A sequential model predictive control method for asynchronous motor in stationary coordinate system

Abstract

This invention proposes a sequential model predictive control method for asynchronous motors in a stationary coordinate system. The steps are as follows: sample the three-phase stator currents (a, b, c) of the asynchronous motor and transform them to the αβ stationary coordinate system; calculate the rotor flux linkage on the α-axis and β-axis of the asynchronous motor; define three β-axis candidate voltages using the DC bus voltage of the three-phase two-level inverter, and predict the three β-axis currents at the next moment; calculate the value of the first cost function; select the β-axis candidate voltage that minimizes the value of the first cost function as the optimal β-axis voltage; select three or two α-axis candidate voltages, predict the α-axis current at the next moment, and calculate the value of the second cost function, selecting the α-axis candidate voltage that minimizes the value of the second cost function as the optimal α-axis voltage; use the obtained optimal β-axis voltage and optimal α-axis voltage as outputs to control the asynchronous motor. This invention significantly reduces the computational complexity of asynchronous motor model predictive control.

Description

Technical Field

[0001] This invention relates to the technical field of power electronics and electric drive, and in particular to a sequential model predictive control method for asynchronous motors in a stationary coordinate system. Background Technology

[0002] With the rapid development of technologies such as new energy vehicles and tunnel boring machines, asynchronous motors have been widely used due to their low cost and high reliability. In recent years, model predictive control technology has received widespread attention and research in asynchronous motor control due to its advantages such as simple principle, easy implementation, and fast dynamic response. However, existing model predictive control often suffers from drawbacks such as computational complexity.

[0003] The literature [MS Mousavi, SA Davari, V. Nekoukar, C. Garcia and J. Rodriguez, "Integral Sliding Mode Observer-Based Ultralocal Model for Finite-Set Model Predictive Current Control of Induction Motor," in IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 10, no. 3, pp. 2912-2922, June 2022] studies a robust model predictive current control method for asynchronous motors. By designing an integral sliding mode observer for disturbance observation, the parameter robustness is improved. However, this method requires simultaneous prediction of α-axis and β-axis currents and evaluation of the cost function, resulting in a large computational burden.

[0004] The literature [MS Mousavi, SA Davari, V. Nekoukar, C. Garcia and J. Rodriguez, "Finite-Set Model Predictive Current Control of Induction Motorsby Direct Use of Total Disturbance," in IEEE Access, vol. 9, pp. 107779-107790, 2021] also studies a robust model predictive current control method for asynchronous motors. By designing a lumped disturbance observer for disturbance observation, the parameter robustness is improved. However, this method also requires simultaneous prediction of α-axis and β-axis currents and evaluation of the cost function, resulting in a large computational load.

[0005] The literature [F. Wang, G. Lin and Y. He, "Passivity-Based Model Predictive Control of Three-Level Inverter-Fed Induction Motor," in IEEE Transactions on Power Electronics, vol. 36, no. 2, pp. 1984-1993, Feb. 2021] studies a model predictive current control method for asynchronous motors based on three-level inverters. This method also requires simultaneous prediction of α-axis and β-axis current and voltage and evaluation of cost functions, resulting in a large computational load.

[0006] The literature [Liu Meng, Lu Ziguang, Yang Shuaishuai. Low-complexity three-vector predictive current control for induction motors [J]. Journal of Guangxi University (Natural Science Edition), 2021, 46 (03): 632-641.] proposes a low-complexity three-vector model predictive control method for induction motors (also known as asynchronous motors). However, this method also requires simultaneous prediction of α-axis and β-axis currents and evaluation of cost functions, resulting in a relatively large computational load.

[0007] The literature [Zhang Yongchang, Yang Haitao. Sensorless Model Predictive Control of Asynchronous Motors [J]. Proceedings of the CSEE, 2014, 34 (15): 2422-2429.] studied sensorless model predictive control of asynchronous motors, but this method still requires simultaneous prediction of α-axis current and β-axis current, and then prediction of flux linkage and torque, which still has the problem of large computational load. Summary of the Invention

[0008] To address the problem that traditional asynchronous motor model prediction methods require simultaneous prediction of α-axis and β-axis currents and cost function evaluation, resulting in high computational complexity, this invention proposes a sequential model predictive control method for asynchronous motors in a stationary coordinate system. This method aims to break away from the conventional model predictive control concept that requires simultaneous prediction of α-axis and β-axis currents and cost function evaluation, achieving sequential model predictive control by first performing β-axis current prediction and cost function evaluation, followed by α-axis current prediction and cost function evaluation, thus significantly reducing the computational complexity of model predictive control.

