Current control method for permanent magnet synchronous motor under high speed and low carrier wave ratio

Abstract

The invention discloses a current control method for a permanent magnet synchronous motor at a high speed and a low carrier wave ratio, and aims to solve the problem that the current instruction tracking performance of a discrete time domain complex vector current controller at the high speed and the low carrier wave ratio is reduced due to insufficient damping of the discrete time domain complex vector current controller. Therefore, the damping ratio of the control system is improved, the influence of disturbance voltage caused by parameter change, back electromotive force change and nonlinearity of the inverter on current instruction tracking can be suppressed, and meanwhile, the problem of current coupling caused by inductance change is avoided; current overshoot can be reduced and current fluctuation caused by interference voltage can be suppressed under the condition that electrical parameters change or are uncertain.

Description

Technical Field

[0001] The invention belongs to the field of permanent magnet synchronous motor control and relates to a current control method of a permanent magnet synchronous motor at high speed and low carrier ratio. Background Art

[0002] High-speed permanent magnet synchronous motors have the advantages of high power density, high efficiency, and strong reliability. They are widely used in aerospace, medical equipment, flywheel energy storage and other fields. High-speed permanent magnet synchronous motors generally adopt high-performance vector control strategies to achieve separate control of the motor's magnetic flux and torque. However, limited by the switching device frequency, as the speed increases, the motor will operate under low carrier ratio conditions. Digital control delays will aggravate the coupling of the AC and DC axis currents, resulting in the inability to accurately track the AC and DC axis current commands, and the dynamic performance of the control system will decrease. It is necessary to design a suitable current decoupling control strategy to achieve complete decoupling of the motor's AC and DC axis currents, thereby comprehensively improving the dynamic performance of the motor at high speeds. Summary of the invention

[0003] The purpose of the present invention is to provide a current control method for a permanent magnet synchronous motor under high speed and low carrier ratio in view of the deficiencies in the prior art, which can effectively improve the current command tracking performance and anti-disturbance performance and reduce current fluctuations.

[0004] The technical solution adopted by the present invention is:

[0005] A current control method for a permanent magnet synchronous motor at high speed and low carrier ratio adopts complex vector active damping and a current prediction method with inductance estimation, introduces a virtual resistor in the active damping complex vector current regulation process, predicts the virtual resistor term feedback current based on an extended state observer model-free current prediction method, and obtains an inductance estimation value in the model-free current prediction method for use in the active damping complex vector current regulation; specifically, the method includes:

[0006] Step 1: Collect the three-phase current and obtain the d-axis current i after coordinate transformation d and q-axis current i q ;

[0007] Step 2: Set the d-axis current i d Input to the d-axis current prediction ESO (extended state observer) and compare it with the command d-axis current After the difference is made, it is input into the active damping complex vector controller; the q-axis current i q Input to the q-axis current prediction ESO and compare it with the q-axis command current After making the difference, it is input into the active damping complex vector controller;

[0008] Step 3: The d-axis current prediction ESO contains an inductance estimator, and the output of the d-axis current prediction ESO is the current prediction value and estimated inductance The active damping complex vector controller outputs the d-axis voltage u d , d-axis voltage u d Subtract the current prediction value With virtual resistance R a The product of the d-axis voltage u′ is obtained. d , d-axis voltage u′ d Returning to the d-axis current prediction ESO; the inductance estimation value of the d-axis current prediction ESO is output to the active damping complex vector controller and the q-axis current prediction ESO;

[0009] Step 4: q-axis current prediction ESO output current prediction value The active damping complex vector controller outputs the q-axis voltage u q , q-axis voltage u q Subtract the current prediction value With virtual resistance R a The product of the q-axis voltage u′ is obtained. q , q-axis voltage u′ q Returning to the q-axis current prediction ESO; the q-axis current prediction ESO uses the inductance estimation value of the inductance estimator;

[0010] Step 5: Substitute u′ d and u′ q After coordinate transformation, the duty cycle signal is output through space vector pulse width modulation to the inverter to drive the motor to operate.

