm of phase A by t. d Considering the complementary effect, the corresponding A-phase lower arm action time is shortened by t. d This ensures that the actual waveform during activation matches the ideal input waveform.

[0055] when i a When the voltage is less than 0, due to the existence of the dead zone, the voltage output has an additional dead zone duration. The compensation method is to shorten the ideal turn-on time of the upper arm of phase A by t. d Considering the complementary effect, the ideal opening time of the lower arm of phase A is extended by td.

[0056] Phase B or Phase C is also based on the same principle, according to the direction of current flow in Phase B (i b >0 or i b <0) or the direction of C-phase current (i c >0 or i c <0), to compensate for the activation time of each of the three phases.

[0057] Figure 5 A schematic flowchart of a method for compensating for the dead-zone effect of a permanent magnet synchronous motor according to an embodiment of the present invention is shown.

[0058] In step S510, the current flow direction (hereinafter also referred to as polarity) of the three phases A, B, and C in the current sector where the synthesized current vector is located is obtained. Therefore, in step S520, the turn-on time of each of the three phases A, B, and C can be compensated according to their respective current flow directions in the current sector, so that the effect in the direction of the compensated voltage vector corresponding to the current sector is increased by a predetermined time. The increased predetermined time can be a relatively fixed duration, for example, 2t. d The scheduled time.

[0059] Therefore, compensation based on the above principles can be achieved according to the current flow direction of each phase A, B, and C.

[0060] Specifically, compensating for the turn-on time of each of the three phases (A, B, and C) within the current sector, based on their respective current flow direction (or polarity), may include: extending the conduction time of the upper bridge power transistor of the phase with positive current flow by t. d The conduction time t of the lower arm bridge power transistor with the current flowing in the positive direction is shortened. d ; and shorten the conduction time t of the upper bridge power transistor that directs current to the negative phase. d Extend the conduction time t of the lower arm bridge power transistor that directs current to the negative phase. d In one embodiment, the power transistor's on-time is extended by t. d This includes: delaying the power transistor's turn-off time by t d / 2, and advance the conduction time by t. d / 2; shorten the power transistor's on-time by t d This includes: advancing the power transistor's turn-off time by t. d / 2, and postpone the conduction time by t d / 2.

[0061] Due to the zero-current clamping caused by the dead-zone effect, the actual current typically crosses the zero point repeatedly, making it difficult to determine the current direction near the zero point. If compensation is performed with an incorrect current direction determination, the dead-zone effect will be further exacerbated. Therefore, determining the current polarity is a prerequisite for correct pulse width compensation. This invention utilizes the synthesized current vector i s The direction of the three-phase current (A, B, and C) can be used to determine the direction of the current flow. In one embodiment, the composite current vector i can be calculated using the d-axis current and q-axis current in a two-phase rotating coordinate system. s To determine the polarity of the current: first calculate the d-axis current i in the two-phase rotating coordinate system. d and q-axis current i q According to the d-axis current i d and q-axis current i q Calculate the composite current vector i s Then calculate the composite current vector i sThe spatial angle δ between the current and the α axis in the two-phase stationary coordinate system is used; then, based on the spatial angle δ, the current sector is determined in the vector circle, thereby determining the flow direction of the three-phase currents ABC. However, this invention is not limited to this; the composite voltage vector U can also be calculated. s The composite current vector i is obtained indirectly through the impedance angle. s The location of the vector circle. The vector circle can be either a voltage vector circle or a current vector circle.

[0062] More specifically, after acquiring the current i a i b and i c Then, a coordinate transformation can be performed to obtain i in the dq two-phase rotating coordinate system. d i q Then, based on i d and i q The relationship between the currents is used to calculate the current synthesis vector i. s The location. Figure 6 It shows i d and i q The relationship between the currents is used to calculate the current synthesis vector i. s One example.

[0063] As shown in the figure, angle ε is the current synthesis vector i s The angle between the rotating coordinate system and the d-axis is equal to the value of i. q with i d The arctangent, i.e., ε = arctan(i q / i d Angle θ is the angle between the d-axis of the rotating coordinate system and the α-axis of the two stationary coordinate systems. The composite vector i s The spatial angle between the α axis and the α axis is δ = ε + θ.