[0009] To achieve the above objectives, the technical solution of the present invention is as follows: a sequential model predictive control method for an asynchronous motor in a stationary coordinate system, comprising the following steps:

[0010] Step 1: Sample the three-phase stator currents abc of the asynchronous motor and transform the three-phase stator currents to the αβ stationary coordinate system to obtain the current on the α axis and the current on the β axis;

[0011] Step 2: Based on the current on the α-axis, the current on the β-axis, and the mathematical model of the asynchronous motor, calculate the rotor flux linkage on the α-axis and the rotor flux linkage on the β-axis of the asynchronous motor.

[0012] Step 3: Define three β-axis candidate voltages using the DC bus voltage of the three-phase two-level inverter. Predict the three β-axis currents at the next moment using the three β-axis candidate voltages, the rotor flux linkage on the α-axis, the rotor flux linkage on the β-axis, and the current on the β-axis.

[0013] Step 4: Calculate the value of the first cost function based on the three β-axis currents and the β-axis current reference value at the next moment;

[0014] Step 5: Compare the values ​​of the three first cost functions obtained, and take the β-axis candidate voltage that minimizes the value of the first cost function as the optimal β-axis voltage;

[0015] Step 6: Based on the obtained optimal β-axis voltage, select 3 or 2 α-axis candidate voltages, and use the α-axis candidate voltages, rotor flux linkage on the β-axis, rotor flux linkage on the α-axis, current on the α-axis, and α-axis current reference value to predict the α-axis current at the next moment and calculate the value of the second cost function. Select the α-axis candidate voltage that minimizes the value of the second cost function as the optimal α-axis voltage.

[0016] Step 7: Use the obtained optimal β-axis voltage and optimal α-axis voltage as outputs to control the asynchronous motor.

[0017] Preferably, the three β-axis alternative voltages U β1 U β2 U β3 respectively satisfy U β1 =0, U β2 = U β3 = , among which, U dc This indicates the DC bus voltage of a three-phase two-level inverter;

[0018] When the optimal β-axis voltage U βopt = U β1 At that time, three alternative voltages for the α-axis are selected as U α11 =0, U α12 = U α13 = ;

[0019] When the optimal β-axis voltage U βopt ≠Uβ1 When selecting two alternative voltages for the α-axis, U α21 = U α22 = .

[0020] Preferably, the current on the α-axis and the current on the β-axis are calculated as follows:

[0021] ;

[0022] Among them, i a i represents the stator current of phase a of the asynchronous motor. b i represents the b-phase stator current of the asynchronous motor. c i represents the c-phase stator current of an asynchronous motor. α i represents the current of the asynchronous motor on the α axis. β This represents the current in the asynchronous motor along the β axis;

[0023] The calculation methods for the rotor flux linkage on the α-axis and the rotor flux linkage on the β-axis of the asynchronous motor are as follows:

[0024] ;

[0025] Where, ψ rα Let ψ be the rotor flux linkage on the α-axis of the asynchronous motor. rβ L represents the rotor flux linkage on the β-axis of the asynchronous motor. m T represents the mutual inductance of an asynchronous motor. r It is the rotor time constant, ω r Let be the rotor electric angular velocity of the asynchronous motor, s represent the Laplace operator, and j represent a 90-degree lead.

[0026] Preferably, the method for predicting the three β-axis currents at the next moment is as follows:

[0027] ;

[0028] Among them, U βm For the β-axis alternative voltages, m = 1, 2, 3; T s ω represents the control period. r The rotor electric angular velocity of the asynchronous motor is given by ψ, where a, b, c, and d are four coefficients. rα Let ψ be the rotor flux linkage on the α-axis of the asynchronous motor. rβ i represents the rotor flux linkage of the asynchronous motor on the β-axis; β This represents the current in the asynchronous motor along the β axis;

[0029] When the optimal β-axis voltage U βopt =U β1 The method for predicting the α-axis current at the next moment is as follows:

[0030] ;

[0031] Among them, U α1j For the α-axis, the alternative voltages are j=1,2,3; i α This represents the current in the asynchronous motor along the α axis;

[0032] When the optimal β-axis voltage U βopt ≠U β1 The method for predicting the α-axis current at the next moment is as follows:

[0033] ;

[0034] Among them, U α2k The alternative voltages for the α-axis are k=1,2.

