A linear hall-based permanent magnet synchronous motor rotor position detection method

Abstract

The application discloses a permanent magnet synchronous motor rotor position detection method based on linear Hall, belongs to the field of motor rotor position detection, installs three linear Hall sensors, collects and corrects the output signals, extracts linear intervals and carries out linearization and splicing, finally synthesizes rotor electric angles, and continuously detects positions, the method has the advantages of no phase delay, high precision, less resource occupation, low cost, and is suitable for high-performance permanent magnet synchronous motor control systems.

Description

Technical Field

[0001] This invention belongs to the field of motor rotor position detection technology, and more specifically, relates to a method for rotor position detection of permanent magnet synchronous motor based on linear Hall effect. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in industrial automation and intelligent manufacturing due to their high efficiency, high power density, and excellent control performance. The accuracy of motor control is highly dependent on the accurate detection of rotor position. Currently, commonly used position detection methods include photoelectric encoders, rotary transformers, and Hall effect sensors. Photoelectric encoders offer high accuracy but are expensive and have poor resistance to harsh environments; rotary transformers are robust and durable but require additional processing circuitry, making the system complex; Hall effect sensors are low-cost but have relatively poor position accuracy and are typically used for startup or coarse positioning. Existing linear Hall effect decoding methods include the arctangent method (CORDIC) and phase-locked loop (PLL). The arctangent method (CORDIC) is sensitive to signal amplitude, DC bias, and magnetic field quality, which can easily cause angle errors. It has strict requirements for magnetic field quality and is prone to position errors, and it has no noise suppression capability. The phase-locked loop (PLL) has high requirements for installation accuracy. Any installation error will cause periodic errors in position estimation, resulting in torque pulsation. The system is complex and difficult to calibrate. When the motor speed changes rapidly, it is prone to delays, which are converted into position errors.

[0003] Therefore, a method for detecting the rotor position of a permanent magnet synchronous motor that consumes few MCU resources, has no phase delay, and has high decoding accuracy is needed. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a rotor position detection method for permanent magnet synchronous motors based on linear Hall effect, which can meet the requirements of low MCU resource consumption, no phase delay and high decoding accuracy.

[0005] The present invention provides a rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect, comprising the following steps: Step 1: Install three linear Hall sensors on the motor body, the first linear Hall sensor... Second linear Hall sensor and the third linear Hall sensor The three linear Hall sensors are evenly distributed in the time domain at a mechanical angle of 30° or an electrical angle of 120°, and are placed on the side of the induction magnetic ring at a consistent distance from the magnetic ring. Step 2: Sampling to obtain the first linear Hall sensor Second linear Hall sensor and the third linear Hall sensor The sensed signal, and the sampled signal is normalized; Step 3: Label the characteristic signal using the first linear Hall sensor. Second linear Hall sensor Or a third linear Hall sensor The index of the sine cycle containing the peak value of any Hall signal is 1, which is used to mark the absolute position of the rotor; Step 4: Correct the signals from the three linear Hall sensors to obtain approximate sine wave signals with consistent amplitude; Step 5: Determine the sector within a single electrical cycle for the three linear Hall sensors that have been acquired, and divide the electrical cycle into six sectors based on the corrected signal; Step 6: Extract the upper and lower intersection points between different signals within a complete electrical cycle, as the start and end points of their respective linear regions. The intersection point number is taken as the first linear Hall sensor as the starting point. Extract the linear interval at the signal intersection point and linearize it. Step 7: Linearization of the single-segment truncated signal, that is, further linearizing the single-segment truncated signal segment into a linear line segment; Step 8: The six linear regions described in Step 5 are spliced ​​together to form a complete electrical cycle. The linear regions are extracted within a complete mechanical cycle and spliced ​​sequentially in a manner where the six linear regions form a 360° electrical angle. Step 9: Convert the spliced ​​signal into rotor electrical angles using fixed coefficients to obtain the rotor's position information; Step 10: Repeat the operation process of steps 5 to 9 within different mechanical cycles to achieve continuous position detection.

[0006] As a further improvement of the present invention, the electrode mentioned in step one is a motor equipped with four pairs of pole sensing magnetic rings, and three linear Hall sensors are placed at the same distance from the sensing magnetic rings and on the side of the sensing magnetic rings.

[0007] As a further improvement of the present invention, the sector determination in step five is based on the amplitude of the three-phase signal.

[0008] As a further improvement of the present invention, the signal correction in step four includes amplitude correction and bias correction, so that the amplitudes of the three-phase signals are consistent and the zero bias is zero.

[0009] As a further improvement of the present invention, the linearization process described in step seven is to process the truncated nonlinear signal segment into a linear line segment through discretization.

