Permanent magnet synchronous motor sensorless control method based on disturbance observer

Abstract

The invention discloses a permanent magnet synchronous motor sensorless control system based on a disturbance observer. Firstly, estimation errors caused by mismatch of motor parameters to effective flux linkage and rotor angles are analyzed; a disturbance observer is designed through a generalized second-order integrator, so that the adverse effect of direct current bias caused by motor parameter mismatch and a pure integrator on the system is compensated. Then, the suppression capability of the provided disturbance observer on direct current bias and parameter mismatch is analyzed; and finally, designing the gain of the observer through a Routh criterion and a stability margin. According to the method, adverse factors brought to the system by parameter mismatch are effectively improved, direct current bias brought by a pure integrator can be suppressed, and the robustness of the system under various working conditions is improved.

Description

Technical Field

[0001] This invention relates to the field of motor drive control technology, and more specifically to a sensorless control method for a permanent magnet synchronous motor based on a disturbance observer. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) have been widely used in industry due to their advantages in efficiency and power density, and have also become a current research hotspot. Vector control technology is the mainstream control technology for PMSMs. This technology improves the control performance of motor current and speed by decoupling the control of the motor's torque current and excitation current. To achieve decoupled control, the drive needs to be equipped with position sensors to obtain rotor position information. However, the signals from position sensors are susceptible to interference from the external environment, especially in harsh environments, and the wiring between the sensor and the controller can also reduce the reliability of the drive system.

[0003] Traditional integrating flux linkage observers obtain stator flux linkage through pure integration, resulting in poor robustness to motor parameter mismatches. Furthermore, due to DC drift caused by integration, the motor is prone to divergence during operation. To address the issues of DC bias and poor robustness associated with pure integrators, it is of great significance to research a flux linkage observer that is highly robust to motor parameters and suppresses DC bias. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a sensorless control method for permanent magnet synchronous motors based on a disturbance observer. This method improves the accuracy of flux linkage and rotor angle estimation, suppresses control system harmonics, and enhances the system's dynamic response speed.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] In a first aspect, the present invention proposes a sensorless control method for a permanent magnet synchronous motor based on a disturbance observer, the method comprising the following steps:

[0007] S1: The impact of motor parameter mismatch on angle estimation in a sensorless control system was analyzed.

[0008] S2: Design a disturbance observer to compensate for the effects of parameter mismatch and system harmonics.

[0009] S3: Analyze the performance of the disturbance observer on harmonic suppression and parameter mismatch compensation.

[0010] S4: Analyze the stability of the disturbance observer and select an appropriate value based on the system bandwidth requirements.

[0011] Furthermore, in step S1, the sensorless model of the permanent magnet synchronous motor using the perturbation observer is...

[0012] First, the voltage model of the permanent magnet synchronous motor in vector form in the αβ coordinate system is as follows:

[0013] u s =R s i s +pψ s

[0014] In the formula, u s =[u sα ,u sβ ] T i s =[i sα i sβ ] T R represents the stator voltage vector and current vector, respectively. s For the stator resistance, ψ s =[ψ sα ,ψ sβ ] T This represents the stator flux linkage vector.

[0015] Among them, the stator magnetic flux ψ s Represented as

[0016]

[0017] In the formula L d ,L q Representing the d-q axis inductance, ψ f Represents permanent magnet flux linkage.

[0018] The effective flux linkage can be obtained by integrating the back electromotive force.

[0019] ψ A =∫(u s -R s i s -L q ·pi s )dt=∫e r dt

[0020] Where e r =[e r_α ,e r_β ] T It is represented as the back electromotive force vector.

[0021] The expression for the effective flux linkage considering parameter mismatch and disturbance compensation can be rewritten as follows:

[0022]

[0023] In the formula This is to compensate for the disturbance in the system. As can be seen from the above formula, parameter mismatch in the motor will introduce a deviation in the estimated effective flux linkage. Without... The compensation may cause excessive deviation in the effective flux estimation, leading to the failure of rotor angle observation.

