A control method of a permanent magnet motor based on MRAS parameter identification

Abstract

The application discloses a kind of permanent magnet synchronous motor control methods based on MRAS parameter identification.A reference model and an adjustable model are established according to the d-q axis current equation of the permanent magnet synchronous motor, an adaptive system is established by MRAS parameter identification method according to the reference model and the adjustable model, the nonlinear part of the adaptive system is extracted from the adaptive system, and an adaptive law is established accordingly.The stator resistance and the stator inductance are calculated using the adaptive law combined with the known d-q axis stator voltage and stator current.The current closed-loop control of the non-flux permanent magnet synchronous motor is realized by using the resistance and inductance parameters through the incremental DPCC controller, and the motor torque control is realized.The speed feedback control of the motor is realized by combining PI control.The method maintains the high dynamic response of incremental DPCC and the advantage of no current static error, and effectively increases the stability of the system by using parameter identification.

Description

Technical Field

[0001] This invention relates to an identification and control method for permanent magnet synchronous motors, and in particular, proposes an improved method for model reference adaptive (MRAS) parameter identification based on a flux-free model of a permanent magnet synchronous motor. The identified parameters are then incorporated into an incremental deadbeat-free current predictive control model to achieve current and speed control of the permanent magnet motor, thereby improving robustness while ensuring high dynamic response. Background Technology

[0002] With the development of permanent magnet materials, permanent magnet synchronous motors (PMSMs) have become widely used in electric vehicles, industrial machinery, water pumps, and other fields due to their advantages such as high efficiency, high power density, simple structure, and high operational reliability. Common motor control methods require high accuracy in the motor's electromagnetic parameters. Mismatches in motor parameters can affect the control performance of PMSMs. Offline measuring instruments, such as LCR meters, have relatively low accuracy in measuring motor parameters. Furthermore, motor parameters may change during actual operation. When the current is high, the stator core saturates, thus reducing the stator inductance. Prolonged operation under these conditions leads to increased motor temperature, increased stator resistance, and a decrease in permanent magnet flux linkage.

[0003] Commonly used online parameter identification control methods include Recursive Least Squares (RLS), Extended Kalman Filter (EKF), and MRAS. RLS requires extensive matrix operations and cannot guarantee the convergence of the estimated parameters. EKF also requires a processor with strong computational capabilities. While the traditional MRAS method can identify stator inductance and stator resistance, flux linkage terms still exist in both the reference and adjustable models, introducing errors. Summary of the Invention

[0004] To address the limitations of existing methods and solve the problems in the background art, this invention proposes a permanent magnet motor control method based on MRAS parameter identification.

[0005] This invention uses the first-order differential form of the dq-axis voltage equation of a permanent magnet synchronous motor as a reference model for stator resistance and stator inductance, independent of the permanent magnet flux linkage, thus avoiding unnecessary flux linkage errors. The resistance and inductance parameters obtained from this parameter identification are applied to incremental deadbeat predictive current control (DPCC) to achieve accurate, fast, and effective closed-loop control of the motor current, thereby better controlling the motor torque.

[0006] The specific process of the method is as follows:

[0007] (1) For a permanent magnet synchronous motor (PMSM) without magnetic flux, an adaptive law is constructed in the form of the first-order differential equation of the model without magnetic flux to identify the parameters and obtain the resistance and inductance parameters.

[0008] (2) Use the resistance and inductance parameters identified in step (1) to perform feedback control on the permanent magnet synchronous motor.

[0009] The resistance and inductance parameters mentioned are the stator resistance of each phase winding of the permanent magnet synchronous motor and the stator inductance at both ends of the winding.

