Three-phase Hall self-learning method for vector control of brushless motor of high-speed chip mounter

Abstract

The invention discloses a three-phase Hall self-learning method for vector control of a brushless motor of a high-speed chip mounter, and the method comprises the steps: constructing a mathematical model under a D-Q-axis coordinate system of the brushless motor, writing open-loop angle data into an array in advance, and obtaining an open-loop angle theta open (index) through the index value look-up table of a sine angle look-up table; and applying voltage on the D axis based on the open-loop angle theta open (index) and dragging the brushless motor to rotate in an open-loop manner, recording the current open-loop angle and a Hall value when the Hall signal is triggered to be interrupted after the motor rotates, and judging the Hall value conversion rule to obtain the open-loop rotation direction of the brushless motor. According to the method, six accurate angles and vector control in each electric period of the brushless motor are obtained through self-learning, line sequence self-learning during connection of the driver and the brushless motor is achieved, the problem of torque fluctuation caused by poor precision during installation of the Hall sensor is solved, and the control performance of the driver is improved.

Description

Technical Field

[0001] This invention relates to the field of SMT high-speed placement machine technology, and in particular to a three-phase Hall self-learning method for vector control of a brushless motor in a high-speed placement machine. Background Technology

[0002] The core function of the film drive mechanism in a high-speed surface mount technology (SMT) feeder is to ensure the uniformity and constant tension of the film winding. The torque smoothness requirement for the feeder's drive motor is higher than its positioning accuracy. In existing control strategies, square wave commutation control, due to its inherent torque ripple and high acoustic noise, cannot meet the smoothness requirements of this application. While vector control based on Hall position interpolation has certain advantages in efficiency and noise control, its rotor position observer, due to the mechanical tolerance of the Hall sensor installation, will produce discontinuous jumps in the estimated value at the electrical angle boundary of the Hall state update. This step change in control quantity ultimately translates into electromagnetic torque ripple, causing motor vibration and additional energy loss. Summary of the Invention

[0003] To address the shortcomings of existing technologies, the purpose of this invention is to provide a three-phase Hall self-learning method for vector control of brushless motors in high-speed pick-and-place machines. This method solves the problems of wiring sequence self-learning when the driver is connected to the feeder motor and torque fluctuation caused by poor accuracy of Hall sensor installation, thereby improving the driver control performance.

[0004] To achieve the above objectives, the technical solution adopted by this invention is: a three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed chip mounter, comprising the following steps:

[0005] Step 1: Construct a mathematical model of the brushless motor in the DQ axis coordinate system, initialize the motor control parameters, including the D-axis current setpoint Id* and the Q-axis current setpoint Iq*, and set the initial index value index=0 and the angle increment step size of the brushless motor during open-loop operation.

[0006] Step 2: Pre-generate a sinusoidal angle lookup table, which contains equally spaced sampled angle values ​​θ for a complete electrical cycle. open The open-loop angle θ is obtained by looking up the sin value corresponding to the current open-loop angle in a table using an incrementing index. open (index);

[0007] Step 3: Use the open-loop angle θ generated in Step 2. open (index) Generates a PWM signal for driving the inverter, producing a signal on the stator of the brushless motor that varies with the open-loop angle θ in the D-axis direction. open (index) A synchronously rotating virtual magnetic field drives the rotor of the brushless motor to rotate.

[0008] Step 4: Control the rotor rotation of the brushless motor and record the open-loop angle and actual rotation direction of the current Hall transition point when the output level of any Hall sensor of the brushless motor changes.

[0009] Step 5: Calculate the actual area of ​​each Hall sector based on the actual rotation direction, Hall jump point, and the open-loop angle corresponding to the Hall jump point; after executing steps 1 to 5, the three-phase Hall self-learning of the brushless motor vector control is automatically completed.

[0010] Step 6: Based on the actual rotation direction, calculate the actual angle sector value corresponding to the current Hall transition point to the next Hall transition point, accurately depict the non-uniform distribution information of each sector in the electrical angle space, and calibrate the obtained actual Hall sector information.

[0011] Step 7: After the brushless motor enters the working state, the control mode of the brushless motor is switched to closed-loop vector control mode. The brushless motor is controlled according to the actual Hall sector information obtained by calibration, and the estimated value of the rotor position is dynamically verified and constrained.