[0064] Obtain the current synthesis vector i s After determining the spatial angle δ, we can then use vector i s Locate the position to find the corresponding current polarity.

[0065] Can be used Figure 7 The vector circle shown is used to determine the polarity of the current. Figure 7This diagram illustrates the current polarity determination based on vector circles. The black arrows represent the positive directions of the three-phase voltage vectors A, B, and C, respectively, while the gray arrows represent the negative directions. The line a1a2 can be made perpendicular to phase A. The right side of the a1a2 line segment represents the positive direction of phase A, where the phase A current is positive; the left side represents the negative direction, where the phase A current is negative. The a1a2 line segment represents the commutation moment at the zero point of the phase A current. A+ indicates the positive direction of the phase A current, and A- indicates the negative direction. This allows us to depict the phase A current distribution.

[0066] Similarly, draw lines b1b2 perpendicular to phase B and c1c2 perpendicular to phase C to depict the current distribution diagrams for phases B and C respectively. Combining the distributions of the three-phase currents together yields... Figure 2 The polarity of the current and the resultant current vector i are shown. s The relationship between the spatial angle ranges and the corresponding relationship between the directions of the compensation voltage vector.

[0067] Angle range ° Current direction Compensation voltage vector direction -30°～30° A+BC- <![CDATA[U4]]> 30°～90° A+B+C- <![CDATA[U6]]> 90°～150° A-B+C- <![CDATA[U2]]> 150°～210° A-B+C+ <![CDATA[U3]]> 210°～270° AB-C+ <![CDATA[U1]]> 270°～330° A+B-C+ <![CDATA[U5]]>

[0068] Table 1 Relationship between current polarity and vector position

[0069] Taking sector II of SVPWM voltage vector synthesis as an example, within the interval 60° to 90°, the polarities of the three-phase currents A, B, and C are positive, positive, and negative, respectively. Based on the above analysis, when the current flow direction (or polarity) is positive, the conduction time of the power management input corresponding to the upper bridge arm is extended by t. d When the current flow direction (or polarity) is negative, the conduction time of the power transistor corresponding to the upper bridge arm will be shortened by t. d Then, the dead zone compensation waveform in the 60° to 90° range of sector II can be obtained. Figure 8 The dead time compensation waveform is shown within the 60° to 90° range of sector II.

[0070] like Figure 8 As shown, U A U B and U C The diagram corresponds to the PWM waveform input of the upper arm power transistors in the three-phase inverter. The thin black line represents the uncompensated waveform, the thick black line represents the compensated actual PWM waveform, and the gray area represents the dead time. As can be seen from the diagram, the upper arms of phases A and B are extended by t on both sides. d / 2 of the action time, the upper arm of phase C shortened by t on both sides. d / 2 duration of action.

[0071] Referring to the seven-segment SVPWM timing calculation formula, the compensated time is:

[0072]

[0073] In the formula, T a T b T c The three-phase activation time before compensation. This refers to the three-phase turn-on time after compensation. Since the current flow direction (or current polarity) is positive, positive, or negative within the [30°~90°] angle range, and according to the vector circle, the direction of the compensation voltage vector corresponding to this angle range is U6, the direction of the compensation voltage vector can be determined as U6. The total operating time of U6 increases by 2t. d The duration of action of U2 remains unchanged, while the duration of action of the zero vector U0 decreases by 2t. d Since U0 and U7 do not actually participate in the force, another implementation method is to increase the total action time of U6 by 2t. d For example, you can keep the B and C phases of the PWM waveform unchanged, and increase the on-time of the A phase by 2t. d It can achieve the same effect as dead zone compensation.

[0074] Within the range of 90° to 120°, the polarities of the three-phase currents A, B, and C are negative, positive, and negative, respectively. The conduction time of the power transistors in the upper bridge arms of phases A and C is reduced by one t. d The effective time should be extended by one t for the upper arm of phase B. d Duration of action. Figure 9 The dead time compensation waveform is shown within the 90° to 120° range of sector II.

[0075] like Figure 9 As shown, U A U B and U C The PWM waveform input corresponds to the upper arm power transistors of the three-phase inverter. Thin black lines represent the uncompensated waveform, thick black lines represent the compensated actual PWM waveform, and the gray area represents the dead time. From... Figure 9 As can be seen from this, the upper bridge arms of phases A and C each shortened by t on both sides. d The action time of / 2, the upper arm of phase B was extended by t on both sides. d / 2 duration of action.