[0011] The active damping complex vector controller is designed with current coupling terms that vary with speed in mind. In theory, it can achieve current control with any low carrier ratio. The active damping method can suppress the influence of disturbance voltage caused by parameter changes, back electromotive force changes, and inverter nonlinearity on current command tracking. The d-axis and q-axis current prediction ESOs are improved model-free current predictions that only require input of motor current and voltage, without requiring parameters such as motor inductance, resistance, and flux linkage. The d-axis current prediction ESO contains an inductance estimator, and the estimated inductance is used in the q-axis current prediction ESO and the active damping complex vector controller.

[0012] The step 1 comprises: using a sampling resistor to sample the three-phase current i a 、i b and i c , through the formula

[0013]

[0014] Perform coordinate transformation to obtain the d-axis current i of the motor in the synchronous rotating coordinate systemd and q-axis current i q . Where θ is the electrical angle.

[0015] The active damping complex vector controller is designed based on the complex vector active damping motor model using the zero-pole cancellation principle, which can be expressed as:

[0016]

[0017] Where K is the current controller bandwidth, R a is a virtual resistor, is the estimated value of motor phase inductance, R s is the motor phase resistance, T s is the control period, ω e is the electrical angular velocity, j is the imaginary operator, and z is the lead operator. The complex vector active damping motor model is a complex vector mathematical model in discrete time domain with virtual resistance added, which is expressed as:

[0018]

[0019] The discrete time domain complex vector mathematical model is expressed as:

[0020]

[0021] The d-axis current prediction ESO includes a d-axis model, and the q-axis current prediction ESO includes a q-axis model. The d-axis model is:

[0022]

[0023] The q-axis model is:

[0024]

[0025] Where k represents the kth moment, k+1 represents the k+1th moment, and f d (k) and f q (k) are the d-axis perturbation and q-axis perturbation respectively, and i d (k) and f d The estimated value of (k), and i q (k) and f q The estimated value of (k), i err_d (k) and i err_q (k) are the d-axis and q-axis stator current errors, respectively, and β1 and β2 are the ESO gains. Estimate the inductance value for the Inductance Estimator.

[0026] The traditional model-free current prediction scheme does not consider the coupling term T that varies with the speed. s ω e (k)i d (k) and T s ω e (k)i q (k), the current prediction performance will be reduced in the dynamic process. The introduction of the coupling term does not increase the additional motor parameters, and the electrical angular velocity ω e (k) is easily obtained by sensors and position-free algorithms. The motor model is:

[0027]

[0028] Where p represents the differential operator, f d represents the d-axis disturbance voltage, f q It represents the q-axis disturbance voltage. The disturbance voltage includes the disturbance voltage caused by resistance mismatch, inductance mismatch and inverter nonlinearity.

[0029] The inductance estimator is:

[0030]

[0031] Where α is the bandwidth of the inductance estimator. The output of the inductance estimator is the inverse of the inductance, and the input is (εd-fd) / ud, ε d is the reference disturbance voltage, and its value is 0. The inductance estimator is designed based on the zero-pole cancellation principle according to the predicted current model. The established predicted current model is:

[0032]

[0033] Its output is Input is

[0034] The advantages of the present invention are:

[0035] (1) The complex vector form is used when establishing the permanent magnet synchronous motor model, which is conducive to the analysis and design of the current controller. The complex vector motor model in the continuous time domain is discretized by the zero-order holder method. The established model is more accurate, and the designed controller can exert good current tracking performance under low carrier ratio conditions.

[0036] (2) The active damping method is similar to adding physical resistance to the controlled object, which can improve the stiffness of the control system and suppress the influence of disturbance voltage caused by parameter changes, back electromotive force changes, and inverter nonlinearity on current command tracking. It can effectively improve the current tracking performance and anti-disturbance performance of the dynamic process.