[0024] Furthermore, due to inverter nonlinearity, uncertainty of initial integral value, and current sampling error, using a pure integrator will cause DC bias and harmonic pollution in the observation flux, making the sensorless system unable to work.

[0025] Furthermore, in S2, a disturbance observer is designed to compensate for the stator flux bias and harmonics caused by motor parameter mismatch and pure integration.

[0026] The second-order generalized integrator (SOGI) is widely used in grid-connected control and motor control due to its flexibility. It can be considered as an orthogonal signal generator with bandpass filtering characteristics, and it does not introduce amplitude attenuation or phase shift at the center frequency. Its transfer function can be expressed as...

[0027]

[0028] In the formula, ω is the center frequency and k is the damping ratio, which determines the filter bandwidth.

[0029] Based on the characteristics of SOGI, the effective flux linkage obtained by the pure integrator is filtered by SOGI, and the difference between this and the stator flux linkage is divided by L. q The estimated stator current is obtained, and the difference between the estimated stator current and the actual stator current is used as a proportional gain k. p After the step, the estimated perturbation in the effective flux can be obtained. By feeding the disturbance back into the back electromotive force, the parameter mismatch and the interference from the pure integrator can be compensated for.

[0030] The transfer function of the disturbance observer obtained through the above design can be expressed as follows:

[0031]

[0032] In the formula, k1 is the damping ratio of SOGI, and k2 is the feedback gain for estimating the disturbance.

[0033] The resulting effective flux linkage is

[0034]

[0035] The transfer function G of the back electromotive force and the effective flux linkage can be obtained. DO (s). By setting the center frequency ω to This allows the observer to achieve frequency adaptation.

[0036] Furthermore, in S3, the ability of the designed disturbance observer to suppress system harmonics and parameter mismatch is further analyzed.

[0037] Suppression capability of DC component: The final value theorem can be obtained by applying the flux linkage obtained using a perturbation observer.

[0038]

[0039] In the formula, e0 is the DC component vector in the back electromotive force. It can be seen that the DC component in the input back electromotive force is completely eliminated after passing through the disturbance observer.

[0040] Operating frequency Amplitude and phase response at: Substitute G DO From (s), we can obtain

[0041]

[0042] The above equation shows that the output amplitude and phase of the designed disturbance observer at the center frequency are integrals of the input back electromotive force, and the effective flux amplitude will not be attenuated.

[0043] Furthermore, in S4, the selected values ​​for the gain parameters in the disturbance observer are given.

[0044] G DO The characteristic equation of the transfer function (s) can be expressed as follows:

[0045]

[0046] Based on the Routh stability criterion, a Routh table can be compiled.

[0047]

[0048] For the system to be stable, the first column of the Routh meter must be all positive numbers; therefore, the center frequency should be positive and set to the absolute value of the rotational speed. And k1 > 0, k2 > 0.

[0049] G DO The open-loop transfer function of (s) is

[0050]

[0051] Based on automatic control theory, the cutoff frequency ω of the proposed disturbance observer is... c and phase margin Defined as:

[0052]

[0053] The expressions for k1 and k2 can be derived from the above formula.

[0054]

[0055] Based on the above formula and the system Bode diagram, k1 and k2 are tuned to 0.707 and 0.5, respectively.

[0056] The beneficial effects of this invention are:

[0057] 1. The disturbance observer proposed in this invention solves the problems of effective flux estimation deviation and rotor angle estimation failure caused by motor parameter mismatch, improves the robustness of the system, and can cope with more complex working conditions.

[0058] 2. The disturbance observer proposed in this invention solves the DC bias problem caused by traditional integrating flux observers, and achieves stable operation of the system over a wide speed range.

[0059] 2. The disturbance observer proposed in this invention has good suppression capability for high-order harmonics of the observed flux linkage and can also effectively reduce harmonics during high-speed operation. Attached Figure Description

[0060] Figure 1 It is a vector graph of the effective flux linkage before and after parameter mismatch.