[0076] Referring to the seven-segment SVPWM timing calculation formula, the compensated time is:

[0077]

[0078] In the formula, T a T b T c This refers to the three-phase activation time before compensation. This refers to the three-phase turn-on time after compensation. Since the current flow direction (or current polarity) is positive, positive, or negative within the [90°~150°] angle range, and according to the vector circle, the direction of the compensation voltage vector corresponding to this angle range is U2, the vector direction of the compensation voltage can be determined as U2. The total operating time of U2 increases by 2t. d The duration of action of U6 remains unchanged, while the duration of action of the zero vector U0 decreases by 2t. d Since U0 and U7 do not actually participate in the force, another implementation method is to increase the total action time of U2 by 2t. d For example, you can keep phases A and C of the PWM waveform unchanged, and increase the on-time of phase B by 2t. d This can achieve an equivalent dead-zone compensation effect. Similarly, PWM compensation schemes for other intervals can be derived. In this invention, equation (1-X) represents the formula for theoretical derivation, while equation (2-X) represents the formula used in actual operation.

[0079] like Figure 7 As shown, the phase B current crosses zero at 30° and 210°; the phase A current crosses zero at 90° and 270°; and the phase C current crosses zero at 150° and 330°. Near the zero-crossing point, the current repeatedly crosses the zero point, which can easily lead to compensation errors and consequently, deterioration of motor control.

[0080] Therefore, the permanent magnet synchronous motor dead-time effect compensation method of the present invention further includes step S530. In step S530, the synthesized current vector i s When entering the near-zero sector where the zero-crossing point of the aforementioned three-phase current is located, the compensation gain k is used to reduce the compensation for the three-phase turn-on time of each of the three phases A, B, and C.

[0081] Here, "near-zero sector" can refer to an angular range near a predetermined angle corresponding to the zero-crossing point (e.g., the aforementioned 30° and 210°, 90° and 270°, 150° and 330°). Specifically, the angular range of -Δd to Δd above and below the predetermined angle corresponding to the zero-crossing point of each phase current can be called the "near-zero sector," where Δd is a predetermined positive angular value. The value of Δd can be arbitrarily agreed upon or an empirical value determined based on experimental data.

[0082] The compensation gain k can be a compensation gain adjustment function, which can be a function that varies with the angle of the synthesized current vector between [predetermined angle -Δd to predetermined angle +Δd] and increases between [-1, 1]. The compensation gain k can be 0 at each of the aforementioned zero-crossing points (the preset angles corresponding to the zero-crossing points are 30° and 210°, 90° and 270°, 150° and 330°). For example, it can be zero within a very small angle range, such as [-1° to 1°], or it can be a monotonically increasing function. Figure 10 An example of a compensation gain function is shown. In the figure, the vertical axis represents the compensation gain k, and the horizontal axis [-Δd to Δd], representing the range of angles requiring compensation, i.e., the angle region covered by the near-zero sector. This curve can be obtained by fitting a quadratic function. In some embodiments, the optimized value of the quadratic function k can be obtained experimentally for different motor models. The composite current vector i is shown when the three-phase currents gradually approach zero. s The angle is located near the preset angle, reducing the compensation gain.

