Frequency adaptive position-sensorless control method for permanent magnet synchronous motor

Abstract

The invention discloses a frequency adaptive position sensorless control method for a permanent magnet synchronous motor in the technical field of permanent magnet synchronous motors, and the method comprises the steps: adding an adaptive vector filter into an internal model of an extended state observer, so as to obtain an estimated counter electromotive force without phase lag; filtering third and fifth harmonics in the estimated counter electromotive force by using an adaptive vector filter; meanwhile, the motor rotating speed and the rotor position angle are estimated through an orthogonal phase-locked loop and are finally fed back to a vector control system; the frequency self-adaptive extended state observer can observe the back electromotive force under a relatively low bandwidth, so that the problem that bandwidth improvement and high-frequency noise filtering of the extended state observer cannot be considered at the same time is solved; according to the method, the frequency is adaptively changed along with the working frequency of the motor, so that the estimation precision of the counter electromotive force in a relatively wide rotating speed range is ensured, and high-performance position-sensorless control of the permanent magnet synchronous motor in the wide rotating speed range is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of permanent magnet synchronous motor control, and particularly relates to a sensorless control method for permanent magnet synchronous motors in a wide speed range. Background Art

[0002] Permanent magnet synchronous motors have the advantages of high efficiency, high reliability, and good dynamic performance, and have broad application prospects in the fields of industrial automation, intelligent household appliances, wind power generation, and new energy vehicles. To achieve high-performance control of permanent magnet synchronous motors, the acquisition of motor speed and rotor position information is crucial. However, mechanical position sensors have the disadvantages of increasing system volume, weight, and cost, and there is a risk of failure in harsh environments such as high temperature and strong vibration, which affects system reliability. The extended state observer method is an effective sensorless control scheme. However, this method has a low-pass filtering characteristic, and the estimated back electromotive force will have a phase lag and multiple harmonics, and the accuracy of motor speed and rotor position estimation is poor. Therefore, it is of great practical significance to study an extended state observer method without phase lag and with good control performance to improve the sensorless control performance of permanent magnet synchronous motors in a wide speed range.

[0003] In order to achieve sensorless control of permanent magnet synchronous motors in a wide speed range, domestic and foreign scholars have conducted relevant research on the extended state observer method. The Chinese invention patent "Predictive Control Method for Permanent Magnet Synchronous Motors Based on Extended State Observer" (Patent No.: CN114172425B) discloses a predictive control method for permanent magnet synchronous motors based on an extended state observer, and designs an extended state observer based on the super-twisting algorithm with a parameter adaptive law, which avoids the prediction error caused by parameter mismatch. The algorithm has simple parameter tuning, does not add additional computational load, and can converge within a finite time. Although the parameter tuning of this method is relatively simple, the model structure is complex and difficult to implement. The Chinese invention patent "Auto-disturbance Rejection Controller Based on Extended State Observer with Finite-Time Convergence" (Patent No.: CN110764418B) discloses an auto-disturbance rejection controller based on an extended state observer with finite-time convergence, including: a tracking differentiator, an extended state observer with finite-time convergence, and a non-linear state error feedback controller. This method expands the total system disturbance into a new state, estimates the total disturbance in finite time and compensates it in real time, which can ensure that the motor still has good dynamic characteristics during low-speed operation. However, the observer parameters are complex and not easy to adjust. Therefore, how to balance the estimation performance of rotor position and speed during low-speed operation of the motor on the basis of the extended state observer method, while reducing the multiple harmonics of the back electromotive force, and realizing sensorless control of permanent magnet synchronous motors in a wide speed range is the main consideration factor of the present invention. Summary of the Invention

[0004] Aiming at the deficiencies existing in the prior art, the present invention provides a frequency adaptive sensorless control method for a permanent magnet synchronous motor, which reduces the phase delay of the back electromotive force estimation of the permanent magnet synchronous motor, eliminates the low-pass filtering characteristic of the extended state observer, and filters out the multiple harmonic components of the estimated back electromotive force obtained by the observer; in addition, the observer frequency changes adaptively with the motor operating frequency, ensuring the estimation accuracy of the back electromotive force within a wide speed range, realizing the sensorless control of the permanent magnet synchronous motor in a wide speed range; improving the performance of the sensorless control of the permanent magnet synchronous motor during transient operation and the estimation accuracy of the motor speed and rotor position angle.