[0012] In a further technical solution, step 1 involves transforming the three-phase stationary coordinate system of the brushless motor into a two-phase rotating coordinate system using Clark transformation and Park transformation. The two-phase rotating coordinate system is the DQ-axis coordinate system.

[0013] In step 2, a complete electrical cycle is an electrical angle of 0° to 360°, and the angle value θ is set to be sampled at equal intervals. open In open-loop operation, the rotor of the brushless motor rotates at a constant speed under open-loop drag. The Sin value corresponding to the current open-loop angle is obtained by looking up a table using an incrementing index value. The index value increments at a fixed frequency to simulate a virtual rotor angle rotating at a constant speed.

[0014] In step 3, a constant voltage Vd is given on the D-axis, and a voltage Vq=0 is given on the Q-axis. The open-loop angle θ generated in step 2 is then used. open (index) performs inverse Parker transformation and space vector pulse width modulation to generate a PWM signal for controlling the driven inverter of the brushless motor.

[0015] In a further technical solution, step 2 directly provides trigonometric function values ​​for the Parker transform and inverse Parker transform through a pre-generated sine angle lookup table, thereby improving real-time calculation efficiency while simulating the angle of a uniformly rotating virtual rotor. The sine angle lookup table adopts a storage space optimization strategy, storing only the reference sine function values ​​in the first quadrant. The floating-point representation of the reference sine function values ​​is converted to a fixed-point representation. By linearly mapping the reference sine function values ​​of 0 to 1 to the integer range of 0 to 32767, implemented in Q15 format, floating-point operations are avoided in the real-time control loop. When obtaining sine function values ​​in the range of 90° to 360°, the reference sine function values ​​stored in the first quadrant are obtained by performing corresponding sign transformations and index address calculations based on the quadrant symmetry principle of trigonometric functions, without additionally storing full-cycle data, thus achieving an optimized balance between storage resources and calculation efficiency.

[0016] In a further technical solution, step 4 involves controlling the rotor of the brushless motor to rotate at a uniform speed. When the output state of any Hall sensor of the brushless motor changes, an external interrupt is triggered on the microcontroller of the brushless motor, with the current moment being the Hall transition point of the Hall transition event. In the interrupt service routine of the external interrupt, the open-loop angle θ at the current moment is immediately recorded. open (H), and simultaneously record the combined Hall state value Habc of the three-phase Hall sensors at this time, and use the combined Hall state value Habc as an index to determine the open-loop angle θ. open (H) stores the data in an array.

[0017] In a further technical solution, step 4 involves continuously recording at least two Hall transition events, analyzing the change sequence of the combined Hall state value Habc, and determining the actual rotation direction of the brushless motor under the current open-loop drive based on the different change patterns of the three-phase Hall signals during forward and reverse rotation (forward rotation is 101-100-110-010-011-001, reverse rotation is 001-011-010-110-100-101).

[0018] In a further technical solution, in step 5, within one complete Hall cycle of the brushless motor, the three-phase Hall sensor of the brushless motor generates six different state values. These six different state values ​​correspond to the boundaries of six Hall sectors, respectively. Based on the open-loop angle θ corresponding to at least six Hall transition points recorded in step 3... open (index), combined with the rotation direction determined in step 4, the actual area of ​​each Hall sector is calculated.

[0019] In a further technical solution, step 1, in the DQ axis coordinate system, the voltage equation of the brushless motor is precisely expressed as: , among which, Ud For direct-axis voltage, U q Let R be the quadrature-axis voltage, R be the stator winding resistance of the brushless motor, and i be the voltage across the quadrature axis. d For direct-axis current, i q For quadrature-axis current, L d For a direct-axis inductor, L q For quadrature axis inductance, W e Let ψ be the angular velocity of the brushless motor. f It is a permanent magnet flux linkage.

[0020] In a further technical solution, in step 7, after the brushless motor switches to the closed-loop vector control mode, during the duration of the given Hall state between any two adjacent Hall transition points, the real-time angle of the brushless motor rotor is estimated by a back EMF observer, a flux linkage observer, or an interpolation algorithm. The real-time angle of the rotor is constrained within the actual area determined by the actual Hall sector information corresponding to the current Hall state.

