Switched reluctance motor rotor position estimation method with feedforward phase-locked loop

Abstract

The application belongs to the field of motor control, and discloses a switched reluctance motor rotor position estimation method with feedforward phase-locked loop, which has the following steps: obtaining the original electrical angle and original mechanical speed of the current sampling moment output by the switched reluctance motor (SMO); feedforward electrical angular velocity estimation; phase deviation calculation and processing; starting gradual progress and initial synchronization; PI control and anti-integral saturation; steady-state speed filtering; position integration, precision correction and angle fusion. The switched reluctance motor sliding mode observer position estimation method can receive the original angle and speed signals containing noise or jitter from the SMO, and solve the problems of slow dynamic response, easy overshoot and large steady-state chattering of the traditional PLL through adaptive PLL and feedforward control, so that the smooth, low-delay and high-precision rotor position and speed are output.

Description

Technical Field

[0001] This invention belongs to the field of motor control and relates to a method for estimating the rotor position of a switched reluctance motor with a feedforward phase-locked loop. Background Technology

[0002] Switched reluctance motors (SRMs) have attracted much attention due to their advantages such as low cost, high fault tolerance, and adaptability to harsh operating conditions. Traditional control relies on mechanical sensors to obtain rotor position, which has drawbacks such as high cost and susceptibility to failure. This has prompted sensorless technology to become the core alternative. Among existing sensorless estimation strategies, sliding mode observers (SMOs) are widely used due to their strong algorithmic robustness and insensitivity to perturbations of internal motor parameters. However, it is important to note that the inherent high-frequency switching characteristics of SMOs can lead to severe chattering and harmonic interference in the estimated position signal. Directly using them for control can exacerbate motor torque ripple and even cause the system to lose synchronization.

[0003] To suppress high-frequency chattering in the output signal of a switched reluctance motor (SMO), cascaded phase-locked loops (PLLs) are commonly used in engineering to extract continuous and smooth rotor position and speed information. However, conventional PLLs face the challenge of balancing filtering depth and dynamic tracking performance when dealing with wide speed ranges and sudden load changes. Literature indicates that conventional fixed-gain PLLs often require reduced loop bandwidth to ensure steady-state accuracy in strong noise environments, which can cause severe phase lag in transient conditions such as rapid acceleration or deceleration. Furthermore, when there is a large deviation in the initial position or the system encounters external disturbances, conventional PI controllers are prone to integral saturation, leading to start-up overshoot and prolonged system oscillations. To address these issues, this invention proposes a position estimation method for switched reluctance motors based on a feedforward adaptive PLL sliding mode observer. By introducing feedforward speed estimation, adaptive frequency conversion filtering, and an anti-integral saturation mechanism, this method effectively resolves the contradiction between the slow dynamic response and steady-state chattering of conventional PLLs, achieving low-delay, high-precision, and smooth tracking of the rotor position and speed of a switched reluctance motor. Summary of the Invention

[0004] This invention proposes a rotor position estimation method for switched reluctance motors with a feedforward phase-locked loop (SMO). It can receive raw angle and speed signals containing noise or jitter from the SMO. Through adaptive PLL and feedforward control, it solves the problems of slow dynamic response, easy overshoot and large steady-state jitter of traditional PLLs, and outputs smooth, low-delay and high-precision rotor position and speed.

[0005] The technical solution of the present invention: A method for rotor position estimation of a switched reluctance motor with a feedforward phase-locked loop, comprising the following steps: Step 1: Obtain the raw electrical angle k of the current time output of the switched reluctance motor (SMO). and the original mechanical rotation speed at the current moment ; Step 2: Estimation of feedforward electric angular velocity; Original mechanical speed Converted to raw electric angular velocity : in, This represents the number of rotor poles of the switched reluctance motor. For the original electric angular velocity The feedforward electric angular velocity is obtained by performing adaptive first-order low-pass filtering. : in, For real-time filtering gain, This is the filter cutoff frequency. Set the sampling period. The filter cutoff angular frequency is adaptively adjusted according to the original mechanical rotation speed. in, , To preset the low-speed dividing speed, This is the cutoff frequency for the low-speed feedforward filter. This is the cutoff frequency for the feedforward filter in the medium-to-high speed range; Step 3: Phase deviation calculation and processing; Calculate the original electrical angle Optimize electrical angle with the previous sampling time Phase deviation between: The phase deviation is subjected to angular periodic principal value processing to ensure it falls within the range of... Equivalent principal value interval: in Represents a symbolic function; The phase bias after principal valueization is then subjected to weighted smoothing: in These are weighting coefficients. .