[0010] Step (1) specifically includes:

[0011] (1.1) Based on the dq-axis current equations of the permanent magnet synchronous motor, a reference model and an adjustable model are established, specifically as follows:

[0012] Firstly, for flux-free permanent magnet synchronous motors, the current differential equation for the flux-free term is established using the following formula and serves as a reference model for MRAS:

[0013]

[0014]

[0015] Among them, i d u represents the actual value of the d-axis stator current. d i represents the stator voltage on the d-axis. q u represents the actual value of the q-axis stator current. q Represents the stator voltage on the q-axis, R is the actual value of the stator resistance, L is the actual value of the stator inductance on the d-axis, t is time, and ω is... e It is the electric angular velocity, d 2 Represents the second derivative;

[0016] By replacing the actual values ​​of the stator resistance and stator inductance in the reference model with estimated values, an adjustable model can be established:

[0017]

[0018]

[0019] in, This represents the estimated value of the d-axis stator current. This represents the estimated value of the q-axis stator current. This is an estimated value for the stator resistance. This is the estimated value of the d-axis stator inductance;

[0020] (1.2) An adaptive system is established based on the reference model and the adjustable model using the MRAS parameter identification method. The adaptive system is divided into a linear part and a nonlinear part. The nonlinear part of the adaptive system is extracted, and then the adaptive law is established using the superstability theory method, as shown below:

[0021]

[0022]

[0023] In the formula, e d and e q These are the output errors of the adjustable model and the reference model, ω, respectively. e It is the electric angular velocity, K fp K fi K gp and K gi The gain parameter represents the adaptive law. u represents the estimated value of the d-axis stator current. d Represents the stator voltage on the d-axis. U represents the estimated value of the q-axis stator current. q Represents the stator voltage on the q-axis. This is an estimated value for the stator resistance. ω represents the estimated value of the stator inductance along the d-axis, where t is time and τ is the time constant; these two are the operators used in the operation. e It is the electrical angular velocity, which is calculated from the angle of the position sensor to obtain the rotational speed; R0 is the initial value of the stator resistance, and L0 is the initial value of the stator inductance. Both are parameters of the permanent magnet synchronous motor itself.

[0024] (1.3) The stator resistance and stator inductance of each phase winding of the permanent magnet synchronous motor are calculated by using the adaptive law combined with the known stator voltage and stator current of the dq axis, and then used as the feedback input of the DPCC controller.

[0025] The adaptive law of this invention is obtained by calculating using the Popov inequality while determining the stability of the linear time-invariant part based on the theory of hyperstability.

[0026] Step (2) specifically involves:

[0027] (2.1) Current closed-loop control of a flux-free permanent magnet synchronous motor is achieved using an incremental DPCC controller based on resistance and inductance parameters, thereby enabling motor torque control:

[0028] First, the resistance and inductance parameters and the known current setpoint of the dq axis are used as inputs to the incremental DPCC controller. At the same time, the three-phase current and rotor position angle of the flux-free permanent magnet synchronous motor are sampled by the sensor. The sampling results are transformed to obtain the actual current value of the dq axis and the electric angular velocity of the motor as another feedback input of the incremental DPCC controller. The voltage setpoint of the dq axis is output by the incremental DPCC controller.

[0029] Next, the voltage setpoint of the dq axis is generated into a corresponding PWM signal by the space vector pulse width modulation module (SVPWM) and then transmitted to the voltage source inverter. The voltage source inverter outputs three-phase voltage and inputs it into the flux-free permanent magnet synchronous motor to perform feedback control and motor torque control.

[0030] (2.2) Combining PI control to achieve motor speed control:

[0031] The speed feedback quantity is obtained by sampling the three-phase current and rotor position angle of the flux-free permanent magnet synchronous motor through sensors. The speed feedback quantity is then input into the speed loop PI controller. The speed loop PI controller takes the difference between the preset speed setpoint and the speed feedback quantity as input, processes it, and outputs the current setpoint of the q-axis to update it as the input of the incremental DPCC controller.

[0032] This invention applies the identified resistance and inductance parameters to incremental deadbeat predictive current control (DPCC).

[0033] The sensor in question is a position sensor.