[0035] Preferably, the first cost function ; where i refβ This is the reference value for the β-axis current.

[0036] Preferably, when the optimal β-axis voltage U βopt =U β1 When, the second cost function is In the formula, i refα This is the reference value for the α-axis current.

[0037] When the optimal β-axis voltage U βopt ≠U β1 When, the second cost function is .

[0038] Preferably, the coefficient ,coefficient ,coefficient ,coefficient , Let represent the leakage inductance coefficient, and satisfy . L m For the mutual inductance of the asynchronous motor, L s L is the stator inductance of the asynchronous motor. r R is the rotor inductance of the asynchronous motor. s R is the stator resistance of the asynchronous motor. r T is the rotor resistance of the asynchronous motor. r Let T be the rotor time constant of the asynchronous motor, satisfying T r =L r / R r .

[0039] Preferably, the β-axis current reference value i refβ From the d-axis current reference value i refd and q-axis current reference value irefq Calculated through coordinate transformation, and satisfying the following: In the formula, θ represents the orientation angle of the control system;

[0040] α-axis current reference value i refα From the d-axis current reference value i refd and q-axis current reference value i refq Calculated through coordinate transformation, and satisfying the following: .

[0041] Preferably, when the optimal β-axis voltage U βopt =U β1 At that time, three alternative voltages for the α-axis are defined as U α11 U α12 U α13 respectively satisfy U α11 =0, U α12 = U α13 = According to the α-axis candidate voltage U α1j Predict the α-axis current i at the next moment α1j Using the predicted three α-axis currents i α1j Calculate the values ​​g of the three second cost functions. 21j The values ​​of the three second cost functions obtained by comparison, g 21j To obtain the value of the second cost function g 21j The candidate voltage for the α-axis that corresponds to the minimum value is the optimal α-axis voltage U. αopt .

[0042] Preferably, when the optimal β-axis voltage U βopt ≠U β1 At that time, two alternative voltages for the α-axis are defined as U α21 U α22 respectively satisfy U α21 = U α22 = And based on the α-axis alternative voltage U α2k (k=1,2) Predict the α-axis current i at the next time step. α2k Based on the predicted two α-axis currents i α2k Calculate the value of the second cost function g 22k ; compared to the values ​​of the two second cost functions g obtained 22k To obtain the value g of the second cost function 22k The candidate voltage for the α-axis that corresponds to the minimum value is the optimal α-axis voltage U. αopt .

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention proposes a low-complexity sequential model predictive control method for asynchronous motors in a stationary coordinate system, which realizes sequential model predictive control of "first performing β-axis current prediction and cost function evaluation, and then performing α-axis current prediction and cost function evaluation", reducing the total number of α-axis and β-axis current predictions from 16 times in the conventional method to 6 or 5 times, which significantly reduces the computational complexity of asynchronous motor model predictive control. Attached Figure Description

[0044] 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, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart of a conventional method.

[0046] Figure 2 This is a flowchart of the present invention.

[0047] Figure 3 This is a schematic diagram showing the positions of the β-axis and α-axis alternative voltages proposed in this invention.

[0048] Figure 4 This is a control block diagram of the present invention.

[0049] Figure 5 The figure shows the simulation results of the conventional 16-iteration model predictive control method, where (a) represents the rotational speed and (b) represents the current.

[0050] Figure 6 The simulation results of the control method with 6 or 5 iterations proposed in this invention are shown in the figure, where (a) is the rotational speed and (b) is the current. Detailed Implementation

[0051] 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.

[0052] Figure 1 This is a flowchart of a conventional method. In the diagram, [U α0 U β0 ]=[0, 0],[U α1 U β1 ]=[ ,0],[U α2 U β2 ]=[ , ], [U α3 U β3 ]=[ , ], [U α4 U β4 ]=[ ,0],[U α5 U β5 ]=[ , ], [U α6 U β6 ]=[ , ], [U α7 U β7 =[0, 0]. From Figure 1 It is evident that the conventional method requires eight α-axis and eight β-axis candidate voltages corresponding to the eight voltage vectors output by the three-phase two-level inverter to perform eight α-axis current predictions and eight β-axis predictions simultaneously. The optimal α-axis voltage and the optimal β-axis voltage are obtained by calculating and comparing the cost function. The total number of iterations required is 16, which is computationally intensive.