[0006] Then, amplitude limiting is performed to obtain the optimized phase deviation: in, This represents the phase deviation after low-pass filtering. The set limit value, Represents the saturation limiting function. The phase deviation is after weighted smoothing; The phase deviation after low-pass filtering is obtained by the following formula: in, For phase deviation filter gain, This is the filter cutoff frequency set for low-pass filtering of phase deviation; Step 4: Initiate gradual and initial synchronization; Set the motor running time to During the motor startup phase, a gradual gain that increases linearly with time is introduced: in, , Preset the startup gradual time; The progressive gain is used to adjust the proportional and integral terms of the PI controller in subsequent step 5, so that the control effect of the phase-locked loop gradually increases during the motor startup phase. When the initial synchronization conditions are met: At that time, compensation is applied to the integral term of the PI controller: in, This is the integral term of the PI controller at the previous sampling time. To preset the initial synchronization time, To preset the initial synchronization phase deviation trigger threshold, The preset initial synchronization integral compensation coefficient; Step 5: PI control and anti-integral saturation; Calculate the relative velocity deviation between the optimized electric angular velocity and the original electric angular velocity at the previous sampling time: in, To prevent the default positive number with a denominator of zero, To optimize electric angular velocity; Calculate the discrete derivative term for optimizing the phase deviation: Constructing a composite control error: in, , and Preset weighting coefficients; The proportional term and integral increment are calculated using a PI controller with asymptotic gain: in, For proportional gain, This is the integral gain; Adaptive adjustment of the integral increment based on optimized phase deviation: in, For positive integral compensation coefficients, The negative integral attenuation coefficient is... Adjust the upper limit threshold for integration. The lower limit threshold is adjusted for integral adjustment; The velocity compensation is obtained from the proportional term, integral term, and integral increment: Adding the feedforward electric angular velocity to the velocity compensation amount, we obtain the estimated electric angular velocity to be limited: Based on the original electric angular velocity The amplitude is dynamically generated to limit the output speed: in, This is the dynamic upper limit coefficient. This is the dynamic upper limit coefficient, which is adjusted according to actual control requirements; Limiting the amplitude of the estimated electric angular velocity yields a preliminary estimated electric angular velocity: When the estimated electric angular velocity to be limited exceeds the limiting boundary, and the composite control error still drives the integral term toward saturation, specifically... hour or hour This triggers the anti-integral saturation mechanism, setting the integral increment to zero and decaying the integral term: in, This is the integral attenuation coefficient; Finally, update the points system and apply a points limit: in, This is the integral limit value; When the motor running time exceeds the steady-state judgment time And the relative velocity deviation is less than the steady-state velocity deviation threshold. hour: Slowly decay the integral term: in, The coefficient for slow decay; Step 6: Steady-state velocity filtering; When the following steady-state condition is met: At that time, the preliminary estimated electric angular velocity obtained in step 5 is... By performing low-pass filtering, the optimized electric angular velocity is obtained: in, To preliminarily estimate the filter gain for electric angular velocity, The cutoff angular frequency is set when using a low-pass filter for the preliminary estimated electrical angular velocity; When the steady-state condition is not met, the optimal electric angular velocity is: Step 7: Position integration, precision correction, and angle fusion; Using the optimized electric angular velocity obtained in step 6, the position is integrated to obtain the integral estimated electric angle: Fine-tuning of the integral estimated electrical angle based on optimized phase deviation: in, This is the positive angle fine-tuning coefficient. This is the negative angle fine-tuning coefficient. For fine-tuning the angle threshold; Use the remainder function Estimated electrical angle after period normalization for: Set angle fusion threshold When the following angle fusion conditions are met: Then, the estimated electrical angle after period normalization is weighted and fused with the original electrical angle to obtain the optimized electrical angle: in, Weighting for angle fusion; When the above angle fusion conditions are not met, the optimized electrical angle is: Finally, the optimized electric angular velocity obtained in step 6 is... Converted to optimized machine speed : .