[0083] In one embodiment, the angle range of -Δd to Δd can be set to, for example, [-5° to 5°]. As mentioned earlier, the B-phase current crosses zero at 30° and 210°; the A-phase current crosses zero at 90° and 270°; and the C-phase current crosses zero at 150° and 330°. At this time, the corresponding near-zero sectors of the A-phase current can be [25° to 35°] and [205° to 215°], the near-zero sectors of the B-phase current can be [85° to 95°] and [265° to 275°], and the near-zero sectors of the C-phase current can be [145° to 155°] and [325° to 335°]. When the synthesized current vector enters the near-zero sector where the zero-crossing point is located, the compensation for reducing the three-phase turn-on time of each of the three phases A, B, and C by compensating for the gain k can be achieved by compensating for the gain k. a T b T c Dead time t d / 2 multiplied by k (i.e., t in equations 2-1 and 2-2) dThe coefficient of / 2 changes from 1 to k), and can include: compensation for the turn-on time of phases A, B, and C by reducing the compensation gain k when entering the near-zero sector where the phase A current zero-crossing point is located (i.e., the composite current vector is enters the range of [25°~35°] and [205°~215°]); compensation for the turn-on time of phases A, B, and C by reducing the compensation gain k when entering the near-zero sector where the phase B current zero-crossing point is located (i.e., the composite current vector is enters the range of [85°~95°] and [265°~275°]); and compensation for the turn-on time of phases A, B, and C by reducing the compensation gain k when entering the near-zero sector where the phase C current zero-crossing point is located (i.e., the composite current vector is enters the range of [85°~95°] and [265°~275°]). Thus, the compensation amount can be reduced near the current zero-crossing point to prevent control deterioration caused by miscompensation. At this point, the incorporation of the compensation gain k as hardware compensation will, to some extent, make up for the losses caused by software compensation.

[0084] As described above Figure 5-10 The software-inserted fixed dead time compensation of the present invention is described in detail. However, the dead time effect includes not only the power device turn-on time error caused by dead time, on / off delay, etc., but also the error caused by the power device turn-on voltage drop U. s and diode voltage drop U d . Figure 11 An example of a waveform diagram showing the voltage output error due to the dead-time effect of phase A is shown. In the figure, U an U is the ideal output voltage of phase A. an1 Considering the impact of conduction time on the output voltage of phase A, U an2 Considering the influence of conduction time and the voltage drop of the power transistor and diode on the A-phase output voltage, the solid line represents the actual output voltage. In other words, Figure 11 The U shown an2 The waveform can be seen as... Figure 4 Considering dead time t in (d) and (e) d and conduction time t d and closing time t off Based on this, further consideration should be given to the on-state voltage drop U of the power transistor. s and diode voltage drop U d The obtained waveform.

[0085] like Figure 11 As shown on the left (a), when the A-phase current i a When >0, the ideal output U an It is turned on at time t1 and turned off at time t3 with an amplitude equal to the power supply voltage U. dc The waveform; considering the dead time t d and conduction time t on and closing time toff Then, output U an2 It is turned on at time t2 and turned off at time t4 with an amplitude equal to the power supply voltage U. dc The waveform; and further considering the power transistor's on-state voltage drop U s and diode voltage drop U d In the case of, according to Figure 3A Analysis of the current freewheeling direction during the dead time shown indicates that: between t0 and t1, VT1 is not conducting, and the current freewheels through diode D4. At this time, the voltage U... an2 =-U d U d The voltage drop across diode D4 is given. Between t1 and t2, VT1 does not immediately conduct due to the dead-time effect, and the current continues to flow through D4. At this time, the voltage is still U. an2 =-U d Between t2 and t3, during the normal A-phase voltage output range, VT1 is turned on, and current flows in through VT1. At this time, the voltage U an2 =U dc -U s That is, the power supply DC voltage U dc Subtract the voltage drop U of power transistor VT1 s This is the output voltage of phase A; between t3 and t4, VT1 is turned off. Due to the dead-time effect, VT1 will not turn off immediately, and current will still flow through VT1. At this time, the voltage U an2 =U dc -U s Between t4 and t5, VT1 is turned off, and at this time the voltage U an2 =-U d .

[0086] Similarly, such as Figure 11 As shown on the left (b), when the A-phase current i a When < 0, the ideal output U an It is turned on at time t1 and turned off at time t3 with an amplitude equal to the power supply voltage U. dc The waveform; considering the dead time t d and conduction time t on and closing time t off Then, output U an2 It is turned on at time t2 and turned off at time t4 with an amplitude equal to the power supply voltage U. dc The waveform; and further considering the power transistor's on-state voltage drop U s and diode voltage drop U d In the case of, according to Figure 3B Analysis of the current freewheeling direction during the dead time shown indicates that:

[0087] Between t0 and t1, VT4 is turned on, and current flows out through VT4. At this time, the voltage U an2 =U s U s The voltage drop across power transistor VT4; between t1 and t2, due to the dead-time effect, VT4 does not turn off immediately, and current continues to flow through VT4. At this time, the voltage U... an2 =U s Between t2 and t3, VT4 is off, and current freewheels through diode D1. At this time, the voltage U an2 =U dc +U d Between t3 and t4, due to the dead-time effect, VT4 will not conduct immediately, and the current will freewheel through D1. At this time, the voltage U an2 =U dc +U d Between t4 and t5, VT4 is on, and at this time the voltage U an2 =U s .