[0053] like Figure 2 As shown, a sequential model predictive control method for an asynchronous motor in a stationary coordinate system is proposed. First, three β-axis candidate voltages are defined, and a first cost function value is calculated using these three β-axis candidate voltages. The β-axis candidate voltage that minimizes the first cost function is identified as the optimal β-axis voltage. Second, based on the optimal β-axis voltage, three or two α-axis candidate voltages are determined, and a second cost function value is calculated using these three or two α-axis candidate voltages. The α-axis candidate voltage that minimizes the second cost function value is identified as the optimal α-axis voltage. Finally, the selected optimal β-axis voltage and optimal α-axis voltage are used as outputs to control the asynchronous motor, thereby achieving low-complexity sequential model predictive control of the asynchronous motor in a stationary coordinate system. This invention reduces the total number of α-axis and β-axis current predictions from 16 to 6 or 5, significantly reducing the computational complexity of asynchronous motor model predictive control. The specific steps of this invention are as follows:

[0054] Step 1: Sample the three-phase stator current i of the asynchronous motor (phases abc). a i b i c and the stator current i a i b i c Transforming to the αβ two-dimensional stationary coordinate system, we obtain the current i α i β .

[0055] Current i α i β The calculation method is as follows:

[0056] ;

[0057] Among them, i a i represents the stator current of phase a of the asynchronous motor. b i represents the b-phase stator current of the asynchronous motor. c i represents the c-phase stator current of an asynchronous motor. α i represents the current of the asynchronous motor on the α axis. β This represents the current on the β axis of the asynchronous motor.

[0058] Step 2: Based on the current i obtained in Step 1 α i β Using the mathematical model of the asynchronous motor, calculate the rotor flux linkage ψ of the asynchronous motor. rα ψ rβ .

[0059] The mathematical model of an asynchronous motor can be expressed as:

[0060] ;

[0061] Where, ω r T is the rotor electric angular velocity of the asynchronous motor. r Let T be the rotor time constant of the asynchronous motor, satisfying T r =L r / R r , ψ rα Let ψ be the rotor flux linkage on the α-axis of the asynchronous motor. rβ Let u be the rotor flux linkage of the asynchronous motor on the β-axis, where a, b, c, and d are four coefficients. α u is the stator voltage of the asynchronous motor on the α-axis. β The coefficient is the stator voltage of the asynchronous motor on the β-axis. ,coefficient ,coefficient ,coefficient , Let represent the leakage inductance coefficient, and satisfy . L m For the mutual inductance of the asynchronous motor, L s L is the stator inductance of the asynchronous motor. r R is the rotor inductance of the asynchronous motor. s R is the stator resistance of the asynchronous motor. r This represents the rotor resistance of the asynchronous motor.

[0062] By rewriting the first two formulas of the above asynchronous motor mathematical model in complex form, we can obtain the rotor flux linkage ψ. rα ψ rβ The calculation method is as follows:

[0063] ;

[0064] Where, ψ rα Let ψ be the rotor flux linkage on the α-axis of the asynchronous motor. rβ L represents the rotor flux linkage on the β-axis of the asynchronous motor. m T represents the mutual inductance of an asynchronous motor. r It is the rotor time constant, satisfying T r =L r / R r L r R represents the rotor inductance of an asynchronous motor. r ω represents the rotor resistance of an asynchronous motor. r Let be the rotor electric angular velocity of the asynchronous motor, s represent the Laplace operator, and j represent a 90-degree lead.

[0065] Step 3: Define three β-axis candidate voltages using the DC bus voltage of the three-phase two-level inverter, and predict the three β-axis currents i at the next moment using these three β-axis candidate voltages. βm (m=1,2,3).

[0066] like Figure 3 As shown, considering that the inverter output has 8 voltage vectors with 3 different β-axis components, 3 β-axis candidate voltages U can be defined to predict the β-axis current. β1 U β2 U β3 respectively satisfy U β1 =0, U β2 = U β3 = , among which, U dc The DC bus voltage of a three-phase two-level inverter is represented by these three components. Using these three components for current prediction is equivalent to predicting the current of all eight vector β-axis components. The goal is to select the optimal β-axis voltage and reduce the number of iterations. The β-axis candidate voltage U is then used. βm Substituting (m=1,2,3) into the β-axis current prediction model shown in the following equation, the β-axis current i at the next moment is predicted. βm ,satisfy:

[0067] ;

[0068] In the formula, T s Indicates the control cycle.

[0069] Step 4: Based on the three β-axis currents i at the next moment βm The first cost function value is calculated using the β-axis current reference value.