[0007] The beneficial effects of this invention are: 1. This invention converts the original mechanical speed output by SMO into the original electrical angular velocity, and obtains the feedforward electrical angular velocity through adaptive first-order low-pass filtering. The feedforward electrical angular velocity is then introduced into the angle tracking PLL, so that the optimized electrical angular velocity can reflect the trend of motor speed change in advance, reduce the dynamic lag caused by relying solely on phase deviation feedback, and improve the position tracking response speed and dynamic estimation accuracy under motor acceleration / deceleration and load disturbance conditions.

[0008] 2. This invention adaptively adjusts the feedforward filter cutoff frequency based on the original mechanical rotation speed, and performs angle period principalization, weighted smoothing, filtering, and amplitude limiting on the phase deviation. This not only maintains good response capability in the low-speed range, but also effectively suppresses high-frequency noise and chattering in the sliding mode observer output in the medium and high-speed range, while avoiding phase jumps at the electrical angle period boundary, thereby improving the continuity, smoothness, and anti-interference capability of the position estimation results.

[0009] 3. In the startup phase, the present invention introduces a gradual gain that increases linearly with time, which gradually enhances the proportional and integral actions of the PI controller and avoids sudden changes in the control quantity at startup. At the same time, when the initial synchronization conditions are met, synchronization compensation is applied to the integral term, so that the estimated electrical angle can quickly approach the original electrical angle. Therefore, it can take into account both the fast synchronization performance and the smooth control performance in the startup phase, and reduce the phase overshoot and speed shock in the startup phase.

[0010] 4. This invention, through dynamic speed limiting and anti-integral saturation mechanisms, stops the integral increment and decays the integral term when the estimated electrical angular velocity exceeds the limiting boundary and the composite control error still drives the integral term toward saturation, effectively suppressing integral saturation. At the same time, it filters the electrical angular velocity under steady-state conditions and weights and fuses the estimated electrical angle with the original electrical angle under steady-state small error conditions, which can reduce steady-state speed fluctuations, reduce long-term integral drift, and improve the steady-state accuracy and reliability of the final output optimized electrical angle, optimized electrical angular velocity, and optimized mechanical speed. Attached Figure Description

[0011] Figure 1This is a block diagram of a feedforward-based adaptive angle tracking PLL system.

[0012] Figure 2 This is a comparison chart of the rotor positions before and after optimization.

[0013] Figure 3 This is a comparison chart of the rotational speed before and after optimization.

[0014] Figure 4 It is a magnified view of the comparison chart of rotational speeds before and after optimization. Detailed Implementation

[0015] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and technical solutions.

[0016] To address the problems of slow dynamic response, large overshoot, and steady-state chattering inherent in traditional sensorless speed control methods for switched reluctance motors based on SMOs, this invention proposes a position estimation method for switched reluctance motors based on a feedforward adaptive PLL using a sliding mode observer. The principle is as follows: Figure 1 As shown. Its embodiment (taking a 12 / 8-pole switched reluctance motor as an example, the number of rotor poles...) =8, sampling time =10us, given speed 500r / min, given torque 1Nm) The specific implementation method is as follows: Step 1: Obtain the output of the sliding mode observer for the switched reluctance motor. and ; Step 2: Estimation of feedforward electric angular velocity.

[0017] Will Converted to .

[0018] right Obtained by adaptive first-order low-pass filtering ,Will Set to 200 r / min, when When the speed is less than 200 r / min, the cutoff angular frequency is set to 600 rad / s; otherwise, it is set to 400 rad / s to reduce acceleration lag. Step 3: Phase deviation calculation and processing.

[0019] calculate ,like Greater than Then perform principal value processing on the angle period to make it fall into The equivalent principal value interval is obtained. .

[0020] when Weighting coefficients at time Weighted smoothing with a value of 0.7 yields the following results: .

[0021] right conduct A low-pass filter of 800 rad / s was applied to remove high-frequency noise, resulting in... and will Limitation at Within, obtain ; Step 4: Initiate gradual and initial synchronization.

[0022] set up A gradual gain, increasing linearly with time, is introduced during the motor startup phase, with a duration of 0.2s. .set up It takes 0.3 seconds. It is 0.02 rad. The value is 10, when the initial synchronization condition is met: Apply step compensation to the integral term of the PI controller at that time. To enhance initial synchronization; Step 5: PI control and anti-integral saturation.