[0021] In a further technical solution, in step 3, during the open-loop dragging stage, the voltage setpoint in the two-phase rotating coordinate system is configured. A constant voltage setpoint Vd is applied to the D-axis corresponding to the direct axis, and the voltage setpoint Vq corresponding to the Q-axis is set to zero. The open-loop angle θ generated in step 2 is then used. open (index) The inverse Parker transform and space vector pulse width modulation are executed sequentially to generate a PWM signal for driving the three-phase inverter. This generates a signal on the stator side of the motor with a spatial direction aligned with the D-axis and a spatial electrical angle that varies with the open-loop angle θ. open (index) A synchronously rotating virtual magnetic field; controlling the rotation and reversal of the virtual magnetic field, with respect to the open-loop angle θ. open (index) implements the following management strategy.

[0022] 1) Angle accumulation: In each current control cycle, the open-loop angle θ is accumulated. open The accumulated value of (index) is incremented once, and the increment Δθ is a configurable non-zero integer.

[0023] 2) Angle calibration, open-loop angle θ open The value range of (index) is linearly mapped to a complete electrical cycle. The angle value range is set from 0 to 65535, corresponding to an electrical angle from 0° to 360°.

[0024] 3) Direction switching, when the open-loop angle θ open When the cumulative increment of (index) reaches the value corresponding to two complete electric cycles, the sign of the increment Δθ is reversed, and the rotation direction of the virtual magnetic field is reversed.

[0025] 4) The process terminates, but continues in both forward and reverse rotations until the open-loop angle θ is reached. open The open-loop dragging process ends when the value of (index) returns to zero.

[0026] In a further technical solution, step 4 involves connecting the external interrupt input pins of the microcontroller to the output signals of the three-phase Hall sensors of the brushless motor, and configuring them to edge-triggered mode; when the rotor of the brushless motor rotates and causes the output level of any Hall sensor to change, the external interrupt is triggered; in the corresponding interrupt service routine, the following operations are performed.

[0027] Step 4.1: Read and record the open-loop angle θ generated in step 2 in real time. open (H)

[0028] Step 4.2: Synchronously read and record the combined Hall state value Habc of the three-phase Hall sensors at this time.

[0029] Step 4.3: Using the combined Hall state value Habc as the index identifier, the open-loop angle value θopen(H) is stored as the associated data in a pre-allocated storage array, thereby establishing the mapping relationship between the Hall state and the corresponding electrical angle.

[0030] Repeatedly execute steps 4.1 to 4.3 above, record at least two Hall state transition events and their corresponding angle data, analyze the change sequence of continuously recorded combined Hall state values ​​Habc, and match them with the preset motor forward and reverse state transition rules to determine the actual rotation direction of the brushless motor under the current open-loop drive.

[0031] After executing steps 1 to 5 and completing the three-phase Hall self-learning for the vector control of the brushless motor, the actual angle sector value can be calculated and the brushless motor can be driven normally after the three-phase Hall self-learning is completed, regardless of how the Hall wiring sequence and the motor power line wiring sequence are connected when the brushless motor is connected to the driver.

[0032] The advantages of this invention compared to existing technologies are as follows: This invention discloses a three-phase Hall position self-learning method for vector control of a brushless motor. First, a mathematical model of the brushless motor in the DQ-axis coordinate system is constructed. Open-loop angle data is pre-written into an array, and the open-loop angle θopen(index) is obtained by looking up the index value of a sine angle lookup table. Based on the open-loop angle θopen(index), a voltage is applied to the D-axis, and the brushless motor is rotated in an open-loop manner. When the motor rotates and the Hall signal triggers an interrupt, the current open-loop angle and Hall value are recorded. Simultaneously, the open-loop rotation direction of the brushless motor is obtained by judging the change law of the Hall value. This invention obtains six precise angles and vector control within each electrical cycle of the brushless motor through self-learning, realizing line sequence self-learning when the driver is connected to the brushless motor. This solves the torque fluctuation problem caused by poor accuracy of Hall sensor installation and improves the driver control performance. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0034] A three-phase Hall effect self-learning method for vector control of brushless motors in high-speed chip mounters. Figure 1 As shown, it includes the following steps 1-7.

[0035] Step 1: Construct a mathematical model of the brushless motor in the DQ axis coordinate system. Transform the three-phase stationary coordinate system (ABC) of the brushless motor into a two-phase rotating coordinate system (DQ axis coordinate system) through Clarke Transform and Park Transform. The two-phase rotating coordinate system is the DQ axis coordinate system. Initialize the motor control parameters, including the D-axis current setpoint Id* and the Q-axis current setpoint Iq*, and set the initial index value index=0 and the angle increment step size of the brushless motor during open-loop operation.