[0010] As a further improvement of the present invention, the first linear Hall sensor, the second linear Hall sensor and the third linear Hall sensor are not saturated at any position of the rotor to prevent the measurement signal from being distorted and unable to proceed to the next step.

[0011] As a further improvement of the present invention, the maximum design speed of the motor in step one is not higher than the maximum limiting speed of the permanent magnet synchronous motor.

[0012] The formula for calculating the maximum speed limit of a permanent magnet synchronous motor is: ; Where P N The number of pole pairs of the motor, F is the maximum signal frequency, F S The sampling frequency.

[0013] As a further improvement of the present invention, the formula for calculating the rotor electrical angle in step nine is as follows: ; Where p is the splicing signal value. The fixed proportional coefficient.

[0014] As a further improvement of the present invention, the three linear Hall sensors are respectively the first linear Hall sensor. Second linear Hall sensor and the third linear Hall sensor The formula for calculating the amplitude correction factor is: , The formula for calculating the bias correction coefficient is as follows: , in , , , , , The first linear Hall sensor Second linear Hall sensor and the third linear Hall sensor The peak and trough values ​​are given by N, where N is the number of sine cycles.

[0015] Compared with the prior art, the advantages of this invention are as follows: the direct decoding method of signal interception and linear splicing avoids complex arctangent or phase-locked loop operations, significantly reducing the occupation of MCU computing resources; at the same time, since there is no lag processing such as filtering, the generated position signal has no phase delay, ensuring the control performance of the motor under high dynamic conditions. The amplitude and bias correction in the early stage effectively overcomes the signal non-idealization problem caused by sensor inconsistency, installation error and magnetic field distortion; the subsequent linearization processing further improves the linearity of the position signal, thereby realizing high-precision rotor position detection; and the use of only linear Hall effect sensors results in low cost. Attached Figure Description

[0016] Figure 1 This invention relates to a first linear Hall sensor for a rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect. Second linear Hall sensor and the third linear Hall sensor A schematic diagram of the actual mechanical installation; Figure 2 This invention relates to a first linear Hall sensor for a rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect. Second linear Hall sensor and the third linear Hall sensor A schematic diagram of the original collected sensor signal after per-unit scaling; Figure 3 This invention relates to a first linear Hall sensor for a rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect. Schematic diagram for finding the maximum peak value within a mechanical cycle; Figure 4 This invention relates to a corrected first linear Hall sensor for a rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect. The corrected second linear Hall sensor The corrected third linear Hall sensor A schematic diagram of the sector curve; Figure 5 This invention relates to a corrected first linear Hall sensor for a rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect. The corrected second linear Hall sensor The corrected third linear Hall sensor A schematic diagram of the curves at the upper and lower intersection points; Figure 6 The first linear Hall sensor, after correction, is used in the rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect, when the motor is running at 200 rpm, according to the present invention. The corrected second linear Hall sensor The corrected third linear Hall sensor A schematic diagram of the electrical angle curve; Figure 7 This invention relates to a rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect sensors. linearization A schematic diagram; Figure 8 The first linear Hall sensor, after correction, is used in the rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect, when the motor is running at 3000 rpm, according to the present invention. The corrected second linear Hall sensor The corrected third linear Hall sensor A schematic diagram of the electrical angle curve.

[0017] Explanation of the labels in the diagram: First linear Hall sensor 1; second linear Hall sensor 2; third linear Hall sensor 3. Detailed Implementation