[0061] Figure 2 This is a structural diagram of a sensorless control method for a permanent magnet synchronous motor based on a disturbance observer, according to an embodiment of the present invention.

[0062] Figure 3 These are Bode plots of the proposed perturbation observers for different values ​​of k1.

[0063] Figure 4 These are Bode plots of the proposed perturbation observers for different values ​​of k2.

[0064] Figure 5 It is a simulation diagram of rotor speed and angle under a given sudden change in rotational speed.

[0065] Figure 6 Motor parameter L q The change is 0.5L q and 2L q The simulation diagram afterward.

[0066] Figure 7 It is the motor parameter R s The change is 2R s The simulation diagram afterward. Detailed Implementation Plan

[0067] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0068] Example 1

[0069] Figure 2 This is a diagram of a sensorless control architecture based on a disturbance observer. This embodiment proposes a sensorless control method for a disturbance observer, which includes the following steps:

[0070] S1: The impact of motor parameter mismatch on angle estimation in a sensorless control system was analyzed.

[0071] S2: Design a disturbance observer to compensate for the effects of parameter mismatch and system harmonics.

[0072] S3: Analyze the performance of the disturbance observer on harmonic suppression and parameter mismatch compensation.

[0073] S4: Analyze the stability of the disturbance observer and select an appropriate value based on the system bandwidth requirements.

[0074] I. The impact of parameter mismatch on angle estimation in sensorless systems

[0075] First, the voltage model of the permanent magnet synchronous motor in vector form in the αβ coordinate system is as follows:

[0076] u s =R s i s +pψ s

[0077] In the formula, u s =[u sα ,u sβ ] T i s =[i sα i sβ ] T R represents the stator voltage vector and current vector, respectively. s For the stator resistance, ψ s =[ψ sα ,ψ sβ ] T This represents the stator flux linkage vector.

[0078] Among them, the stator magnetic flux ψ s Represented as

[0079]

[0080] In the formula L d ,L q Representing the d-q axis inductance, ψ f Represents permanent magnet flux linkage.

[0081] The effective flux linkage can be obtained by integrating the back electromotive force.

[0082] ψ A =∫(u s -R s i s -L q ·pi s )dt=∫e r dt

[0083] Where e r =[e r_α ,e r_β ] T It is represented as the back electromotive force vector.

[0084] The expression for the effective flux linkage considering parameter mismatch and disturbance compensation can be rewritten as follows:

[0085]

[0086] In the formula This is to compensate for the disturbance in the system. As can be seen from the above formula, parameter mismatch in the motor will introduce a deviation in the estimated effective flux linkage. Without... The compensation may cause excessive deviation in the effective flux estimation, leading to the failure of rotor angle observation.

[0087] Figure 1 The values ​​of ΔR are given under ideal conditions and under conditions of motor parameter mismatch. s =R s ΔL q =-0.5L q The vector diagram of the effective flux linkage shows that the effective flux linkage cannot be aligned with the d-axis, resulting in a large angular error. Furthermore, due to inverter nonlinearity, uncertainty of the initial integral value, and current sampling error, using a pure integrator will cause DC bias and harmonic pollution in the observed flux linkage, making the sensorless system unable to operate.

[0088] II. Principles and Design of Disturbance Observers

[0089] The second-order generalized integrator (SOGI) is widely used in grid-connected control and motor control due to its flexibility. It can be considered as an orthogonal signal generator with bandpass filtering characteristics, and it does not introduce amplitude attenuation or phase shift at the center frequency. Its transfer function can be expressed as...

[0090]

[0091] In the formula, ω is the center frequency and k is the damping ratio, which determines the filter bandwidth.

[0092] Based on the characteristics of SOGI, the effective flux linkage obtained by the pure integrator is filtered by SOGI, and the difference between this and the stator flux linkage is divided by L. q The estimated stator current is obtained, and the difference between the estimated stator current and the actual stator current is used as a proportional gain k. p After the step, the estimated perturbation in the effective flux can be obtained. By feeding the disturbance back into the back electromotive force, the parameter mismatch and the interference from the pure integrator can be compensated for.