[0070] The three β-axis currents i predicted in step 3 βm Substitute the values ​​into the first cost function shown in the following equation to calculate the value of the first cost function g. 1m ,satisfy:

[0071] ;

[0072] In the formula, g 1m Alternate voltage U for the β-axis βm The first cost function value corresponding to (m=1,2,3), i refβ The β-axis current reference value can be derived from the d-axis current reference value i. refd and q-axis current reference value i refq Calculated through coordinate transformation, and satisfying the following:

[0073] ;

[0074] In the formula, θ represents the orientation angle of the control system, which satisfies: t represents time. It is usually obtained from the rated excitation current of the asynchronous motor. It is obtained from the output of the speed closed loop.

[0075] Step 5: Compare the three first cost functions g obtained in Step 4. 1m The value of (m=1,2,3) is obtained to make the first cost function value g. 1m The candidate voltage for the β-axis that is at its minimum is taken as the optimal β-axis voltage U. βopt ;

[0076] By executing steps 3-5, the optimal β-axis component among the eight voltage vectors of the inverter can be determined through three current predictions and the first cost function evaluation. This reduces the number of iterations and paves the way for subsequent determination of the optimal α-axis voltage U. αopt Laying the foundation.

[0077] Step 6: Based on the optimal β-axis voltage U obtained in Step 5 βopt Select two or three α-axis candidate voltages, and use these candidate voltages to predict the α-axis current at the next time step and calculate the second cost function. Select the α-axis candidate voltage that minimizes the second cost function as the optimal α-axis voltage U. αopt Specifically:

[0078] When the optimal β-axis voltage U βopt =U β1 Then, execute steps S611, S612, and S613:

[0079] Step S611, as follows Figure 3 As shown, when the optimal β-axis voltage U βopt =U β1 At that time, the corresponding voltage vectors are u0, u4, and u1, and the α-axis voltages corresponding to these three voltage vectors are 0, u4, and u1, respectively. , Therefore, the three α-axis candidate voltages are defined as U... α11 U α12 U α13 respectively satisfy U α11 =0, U α12 = U α13 = And set the α-axis alternative voltage U α1j Substituting j=1,2,3 into the α-axis current prediction model shown in the following equation, we can predict the α-axis current i at the next moment. α1j ,satisfy:

[0080] .

[0081] These three variables are defined because the α-axis voltages of the three voltage vectors corresponding to the optimal β-axis voltage are these three voltages. Defining these three voltages is mainly for subsequent selection of the optimal α-axis voltage U. αopt Combining the selected optimal β-axis voltage U βopt and the optimal α-axis voltage U αopt This allows for the selection of the optimal vector from the inverter's eight voltage vectors, significantly reducing the number of iterations. Conventional methods use the inverter's eight voltage vectors to simultaneously predict the α-axis and β-axis currents, requiring eight predictions for each axis, totaling 16 iterations. The method of this invention performs a maximum of six iterations (three for the β-axis and three for the α-axis).

[0082] Step S612: Calculate the three α-axis currents i predicted in step S611. α1j Substitute (j=1,2,3) into the following formula to calculate the second cost function value g. 21j ,satisfy:

[0083] ;

[0084] In the formula, g 21j Alternate voltage U for the α-axis α1j The second cost function value corresponding to (j=1,2,3), i refα The α-axis current reference value can be derived from the d-axis current reference value i. refd and q-axis current reference value i refq After coordinate transformation calculation, it is found that:

[0085] .

[0086] Step S613: Compare the three second cost function values ​​g obtained in step S612. 21j To obtain the value of the second cost function g 21j The candidate voltage for the α-axis that corresponds to the minimum value is the optimal α-axis voltage U. αopt ;

[0087] When the optimal β-axis voltage U βopt ≠U β1 Then, execute steps S621, S622, and S623:

[0088] Step S621, as follows Figure 3 As shown, when the optimal β-axis voltage U βopt ≠U β1 When this is the case, it means that the optimal β-axis voltage U βopt equal to U β2 or U β3 The corresponding voltage vectors are u2, u3, u5, and u6, where the α-axis voltages corresponding to voltage vectors u3 and u5 are both... The α-axis voltages corresponding to the two voltage vectors u2 and u6 are both Therefore, two α-axis candidate voltages are defined here as U... α21 U α22 respectively satisfy U α21 = U α22 = And set the α-axis alternative voltage U α2k Substituting (k=1,2) into the α-axis current prediction model shown in the following equation, predict the α-axis current i at the next moment. α2k ,satisfy:

[0089] ;

[0090] These two variables are defined because the α-axis voltages of the four voltage vectors corresponding to the optimal β-axis voltage are these two voltages. Defining these two voltages is primarily for subsequent selection of the optimal α-axis voltage U. αopt Combining the selected optimal β-axis voltage U βopt and the optimal α-axis voltage U αopt This allows the optimal vector to be selected from the eight voltage vectors of the inverter, and significantly reduces the number of iterations.

