A memory motor model predictive current control method with constant switching frequency

Abstract

The present application belongs to the technical field of memory motor control, and discloses a memory motor model predictive current control method for realizing constant switching frequency, comprising the following steps: collecting memory motor speed, rotor position, three-phase winding current and inverter bus voltage; calculating winding current dq axis component; calculating q-axis current given value, obtaining d-axis current given value; updating direct and quadrature axis inductance and permanent magnet flux linkage in the memory motor current prediction model; selecting the optimal value from six non-zero voltage vectors; calculating the duty cycle of the optimal non-zero voltage vector and the duty cycle of the zero vector, and obtaining the inverter three-phase bridge arm switching tube control signal. By inserting zero vectors, the system steady-state performance is improved and the inverter switching frequency is fixed in the mode of acting one non-zero voltage vector and two zero voltage vectors in each control cycle, which overcomes the defects of traditional model predictive current control, such as dependence on accurate mathematical model, poor steady-state performance, and non-fixed switching frequency.

Description

Technical Field

[0001] This invention belongs to the field of memory motor control technology, specifically relating to a memory motor model predictive current control method for achieving constant switching frequency. Background Technology

[0002] Compared to traditional permanent magnet synchronous motors, memory motors (MMMs) offer controllable permanent magnet flux linkage, facilitating wide-range speed regulation and achieving a broad, efficient operating range. Therefore, they have garnered widespread research and attention from scholars both domestically and internationally. Since the concept of memory motors emerged in 2001, significant progress has been made in their design and optimization techniques over the past two decades. In recent years, researchers have begun to focus on the control technology of memory motors. The unique magnet regulation characteristics of memory motors place extremely high demands on the robustness, speed, and accuracy of their control strategies. However, existing publications primarily utilize PI controllers and linear active disturbance rejection controllers to control memory motors, which present challenges such as limited control bandwidth and complex controller parameter selection.

[0003] Unlike traditional linear control, model predictive control (MDC) offers significant advantages in control bandwidth and parameter design, making it a research hotspot in power electronic converters and motor drives in recent years. MDC is essentially an online optimization algorithm based on an accurate mathematical model, demanding extremely high precision in model parameters. However, changes in the magnetization state of a memory motor inevitably lead to variations in parameters such as inductance and flux linkage, making it difficult to meet the precision requirements of MDC. Furthermore, traditional MDC algorithms select only one voltage vector in each control cycle, resulting in poor steady-state performance and inconsistent inverter switching frequencies. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a model predictive current control method for memory motors that achieves constant switching frequency, thereby solving the defects of existing technologies such as reliance on precise mathematical models, poor steady-state performance, and inconsistent switching frequency.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A method for predictive current control of a memory motor model to achieve constant switching frequency includes the following steps:

[0007] The system collects and stores the motor speed, motor rotor position, three-phase winding current, and inverter bus voltage; and uses the Park transform module to calculate the dq-axis components of the winding current.

[0008] The q-axis current setpoint is calculated using a speed controller, and the d-axis current setpoint is obtained using a magnetization state controller.

[0009] Based on the comparison between the magnetization state and the parameter storage table, the quadrature and direct axis inductances and permanent magnet flux linkages in the memory motor current prediction model are updated;

[0010] Using the value function, the optimal non-zero voltage vector is selected from six non-zero voltage vectors;

[0011] The duty cycle calculation module is used to calculate the duty cycle of the optimal non-zero voltage vector and the duty cycle of the zero vector. Then, the control signal of the inverter three-phase bridge arm switching transistor is obtained through the control signal generation module.

[0012] Furthermore, the dq-axis components of the winding current are calculated using the Park transform module, as shown in the following equation:

[0013]

[0014] In the formula, θ e To memorize the rotor position of the motor, i a i b i c For the three-phase winding current, i d i q Let d be the q-axis component of the winding current.

[0015] Furthermore, the q-axis current setpoint is calculated using the speed controller. As shown in the following formula:

[0016]

[0017] In the formula, k p k is the proportionality coefficient. i n is the integral coefficient. * This is the setpoint for the rotational speed.

[0018] Furthermore, the d-axis current setpoint is obtained using a magnetization state controller;

[0019] Based on the given speed n of the memory motor * Given a reference magnetization state MS within the specified speed range * As shown in the following formula:

[0020]

[0021] In the formula, n ts This is the rotational speed threshold.

[0022] Based on the current magnetization state MS and the reference magnetization state MS * Determine the reference value of the d-axis current. As shown in the following formula:

[0023]

[0024] In the formula, idp1 (t) and i dp2 (t) represents a trapezoidal pulse wave.

[0025] Furthermore, the parameter storage table contains the direct and quadrature axis inductance L of the memory motor under different magnetization states. d (MS), L q (MS) and permanent magnet flux ψ PM (MS).