[0005] The object of the present invention is achieved as follows: A frequency adaptive sensorless control method for a permanent magnet synchronous motor, comprising the following steps:

[0006] Step 1) Design of a frequency adaptive extended state observer: Rewrite the voltage equations of the dq axes of the motor, obtain the voltage equations and back electromotive force of the αβ axes through coordinate transformation, and then expand the external unknown disturbance into a new state to construct an extended state observer; add an adaptive vector filter to the internal model of the extended state observer to replace the integral link, use the αβ axis voltage and current as inputs, and obtain the estimated back electromotive force without phase lag using the frequency adaptive extended state observer.

[0007] Step 2) Estimation of the motor speed and rotor position: Use the estimated back electromotive force obtained in Step 1) as the input, and use a normalized orthogonal phase-locked loop to extract the position information contained in the back electromotive force, and finally obtain the motor speed and rotor position;

[0008] Step 3) Vector control operation of the motor without a position sensor: Use the motor speed estimated in Step 2) as the feedback, compare it with the given speed, obtain the input of the current loop through a proportional-integral controller, then perform coordinate transformation on the sampled three-phase current of the motor, subtract it from the given current, pass through two proportional-integral controllers, and output to the space vector pulse width modulator, finally establishing a double closed-loop speed regulation system for the permanent magnet synchronous motor with speed as the outer loop and current as the inner loop.

[0009] Further, the specific construction steps of the extended state observer in Step 1) include:

[0010] The voltage equations of the dq axes of the permanent magnet synchronous motor are rewritten as

[0011]

[0012] where, u d 、u q and i d 、i q respectively represent the stator voltages and currents on the dq axes, and R sdenotes the stator resistance, L d and L q denote the stator inductances on the dq axes, p is the differential operator, ω e and ψ f denote the electrical angular velocity and the permanent magnet flux linkage of the motor respectively; Coordinate transformation of the above equation gives:

[0013]

[0014] where, u α and u β and i α and i β denote the stator voltages and currents on the αβ axes respectively, e α and e β are the back electromotive forces, defined as follows:

[0015]

[0016] where, θ e is the rotor position angle of the motor. According to the main idea of the extended state observer, the system input and the lumped disturbance are expressed as:

[0017]

[0018] The corresponding extended state observer is established as follows:

[0019]

[0020] where, z 1 and z 2 denote the estimated values of the stator current and the unknown disturbance respectively, e 1 denotes the estimation error, β 1 and β 2 are the observer gains and are selected to be the same on the αβ axes;

[0021] When the extended state observer becomes stable, the estimation error will tend to zero, the estimated value of the unknown disturbance will converge, and then the estimated value of the back electromotive force will be obtained.

[0022] Furthermore, the specific design steps of the adaptive vector filter in step 1) include:

[0023] The transfer function of the scalar form of the adaptive vector filter is:

[0024]

[0025] where, [x α x β T and [y α y β ​​T respectively represent the input vector and the desired output vector, ω 0 is the resonance frequency, k f is the gain of the adaptive vector filter;

[0026] Then, through the concept of complex vectors, introduce

[0027]

[0028] Then the vector form transfer function of the adaptive vector filter is:

[0029] .