[0036] In the DQ-axis coordinate system, the voltage equation of the brushless motor is precisely expressed as follows: , among which, U d For direct-axis voltage, U q Let R be the quadrature-axis voltage, R be the stator winding resistance of the brushless motor, and i be the voltage across the quadrature axis. d For direct-axis current, i q For quadrature-axis current, L d For a direct-axis inductor, L q For quadrature axis inductance, W e Let ψ be the angular velocity of the brushless motor. f It is a permanent magnet flux linkage.

[0037] Step 2: Pre-generate a sinusoidal angle lookup table, which contains equally spaced sampled angle values ​​θ for a complete electrical cycle. open A complete electrical cycle is an electrical angle of 0° to 360°, with the angle value θ sampled at equal intervals. open In open-loop operation, the brushless motor's rotor rotates at a constant speed under open-loop drag. The sin value corresponding to the current open-loop angle is obtained by looking up a table using an incrementing index value. This index value increments at a fixed frequency, simulating a virtual rotor angle rotating at a constant speed, thus yielding the open-loop angle θ. open (index).

[0038] Specifically, by using a pre-generated sinusoidal angle lookup table, trigonometric function values ​​are directly provided for the Parker transform and inverse Parker transform, thereby improving real-time computation efficiency while simulating the angle of a uniformly rotating virtual rotor.

[0039] The sine angle lookup table employs a storage space optimization strategy, storing only the reference sine function values ​​within the first quadrant (0° to 90°). It converts the floating-point representation of the reference sine function values ​​to a fixed-point representation by linearly mapping the reference sine function values ​​(0 to 1) to the integer range of 0 to 32767, implemented in Q15 format, thus avoiding floating-point operations in the real-time control loop. When retrieving sine function values ​​within the range of 90° to 360°, the quadrant symmetry principle of trigonometric functions is used to perform corresponding sign transformations and index address calculations on the reference sine function values ​​stored in the first quadrant, without additionally storing full-cycle data, achieving an optimized balance between storage resources and computational efficiency.

[0040] Step 3: Apply a constant voltage Vd to the D-axis and a voltage Vq=0 to the Q-axis, using the open-loop angle θ generated in Step 2. open (index) Perform inverse Parker transformation and space vector pulse width modulation to generate a PWM signal for controlling the inverter driven by the brushless motor, producing a signal on the stator of the brushless motor that varies with the open-loop angle θ in the D-axis direction. open (index) A synchronously rotating virtual magnetic field drives the rotor of the brushless motor to rotate.

[0041] During the open-loop drag phase, the voltage setpoints in the two-phase rotating coordinate system are configured. A constant voltage setpoint Vd is applied to the D-axis corresponding to the direct axis, and the voltage setpoint Vq corresponding to the Q-axis is set to zero. The open-loop angle θ generated in step 2 is then used. open (index) The inverse Parker transform and space vector pulse width modulation are executed sequentially to generate a PWM signal for driving the three-phase inverter. This generates a signal on the stator side of the motor with a spatial direction aligned with the D-axis and a spatial electrical angle that varies with the open-loop angle θ. open(index) A synchronously rotating virtual magnetic field drives the rotor of the brushless motor to rotate.

[0042] To control the rotation and reversal of the virtual magnetic field, the open-loop angle θ is... open (index) implements the following management strategy.

[0043] 1) Angle accumulation: In each current control cycle, the open-loop angle θ is accumulated. open The accumulated value of (index) is incremented once, and the increment Δθ is a configurable non-zero integer.

[0044] 2) Angle calibration, open-loop angle θ open The numerical range of (index) is linearly mapped to a complete electrical cycle (0° to 360° electrical angle). As a preferred embodiment, the angle numerical range is set to 0 to 65535, corresponding to an electrical angle of 0° to 360°.

[0045] 3) Direction switching, when the open-loop angle θ open When the cumulative increment of (index) reaches a value corresponding to two complete electrical cycles (720° electrical angle), the sign of the increment Δθ is reversed, and the rotation direction of the virtual magnetic field is reversed.

[0046] 4) The process terminates, but continues in both forward and reverse rotations until the open-loop angle θ is reached. open The open-loop dragging process ends when the value of (index) returns to zero.

[0047] Step 4: Control the rotor rotation of the brushless motor and record the open-loop angle and actual rotation direction (Direction) of the current Hall transition point when the output level of any Hall sensor of the brushless motor changes.