[0034] The adaptive law of this invention is the relationship between the first-order differential forms of the voltage and current of the dq axis input in the MRAS parameter identification part and the output stator resistance and stator inductance. Using the adaptive law, the stator resistance and stator inductance can be calculated based on the known dq axis voltage and current, and used as the feedback input of the DPCC controller to realize the closed loop of the permanent magnet synchronous motor current loop in the control method of this invention.

[0035] The principle and process of this invention are as follows:

[0036] (1) Establish a reference model and an adjustable model. The current differential equation without permanent magnet flux linkage is taken as the reference model:

[0037]

[0038]

[0039] Among them, i d u represents the actual value of the d-axis stator current. d i represents the stator voltage on the d-axis. q u represents the actual value of the q-axis stator current. q Represents the stator voltage on the q-axis, R is the actual value of the stator resistance, L is the actual value of the stator inductance on the d-axis, t is time, and ω is... e It is electric angular velocity.

[0040] By replacing the actual values ​​of the stator resistance and stator inductance in the reference model with the estimated values, an adjustable model is obtained:

[0041]

[0042]

[0043] in, u represents the estimated value of the d-axis stator current. d Represents the stator voltage on the d-axis. U represents the estimated value of the q-axis stator current. q Represents the stator voltage on the q-axis. This is an estimated value for the stator resistance. Here is the estimated value of the d-axis stator inductance, where t is time and ω is the inductance. e It is electric angular velocity;

[0044] Subtracting the upper and lower equations of the adjustable model and the reference model yields two output errors:

[0045]

[0046]

[0047] For ease of subsequent calculations, they are arranged in matrix form:

[0048]

[0049] Among them, e d and e q It is the output error of the adjustable model and the reference model, i d i represents the actual value of the d-axis stator current. q R represents the actual value of the q-axis stator current, R represents the actual value of the stator resistance, and L represents the actual value of the d-axis stator inductance. u represents the estimated value of the d-axis stator current. d Represents the stator voltage on the d-axis. U represents the estimated value of the q-axis stator current. q Represents the stator voltage on the q-axis. This is an estimated value for the stator resistance. Here is the estimated value of the d-axis stator inductance, where t is time and ω is the inductance. e It is electric angular velocity.

[0050] (2) An adaptive system is established based on the reference model and the adjustable model using the MRAS parameter identification method. The adaptive system is divided into a linear part and a nonlinear part. The nonlinear part of the adaptive system is extracted from it, and then the adaptive law is established using the superstability theory method. Finally, the MRAS adaptive law can be obtained:

[0051]

[0052]

[0053] In the formula, e d and e q It is the output error between the adjustable model and the reference model, ψ f It is a permanent magnet flux linkage, ω e It is the electric angular velocity, K fp K fi K gp and K gi The gain parameter represents the adaptive law. u represents the estimated value of the d-axis stator current. d Represents the stator voltage on the d-axis. U represents the estimated value of the q-axis stator current. q Represents the stator voltage on the q-axis. This is an estimated value for the stator resistance. ω represents the estimated value of the stator inductance along the d-axis, where t is time and τ is the time constant; these two are the operators used in the operation. e It is the electrical angular velocity, which is obtained by calculating the rotational speed based on the angle of the position sensor; R0 is the initial value of the stator resistance, and L0 is the initial value of the stator inductance. These two are parameters of the permanent magnet synchronous motor itself, which makes the parameter identification system asymptotically stable.

[0054] Because the control cycle of the motor is much smaller than the mechanical time constant, therefore ω e It remains constant within adjacent control cycles. Furthermore, the motor parameters gradually change with the electrical variables. ω e R, L and ψ f It remains unchanged in a dynamic system, and is:

[0055]

[0056]

[0057]

[0058]

[0059] (3) The resistance and inductance parameters obtained by parameter identification are applied to incremental deadbeat current predictive control (DPCC), and the specific control equation is as follows; This control method does not require permanent magnet flux parameters. When the motor parameters are accurate, it has high dynamic performance with current delay of two control cycles and error-free following of the given value. It can remain stable as long as the inductance parameter does not exceed twice the actual value, and the current is statically error-free even when there are errors in the resistance and inductance parameters.