[0018] Specific Implementation Example 1: Please refer to... Figure 1-8 This invention relates to a rotor position detection method for a permanent magnet synchronous motor based on linear Hall effect sensors. It utilizes an MCU with AD sampling capabilities and establishes an external signal sampling circuit. The MCU's AD function is initialized to configure three linear Hall effect sensors, with the sampling frequency matching the motor control loop period. Four pairs of induction magnetic rings are mounted on the motor body's sensors, with the three linear Hall effect sensors positioned 4.5mm away from the induction magnetic rings and placed on the side of the induction magnetic rings. When the magnetic pole is directly facing the linear Hall effect sensor, a peak voltage will be induced. Includes the following steps: Step 1: The first linear Hall sensor 1, the second linear Hall sensor 2, and the third linear Hall sensor 3 are uniformly distributed in the time domain at a mechanical angle of 30°. Step 2: Sequentially sample the first linear Hall sensor 1, the second linear Hall sensor 2, and the third linear Hall sensor 3, converting the analog voltage signals into digital signals. The resulting digital signals are as follows: , , And will be standardized according to consistent standards; Step 3: For the first linear Hall sensor 1 The peak values ​​of multiple sinusoidal periods are sampled sequentially to determine the absolute position; In this implementation example, sampling is performed on four mechanical cycles. The peak value is the most basic measurement data. In this embodiment, a 4-level design is adopted, so a total of 16 consecutive peak data are sampled. The peak value is synchronously averaged to calculate the average peak voltage within each electrical cycle. The calculation method is as shown in Formula 1: ; Search The maximum value is used as the characteristic signal If the index of the sine cycle is 1 (Index=1), then each linear Hall will have four complete sine wave cycles within one mechanical cycle, which are marked in chronological order as index 2 (Index=2), index 3 (Index=3), index 4 (Index=4), and so on in a cycle. Step 4: Sample the data from the first linear Hall sensor 1. Second linear Hall sensor 2 and the third linear Hall sensor 3 The peak and trough values ​​are used for coefficient calculation and signal correction. Data from 64 mechanical cycles are recorded sequentially as the basis for correction. After removing suspicious data, the average value is used for calculation to improve the accuracy of the correction coefficient. Linear Hall effect sensors are arranged sequentially, and the sampled data are as follows: , , , , , Where N is the number of sine cycles, and N = 64 x 4 = 256; the collected peak and valley values ​​are processed, and the amplitude correction coefficient of each linear Hall is calculated. and bias correction coefficient The signal is subjected to equal amplitude processing, and the amplitude of all signals is equalized to A. The collected data is then processed. The amplitude correction coefficient is calculated as shown in Formula 2, and the bias correction coefficient is calculated as shown in Formula 3. ; After amplitude equalization processing, the signal is corrected as follows: ; in: First linear Hall sensor 1 The coefficient for amplitude correction to A, For the second linear Hall sensor 2 The coefficient for amplitude correction to A, For the third linear Hall sensor 3 The coefficient for amplitude correction to A; For the first linear Hall sensor 1 Amplitude correction to a bias of 0 at the midpoint. For the second linear Hall sensor 2 Amplitude correction to a bias of 0 at the midpoint. For the third linear Hall sensor 3 Amplitude correction to a bias where the midpoint is 0; For the first linear Hall sensor 1 Corrected data, For the second linear Hall sensor 2 Corrected data, For the third linear Hall sensor 3 Corrected data; Step 5: Analyze the data corrected by the first linear Hall sensor 1 as described in Step 4. Data corrected by the second linear Hall sensor 2 And the corrected data from the third linear Hall sensor 3 The amplitude is used to calibrate the sector within a single electrical cycle to obtain the absolute position within a single mechanical cycle; the calibration method is as shown in Formula 5, and the sector calculation method is as shown in Formula 6: ; Within a complete electrical angle cycle, six sectors are divided sequentially, and the sequencing occurs as follows: ; in: for and The judgment coefficient, for and The judgment coefficient, for and The judgment coefficient; Step Six: Correct the first linear Hall sensor 1 2. Corrected second linear Hall sensor and the corrected third linear Hall sensor 3 Perform the upper crossover point between two-phase linear Hall effect sensors. and its lower intersection Sampling markers; In theory: ; Six linear segments were obtained within a complete electrical angle period, and were labeled as follows: ; Step 7: The signals from Line 1 to Line 6 described in Step 6 are not perfectly linear curves. Therefore, the extracted signals are further linearized to obtain... To improve the final decoding effect, Line1 is discretized into M points according to the current amplitude, and each point is determined by... The values ​​of the linear function formed; in: ; Step 8: Signal Arranged sequentially within one electrical cycle, the coefficient P related to the electrical angle is obtained, and its value is determined as shown in Formula 8: ; Step 9: The difference between the maximum and minimum values ​​of P is 12 times. Its corresponding electrical angle is ; Then the coefficient for: ; P multiplied by a fixed coefficient The electrical angle of the rotor is then obtained. The calculation of the electrical angle is as shown in Formula 10: ; Step 10: Repeat steps 5 to 9 within different mechanical cycles to obtain complete and continuous rotor position information.

[0019] This method obtains accurate rotor information by intercepting the original signal, and with the help of correction techniques, it can achieve extremely high position accuracy; thus, it achieves high-precision control at extremely low cost.

[0020] In a further embodiment, after using this method for position decoding, the effect is verified: the motor is controlled at 200 RPM and 3000 RPM respectively using FOC control mode, and the data is compared. Figure 6 and Figure 8 The decoding effect of the position information was displayed. At the lowest and highest speeds of the test motor, there was no phase delay, and the decoded rotor position information was complete and continuous.