[0048] The microcontroller controls the brushless motor's rotor to rotate at a constant speed. An external interrupt is triggered on the brushless motor's microcontroller when the output state of any Hall sensor changes. The current moment is the Hall transition point of the Hall transition event. In the interrupt service routine of the external interrupt, the open-loop angle θ at the current moment is immediately recorded. open (H), and simultaneously record the combined Hall state value Habc of the three-phase Hall sensors at this time, and use the combined Hall state value Habc as an index to determine the open-loop angle θ. open (H) stores the data in an array.

[0049] Record at least two Hall transition events continuously, analyze the change sequence of the combined Hall state value Habc, and based on the different change patterns of the three-phase Hall signals during forward and reverse rotation (forward rotation is 101-100-110-010-011-001, reverse rotation is 001-011-010-110-100-101), determine the actual rotation direction (Direction) of the brushless motor under the current open-loop drive based on the change patterns.

[0050] Connect the external interrupt input pins of the microcontroller to the output signals of the three-phase Hall sensors of the brushless motor, and configure them to edge-triggered mode; when the rotor of the brushless motor rotates and causes the output level of any Hall sensor to change, the external interrupt is triggered; in the corresponding interrupt service routine, the following operations are performed.

[0051] Step 4.1: Read and record the open-loop angle θ generated in step 2 in real time. open (H)

[0052] Step 4.2: Synchronously read and record the combined Hall state value Habc of the three-phase Hall sensors at this time.

[0053] Step 4.3: Using the combined Hall state value Habc as the index identifier, the open-loop angle value θopen(H) is stored as the associated data in a pre-allocated storage array, thereby establishing the mapping relationship between the Hall state and the corresponding electrical angle.

[0054] Repeatedly execute steps 4.1 to 4.3 above, record at least two Hall state transition events and their corresponding angle data, analyze the change sequence of continuously recorded combined Hall state values ​​Habc, and match them with the preset motor forward and reverse state transition rules to determine the actual rotation direction Direction of the brushless motor under the current open-loop drag.

[0055] Step 5: Calculate the actual area of ​​each Hall sector based on the actual rotation direction, Hall transition points, and the corresponding open-loop angles. Within one complete Hall cycle of the brushless motor, the three-phase Hall sensors generate six different state values. These six different state values ​​correspond to the boundaries of the six Hall sectors. Theoretically, these six different state values ​​cover six uniform 60° regions. However, due to Hall mounting accuracy issues, these six regions are not uniform 60° regions. The actual area is determined based on the open-loop angle θ corresponding to at least six Hall transition points recorded in Step 3. open (index), such as the open-loop angle θ obtained by recording 6 times. open (H1) to θ open(H6), combined with the rotation direction obtained in step 4, the actual area of ​​each Hall sector is calculated.

[0056] Executing steps 1 to 5 automatically completes the three-phase Hall self-learning of the brushless motor vector control. When the brushless motor is connected to the driver, regardless of how the Hall wiring sequence and the motor power line wiring sequence are connected, the actual angle sector value can be calculated after the three-phase Hall self-learning is completed, and the brushless motor can be driven normally.

[0057] Step 6: Based on the actual rotation direction, calculate the actual angle sector value corresponding to the current Hall transition point to the next Hall transition point, accurately depict the non-uniform distribution information of each sector in the electrical angle space, and calibrate the obtained actual Hall sector information.

[0058] Step 7: After the brushless motor enters the working state, the control mode of the brushless motor is switched to closed-loop vector control mode. The brushless motor is controlled according to the actual Hall sector information obtained by calibration, and the estimated value of the rotor position is dynamically verified and constrained.

[0059] After the brushless motor switches to closed-loop vector control mode, during the duration of the given Hall state between any two adjacent Hall transition points, the real-time angle of the brushless motor rotor is estimated by the back EMF observer, flux linkage observer, or interpolation algorithm. The real-time angle of the rotor is constrained within the actual area determined by the actual Hall sector information corresponding to the current Hall state.