[0026] Furthermore, the optimal non-zero voltage vector is selected from the six non-zero voltage vectors, specifically including the following steps:

[0027] Using a current prediction model, calculate six non-zero voltage vectors V. i The predicted current value i corresponding to (i = 1 to 6) s (k+1|V i The current prediction model is as follows:

[0028]

[0029] In the formula, i d (k+1|V i ) and i q (k+1|V i ) are respectively the voltage vector V i The corresponding predicted dq-axis current, i d (k) and i q (k) represent the sampled values ​​of the dq-axis currents at the current moment, T s For the sampling period, u d (V i ) and u q (V i ) are respectively the voltage vector V i The corresponding dq-axis voltage, R is the winding internal resistance, ω e This is for memorizing the electric angular velocity of the motor.

[0030] Current prediction value i s (k+1|V i Specifically:

[0031] i s (k+1|V i )=[i d (k+1|V i )i q (k+1|V i )] T

[0032] The effect of the six non-zero voltage vectors is evaluated using a value function to obtain g(V) i The value function is as follows:

[0033]

[0034] Then, by comparing g(V) i The optimal non-zero voltage vector V is obtained. opt The optimal non-zero voltage vector V opt satisfy:

[0035]

[0036] Further, the duty cycle of the optimal non-zero voltage vector and the duty cycle of the zero vector are calculated, and then the control signals of the inverter's three-phase bridge arm switching transistors are obtained through the control signal generation module. Specifically, this includes the following steps:

[0037] Calculate the zero vector value function value g(V) z )=g(V0)=g(V7);

[0038] Calculate the optimal non-zero voltage vector V opt duty cycle d opt With zero vector V z duty cycle d z Specifically:

[0039]

[0040] Based on the allocation of the optimal non-zero voltage vector V opt With duty cycle d opt d z Using the control signal generation module, the duty cycle d of the three-phase bridge arms is allocated. a d b With d c Specifically:

[0041] If the optimal non-zero vector V opt =V1, then:

[0042]

[0043] If the optimal non-zero vector V opt =V2, then:

[0044]

[0045] If the optimal non-zero vector V opt =V3, then:

[0046]

[0047] If the optimal non-zero vector V opt =V4, then:

[0048]

[0049] If the optimal non-zero vector V opt =V5, then:

[0050]

[0051] If the optimal non-zero vector V opt =V6, then:

[0052]

[0053] Based on the duty cycle d of the three-phase bridge arm a d b With d c The three-phase bridge arm control signals are obtained.

[0054] The beneficial effects of this invention are:

[0055] 1. The method provided by this invention can significantly improve the steady-state performance of the system, reduce current and electromagnetic torque ripple, and ensure that the inverter switching frequency and sampling frequency are consistent. At the same time, the dynamic performance of the method provided by this invention is comparable to that of the traditional MPCC algorithm, and it can quickly track the reference values ​​of rotational speed and magnetic current.

[0056] 2. This method considers the changes in the parameters of the memory motor under different magnetization states. By inserting zero vectors and applying one non-zero voltage vector and two zero voltage vectors in each control cycle, the steady-state performance of the system is improved and the inverter switching frequency is fixed. This overcomes the shortcomings of traditional model predictive current control, such as reliance on accurate mathematical models, poor steady-state performance, and inconsistent switching frequency. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a control block diagram of the present invention;

[0059] Figure 2 This is a schematic diagram of the trapezoidal pulse wave of the magnetizing current of the present invention;

[0060] Figure 3 The linear diagram of the A-phase stator current waveform for predictive current control in this invention and the traditional model is shown.

[0061] Figure 4 The diagram shows the linear waveforms of the dq-axis currents predicted by the present invention and the traditional model for current control.

[0062] Figure 5 The linear graph of the electromagnetic torque waveform for predictive current control in this invention and the traditional model is shown.

[0063] Figure 6 This is a linear graph of the motor speed waveform for predictive current control based on the present invention and a traditional model;

[0064] Figure 7 This is a linear graph of the switching frequency waveform for predictive current control based on the present invention and a conventional model. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] Combination Figure 1-7 As shown, this invention provides a method for predictive current control of a memory motor model to achieve a constant switching frequency, comprising the following steps:

[0067] Step S1: Collect and memorize the motor speed n and the motor rotor position θ. e Three-phase winding current i a i b i c Inverter bus voltage V dc The Park transform module is used to calculate the dq-axis components i of the winding current. d with i q The formula is as follows:

[0068]

[0069] Step S2: Calculate the q-axis current setpoint using the speed controller. The d-axis current setpoint is obtained using a magnetization state controller.

[0070] Specifically, step S2 can be divided into the following steps:

[0071] Step S21: Calculate the q-axis current setpoint using the speed controller. As shown in the following formula:

[0072]

[0073] In the formula, k p k is the proportionality coefficient. i n is the integral coefficient.* This is the setpoint for the rotational speed.