[0091] Step S622: Calculate the two α-axis currents i predicted in step S621. α2k Substitute these values ​​into the following formula to calculate the second cost function g. 22k The value satisfies:

[0092]

[0093] In the formula, g 22k Alternate voltage U for the α-axis α2k The second cost function value corresponding to (k=1,2), i refα The α-axis current reference value can be derived from the d-axis current reference value i. refd and q-axis current reference value i refq After coordinate transformation calculation, it is found that:

[0094] .

[0095] Step S623: Compare the two second cost function values ​​g obtained in step S622. 22k (k=1,2), obtain the second cost function g 22k The candidate voltage for the α-axis that corresponds to the minimum value is the optimal α-axis voltage U. αopt ;

[0096] It should be noted that, as can be seen from steps 3, 4, and 5, the optimal β-axis voltage U βopt There are 3 beta-axis alternative voltages U β1 U β2 U β3 One of them, when U βopt ≠U β1 When this is the case, it means that the optimal β-axis voltage U βopt equal to U β2 or U β3 , Figure 3 This diagram illustrates the distribution of the eight basic voltage vectors at the inverter output. The positions of the β-axis and α-axis candidate voltages proposed in this invention are marked in the diagram.

[0097] Depend on Figure 3 It can be seen that the optimal β-axis voltage U β2 The corresponding voltage vectors are u5 and u6, and the optimal β-axis voltage U β3 The corresponding voltage vectors are u2 and u3, and the α-axis voltages corresponding to voltage vectors u3 and u5 are both U. α21 = The α-axis voltages corresponding to voltage vectors u2 and u6 are both U. α22 = Therefore, when the optimal β-axis voltage U βopt ≠U β1 At that time, i.e., U βopt =U β2 or U β3 When the corresponding four vectors have the same two α-axis voltage components, steps S621, S622, and S623 only need to be executed to select the optimal α-axis voltage U. αopt . Figure 3 The voltage vectors u0, u1, u2, u3, u4, u5, u6, and u7 represent the eight basic voltage vectors output by the two-level inverter.

[0098] Step 7: Apply the optimal β-axis voltage U obtained in Step 5. βopt And the optimal α-axis voltage U obtained in step 6 αopt As output, it is used to control the asynchronous motor, thereby realizing low-complexity sequential model predictive control of the asynchronous motor in the stationary coordinate system.

[0099] By performing step 6, the optimal α-axis component U among the eight voltage vectors of the inverter can be determined through three or two current predictions and a second cost function evaluation. αopt As can be seen from steps 3-5, the method proposed in this invention requires only 5 or 6 current predictions to obtain the optimal α-axis component U. αopt and the optimal β-axis voltage U βopt Finally, by executing step 7, model predictive control of the asynchronous motor can be achieved.

[0100] Figure 3 This is a schematic diagram showing the positions of the β-axis and α-axis candidate voltages proposed in this invention. When the β-axis candidate vector is U... β1 At that time, three alternative voltages for the α-axis are defined as U α11 U α12 U α13 When the β-axis alternative voltage is U β2 or U β3 At that time, two alternative voltages for the α-axis are defined as U α21 U α22 This definition allows the method proposed in this invention to first perform three β-axis current predictions, a first cost function calculation, and optimal β-axis voltage selection, followed by three or two α-axis current predictions, a second cost function calculation, and optimal α-axis voltage selection. This overcomes the drawback of conventional methods that require 16 current predictions, necessitating simultaneous 8 β-axis current predictions, 8 α-axis current predictions, and 8 cost function calculations using the inverter's 8 voltage vectors to obtain both optimal α-axis and optimal β-axis voltages. This invention reduces the number of iterations by predicting the β-axis and α-axis currents separately.