[0088] In summary, when i a When >0, the output voltage U of phase A an for:

[0089]

[0090] when i a When <0, the output voltage U of phase A an for:

[0091]

[0092] In the formula, VT1 = 1 represents that the upper tube of phase A (the power tube of the upper bridge arm) is turned on, and VT1 = 0 represents that the lower tube of phase A (the power tube of the lower bridge arm) is turned on.

[0093] Output voltage U jn (j = a, b, c) is determined by the power transistor voltage drop, the diode forward voltage drop, and the current direction, assuming one switching cycle T. s If the current direction does not change, then the output voltage of phase j can be summarized as follows:

[0094]

[0095] In the formula, D j This represents the duty cycle of the upper arm of phase j within 1 Ts.

[0096] PMSM can be equivalent to a three-phase symmetrical load, where o is the neutral point. In a three-phase symmetrical load, we have:

[0097] U ao +Ubo +U co =0 (1-10)

[0098] In the formula, U ao U bo U co These are the phase voltages of phases A, B, and C, respectively. (Reference) Figure 1 We can obtain:

[0099] U jn =U jo +U on (1-11)

[0100] In the formula, U on This is the voltage from the load neutral point to point n.

[0101] From equations (1-10) and (1-11), we can obtain:

[0102]

[0103] For ease of theoretical analysis, the smaller value (U) can be used. d -U s This is ignored. Therefore, substituting equation (1-12) into the expression for the voltage of phase A yields:

[0104]

[0105] In the formula, T a T b T c These represent the conduction times of phases A, B, and C, respectively. When dead-time effects are not considered, the expected value of phase voltage in phase A is... for:

[0106]

[0107] Substituting equation (1-6) into equation (1-13), we get:

[0108]

[0109]

[0110] From equations (1-14) and (1-15), the phase voltage error of phase A caused by the dead zone effect is:

[0111]

[0112] in:

[0113]

[0114] Similarly, the phase voltage error between phases B and C is:

[0115]

[0116]

[0117] In summary, the voltage deviation caused by the dead-time effect mainly consists of two parts: one part is the inserted fixed dead time (see equation (1-4) M=(t)). off -t on -t d Part of it is the nonlinear characteristics of power devices, U d and U s The resulting pressure drop. Therefore, according to Figure 4 The method shown, which simply uses fixed-time compensation, is unlikely to achieve optimal compensation results. A preferred embodiment introduces a compensation gain k for partial hardware compensation and mitigates the impact of current zero-crossing. In another preferred embodiment, the present invention can employ hardware feedforward compensation as described below to reduce the impact of dead-time effects. In a further embodiment, a combination of software-inserted fixed dead-time and hardware feedforward compensation can be used to further reduce the impact of dead-time effects.

[0118] The hardware feedforward compensation method used in this invention calculates and compensates for error voltage through online dead-time compensation. Its basic principle is to compare the given voltage with the voltage output by the hardware circuit, and then compensate for the deviation between the given voltage and the actual output voltage.

[0119] Directly measuring the three-phase output voltage requires high-precision and rapid-response hardware circuitry. Furthermore, the three-phase output voltage contains significant noise; filtering would cause signal amplitude attenuation and phase delay, making compensation difficult. Therefore, the hardware feedforward compensation of this invention converts the error signal into a two-phase rotating coordinate system (i.e., obtains the d-axis voltage U). d and q-axis voltage U q This method enables the conversion of errors from vectors to scalars, simplifying the implementation and eliminating the need for additional hardware.

[0120] In one embodiment, the dead-zone compensation method of the present invention may further include: based on the d-axis current i of the previous cycle... d q-axis current i q d-axis voltage U d q-axis voltage U q Calculate the d-axis compensation voltage U for the current cycle based on the rotor speed. d,com and q-axis compensation voltage U q,com ; and the d-axis compensation voltage U using the current cycle d，com and q-axis compensation voltage U q，comHardware compensation is performed on the d-axis and q-axis voltages for the current cycle.