[0093] The transfer function of the disturbance observer obtained through the above design can be expressed as follows:

[0094]

[0095] In the formula, k1 is the damping ratio of SOGI, and k2 is the feedback gain for estimating the disturbance.

[0096] The resulting effective flux linkage is

[0097]

[0098] The transfer function G of the back electromotive force and the effective flux linkage can be obtained. DO (s). By setting the center frequency ω to This allows the observer to achieve frequency adaptation. Figure 2 Block diagram of the designed disturbance observer

[0099] III. Performance Analysis of the Proposed Disturbance Observer

[0100] Suppression capability of DC component: The final value theorem can be obtained by applying the flux linkage obtained using a perturbation observer.

[0101]

[0102] In the formula, e0 is the DC component vector in the back electromotive force. It can be seen that the DC component in the input back electromotive force is completely eliminated after passing through the disturbance observer.

[0103] Operating frequency Amplitude and phase response at: Substitute G DO From (s), we can obtain

[0104]

[0105] The above equation shows that the output amplitude and phase of the designed disturbance observer at the center frequency are integrals of the input back electromotive force, and the effective flux amplitude will not be attenuated.

[0106] IV. Provide the selected values ​​for the gain parameter in the disturbance observer.

[0107] G DO The characteristic equation of the transfer function (s) can be expressed as follows:

[0108]

[0109] Based on the Routh stability criterion, a Routh table can be compiled.

[0110]

[0111] For the system to be stable, the first column of the Routh meter must be all positive numbers; therefore, the center frequency should be positive and set to the absolute value of the rotational speed. And k1 > 0, k2 > 0.

[0112] G DO The open-loop transfer function of (s) is

[0113]

[0114] Based on automatic control theory, the cutoff frequency ω of the proposed disturbance observer is... c and phase margin Defined as:

[0115]

[0116] The expressions for k1 and k2 can be derived from the above formula.

[0117]

[0118] Figure 3 Figure 4 shows the Bode plot of the system when k1 and k2 have different values. It can be seen that when the value of k1 is larger, the center frequency of the system begins to shift and the filtering capability also begins to decrease. When the value is smaller, the system bandwidth decreases, resulting in a slow dynamic response. The value characteristics of k2 are the opposite of k1. Considering the balance between the filtering performance and dynamic response of the system, and according to the tuning principle of the above formula, k1 and k2 are set to 0.707 and 0.5 respectively.

[0119] In this embodiment of the invention, a positionless control method for permanent magnet synchronous motors based on a disturbance observer is used. By using the proposed method, the estimation accuracy of the rotor angle is improved, and it has good robustness under the condition of motor parameter mismatch. Figure 5 The simulation results are for a given sudden change in rotational speed (100 r / min - 400 r / min - 6800 r / min). The system has good dynamic and steady-state performance. Figure 6 This is a case of motor inductance mismatch (L) q -0.5L q -2L q Simulation results under ( ); Figure 7 Motor resistance mismatch (R) s -2R s The simulation results show that when the parameters change significantly, the estimated rotor angle and speed will deviate slightly, but within an acceptable range, the system can operate stably.

Claims

1. A sensorless control method for a permanent magnet synchronous motor based on a disturbance observer, comprising the following steps: S1: The impact of motor parameter mismatch on angle estimation in a sensorless control system was analyzed. S2: Design a disturbance observer to compensate for the effects of parameter mismatch and system harmonics. S3: Analyze the performance of the disturbance observer on harmonic suppression and parameter mismatch compensation. S4: Analyze the stability of the disturbance observer and select an appropriate value based on the system bandwidth requirements.