[0060]

[0061]

[0062]

[0063]

[0064]

[0065] Among them, T s Indicates the control cycle. and The identification results for the stator resistance and the d-axis and q-axis inductances are shown respectively. e (k) represents the electric angular velocity feedback value at time k; Represents the incremental current along the d and q axes at time k; This represents the d-axis and q-axis current feedback values ​​at time k; Represents the estimated incremental current values ​​along the d and q axes at time k+1; This represents the incremental voltage inputs along the d and q axes during the k-th control cycle. Represents the estimated d-axis and q-axis current values ​​at time k+1; This represents the given values ​​of the d-axis and q-axis currents at time k; This represents the d-axis and q-axis voltage setpoints for the (k+1)th control cycle.

[0066] The beneficial effects of this invention are:

[0067] This invention proposes an improved reference model and an adjustable model that do not rely on permanent magnet flux linkage. It uses Popov's ultrastability theory to obtain an adaptive law for motor parameter identification, thus avoiding unnecessary flux linkage errors.

[0068] Based on this, the identified resistance and inductance parameters are applied to incremental deadbeat current predictive control (DPCC). This method maintains the advantages of high dynamic response and no steady-state current error of incremental DPCC, while the use of parameter identification effectively increases the stability of the system. Attached Figure Description

[0069] Figure 1 This is a logic block diagram of the control method described in this invention applied to a permanent magnet synchronous motor control system;

[0070] Figure 2 This is a block diagram of the MRAS for stator inductance and stator resistance identification according to the present invention. Detailed Implementation

[0071] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0072] like Figure 1As shown, in the control system of the flux-free permanent magnet synchronous motor of the present invention, the flux-free permanent magnet synchronous motor is connected to a voltage source inverter. A position sensor for detecting the rotor position angle is installed on the flux-free permanent magnet synchronous motor. The position sensor and the speed conversion calculation module are connected to the input terminal of the incremental DPCC controller. The input terminal of the incremental DPCC controller is connected to a PI controller. The output terminal of the incremental DPCC controller is connected to the voltage source inverter after passing through the inverse Park transformation module and the space vector pulse width modulation module in sequence.

[0073] The embodiments of the present invention and their implementation process are as follows:

[0074] (1) The first-order differential equation of the permanent magnet synchronous motor (PMSM) model without magnetic flux is used as the reference model for MRAS parameter identification.

[0075] (2) Apply the identified resistance and inductance parameters to incremental deadbeat predictive current control (DPCC).

[0076] (3) Incremental DPCC is used to achieve closed-loop control of motor current, thereby controlling motor torque. The incremental DPCC controller takes the dq-axis current setpoint as input and samples the three-phase current and rotor position angle of the permanent magnet synchronous motor through sensors. The actual value of dq-axis current and the electric angular velocity of the motor obtained after sampling are used as the feedback input of the controller. The dq-axis voltage setpoint output by the controller is generated into a corresponding PWM signal by the space vector pulse width modulation (SVPWM) module and sent to the inverter. The voltage output by the inverter is applied to the permanent magnet synchronous motor for feedback control.

[0077] (4) Motor speed control is achieved through proportional-integral (PI) control. The speed feedback is obtained by calculation based on the sampled values ​​of the position sensor; the speed loop PI controller takes the difference between the speed setpoint and the feedback as input, and outputs the q-axis current setpoint as the input of the incremental DPCC.

[0078] The MRAS parameter identification content in (1) above is as follows:

[0079] Based on the dq-axis current equations of the permanent magnet synchronous motor, a reference model and an adjustable model are established, specifically as follows:

[0080] Firstly, for flux-free permanent magnet synchronous motors, the current differential equation for the flux-free term is established using the following formula and serves as a reference model for MRAS:

[0081]

[0082]

[0083] Among them, id u represents the actual value of the d-axis stator current. d i represents the stator voltage on the d-axis. q u represents the actual value of the q-axis stator current. q Represents the stator voltage on the q-axis, R is the actual value of the stator resistance, L is the actual value of the stator inductance on the d-axis, t is time, and ω is... e It is electric angular velocity.