[0101] like Figure 4 As shown in the control block diagram, the outer speed loop is determined by the set speed reference value n. ref and the measured speed feedback value n r It is constructed using a proportional-integral (PI) controller, and its output is the torque current reference value i. qref i dref and i qrefThe current reference value is jointly provided by the model predictive control method proposed in this invention. The inner current loop executes the model predictive control method proposed in this invention, and the output is the optimal β-axis voltage U. βopt and the optimal α-axis voltage U αopt Finally, the optimal β-axis voltage U is... βopt Optimal α-axis voltage U αopt The corresponding switch states are used to control the three-phase two-level inverter, thereby achieving the goal of asynchronous motor current control. Figure 4 In, n p This represents the number of pole pairs of the motor, and the speed feedback value n. r It is the measured electric angular velocity ω of the motor rotor. r The calculated synchronization angular frequency ω s From the rotor electric angular velocity ω r The angle θ of the rotor magnetic field is calculated from the slip frequency and the synchronous angular frequency ω. s The points are obtained through integration.

[0102] To verify the effectiveness of the present invention, a simulation was conducted, and the motor parameters used in the simulation are shown in Table 1.

[0103] Table 1 Asynchronous Motor Parameters

[0104]

[0105] During the simulation, the motor reference speed was set to 1500 r / min. At 0.4 s, the load torque suddenly increased from 0 Nm to 15 Nm. The excitation current reference value i... dref Set to 2.7637A, the speed loop output is the torque current reference value i. qref DC side voltage U of a three-phase two-level inverter dc The voltage is 400V. The proportional coefficient of the speed outer loop proportional-integral controller is 0.15, and the integral coefficient is 7. The current inner loop employs a conventional 16-iteration model predictive control method and a low-complexity sequential model predictive control method for asynchronous motors in a stationary coordinate system with 6 or 5 iterations proposed in this invention. The simulation results of the conventional method are as follows: Figure 5 As shown, the simulation results of the method proposed in this invention are as follows: Figure 6 As shown. Comparison Figure 5 and Figure 6 As can be seen, both conventional methods and the method proposed in this invention can achieve rapid speed tracking and disturbance rejection control, verifying the effectiveness of the method proposed in this invention. Therefore, the computational complexity of the model predictive control method proposed in this invention is significantly reduced, while still achieving high-performance current control of asynchronous motors, demonstrating the superiority of the method proposed in this invention.

[0106] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A predictive control method for a sequential model of an asynchronous motor in a stationary coordinate system, characterized in that, The steps are as follows: Step 1: Sample the three-phase stator currents abc of the asynchronous motor and transform the three-phase stator currents to the αβ stationary coordinate system to obtain the current on the α axis and the current on the β axis; Step 2: Based on the current on the α-axis, the current on the β-axis, and the mathematical model of the asynchronous motor, calculate the rotor flux linkage on the α-axis and the rotor flux linkage on the β-axis of the asynchronous motor. Step 3: Define three β-axis candidate voltages using the DC bus voltage of the three-phase two-level inverter. Predict the three β-axis currents at the next moment using the three β-axis candidate voltages, the rotor flux linkage on the α-axis, the rotor flux linkage on the β-axis, and the current on the β-axis. Step 4: Calculate the value of the first cost function based on the three β-axis currents and the β-axis current reference value at the next moment; Step 5: Compare the values ​​of the three first cost functions obtained, and take the β-axis candidate voltage that minimizes the value of the first cost function as the optimal β-axis voltage; Step 6: Based on the obtained optimal β-axis voltage, select 3 or 2 α-axis candidate voltages, and use the α-axis candidate voltages, rotor flux linkage on the β-axis, rotor flux linkage on the α-axis, current on the α-axis, and α-axis current reference value to predict the α-axis current at the next moment and calculate the value of the second cost function. Select the α-axis candidate voltage that minimizes the value of the second cost function as the optimal α-axis voltage. Step 7: Use the obtained optimal β-axis voltage and optimal α-axis voltage as outputs to control the asynchronous motor.

2. The method for predictive control of asynchronous motors in a stationary coordinate system according to claim 1, characterized in that, The three β-axis alternative voltages U β1 U β2 U β3 respectively satisfy U β1 =0, U β2 = U β3 = , among which, U dc This indicates the DC bus voltage of a three-phase two-level inverter; When the optimal β-axis voltage U βopt = U β1 At that time, three alternative voltages for the α-axis are selected as U α11 =0, U α12 = U α13 = ; When the optimal β-axis voltage U βopt ≠U β1 When selecting two alternative voltages for the α-axis, U α21 = U α22 = .