[0037] (3) The ESO-based model-free current prediction method considers the motor resistance voltage drop, inductance voltage drop and back-EMF component as disturbances during modeling, which can avoid the impact of motor parameter mismatch on the predicted current. The designed inductance estimator can adjust the inductance value of the entire current controller online in real time to avoid current overshoot and oscillation caused by parameter changes in the dynamic process. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The structural block diagram of the complex vector current control system of permanent magnet synchronous motor predictive current active damping;

[0039] Figure 2 Active damping complex vector current control block diagram;

[0040] Figure 3 Active damping complex vector motor model;

[0041] Figure 4 d-axis current prediction ESO structure diagram;

[0042] Figure 5 q-axis current prediction ESO structure diagram;

[0043] Figure 6 Inductance estimation performance at different speeds;

[0044] Figure 7 The current command tracking comparison experimental waveforms of three methods: proportional integral (PI) control, complex vector current control (CVCR) and predictive current active damping complex vector control (PC-ADCV) of the present invention;

[0045] Figure 8 Comparative experimental waveforms of current disturbance suppression using three methods: proportional integral (PI) control, complex vector current control (CVCR) and predictive current active damping complex vector control (PC-ADCV) of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below in conjunction with the examples of the present invention. The described examples are part of the examples of the present invention, not all of the examples. The present invention can be implemented in other ways, and those skilled in the art can make similar improvements without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below. All other embodiments obtained without making creative work belong to the protection scope of the present invention.

[0047] Example 1

[0048] like Figure 1As shown, a current control method for a permanent magnet synchronous motor at high speed and low carrier ratio, the method comprising:

[0049] Step 1: The inverter converts DC power into AC power to drive the motor. Figure 1 PMSM stands for permanent magnet synchronous motor, and then the three-phase stator current i of PMSM is collected. a 、i b 、i c , through the formula

[0050]

[0051] Transformation to obtain the d-axis current i d and q-axis current i q ; where θ is the electrical angle, which can be obtained through an encoder or position sensorless technology.

[0052] Step 2: Set the d-axis current i d The d-axis current prediction input is used by the ESO and is compared with the commanded d-axis current After the difference is made, it is input into the active damping complex vector controller; the q-axis current i q Input to the q-axis current prediction ESO and compare it with the q-axis command current After the difference is made, it is input into the active damping complex vector controller; wherein, the structure of the active damping complex vector controller is as follows Figure 2 As shown, i d and i q Use i dq =i d +ji q In the form of a complex vector, u d and u q Use u dq =u d +ju q It can be expressed in complex vector form to simplify the analysis and design of the controller. It represents 1.5 control cycle delay compensation. The active damping complex vector controller is designed based on the complex vector active damping motor model using the zero-pole cancellation principle and can be expressed as:

[0053]

[0054] Where K is the current controller bandwidth, and the recommended value is 0.25 to achieve current step response without overshoot. a is a virtual resistor, is the estimated value of motor phase inductance, R s is the motor phase resistance, T s is the control period, ω eis the electrical angular velocity, j is the imaginary operator, and z is the lead operator. The active damping complex vector motor model is obtained by adding virtual resistance to the discrete time domain complex vector mathematical model, such as Figure 3 As shown, the active damping complex vector motor model can be expressed as:

[0055]

[0056] The discrete time domain complex vector mathematical model is expressed as:

[0057]

[0058] The discrete time domain complex vector mathematical model is a discretization of the continuous time domain complex vector mathematical model by the method of zero-order holder, and the digital control delay caused by the PWM effect is considered. The Taylor formula (e AT ≈1+AT) to simplify. The model before simplification is:

[0059]

[0060] The continuous time domain complex vector mathematical model is:

[0061]

[0062] Step 3: d-axis current prediction ESO output current prediction value and estimated inductance The active damping complex vector controller outputs the d-axis voltage u d , d-axis voltage u d Subtract the current prediction value With virtual resistance R a The product of the d-axis voltage u′ is obtained. d , d-axis voltage u′ d Return to the d-axis current prediction ESO; q-axis current prediction ESO output current prediction value The active damping complex vector controller outputs the q-axis voltage u q , q-axis voltage u q Subtract the current prediction value With virtual resistance R a The product of the q-axis voltage u′ is obtained. q , q-axis voltage u′ q Return to the q-axis current prediction ESO. The d-axis current prediction ESO structure diagram is as follows Figure 4 As shown, it includes d-axis disturbance observation and current prediction, and includes an inductance estimator for estimating the inductance. The d-axis current prediction ESO includes a d-axis model, and the d-axis model is:

[0063]