Claims

1. A three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine, characterized in that: Includes the following steps, Step 1: Construct a mathematical model of the brushless motor in the DQ axis coordinate system, initialize the motor control parameters, including the D-axis current setpoint Id* and the Q-axis current setpoint Iq*, and set the initial index value index=0 and the angle increment step size of the brushless motor during open-loop operation. Step 2: Pre-generate a sinusoidal angle lookup table, which contains equally spaced sampled angle values ​​θ for a complete electrical cycle. open The open-loop angle θ is obtained by looking up the sin value corresponding to the current open-loop angle in a table using an incrementing index. open (index); Step 3: Use the open-loop angle θ generated in Step 2. open (index) Generates a PWM signal for driving the inverter, producing a signal on the stator of the brushless motor that varies with the open-loop angle θ in the D-axis direction. open (index) A synchronously rotating virtual magnetic field drives the rotor of the brushless motor to rotate. Step 4: Control the rotor rotation of the brushless motor and record the open-loop angle and actual rotation direction of the current Hall transition point when the output level of any Hall sensor of the brushless motor changes. Step 5: Calculate the actual area of ​​each Hall sector based on the actual rotation direction, Hall jump point, and the open-loop angle corresponding to the Hall jump point; execute steps 1 to 5 to complete the three-phase Hall self-learning of the brushless motor vector control; Step 6: Based on the actual rotation direction, calculate the actual angle sector value corresponding to the current Hall transition point to the next Hall transition point, accurately depict the non-uniform distribution information of each sector in the electrical angle space, and calibrate the obtained actual Hall sector information. Step 7: After the brushless motor enters the working state, the control mode of the brushless motor is switched to closed-loop vector control mode. The brushless motor is controlled according to the actual Hall sector information obtained by calibration, and the estimated value of the rotor position is dynamically verified and constrained.

2. The three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to claim 1, characterized in that: In step 1, the three-phase stationary coordinate system of the brushless motor is transformed into a two-phase rotating coordinate system through Clark transformation and Park transformation. The two-phase rotating coordinate system is the DQ axis coordinate system. In step 2, a complete electrical cycle is an electrical angle of 0° to 360°, and the angle value θ is set to be sampled at equal intervals. open In open-loop operation, the rotor of the brushless motor rotates at a constant speed under open-loop drag. The sin value corresponding to the current open-loop angle is obtained by looking up a table through an incrementing index value. The index value increments at a fixed frequency to simulate a virtual rotor angle that rotates at a constant speed. In step 3, a constant voltage Vd is given on the D-axis, and a voltage Vq=0 is given on the Q-axis. The open-loop angle θ generated in step 2 is then used. open (index) performs inverse Parker transformation and space vector pulse width modulation to generate a PWM signal for controlling the driven inverter of the brushless motor.

3. The three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to claim 2, characterized in that: In step 2, trigonometric function values ​​are directly provided for the Parker transform and inverse Parker transform through a pre-generated sinusoidal angle lookup table, which improves real-time calculation efficiency while simulating the angle of a uniformly rotating virtual rotor. The sine angle lookup table employs a storage space optimization strategy, storing only the reference sine function values ​​within the first quadrant. It converts the floating-point representation of these reference sine function values ​​to a fixed-point representation by linearly mapping the reference sine function values ​​(0 to 1) to the integer range of 0 to 32767, implemented in Q15 format, thus avoiding floating-point operations in the real-time control loop. When retrieving sine function values ​​within the range of 90° to 360°, the quadrant symmetry principle of trigonometric functions is used to perform corresponding sign transformations and index address calculations on the reference sine function values ​​stored in the first quadrant, without additionally storing full-cycle data, achieving an optimized balance between storage resources and computational efficiency.

4. The three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to claim 1, characterized in that: In step 4, the rotor of the brushless motor is controlled to rotate at a constant speed. When the output state of any Hall sensor of the brushless motor changes, an external interrupt is triggered on the microcontroller of the brushless motor. The current moment is the Hall transition point of the Hall transition event. In the interrupt service routine of the external interrupt, the open-loop angle θ at the current moment is immediately recorded. open (H), and simultaneously record the combined Hall state value Habc of the three-phase Hall sensors at this time, and use the combined Hall state value Habc as an index to determine the open-loop angle θ. open (H) stores the data in an array.

5. The three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to claim 1, characterized in that: In step 4, at least two Hall transition events are continuously recorded, and the change sequence of the combined Hall state value Habc is analyzed. Based on the different change patterns of the three-phase Hall signals during forward and reverse rotation (forward rotation is 101-100-110-010-011-001, reverse rotation is 001-011-010-110-100-101), the actual rotation direction of the brushless motor under the current open-loop drive is determined.