[0023] Calculate relative velocity deviation ,calculate Discrete derivative term Construct a composite control error: Construct a PI controller with asymptotic gain, and adjust the proportional gain. Set to 150, integral gain Set to 650, combined and Calculate the proportional term and integral increment .

[0024] set up , , , ,according to right Perform adaptive adjustment and update in real time .

[0025] calculate , .

[0026] set up It is 1.01. It is 0.99, for By limiting the amplitude, we can obtain .

[0027] set up It is 0.9. For 150, when or At this time, the anti-integral saturation mechanism is triggered: , Then update the points system and apply a points limit: .

[0028] set up For 1 second, It is 0.01. The value is 0.99, when the steady-state condition is met: At that time, the integral term is gradually decayed: .

[0029] Step 6: Steady-state velocity filtering.

[0030] set up The value is 250 rad / s. When the steady-state condition is met, for When low-pass filtering is performed and the steady-state condition is not met, no processing is performed, and the final result is obtained. .

[0031] Step 7: Position integration and precision calibration.

[0032] calculate ,set up It is 0.025. It is 0.025. It is 0.03 rad, according to right Make minor adjustments: right Periodic normalization process .

[0033] set up If the value is 0.05 rad, then the fusion condition is: set up The value is 0.02. Based on this fusion condition, the calculation is... : Finally Converted to And update the state variables for use in the next cycle; Depend on Figure 2 It can be seen that the electrical angle output by the SMO exhibits some jitter, and the traditional PLL output shows significant phase lag. However, by employing the adaptive angle tracking PLL of this invention, the electrical angle can better follow the changes in the original position while maintaining good smoothness. Figure 3 and Figure 4 It can be seen that the SMO speed has obvious high-frequency jitter. The traditional PLL response is relatively lagging and fluctuates. After adopting the method of the present invention, the speed can reach a stable state more quickly and remain near the given speed. The steady-state speed fluctuation is significantly reduced and the output is smoother.

[0034] The parameters with specific numerical values ​​given in the above embodiments are provided for the purpose of illustrating the technical solution of the present invention. Those skilled in the art can adjust these parameters according to the structural parameters of the switched reluctance motor, the operating speed range, the sampling period, the noise level of the sliding mode observer, and the dynamic response requirements of the control system. Such adjustments do not affect the essence of the technical solution of the present invention.

Claims

1. A method for estimating the rotor position of a switched reluctance motor with a feedforward phase-locked loop, characterized in that, The steps are as follows: Step 1: Obtain the raw electrical angle at the current sampling time k of the switched reluctance motor (SMO). and original mechanical speed ; Step 2: Estimation of feedforward electric angular velocity; Step 3: Phase deviation calculation and processing; Step 4: Initiate gradual and initial synchronization; Step 5: PI control and anti-integral saturation; Step 6: Steady-state velocity filtering; Step 7: Position integration, precision correction and angle fusion.

2. The rotor position estimation method for a switched reluctance motor with a feedforward phase-locked loop according to claim 1, characterized in that, The specific implementation process of step 2 is as follows: Original mechanical speed Converted to raw electric angular velocity : in, This represents the number of rotor poles of the switched reluctance motor. For the original electric angular velocity The feedforward electric angular velocity is obtained by performing adaptive first-order low-pass filtering. : in, For real-time filtering gain, This is the filter cutoff frequency. Set the sampling period. The filter cutoff frequency is adaptively adjusted based on the original mechanical rotation speed. in, , To preset the low-speed dividing speed, This is the cutoff frequency for the low-speed feedforward filter. This is the cutoff frequency for the feedforward filter in the medium-to-high speed range.

3. The rotor position estimation method for a switched reluctance motor with a feedforward phase-locked loop according to claim 2, characterized in that, The specific implementation process of step 3 is as follows: Calculate the original electrical angle Optimize electrical angle with the previous sampling time Phase deviation between: The phase deviation is subjected to angular periodic principal value processing to ensure it falls within the range of... Equivalent principal value interval: in, Represents a symbolic function; The phase deviation after principal-valued angular periodization is then subjected to weighted smoothing: in, These are weighting coefficients. ; Then, amplitude limiting is performed to obtain the optimized phase deviation: in, This represents the phase deviation after low-pass filtering. The set limit value, Represents the saturation limiting function. The phase deviation is after weighted smoothing; The phase deviation after low-pass filtering is obtained by the following formula: in, The filter gain is the phase deviation. This is the filter cutoff frequency set for low-pass filtering of phase deviation.