[0030] Furthermore, the specific design steps of the frequency adaptive extended state observer in step 1) include:

[0031] Embed the adaptive vector observer into the internal model of the extended state observer to replace the integral link. The frequency adaptive extended state observer is expressed as

[0032]

[0033]

[0034] where α and β represent the estimated values of the stator currents on the αβ axes, ω e is the electrical angular velocity of the motor, β 1 and β 2 are the gains of the frequency adaptive extended state observer; Using the scale and bandwidth parameter tuning method, the gain parameters are parameterized as follows:

[0035]

[0036] where ω c represents the bandwidth of the frequency adaptive extended state observer.

[0037] Furthermore, the specific steps of step 2) include:

[0038] The phase-locked loop extracts the rotor speed and position information from the estimated back electromotive force, including a phase detector, a loop filter, and a voltage-controlled oscillator; The phase-locked loop automatically synchronizes the output signal with the input signal through closed-loop regulation, thereby indirectly extracting the phase information of the input signal, and then estimating the fundamental frequency signal by adjusting the error through a proportional-integral controller, and integrating the fundamental frequency to obtain the estimated rotor position; The bandwidth of the phase-locked loop changes with the operating state of the motor. When the motor switches to wide-speed range operation, different phase-locked loop gains need to be calculated; Using an orthogonal phase-locked loop with back electromotive force normalization; Its equivalent position error can be expressed as:

[0039]

[0040] The rotor position error is used to obtain the motor speed through a PI controller; the rotor position angle is obtained by integrating the estimated rotor speed; the closed-loop transfer function of the back-electromotive force normalized quadrature phase-locked loop is expressed as:

[0041]

[0042] where k p and k i are the proportional gain and integral gain of the PI controller, respectively.

[0043] Furthermore, the specific steps of step 3 include:

[0044] After comparing the estimated motor speed obtained in the previous step with the given speed, the current given value i q * is obtained through a proportional-integral controller. At the same time, the three-phase current of the permanent magnet synchronous motor is sampled, and the dq-axis currents are obtained after coordinate transformation. The current feedback values i d and i q are subtracted from the current given value i d * and i q * respectively. After that, the voltage u α and u β are obtained through a proportional-integral controller and an inverse Park transformation; then, a PWM wave is synthesized through space vector pulse width modulation and output to the three-phase inverter; finally, a double closed-loop vector control system of the permanent magnet synchronous motor with speed as the outer loop and current as the inner loop is established.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0046] 1) The frequency adaptive extended state observer proposed by the present invention eliminates the low-pass filtering characteristic of the extended state observer and reduces the phase lag of the estimated back-electromotive force;

[0047] 2) The present invention proposes an adaptive scheme to make the frequency of the extended state observer change adaptively with the motor operating frequency, realizing high-performance sensorless control of the permanent magnet synchronous motor in a wide speed range;

[0048] 3) The present invention uses an adaptive vector filter to filter out the multiple harmonic components of the estimated back-electromotive force and reduces the estimation errors of the motor speed and rotor position;

[0049] 4) The present invention extracts position information from the back electromotive force of the rotor by using an improved extended state observer, eliminating the mechanical position sensor, reducing the volume and cost of the system, and enhancing the stability of the system. Description of the Drawings

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without creative efforts.

[0051] Figure 1 It is the overall block diagram of the sensorless control of the permanent magnet synchronous motor.

[0052] Figure 2 It is the block diagram of the extended state observer.

[0053] Figure 3 It is the block diagram of the frequency adaptive extended state observer.

[0054] Figure 4 It is the comparison diagram of the true position angle and the estimated position angle.

[0055] Figure 5 It is the position error diagram of the permanent magnet synchronous motor.

[0056] Figure 6 It is the comparison diagram of the true rotational speed and the estimated rotational speed.