6. The three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to claim 1, characterized in that: In step 5, within one complete Hall cycle of the brushless motor, the three-phase Hall sensor of the brushless motor generates six different state values. These six different state values ​​correspond to the boundaries of six Hall sectors, respectively. Based on the open-loop angle θ corresponding to at least six Hall transition points recorded in step 3... open (index), combined with the rotation direction determined in step 4, the actual area of ​​each Hall sector is calculated.

7. The three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to claim 1, characterized in that: In step 1, the voltage equation of the brushless motor in the DQ axis coordinate system is precisely expressed as follows: ， Among them, U d For direct-axis voltage, U q Let R be the quadrature-axis voltage, R be the stator winding resistance of the brushless motor, and i be the voltage across the quadrature axis. d For direct-axis current, i q For quadrature-axis current, L d For a direct-axis inductor, L q For quadrature axis inductance, W e Let ψ be the angular velocity of the brushless motor. f It is a permanent magnet flux linkage.

8. The three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to claim 1, characterized in that: In step 7, after the brushless motor switches to closed-loop vector control mode, during the duration of the given Hall state between any two adjacent Hall transition points, the real-time angle of the brushless motor rotor is estimated by a back EMF observer, a flux linkage observer, or an interpolation algorithm. The real-time angle of the rotor is constrained within the actual area determined by the actual Hall sector information corresponding to the current Hall state.

9. A three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to any one of claims 2 to 8, characterized in that: In step 3, during the open-loop drag phase, the voltage setpoint in the two-phase rotating coordinate system is configured: A constant voltage setpoint Vd is applied to the D-axis corresponding to the direct axis, and the voltage setpoint Vq corresponding to the Q-axis is set to zero. The open-loop angle θ generated in step 2 is then used. open (index) The inverse Parker transform and space vector pulse width modulation are executed sequentially to generate a PWM signal for driving the three-phase inverter. This generates a signal on the stator side of the motor with a spatial direction aligned with the D-axis and a spatial electrical angle that varies with the open-loop angle θ. open (index) A synchronously rotating virtual magnetic field; controlling the rotation and reversal of the virtual magnetic field, with respect to the open-loop angle θ. open (index) implements the following management strategy. 1) Angle accumulation: In each current control cycle, the open-loop angle θ is accumulated. open The accumulated value of (index) is incremented once, and the increment Δθ is a configurable non-zero integer; 2) Angle calibration, open-loop angle θ open The numerical range of (index) is linearly mapped to a complete electrical cycle. The angle numerical range is set from 0 to 65535, corresponding to an electrical angle from 0° to 360°. 3) Direction switching, when the open-loop angle θ open When the cumulative increment of (index) reaches the value corresponding to two complete electric cycles, the sign of the increment Δθ is reversed, and the rotation direction of the virtual magnetic field is reversed. 4) The process terminates, but continues in both forward and reverse rotations until the open-loop angle θ is reached. open The open-loop dragging process ends when the value of (index) returns to zero.

10. The three-phase Hall effect self-learning method for vector control of a brushless motor in a high-speed pick-and-place machine according to claim 9, characterized in that: In step 4, the external interrupt input pins of the microcontroller are connected to the output signals of the three-phase Hall sensors of the brushless motor, and configured to edge-triggered mode. When the rotor of the brushless motor rotates and causes a jump in the output level of any Hall sensor, the external interrupt is triggered. In the corresponding interrupt service routine, the following operations are performed: Step 4.1: Read and record the open-loop angle θ generated in step 2 in real time. open (H); Step 4.2: Synchronously read and record the combined Hall state value Habc of the three-phase Hall sensors at this time; Step 4.3: Using the combined Hall state value Habc as the index identifier, the open-loop angle value θopen(H) is stored as the associated data in a pre-allocated storage array, thereby establishing the mapping relationship between the Hall state and the corresponding electrical angle. Repeatedly execute steps 4.1 to 4.3 above, record at least two Hall state transition events and their corresponding angle data, analyze the change sequence of continuously recorded combined Hall state values ​​Habc, and match them with the preset motor forward and reverse state transition rules to determine the actual rotation direction of the brushless motor under the current open-loop drag. After executing steps 1 to 5, the three-phase Hall self-learning of the brushless motor vector control is automatically completed. When the brushless motor is connected to the driver, regardless of how the Hall wiring sequence and the motor power line wiring sequence are connected, the actual angle sector value can be calculated after the three-phase Hall self-learning is completed, and the brushless motor can be driven normally.

Images