[0121] Specifically, the d-axis current i of the previous cycle d q-axis current i q The rotational speed *n* can be used to calculate the theoretical d-axis voltage and the theoretical q-axis voltage. The theoretical d-axis voltage and the actually acquired d-axis voltage *U* from the previous cycle... d The d-axis voltage error value between the two values ​​is used as the d-axis compensation voltage U for the current cycle. d，com The theoretical q-axis voltage and the actual q-axis voltage U from the previous cycle... q The q-axis voltage error value between the two values ​​is then used as the q-axis compensation voltage U for the current period. q，com .

[0122] Specifically, in the dq coordinate system, the PMSM voltage equation is expressed as follows:

[0123]

[0124] In the formula, R s L is the stator resistance. d For the d-axis inductance, L q For q-axis inductance, ω e Let ψ be the electric angular velocity (which can be obtained from the rotational speed n). f It is a magnetic flux.

[0125] Since there is a difference between the actual and theoretical values ​​of the d-axis and q-axis voltages, if U d and U q If we consider them as the actual d-axis voltage and q-axis voltage respectively, then we can obtain the following formula that takes into account the d / q-axis error:

[0126]

[0127] In the formula, U d，err U q，err These are the d-axis and q-axis error voltages, respectively. A d-axis and q-axis compensation voltage U is then applied. d，com U q，com By discretizing it, we can obtain:

[0128]

[0129] According to equation (2-3), the current compensation voltage U d，com (x), U q，com (x) and error voltage U d，err (x), U q，errWhen (x) are equal, the interference caused by the dead zone factor can be eliminated. Since the switching period is usually extremely short, it can be assumed that the change in error voltage is close to zero within one sampling period, that is, the error U of the current period is zero. d，err (x), U q，err (x) is compared with the error U of the previous period. d，err (x-1), U q，err (x-1) are equal, therefore if the compensation voltage U of the current cycle is equal... d，com (x), U q，com (x) can be compared with the error U of the previous period. d，err (x-1), U q，err If (x-1) is equal, the error of the current period can be approximately compensated. Therefore, we can obtain:

[0130]

[0131] Furthermore, since the change in current is relatively small when the motor is stable, equation (2-4) can be simplified to:

[0132]

[0133] Therefore, this invention can utilize the d-axis current id and q-axis current i based on the previous cycle. q The theoretical d / q-axis voltage is obtained from the rotational speed n, where the theoretical d-axis voltage = R s i d (x-1)- ω e L q i q (x-1), theoretical q-axis voltage = R s i q (x-1)+ω e L d i d (x-1)+ω e ψ f Then, the actual d / q axis voltage U from the previous cycle is collected. d (x-1) / U q The difference between (x-1) and the previously calculated theoretical d / q axis voltage is used as dynamic compensation, i.e., as the compensation voltage U for the current cycle. d，com (x), U q，com (x) is used to compensate for the voltage of the d / q axis voltage of the current cycle (i.e., the voltage sent to the Park inverse transformation module) in real time.

[0134] In one embodiment, the rotational speed n (i.e., the electrical angular velocity ω) can be considered as... e In another embodiment, the theoretical d / q-axis voltage is approximately constant between adjacent cycles. It can be calculated based on the d / q-axis current and rotational speed of the previous cycle: Theoretical d-axis voltage = Rs i d (x-1)-ω e L q i q (x-1), theoretical q-axis voltage = R s i q (x-1)+ω e L d i d (x-1)+ω e ψ f , where i d (x-1) and i q (x-1) represent the d / q axis currents of the previous cycle, ω e R is the electric angular velocity of the previous cycle (which can be obtained from the rotational speed n of the previous cycle), and the rest are constants. s L is the stator resistance. d For the d-axis inductance, L q For q-axis inductance, ψ f It is a magnetic flux.

[0135] This hardware feedforward scheme can be used independently to achieve dead-time compensation, or it can be used as an auxiliary scheme to compensate for the shortcomings of simply inserting a fixed dead-time in software, thus avoiding the adverse effects of current zero-crossing to a certain extent. FFT analysis through simulation experiments shows that, compared to the effect of simply inserting a fixed compensation dead-time in software, supplementing with hardware feedforward compensation can significantly reduce the total harmonic distortion (THD) value, especially showing excellent results in reducing the 5th and 7th harmonics, reducing the total harmonic distortion by more than half.