[0084] By replacing the actual values ​​of the stator resistance and stator inductance in the reference model with estimated values, an adjustable model can be established:

[0085]

[0086] in, u represents the estimated value of the d-axis stator current. d Represents the stator voltage on the d-axis. U represents the estimated value of the q-axis stator current. q Represents the stator voltage on the q-axis. This is an estimated value for the stator resistance. Here is the estimated value of the d-axis stator inductance, where t is time and ω is the inductance. e It is electric angular velocity.

[0087] An adaptive system is established based on the reference model and the adjustable model using the MRAS parameter identification method. The adaptive system is divided into a linear part and a nonlinear part. The nonlinear part of the adaptive system is extracted, and then the adaptive law is established using the superstability theory method, as shown below:

[0088]

[0089]

[0090] In the formula, e d and e q It is the output error between the adjustable model and the reference model, ψ f It is a permanent magnet flux linkage, ω e It is the electric angular velocity, K fp K fi K gp and K gi The gain parameter represents the adaptive law. u represents the estimated value of the d-axis stator current. d Represents the stator voltage on the d-axis. U represents the estimated value of the q-axis stator current. q Represents the stator voltage on the q-axis. This is an estimated value for the stator resistance. ω represents the estimated value of the stator inductance along the d-axis, where t is time and τ is the time constant; these two are the operators used in the operation. e It is the electrical angular velocity, which is calculated from the angle of the position sensor to obtain the rotational speed; R0 is the initial value of the stator resistance, and L0 is the initial value of the stator inductance. Both are parameters of the permanent magnet synchronous motor itself.

[0091] Based on the establishment of the reference model and the derivation of the adaptive rate, a block diagram for MRAS parameter identification can be established, such as... Figure 2 As shown. In the MRAS parameter identification module, the stator voltage u on the d-axis and q-axis is... d u q As known inputs to the reference model and the adjustable model, after calculation, the reference model outputs the actual values ​​of the stator current i on the d-axis and q-axis. d and i q The adjustable model outputs estimated values ​​of the stator current on the d-axis and q-axis. and Obtain e by subtracting them separately d and e q This refers to the output error between the adjustable model and the reference model. In e d and e q Based on this, the adaptive rate is derived using the superstability theory, and the estimated value of the stator resistance is obtained by simultaneously solving the adaptive rate equation. d-axis stator inductance estimate Then input it back into the adjustable model for the next round of calculation.

[0092] The method of this invention maintains the advantages of high dynamic response and no current steady-state error of incremental DPCC, while using parameter identification to accurately identify the stator inductance at both ends of the stator winding of the permanent magnet synchronous motor and the stator resistance of each phase, and avoids the error caused by the flux linkage term, thus effectively increasing the stability of the system.

[0093] Specific application examples are explained.

[0094] To verify the reliability of this method, relevant simulation experiments were conducted. The parameters of the surface-mounted permanent magnet synchronous motor used as an example in the experiment are shown in Table 1 below.

[0095] Table 1 Motor Parameters

[0096] Extreme logarithm 4 Stator resistance 2.45Ω Stator inductor 7.68mH Magnetic Link 0.6165Wb bus voltage 150V Rated torque 1.5Nm Rated speed 600rpm Moment of inertia <![CDATA[0.000031kg*m 2 ]]> effective current 1.5A

[0097] Based on the motor parameters in this table, the effectiveness of the invention can be demonstrated by building and simulating the control system in MATLAB / Simulink according to the system shown in the abstract figures.