[0057] Figure 7 It is the rotational speed error diagram of the permanent magnet synchronous motor. Detailed Embodiments

[0058] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0059] As Figure 1 shown, a frequency adaptive sensorless control method for a permanent magnet synchronous motor includes:

[0060] Step 1, the design of the frequency adaptive extended state observer;

[0061] The voltage equations of the dq axes of the permanent magnet synchronous motor can be rewritten as

[0062]

[0063] Among them, u d , u q and i d , i q respectively represent the stator voltage and current on the dq axes, R s and L respectively represent the stator resistance and inductance, p is the differential operator, ω e and ψ f respectively represent the electrical angular velocity of the motor and the permanent magnet flux linkage. Coordinate transformation of the above formula gives

[0064]

[0065] Among them, u α , u β and i α , i β respectively represent the stator voltage and current on the αβ axes, e α and e β are the back electromotive forces, defined as follows:

[0066]

[0067] According to the main idea of the extended state observer, the system input and lumped disturbance can be expressed as:

[0068]

[0069] Then, the corresponding extended state observer is established as follows:

[0070]

[0071] Among them, z 1 and z 2 respectively represent the estimated values of the stator current and the unknown disturbance, e 1 represents the estimation error, β 1 and β 2 are the observer gains and are selected to be the same on the αβ axes.

[0072] The structure diagram of the extended state observer is as shown in Figure 2 . When the observer becomes stable, the estimation error will tend to zero, the estimated value of the unknown disturbance will converge, and then, the estimated value of the back electromotive force can be obtained.

[0073] The adaptive vector filter will bring amplitude attenuation or phase shift to the target component, but will suppress other components of the input signal and has good positive and negative frequency selectivity. Its transfer function in scalar form is:

[0074]

[0075] Among them, [x α xβ T and [y α y β T represent the input vector and the desired output vector respectively, and ω 0 is the resonance frequency, and k f is the gain of the adaptive vector filter.

[0076] Then, through the concept of complex vectors,

[0077]

[0078] the vector form transfer function of the adaptive vector filter can be obtained as:

[0079]

[0080] As can be seen from the above formula, when k f is larger, the filter bandwidth becomes wider, but the selectivity is weaker, and more high-frequency noise interference will be introduced. Therefore, the design of k f should be reasonable to ensure good dynamic response and filtering performance at the same time. If the frequency of the fundamental back electromotive force can be obtained in real time and set as the resonance frequency ω 0 , the third and fifth harmonics of the estimated back electromotive force can be filtered out by using the adaptive vector filter, so as to realize the adaptive change of the frequency with the motor operating frequency and ensure the estimation accuracy of the back electromotive force in a wide speed range.

[0081] The adaptive vector observer is implanted into the internal model of the extended state observer to replace the integral link. The structural block diagram of the frequency adaptive extended state observer is as Figure 3 shown and can be expressed as

[0082]

[0083]

[0084] where, α and β represent the estimated values of the αβ-axis stator currents, and ω e is the electrical angular velocity of the motor. As can be seen from the above formula, the frequency adaptive extended state observer no longer has the low-pass filtering characteristic, and the estimated back electromotive force will not produce a phase lag, so compensation is not required. β 1 and β 2 are the gains of the frequency adaptive extended state observer and can be adjusted independently of the system parameters, ensuring high robustness of the disturbance estimation. For the convenience of parameter adjustment and theoretical analysis, the scale and bandwidth parameter tuning method is adopted. The gain parameters are parameterized as follows:​​

[0085]

[0086] Among them, ω c represents the bandwidth of the frequency adaptive extended state observer. A higher ω c helps to improve the system response rate, but will increase the sensitivity of the observer to noise.

[0087] Furthermore, the implementation process of the said step 2 is as follows:

[0088] The phase-locked loop can extract the rotor speed and position information from the estimated back electromotive force. It includes a phase detector, a loop filter, and a voltage-controlled oscillator. The main principle of the phase-locked loop is to automatically synchronize the output signal with the input signal through closed-loop regulation, thereby indirectly extracting the phase information of the input signal. Then, the proportional-integral controller is used to adjust the error to estimate the fundamental frequency signal, and the fundamental frequency is integrated to obtain the estimated rotor position. The bandwidth of the phase-locked loop varies with the operating state of the motor. When the motor switches to wide-speed-range operation, different phase-locked loop gains need to be calculated. Therefore, in order to eliminate the influence of the change in the phase-locked loop bandwidth on the speed and rotor position estimation accuracy during wide-speed-range operation of the motor, an orthogonal phase-locked loop with back electromotive force normalization is adopted. Its equivalent position error can be expressed as:

[0089]

[0090] The rotor position error is used to obtain the motor speed through a PI controller. By integrating the estimated rotor speed, the rotor position angle can be obtained. The closed-loop transfer function of the orthogonal phase-locked loop with back electromotive force normalization can be expressed as:

[0091]

[0092] Among them, k p and k i are the proportional gain and integral gain of the PI controller respectively. The larger the bandwidth of the phase-locked loop, the faster its dynamic performance and the faster the tracking speed of the motor speed, but it will reduce the filtering performance and introduce more high-frequency noise interference to the estimated speed and rotor position. Therefore, the proportional gain and integral gain should be reasonably selected to ensure its estimation performance.

[0093] Furthermore, the implementation process of the said step 3 is as follows:

[0094] After comparing the estimated motor speed obtained in the previous step with the given speed, the current given value i q * is obtained through a proportional-integral controller. At the same time, the three-phase current of the permanent magnet synchronous motor is sampled, and the dq-axis currents are obtained after coordinate transformation. The current feedback values i d and i qWith the current reference value i d * and i q * After taking the difference, the voltage u α and u β are obtained after passing through a proportional-integral controller and an inverse Park transformation. Then, through space vector pulse width modulation, a PWM wave is synthesized and output to a three-phase inverter. Finally, a double closed-loop vector control system for a permanent magnet synchronous motor with speed as the outer loop and current as the inner loop is established.

[0095] Figure 4 and Figure 5 are respectively the comparison diagram of the actual position angle and the estimated position angle of the permanent magnet synchronous motor and the position error diagram. It can be seen from the figure that the estimated position angle and the actual position angle of the motor in the proposed method are basically coincident, and the position error is only 0.02 rad, showing a good position estimation effect.

[0096] Figure 6 and Figure 7 are respectively the comparison diagram of the actual speed and the estimated speed of the permanent magnet synchronous motor and the speed error diagram. It can be seen from the figure that the estimated speed of the motor can well follow the reference speed, and the speed error is only 10 r / min.

[0097] The description of the above embodiments is only used to help understand the method of the present invention and its core idea. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. A method for frequency adaptive position sensorless control of a permanent magnet synchronous motor, characterized in that: The following steps are involved: Step 1) Design of frequency-adaptive extended state observer: rewrite the voltage equation of the motor dq axis, obtain the αβ axis voltage equation and back EMF through coordinate transformation, and then expand the external unknown disturbance into a new state to construct an extended state observer; add an adaptive vector filter to the internal model of the extended state observer to replace the integral link, take the αβ axis voltage and current as input, and use the frequency-adaptive extended state observer to obtain the estimated back EMF without phase lag. Step 2) Estimation of motor speed and rotor position: Taking the estimated back electromotive force obtained in step 1) as input, a normalized orthogonal phase-locked loop is used to extract the position information contained in the back electromotive force, and finally the motor speed and rotor position are obtained; Step 3) Motor vector control operation without position sensor: The motor speed estimated in step 2) is used as feedback, compared with the given speed, and then the input of the current loop is obtained through the proportional-integral controller. Then the sampled three-phase current of the motor is transformed into coordinates, subtracted from the given current, and output to the space vector pulse width modulator through two proportional-integral controllers. Finally, a dual closed-loop permanent magnet synchronous motor speed control system with speed as the outer loop and current as the inner loop is established.