[0136] The present invention can also be implemented as a permanent magnet synchronous motor, including a PMSM motor and a vector control circuit for performing FOC (Free-Oriented Control) on the motor. The vector control circuit may include, for example... Figure 1 All modules are shown. In addition, the vector control circuit may further include: a fixed-time compensation module located between the SVPWM and the three-phase inverter bridge, used to perform the above-described combination. Figure 5 The software fixed-time compensation method is used to compensate for the three-phase turn-on time of each of the three phases A, B, and C.

[0137] Furthermore, when hardware feedforward compensation is required as a supplement to fixed-time compensation, the vector control circuit may also include: a hardware feedforward compensation module, used to perform hardware feedforward compensation based on the d-axis current i of the previous cycle. d q-axis current i q d-axis voltage U d q-axis voltage U q (In one embodiment, rotor speed n is further included) Calculate the d-axis compensation voltage and q-axis compensation voltage for the current cycle, and perform hardware compensation on the d-axis voltage and q-axis voltage for the current cycle.

[0138] Figure 12 A schematic diagram of a permanent magnet synchronous motor according to an embodiment of the present invention is shown. As shown in the figure, compared to Figure 1 The permanent magnet synchronous motor 1200 of the present invention may additionally include a fixed-time compensation module 1210 located between the SVPWM module and the three-phase inverter bridge, for performing the above-described combination. Figure 5 The aforementioned software fixed-time compensation method, namely, for T a T b T c The conduction and shutdown are performed by t d The fixed time compensation is 2 / 2, and the fixed time compensation can be reduced near the zero crossing.

[0139] Furthermore, the permanent magnet synchronous motor 1200 may also include a hardware feedforward compensation module 1220, which acquires in real time the d-axis current i of the previous cycle. d q-axis current i q d-axis voltage U d q-axis voltage U q The value of the rotor speed n is used to calculate the d-axis compensation voltage U. d，com and q-axis compensation voltage U q，com The aforementioned d-axis compensation voltage U d，com and q-axis compensation voltage U q，com The output q-axis and d-axis voltages U are fed in and regulated by the q-axis current loop and d-axis current loop. d and U q To perform the inverse Park transform.

[0140] The method for compensating for the dead time effect of a permanent magnet synchronous motor according to the present invention, and the permanent magnet synchronous motor using the method, have been described in detail above with reference to the accompanying drawings. The software-inserted fixed dead time of the present invention can provide fixed compensation for the dead time, and by introducing a compensation gain k to reduce the fixed compensation near the zero-crossing point, it avoids problems caused by the current zero-crossing. Furthermore, the software-fixed dead time compensation of the present invention can be supplemented by a novel hardware feedforward dead time compensation scheme. This scheme can dynamically compensate for the dq-axis voltage by real-time reading of the dq-axis voltage, current, and speed. The above-mentioned hardware-assisted compensation scheme does not require additional hardware, has a small computational load, and when combined with the software-fixed dead time compensation method, can greatly reduce current harmonics and improve motor control performance.

[0141] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for compensating for dead-time effect in a permanent magnet synchronous motor, comprising: Obtain the direction of the three-phase currents (A, B, and C) in the current sector where the composite current vector is located; Based on the current flow direction of each of the three phases A, B, and C in the current sector, the turn-on time of each of the three phases A, B, and C is compensated so that the effect of the compensation voltage vector in the current sector is increased by a predetermined time. as well as When the synthesized current vector enters the near-zero sector where the zero-crossing point of the three-phase currents ABC is located, the compensation gain k is used to reduce the compensation for the three-phase turn-on time of each of the three phases ABC. The method further includes: Calculate the d-axis compensation voltage and q-axis compensation voltage for the current cycle based on the d-axis current, q-axis voltage, d-axis voltage, q-axis voltage, and rotor speed collected in the previous cycle; and Hardware compensation is performed on the d-axis and q-axis voltages of the current period using the d-axis compensation voltage and q-axis compensation voltage of the current period. The hardware compensation serves as a supplement to the compensation for the three-phase turn-on time.