[0064] The q-axis current predicts the ESO structure diagram as shown in Figure 5 As shown, it includes q-axis disturbance observation and current prediction, and the q-axis current prediction ESO includes a q-axis model, and the q-axis model is:

[0065]

[0066] Where k represents the kth moment, k+1 represents the k+1th moment, and f d (k) and f q (k) are the d-axis perturbation and q-axis perturbation respectively, and i d (k) and f d The estimated value of (k), and i q (k) and f q The estimated value of (k), i err_d (k) and i err_q (k) are the stator current errors of the d-axis and q-axis respectively. β1 and β2 are ESO gains, and the recommended value range is 1500~2500rad / s. The inductance estimator estimates the inductance value. The motor model is:

[0067]

[0068] Where p represents the differential operator, b d and b q It represents the disturbance voltage caused by parameter mismatch, which can be specifically expressed as:

[0069]

[0070] This includes the disturbance voltage caused by resistance mismatch, inductance mismatch, flux linkage mismatch, and the disturbance voltage λ caused by inverter nonlinearity d and λ q Δ represents the difference between the actual value and the controller parameter. The mathematical model is further simplified as:

[0071]

[0072] in

[0073]

[0074] The inductance estimator is:

[0075]

[0076] Where α is the bandwidth of the inductance estimator, ranging from 0 to 1. The output of the inductance estimator is the inverse of the inductance, and the input is (εd-fd) / ud, ε d is the reference disturbance voltage, and its value is 0. The inductance estimator is designed based on the zero-pole cancellation principle according to the predicted current model. The established predicted current model is:

[0077]

[0078] Its output is Input is

[0079] Step 4: Substitute u′ d and u′ q After coordinate transformation, the duty cycle signal is output through space vector pulse width modulation to the inverter to drive the motor to operate.

[0080] The working process and working principle of the predictive current active damping complex vector control (PC-ADCV) of the present invention are as follows: the transformed AC and DC axis currents are respectively subtracted from the command AC and DC axis currents, and the control voltage is output through the active damping complex vector controller. The control voltage is subtracted from the product of the predicted current output by the predicted current ESO and the virtual resistance to obtain the total voltage, which is input into the inverter and then input into the motor after the inverter conversion. In this scheme, since the active damping complex vector controller adopts a discrete time domain motor model design and takes into account the coupling voltage that changes with the speed, the current command can be accurately tracked during the dynamic process even at a low carrier ratio. The active damping method can also improve the system stiffness and suppress the influence of parameter mismatch, back electromotive force change, and inverter nonlinearity. At the same time, the predicted current ESO does not use motor parameters and can output an estimated inductance value for use in the active damping complex vector controller, thereby avoiding current coupling caused by inductance mismatch.

[0081] In order to verify the performance of the current control of the present invention, a software and hardware control system was built, and an experiment was conducted using a high-speed permanent magnet motor for a vacuum cleaner. The motor parameters are as follows: rated voltage 25.2V, rated power 500W, rated speed 80krpm, 1 pole pair, phase resistance 0.033Ω, phase inductance 25uH. The switching frequency and current sampling frequency of the experiment are both 10kHz, the dead time is 0.5us, and the inductive algorithm is used. Figure 6 The experimental waveforms of the inductance estimation performance of the PC-ADCV method at different speeds. As the current increases, the inductance becomes saturated and its value decreases. In order to further verify the performance of the proposed method, the speed is set to 60krpm. Figure 7 and Figure 8 The following are the comparison diagrams of current command tracking performance and disturbance suppression performance of the three methods: PI, CVCR and PC-ADCV. The figure includes the q-axis command current and q-axis feedback current i q , d-axis command current and d-axis feedback current i d The current command tracking experiment sets the q-axis command current to step from 3A to 15A. Figure 7 It can be seen that when the q-axis command current step of the PI method occurs, both the d-axis and q-axis currents have overshoots, which is caused by current cross-coupling. The CVCR method takes current cross-coupling into account when modeling. Compared with the PI control method, the overshoot is reduced, but due to the low damping characteristics, the d-axis current still has a large overshoot. The PC-ADCV method can effectively reduce the current overshoot caused by inductance mismatch by introducing virtual resistance and estimating the inductance in real time. The disturbance suppression experiment is set to suddenly add the q-axis disturbance voltage u q_dis =3V, corresponding to Figure 8 It can be seen that the PI and CVCR methods need a long time to recover to the steady state after the disturbance is applied, and the PC-ADCV method can effectively reduce the fluctuation of the AC and DC axis currents and only takes a short adjustment time to reach the stable state.