[0074] Step S22: Based on the given speed n of the memory motor * Given a reference magnetization state MS within the specified speed range * As shown in the following formula:

[0075]

[0076] In the formula, n ts The rotational speed threshold can be obtained through finite element analysis or experiments. In this embodiment, n ts It is 800 revolutions per minute.

[0077] Step S23: Based on the current magnetization state MS and the reference magnetization state MS obtained in step S21 * The reference value for the d-axis current is determined as follows:

[0078]

[0079] In the formula, i dp1 (t) and i dp2 (t) is a trapezoidal pulse wave, the amplitude of which can be determined by finite element method or experimental method.

[0080] In this embodiment, as Figure 2 As shown, i dp1 The amplitude of (t) is -25.4A, i dp2 The amplitude of (t) is 32A, and the excitation phase, hold phase and recovery phase are all 10ms.

[0081] Step S3: Based on the comparison between the magnetization state and the parameter storage table, adjust the quadrature and direct axis inductance L in the memory motor current prediction model. d (MS), L q (MS), permanent magnet flux ψ PM (MS) is updated. Specifically, the parameter storage table contains the direct and quadrature axis inductance L of the memory motor under different magnetization states. d (MS), L q (MS), permanent magnet flux ψ PM (MS). This can be obtained through finite element analysis or experimental methods. In this embodiment, the parameter storage table is shown in Table 1.

[0082] Table 1

[0083]

[0084] Step S4: Using the value function, from the 6 non-zero voltage vectors V i Select the optimal non-zero voltage vector V from (i = 1 to 6). opt .

[0085] Specifically, firstly, using a current prediction model, six non-zero voltage vectors V are calculated. i The predicted current value i corresponding to (i = 1 to 6) s (k+1|V i The current prediction model is as follows:

[0086]

[0087] In the formula, i d (k+1|V i ) and i q (k+1|V i ) are respectively the voltage vector V i The corresponding predicted dq-axis current, i d (k) and i q (k) represent the sampled values ​​of the dq-axis currents at the current moment, T s For the sampling period, u d (V i ) and u q (V i ) are respectively the voltage vector V i The corresponding dq-axis voltage, R is the winding internal resistance, ω e To store the electric angular velocity of the motor, the predicted current value i s (k+1|V i Specifically:

[0088] i s (k+1|V i )=[i d (k+1|V i )i q (k+1|V i )] T (4)

[0089] Secondly, the effect of the six non-zero voltage vectors is evaluated using a value function to obtain g(V i The value function is specifically:

[0090]

[0091] In the formula, This is the given value for the dq-axis current.

[0092] Then, by comparing g(V) i The optimal non-zero voltage vector V is obtained. opt The optimal non-zero voltage vector V opt satisfy:

[0093]

[0094] Step S5: Calculate the optimal non-zero voltage vector V using the duty cycle calculation module. opt duty cycle d opt With zero vector V z duty cycle d z Then, the control signals of the inverter's three-phase bridge arm switching transistors are obtained through the control signal generation module.

[0095] Specifically, step S5 can be divided into the following steps:

[0096] Step S51: Calculate the zero vector value function value g(V) z )=g(V0)=g(V7);

[0097] Step S52: Calculate the optimal non-zero voltage vector V opt duty cycle d opt With zero vector V z duty cycle d z Specifically:

[0098]

[0099] Step S53: Based on the allocation of the optimal non-zero voltage vector V opt With duty cycle d opt d z Using the control signal generation module, the duty cycle d of the three-phase bridge arms is allocated. a d b With d c Specifically:

[0100] If the optimal non-zero vector V opt =V1, then:

[0101]

[0102] If the optimal non-zero vector V opt =V2, then:

[0103]

[0104] If the optimal non-zero vector V opt =V3, then:

[0105]

[0106] If the optimal non-zero vector V opt =V4, then:

[0107]

[0108] If the optimal non-zero vector V opt =V5, then:

[0109]

[0110] If the optimal non-zero vector V opt =V6, then:

[0111]

[0112] Step S54: Based on the calculated duty cycle d a d b With d c The control signals for the three-phase bridge arms of the inverter are obtained.

[0113] The simulation conditions for this embodiment of the invention are as follows: the load torque is given as 3 N·m, the reference speed is initially given as 600 rpm, it linearly increases to 1000 rpm at 0.3s, and linearly decreases to 600 rpm at 0.7s, with a sampling frequency of 10 kHz.

[0114] Simulation results are as follows Figure 3-7 As shown, compared with the traditional MPCC algorithm, the method provided by this invention can significantly improve the steady-state performance of the system, reduce current and electromagnetic torque ripple, and ensure that the inverter switching frequency and sampling frequency are consistent. At the same time, the dynamic performance of the method provided by this invention is comparable to that of the traditional MPCC algorithm, and it can quickly track the reference values ​​of rotational speed and magnetic current.