Claims

1. A control method for a permanent magnet synchronous motor based on MRAS parameter identification, comprising the following steps: (1) For a permanent magnet synchronous motor without magnetic flux, an adaptive law is constructed in the form of the first-order differential equation of the model without magnetic flux to identify the parameters and obtain the resistance and inductance parameters. Step (1) specifically includes: (1.1) Based on the dq-axis current equations of the permanent magnet synchronous motor, a reference model and an adjustable model are established, specifically as follows: Firstly, for a flux-free permanent magnet synchronous motor, the current differential equation for the flux-free term is established as follows and used as a reference model: in, This represents the actual value of the d-axis stator current. Represents the stator voltage on the d-axis. This represents the actual value of the q-axis stator current. Represents the stator voltage on the q-axis. This is the actual value of the stator resistance. This is the actual value of the d-axis stator inductance. For time, It is the electric angular velocity, d 2 Represents the second derivative; By replacing the actual values ​​of the stator resistance and stator inductance in the reference model with estimated values, an adjustable model can be established: in, This represents the estimated value of the d-axis stator current. This represents the estimated value of the q-axis stator current. This is an estimated value for the stator resistance. This is the estimated value of the d-axis stator inductance; (1.2) An adaptive system is established based on the reference model and the adjustable model using the MRAS parameter identification method. The nonlinear part of the adaptive system is extracted from it, and then the adaptive law is established using the superstability theory method, as shown below: In the formula, and These are the output errors of the adjustable model and the reference model, respectively. , , and The gain parameter represents the adaptive law. This represents the estimated value of the d-axis stator current. Represents the stator voltage on the d-axis. This represents the estimated value of the q-axis stator current. Represents the stator voltage on the q-axis. This is an estimated value for the stator resistance. This is the estimated value of the d-axis stator inductance. For time, It is a time constant; It is electric angular velocity; This is the initial value of the stator resistance. Initial value of stator inductance; (1.3) The stator resistance and stator inductance at both ends of each phase winding of the permanent magnet synchronous motor are calculated by using the adaptive law in combination with the known stator voltage and stator current of the dq axis. (2) Use the resistance and inductance parameters identified in step (1) to perform feedback control on the permanent magnet synchronous motor.

2. The permanent magnet synchronous motor control method based on MRAS parameter identification according to claim 1, characterized in that: The resistance and inductance parameters mentioned are the stator resistance of each phase winding of the permanent magnet synchronous motor and the stator inductance at both ends of the winding.

3. The permanent magnet synchronous motor control method based on MRAS parameter identification according to claim 1, characterized in that: Step (2) specifically involves: (2.1) Current closed-loop control of a flux-free permanent magnet synchronous motor is achieved using an incremental DPCC controller based on resistance and inductance parameters, thereby enabling motor torque control: First, the resistance and inductance parameters and the current setpoint of the dq axis are used as inputs to the incremental DPCC controller. At the same time, the three-phase current and rotor position angle of the flux-free permanent magnet synchronous motor are sampled by the sensor. The sampling results are transformed to obtain the actual current value of the dq axis and the electric angular velocity of the motor as another input to the incremental DPCC controller. The voltage setpoint of the dq axis is output by the incremental DPCC controller. Next, the voltage setpoint of the dq axis is generated into a corresponding PWM signal by the space vector pulse width modulation module and then transmitted to the voltage source inverter. The voltage source inverter outputs three-phase voltage and inputs it into the flux-free permanent magnet synchronous motor to perform feedback control on the permanent magnet synchronous motor. (2.2) Combining PI control to achieve motor speed control: The speed feedback is obtained by sampling the three-phase current and rotor position angle of the flux-free permanent magnet synchronous motor through sensors. The speed feedback is then input into the PI controller. The PI controller takes the difference between the speed setpoint and the speed feedback as input, processes it, and outputs the current setpoint of the q-axis as an update, which is then used as the input of the incremental DPCC controller.

4. The permanent magnet synchronous motor control method based on MRAS parameter identification according to claim 3, characterized in that: The sensor in question is a position sensor.

Images