2. A method for frequency adaptive position sensorless control of a permanent magnet synchronous motor according to claim 1, characterized in that: The specific construction steps of the extended state observer in step 1) include: The voltage equation of the permanent magnet synchronous motor dq axis is rewritten as Among them, u d 、u q and i d 、i q Represent the stator voltage and current on the dq axis respectively, R s Indicates stator resistance, L d , L q represents the stator inductance on the dq axis, p is the differential operator, ω e and ψ f Represent the electrical angular velocity of the motor and the permanent magnet flux respectively; transform the above equation into: Among them, u α 、u β and i α 、i β They represent the stator voltage and current on the α and β axes respectively, and e α and e β is the back electromotive force, defined as follows: Among them, θ e is the motor rotor position angle. According to the main idea of ​​the extended state observer, the system input and lumped disturbance are expressed as: The corresponding extended state observer is established as follows: Among them, z1 and z2 represent the estimated values ​​of stator current and unknown disturbance respectively, e1 represents the estimation error, β1 and β2 are the observer gains selected to be the same on the αβ axis; When the extended state observer becomes stable, the estimation error will tend to zero, the estimated value of the unknown disturbance will converge, and then the estimated value of the back EMF is obtained.

3. A method for frequency adaptive position sensorless control of a permanent magnet synchronous motor according to claim 2, characterized in that: The specific design steps of the adaptive vector filter in step 1) include: The transfer function of the adaptive vector filter in scalar form is: Among them, [x α x β ] T and [y α y β ] T denote the input vector and the desired output vector respectively, ω0 is the resonant frequency, k f is the gain of the adaptive vector filter; Then, through the concept of complex vector, we introduce Then the vector form transfer function of the adaptive vector filter is: 。 4. A method for frequency adaptive position sensorless control of a permanent magnet synchronous motor according to claim 3, characterized in that: The specific design steps of the frequency adaptive extended state observer in step 1) include: The adaptive vector observer is implanted into the internal model of the extended state observer to replace the integral link. The frequency adaptive extended state observer is expressed as in, α and β represents the estimated value of the αβ axis stator current, ω e is the electrical angular velocity of the motor, β1 and β2 are the gains of the frequency adaptive extended state observer; the scale and bandwidth parameter tuning method is used, and the gain parameterization is as follows: Among them, ω c represents the bandwidth of the frequency adaptive extended state observer.

5. A method for frequency adaptive position sensorless control of a permanent magnet synchronous motor according to claim 4, characterized in that: The specific steps of step 2) include: The phase-locked loop extracts the rotor speed and position information from the estimated back-EMF, and includes a phase detector, a loop filter, and a voltage-controlled oscillator. The phase-locked loop automatically synchronizes the output signal with the input signal through closed-loop regulation, thereby indirectly extracting the phase information of the input signal, and then estimates the baseband signal by adjusting the error through the proportional-integral controller, and integrates the baseband to obtain the estimated rotor position. The bandwidth of the phase-locked loop varies with the operating state of the motor. When the motor switches to a wide speed range, different phase-locked loop gains need to be calculated. An orthogonal phase-locked loop with normalized back-EMF is used. Its equivalent position error can be expressed as: The rotor position error is used to obtain the motor speed through the PI controller; the rotor position angle is obtained by integrating the estimated rotor speed; the orthogonal phase-locked loop closed-loop transfer function of the normalized back-EMF is expressed as: Among them, k p and k i are the proportional gain and integral gain of the PI controller respectively.

6. A method for frequency adaptive position sensorless control of a permanent magnet synchronous motor according to claim 5, characterized in that: The specific steps of step 3 include: After comparing the estimated motor speed obtained in the previous step with the given speed, the current given value i is obtained through the proportional integral controller. q * At the same time, the three-phase current of the permanent magnet synchronous motor is sampled, and the current of the dq axis is obtained after coordinate transformation. The current feedback value i d and i q With the current given value i d * and i q * After the difference is made, the voltage u is obtained after the proportional integral controller and the inverse Park transformation. α and u β ; Then, the PWM wave is synthesized through space vector pulse width modulation and output to the three-phase inverter; Finally, a double closed-loop permanent magnet synchronous motor vector control system with speed as the outer loop and current as the inner loop is established.

Images