[0115] This method considers the changes in the parameters of the memory motor under different magnetization states. By inserting zero vectors and applying one non-zero voltage vector and two zero voltage vectors in each control cycle, the steady-state performance of the system is improved and the inverter switching frequency is fixed. This overcomes the shortcomings of traditional model predictive current control, such as reliance on accurate mathematical models, poor steady-state performance, and inconsistent switching frequency.

[0116] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0117] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the present invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for predictive current control of a memory motor model to achieve constant switching frequency, characterized in that, Includes the following steps: The system collects and stores the motor speed, rotor position, three-phase winding current, and inverter bus voltage; and uses the Park transform module to calculate the winding current. dq Axial components; Calculation using speed controller q The shaft current setpoint is obtained using the magnetization state controller. d Shaft current setpoint; Based on the comparison between the magnetization state and the parameter storage table, the quadrature and direct axis inductances and permanent magnet flux linkages in the memory motor current prediction model are updated; Using the value function, the optimal non-zero voltage vector is selected from six non-zero voltage vectors; The duty cycle calculation module is used to calculate the duty cycle of the optimal non-zero voltage vector and the duty cycle of the zero vector. Then, the control signal of the inverter three-phase bridge arm switching transistor is obtained through the control signal generation module. The parameter storage table contains the direct and quadrature axis inductances of the memory motor under different magnetization states. L d ( MS ), L q ( MS ) and permanent magnet flux ( MS );in, MS This represents the current magnetization state. The selection of the optimal non-zero voltage vector from the six non-zero voltage vectors includes the following steps: Using a current prediction model, calculate six non-zero voltage vectors. V i Corresponding current prediction value i s ( k +1| V i ), i = 1~6, the current prediction model is as follows: In the formula, i d ( k +1| V i )and i q ( k +1| V i () are voltage vectors V i corresponding dq Predicted shaft current value i d ( k )and i q ( k ) are respectively dq The sampled value of the shaft current at the current moment. T s The sampling period is u d ( V i )and u q ( V i () are voltage vectors V i corresponding dq shaft voltage, R For the winding internal resistance, ω e To store the electric angular velocity of the motor; Current prediction value i s ( k +1| V i Specifically: The effect of the six non-zero voltage vectors was evaluated using a value function. g ( V i The value function is as follows: Then, through comparison g ( V i ), thus obtaining the optimal non-zero voltage vector. V opt The optimal non-zero voltage vector V opt satisfy: 。 2. The method for predictive current control of a memory motor model with constant switching frequency according to claim 1, characterized in that, Calculating winding current using the Park transform module dq The axial components are as follows: In the formula, θ e To memorize the position of the motor rotor, i a , i b , i c This refers to the three-phase winding current. i d , i q For winding current dq Axial components.

3. The method for predictive current control of a memory motor model with constant switching frequency according to claim 1, characterized in that, Calculation using speed controller q Shaft current setpoint ; as shown in the following formula: In the formula, k p This is the proportionality coefficient. k i The integral coefficient is... n * This is the setpoint for the rotational speed.

4. The method for predictive current control of a memory motor model with constant switching frequency according to claim 1, characterized in that, Obtained using magnetization state controller d Shaft current setpoint; Based on the speed given by the memory motor Given a reference magnetization state within the specified speed range As shown in the following formula: In the formula, n ts This is the rotational speed threshold; Based on the current magnetization state MS Compared with the reference magnetization state ,Sure d Shaft current reference value As shown in the following formula: In the formula, i dp1 ( t )and i dp2 ( t () is a trapezoidal pulse wave.

5. The method for predictive current control of a memory motor model with constant switching frequency according to claim 1, characterized in that, The duty cycle of the optimal non-zero voltage vector and the duty cycle of the zero vector are calculated. Then, the control signals of the inverter's three-phase bridge arm switching transistors are obtained through the control signal generation module. The specific steps include: Calculate the zero vector value function value g ( V z ) = g ( V 0) = g ( V 7); Calculate the optimal non-zero voltage vector V opt duty cycle d opt With zero vector V z duty cycle d z Specifically: Based on the allocation of the optimal non-zero voltage vector V opt Duty cycle d opt , d z The control signal generation module is used to allocate the duty cycle of the three-phase bridge arms. d a , d b and d c Specifically: If the optimal nonzero vector V opt = V 1, then: If the optimal nonzero vector V opt = V 2, then: If the optimal nonzero vector V opt = V 3, then: If the optimal nonzero vector V opt = V 4, then: If the optimal nonzero vector V opt = V 5, then: If the optimal nonzero vector V opt = V 6, then: Based on the duty cycle of the three-phase bridge arm d a , d b and d c The three-phase bridge arm control signals are obtained.

Images