[0021] In a further embodiment, according to Shannon's sampling theorem, the sampling frequency needs to be greater than twice the maximum frequency of the signal. In this embodiment, the sampling frequency is set to... Number of pole pairs P of motor N Since it has 4 stages, the maximum speed limit of the permanent magnet synchronous motor is: ; In this implementation example, the motor is designed for a maximum speed of 3000 RPM, which is far below the maximum speed limit. This is sufficient to ensure that the original analog signal is recovered without distortion.

[0022] In a further embodiment, the first linear Hall sensor 1, the second linear Hall sensor 2, and the third linear Hall sensor 3 are not saturated at any position on the rotor to prevent the measurement signal from being distorted and thus preventing further operation.

[0023] The above content and structure describe the basic principles, main features, and advantages of the product of this invention. Those skilled in the art should understand that the embodiments and descriptions above are only illustrative of the principles of this invention. Various changes and modifications can be made to this invention without departing from its spirit and scope. All such changes and modifications fall within the scope of this invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting the rotor position of a permanent magnet synchronous motor based on linear Hall effect, characterized in that: Includes the following steps: Step 1: Install three linear Hall sensors on the motor body, which are evenly distributed in the time domain at a mechanical angle of 30° or an electrical angle of 120°. Step 2: Sample the sensing signals from each Hall sensor and standardize them; Step 3: Mark the characteristic signal. Use the sine cycle index of the peak value of any Hall signal from the three linear Hall sensors as 1 to complete the determination of the rotor's absolute position. Step 4: Perform signal correction on the three acquired linear Hall sensors. The signal correction includes amplitude correction and bias correction to make the amplitude of the three-phase signals consistent and the zero bias zero, so as to obtain an approximate sine wave signal with consistent amplitude. Step 5: Determine the sector within a single electrical cycle for the three linear Hall sensors. Use the amplitude of the three-phase signal as the basis for dividing the six sectors. Divide the six sectors within the electrical cycle according to the corrected signal. Step 6: Extract the linear interval at the signal crossover point and linearize it. The signal crossover point is the upper crossover point and the lower crossover point between the two-phase linear Hall effect sensors. Step 7: Linearization of the single-segment truncated signal, that is, further linearizing the single-segment truncated signal segment into a linear line segment; Step 8: Splice the six sectors described in Step 5 into a complete electrical cycle, extract the linear region within a complete mechanical cycle, and sequentially splice the six linear regions in a manner that each represents a 360-degree electrical angle. The linear region is the area formed by the linear line segments obtained in Step 7. Step 9: Convert the spliced ​​signal into rotor electrical angles using fixed coefficients to obtain the rotor's position information; Step 10: Repeat the operation process of steps 5 to 9 within different mechanical cycles to achieve continuous position detection.

2. The method for rotor position detection of a permanent magnet synchronous motor based on linear Hall effect as described in claim 1, characterized in that: The motor mentioned in step one is a motor equipped with a 4-pair induction magnetic ring. The three linear Hall sensors are placed on the side of the induction magnetic ring at the same distance from the induction magnetic ring.

3. The method for rotor position detection of a permanent magnet synchronous motor based on linear Hall effect as described in claim 1, characterized in that: The linearization step in step seven involves processing the truncated nonlinear signal segment into a linear line segment using discretization techniques.

4. The method for rotor position detection of a permanent magnet synchronous motor based on linear Hall effect as described in claim 1, characterized in that: The maximum design speed of the motor mentioned in step one shall not exceed the maximum limiting speed of the permanent magnet synchronous motor. The formula for calculating the maximum limiting speed of the permanent magnet synchronous motor is as follows: ； in Where is the number of pole pairs of the motor, and F is the maximum signal frequency. The sampling frequency.

5. The method for rotor position detection of a permanent magnet synchronous motor based on linear Hall effect as described in claim 1, characterized in that: The formula for calculating the rotor electrical angle mentioned in step nine is as follows: ； Where p is the splicing signal value. The fixed coefficient is mentioned above.

6. The method for rotor position detection of a permanent magnet synchronous motor based on linear Hall effect as described in claim 1, characterized in that: The three linear Hall sensors are labeled as follows: The first linear Hall sensor (1), labeled as The second linear Hall sensor (2) and labeled as The formula for calculating the amplitude correction coefficient of the third linear Hall sensor (3) is as follows: ， The formula for calculating the bias correction coefficient is as follows: ， in , , , , , These are the first linear Hall sensor (1). , Second linear Hall sensor (2) And the third linear Hall sensor (3) The peak and trough values ​​are N, where N is the number of sine cycles.

7. The method for rotor position detection of a permanent magnet synchronous motor based on linear Hall effect as described in claim 1, characterized in that: The first linear Hall sensor (1), the second linear Hall sensor (2), and the third linear Hall sensor (3) are not saturated at any position on the rotor.

Images