2. The control method according to claim 1, characterized in that, In step S1, the sensorless model of the permanent magnet synchronous motor using the disturbance observer is as follows: First, the voltage model of the permanent magnet synchronous motor in vector form in the αβ coordinate system is as follows: u s =R s i s +pψ s In the formula, u s =[u sα ,u sβ ] T i s =[i sα i sβ ] T R represents the stator voltage vector and current vector, respectively. s For the stator resistance, ψ s =[ψ sα ,ψ sβ ] T This represents the stator flux linkage vector. Among them, the stator magnetic flux ψ s Represented as In the formula L d ,L q Representing the d-q axis inductance, ψ f Represents permanent magnet flux linkage. The effective flux linkage can be obtained by integrating the back electromotive force. Where e r =[e r_α ,e r_β ] T It is represented as the back electromotive force vector. The expression for the effective flux linkage considering parameter mismatch and disturbance compensation can be rewritten as follows: In the formula This is to compensate for the disturbance in the system. As can be seen from the above formula, parameter mismatch in the motor will introduce a deviation in the estimated effective flux linkage. Without... The compensation may cause excessive deviation in the effective flux estimation, leading to the failure of rotor angle observation. Furthermore, due to inverter nonlinearity, uncertainty of initial integral value, and current sampling error, using a pure integrator will cause DC bias and harmonic pollution in the observation flux, making the sensorless system unable to work.

3. The control method according to claim 1, characterized in that, In step S2, a disturbance observer is designed to compensate for the stator flux bias and harmonics caused by motor parameter mismatch and pure integration: The second-order generalized integrator (SOGI) is widely used in grid-connected control and motor control due to its flexibility. It can be considered as an orthogonal signal generator with bandpass filtering characteristics, and it does not introduce amplitude attenuation or phase shift at the center frequency. Its transfer function can be expressed as... In the formula, ω is the center frequency and k is the damping ratio, which determines the filter bandwidth. Based on the characteristics of SOGI, the effective flux linkage obtained by the pure integrator is filtered by SOGI, and the difference between this and the stator flux linkage is divided by L. q The estimated stator current is obtained, and the difference between the estimated stator current and the actual stator current is used as a proportional gain k. p After the step, the estimated perturbation in the effective flux can be obtained. By feeding the disturbance back into the back electromotive force, the parameter mismatch and the interference from the pure integrator can be compensated for. The transfer function of the disturbance observer obtained through the above design can be expressed as follows: In the formula, k1 is the damping ratio of SOGI, and k2 is the feedback gain for estimating the disturbance. The resulting effective flux linkage is The transfer function G of the back electromotive force and the effective flux linkage can be obtained. DO (s). By setting the center frequency ω to This allows the observer to achieve frequency adaptation.

4. The control method according to claim 1, characterized in that, In step S3, the ability of the designed disturbance observer to suppress system harmonics and parameter mismatch is further analyzed. Suppression capability of DC component: The final value theorem can be obtained by applying the flux linkage obtained using a perturbation observer. In the formula, e0 is the DC component vector in the back electromotive force. It can be seen that the DC component in the input back electromotive force is completely eliminated after passing through the disturbance observer. Operating frequency Amplitude and phase response at: Substitute G DO From (s), we can obtain The above equation shows that the output amplitude and phase of the designed disturbance observer at the center frequency are integrals of the input back electromotive force, and the effective flux amplitude will not be attenuated.

5. The control method according to claim 1, characterized in that, In step S4, the selected values ​​for the gain parameter in the disturbance observer are given: G DO The characteristic equation of the transfer function (s) can be expressed as follows: Based on the Routh stability criterion, a Routh table can be compiled. For the system to be stable, the first column of the Routh meter must be all positive numbers; therefore, the center frequency should be positive and set to the absolute value of the rotational speed. And k1 > 0, k2 > 0. G DO The open-loop transfer function of (s) is Based on automatic control theory, the cutoff frequency ω of the proposed disturbance observer is... c and phase margin Defined as: The expressions for k1 and k2 can be derived from the above formula. Based on the above formula and the system Bode diagram, k1 and k2 are tuned to 0.707 and 0.5, respectively.

Images