4. The rotor position estimation method for a switched reluctance motor with a feedforward phase-locked loop according to claim 3, characterized in that, The specific implementation process of step 4 is as follows: Set the motor running time to During the motor startup phase, a gradual gain that increases linearly with time is introduced: in, , Preset the startup gradual time; The progressive gain is used to adjust the proportional and integral terms of the PI controller in subsequent step 5, so that the control effect of the phase-locked loop gradually increases during the motor startup phase. When the initial synchronization conditions are met: At that time, compensation is applied to the integral term of the PI controller: in, This is the integral term of the PI controller at the previous sampling time. To preset the initial synchronization time, To preset the initial synchronization phase deviation trigger threshold, This is the preset initial synchronization integral compensation coefficient.

5. The rotor position estimation method for a switched reluctance motor with a feedforward phase-locked loop according to claim 4, characterized in that, The specific implementation process of step 5 is as follows: Calculate the relative velocity deviation between the optimized electric angular velocity and the original electric angular velocity at the previous sampling time: in, To prevent the default positive number with a denominator of zero, To optimize electric angular velocity; Calculate the discrete derivative term for optimizing the phase deviation: Constructing a composite control error: in, , and Preset weighting coefficients; The proportional term and integral increment are calculated using a PI controller with asymptotic gain: in, For proportional gain, This is the integral gain; Adaptive adjustment of the integral increment based on optimized phase deviation: in, For positive integral compensation coefficients, The negative integral attenuation coefficient is... Adjust the upper limit threshold for integration. The lower limit threshold is adjusted for integral adjustment; The velocity compensation is obtained from the proportional term, integral term, and integral increment: Adding the feedforward electric angular velocity to the velocity compensation amount, we obtain the estimated electric angular velocity to be limited: Based on the original electric angular velocity The amplitude is dynamically generated to limit the output speed: in, This is the dynamic upper limit coefficient. This is the lower limit coefficient for dynamic amplitude limiting, which is adjusted according to actual control requirements; Limiting the amplitude of the estimated electric angular velocity yields a preliminary estimated electric angular velocity: When the estimated electric angular velocity to be limited exceeds the limiting boundary, and the composite control error still drives the integral term toward saturation, specifically... hour or hour This triggers the anti-integral saturation mechanism, setting the integral increment to zero and decaying the integral term: in, This is the integral attenuation coefficient; Finally, update the points system and apply a points limit: in, This is the integral limit value; When the motor running time exceeds the steady-state judgment time And the relative velocity deviation is less than the steady-state velocity deviation threshold. hour: Slowly decay the integral term: in, This is a slow decay coefficient.

6. The rotor position estimation method for a switched reluctance motor with a feedforward phase-locked loop according to claim 5, characterized in that, The specific implementation process of step 6 is as follows: When the following steady-state condition is met: At that time, the preliminary estimated electric angular velocity obtained in step 5 is... By performing low-pass filtering, the optimized electric angular velocity is obtained: in, To preliminarily estimate the filter gain for electric angular velocity, The cutoff angular frequency is set when using a low-pass filter for the preliminary estimated electrical angular velocity; When the steady-state condition is not met, the optimal electric angular velocity is: 。 7. The rotor position estimation method for a switched reluctance motor with a feedforward phase-locked loop according to claim 6, characterized in that, The specific implementation process of step 7 is as follows: Using the optimized electric angular velocity obtained in step 6, the position is integrated to obtain the integral estimated electric angle: Fine-tuning of the integral estimated electrical angle based on optimized phase deviation: in, This is the positive angle fine-tuning coefficient. This is the negative angle fine-tuning coefficient. For fine-tuning the angle threshold; Use the remainder function Estimated electrical angle after period normalization for: Set angle fusion threshold When the following angle fusion conditions are met: Then, the estimated electrical angle after period normalization is weighted and fused with the original electrical angle to obtain the optimized electrical angle: in, Weighting for angle fusion; When the above angle fusion conditions are not met, the optimized electrical angle is: Finally, the optimized electric angular velocity obtained in step 6 is... Converted to optimized machine speed : 。

Images