2. The method as described in claim 1, wherein, Based on the current flow direction of each of the three phases A, B, and C in the current sector, compensation for the three-phase turn-on time of each of the three phases A, B, and C includes: Extend the conduction time of the upper bridge power transistor in the phase with positive current flow by t. d The conduction time t of the lower arm bridge power transistor with the current flowing in the positive direction is shortened. d ;as well as Shorten the conduction time t of the upper bridge power transistor in the phase with negative current flow. d Extend the conduction time t of the lower arm bridge power transistor that directs current to the negative phase. d .

3. The method as described in claim 2, wherein, Extend the power transistor's on-time by t d include: Delay the turn-off time of the power transistor by t d / 2, and advance the conduction time by t. d / 2; Shorten the power transistor's on-time t d include: Advance the turn-off time of the power transistor by t d / 2, and postpone the conduction time by t d / 2.

4. The method of claim 1, wherein, The near-zero sector includes: The predetermined angle range of -Δd to Δd corresponds to the zero-crossing point of each phase current, where Δd is a predetermined positive angle value.

5. The method of claim 4, wherein, The compensation gain k is a function that varies with the angle of the synthesized current vector between [predetermined angle - Δd ~ predetermined angle + Δd] and increases between [-1, 1].

6. The method of claim 1, wherein, The compensation gain k is 0 at each zero-crossing point of the angle of the synthesized current vector.

7. The method of claim 1, wherein, When the synthesized current vector enters the near-zero sector where the zero-crossing point of the ABC three-phase current is located, the compensation for reducing the three-phase turn-on time of each of the ABC phases by the compensation gain k includes: When entering the near-zero sector where the current of phase A crosses zero, the compensation gain k is multiplied by the compensation amounts of phases A, B, and C respectively; When entering the near-zero sector where the B-phase current zero-crossing point is located, the compensation gain k is multiplied by the compensation amounts of phases A, B, and C respectively; and When entering the near-zero sector where the current of phase C crosses zero, the compensation gain k is multiplied by the compensation amounts of phases A, B, and C respectively.

8. The method of claim 1, wherein, The current sector where the synthesized current vector is located includes: Calculate the d-axis current and q-axis current in a two-phase rotating coordinate system; The composite current vector is calculated based on the d-axis current and the q-axis current. Calculate the spatial angle between the combined current vector and the α-axis in the two-phase stationary coordinate system; and The current sector is determined in the current vector circle based on the spatial angle.

9. The method of claim 1, wherein, The calculation of the d-axis compensation voltage and q-axis compensation voltage for the current cycle based on the d-axis current, q-axis voltage, d-axis voltage, q-axis voltage, and rotor speed of the previous cycle includes: The d-axis voltage error value between the acquired d-axis voltage of the previous cycle and the theoretical d-axis voltage calculated based on the d-axis current and q-axis current of the previous cycle is obtained and used as the d-axis compensation voltage for the current cycle; and The q-axis voltage error value between the collected q-axis voltage of the previous cycle and the theoretical q-axis voltage calculated based on the d-axis current and q-axis current of the previous cycle is obtained and used as the q-axis compensation voltage for the current cycle.

10. A permanent magnet synchronous motor, comprising: Electric motor; A vector control circuit, the vector control circuit comprising: A software time compensation module located between the SVPWM and the three-phase inverter bridge is used to obtain the current flow direction of the three-phase ABC currents in the current sector where the synthesized current vector is located; based on the current flow direction of each of the three phases ABC in the current sector, it compensates for the three-phase turn-on time of each of the three phases ABC so that the effect of the compensation voltage vector in the direction corresponding to the current sector is increased by a predetermined time; and when the synthesized current vector enters the near-zero sector where the zero-crossing point of the three-phase current of the three phases ABC is located, it reduces the compensation for the three-phase turn-on time of each of the three phases ABC by using a compensation gain k. The hardware feedforward compensation module is used to calculate the d-axis compensation voltage and q-axis compensation voltage for the current cycle based on the collected d-axis current, q-axis voltage, d-axis voltage, q-axis voltage, and rotor speed of the previous cycle, so as to perform hardware compensation on the d-axis voltage and q-axis voltage for the current cycle. The hardware compensation serves as a supplement to the compensation for the three-phase turn-on time.

Images