[0082] The above examples are only used to illustrate the technical solutions of the present invention, but not to limit them. A person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, any technical solution obtained by equivalent substitution shall fall within the protection scope of the present invention.

Claims

1. A current control method for a permanent magnet synchronous motor at high speed and low carrier ratio, characterized in that: The current prediction method using complex vector active damping and inductance estimation introduces a virtual resistor in the active damping complex vector current regulation process, predicts the virtual resistor item feedback current based on a model-free current prediction method of an extended state observer, and obtains an inductance estimation value in the model-free current prediction method for use in the active damping complex vector current regulation.

2. The current control method of a permanent magnet synchronous motor at high speed and low carrier ratio according to claim 1 is characterized in that: The following steps are involved: Step 1: Collect the three-phase current of the permanent magnet synchronous motor and obtain the d-axis current i after coordinate transformation d and q-axis current i q ; Step 2: Set the d-axis current i d Input to the d-axis current prediction ESO, and at the same time the d-axis current i d and d-axis command current After the difference is made, it is input into the active damping complex vector controller; the q-axis current i q Input to the q-axis current prediction ESO, and at the same time the q-axis current i q With q-axis command current After making the difference, it is input into the active damping complex vector controller; Step 3: d-axis current prediction The ESO contains an inductance estimator whose output is the current prediction value and estimated inductance The active damping complex vector controller outputs the d-axis voltage u d , d-axis voltage u d Subtract the current prediction value With virtual resistance R a The product of the d-axis voltage u′ is obtained. d Then return to the d-axis current prediction ESO; the d-axis current prediction ESO converts the inductance estimation value Output to active damping complex vector controller and q-axis current prediction ESO; Step 4: q-axis current prediction ESO output current prediction value The active damping complex vector controller outputs the q-axis voltage u q , q-axis voltage u q Subtract the current prediction value With virtual resistance R a The product of the q-axis voltage u′ is obtained. q Then return to the q-axis current prediction ESO; Step 5: Substitute u′ d and u′ q After coordinate transformation, the duty cycle signal is output through space vector pulse width modulation to the inverter to drive the motor to operate.

3. The current control method of a permanent magnet synchronous motor at high speed and low carrier ratio according to claim 2 is characterized in that: The active damping complex vector controller is designed based on the complex vector active damping motor model using the zero-pole cancellation principle, and its transfer function is expressed as: Where K is the current controller bandwidth, R a is a virtual resistor, is the estimated value of motor phase inductance, R s is the motor phase resistance, T s is the control period, ω e is the electrical angular velocity, j is the imaginary operator, and z is the lead operator; the complex vector active damping motor model is obtained by adding a virtual resistor to the discrete time domain complex vector mathematical model, wherein the complex vector active damping motor model is: The discrete time domain complex vector mathematical model is:

4. The current control method of a permanent magnet synchronous motor at high speed and low carrier ratio according to claim 2 is characterized in that: The d-axis current prediction ESO includes a d-axis model, and the q-axis current prediction ESO includes a q-axis model. The d-axis model is: The q-axis model is: Where k represents the kth moment, k+1 represents the k+1th moment, and f d (k) and f q (k) are d-axis perturbation and q-axis perturbation respectively, and i d (k) and f d The estimated value of (k), and i q (k) and f q The estimated value of (k), i err_d (k) and i err_q (k) are the d-axis and q-axis stator current errors, β1 and β2 are ESO gains, Estimate the inductance value for the Inductance Estimator.

5. The current control method of a permanent magnet synchronous motor at high speed and low carrier ratio according to claim 4 is characterized in that: The motor model is: Where p represents the differential operator, f d represents the d-axis disturbance, f q represents the q-axis perturbation.

6. The current control method of a permanent magnet synchronous motor at high speed and low carrier ratio according to claim 2 is characterized in that: The inductance estimator is: where α is the inductance estimator bandwidth.

Images