3. The method for predictive control of asynchronous motors in a stationary coordinate system according to claim 2, characterized in that, The calculation methods for the current on the α-axis and the current on the β-axis are as follows: ； where i a represents the a-phase stator current of the asynchronous motor, i b represents the b-phase stator current of the asynchronous motor, i c represents the c-phase stator current of the asynchronous motor, i α represents the current of the asynchronous motor on the α-axis, i β represents the current of the asynchronous motor on the β-axis; The calculation methods for the rotor flux linkage on the α-axis and the rotor flux linkage on the β-axis of the asynchronous motor are as follows: ； where ψ rα is the rotor flux of the asynchronous machine on the α-axis, ψ rβ is the rotor flux of the asynchronous machine on the β-axis, L m denotes the mutual inductance of the asynchronous machine, T r is the rotor time constant, ω r is the rotor electrical angular velocity of the asynchronous machine, s denotes the Laplace operator and j denotes a phase lead of 90 degrees.

4. The method for predictive control of an asynchronous motor in a stationary coordinate system according to claim 2 or 3, characterized in that, The method for predicting the three β-axis currents at the next moment is as follows: ； where U βm is the alternative voltage on the β axis, m = 1, 2, 3; T s denotes the control period, ω r is the rotor electrical angular velocity of the asynchronous machine, a, b, c, d are four coefficients, ψ rα is the rotor flux of the asynchronous machine on the α axis, ψ rβ is the rotor flux of the asynchronous machine on the β axis; i β denotes the current of the asynchronous machine on the β axis; When the optimal β-axis voltage U βopt =U β1 The method for predicting the α-axis current at the next moment is as follows: ； Among them, U α1j For the α-axis, the alternative voltages are j=1,2,3; i α This represents the current in the asynchronous motor along the α axis; When the optimal β-axis voltage U βopt ≠U β1 The method for predicting the α-axis current at the next moment is as follows: ； Among them, U α2k The alternative voltages for the α-axis are k=1,2.

5. The sequential model predictive control method for asynchronous motors in a stationary coordinate system according to claim 4, characterized in that, The first cost function ; where i refβ This is the reference value for the β-axis current.

6. The method for predictive control of an asynchronous motor in a stationary coordinate system according to claim 5, characterized in that, When the optimal β-axis voltage U βopt =U β1 When, the second cost function is In the formula, i refα This is the reference value for the α-axis current. When the optimal β-axis voltage U βopt ≠U β1 When, the second cost function is .

7. The method for predictive control of an asynchronous motor in a stationary coordinate system according to claim 6, characterized in that, The coefficient ,coefficient ,coefficient ,coefficient , Let represent the leakage inductance coefficient, and satisfy . L m For the mutual inductance of the asynchronous motor, L s L is the stator inductance of the asynchronous motor. r R is the rotor inductance of the asynchronous motor. s R is the stator resistance of the asynchronous motor. r T is the rotor resistance of the asynchronous motor. r Let T be the rotor time constant of the asynchronous motor, satisfying T r =L r / R r .

8. The method for sequential model predictive control of an asynchronous motor in a stationary coordinate system according to claim 7, characterized in that, The β-axis current reference value i refβ From the d-axis current reference value i refd and q-axis current reference value i refq Calculated through coordinate transformation, and satisfying the following: In the formula, θ represents the orientation angle of the control system; α-axis current reference value i refα From the d-axis current reference value i refd and q-axis current reference value i refq Calculated through coordinate transformation, and satisfying the following: .

9. The method for sequential model predictive control of an asynchronous motor in a stationary coordinate system according to any one of claims 6-8, characterized in that, When the optimal β-axis voltage U βopt =U β1 At that time, three alternative voltages for the α-axis are defined as U α11 U α12 U α13 respectively satisfy U α11 =0, U α12 = U α13 = According to the α-axis candidate voltage U α1j Predict the α-axis current i at the next moment α1j Using the predicted three α-axis currents i α1j Calculate the values ​​g of the three second cost functions. 21j The values ​​of the three second cost functions obtained by comparison, g 21j To obtain the value of the second cost function g 21j The candidate voltage for the α-axis that corresponds to the minimum value is the optimal α-axis voltage U. αopt .

10. The method for sequential model predictive control of an asynchronous motor in a stationary coordinate system according to any one of claims 6-8, characterized in that, When the optimal β-axis voltage U βopt ≠U β1 At that time, two alternative voltages for the α-axis are defined as U α21 U α22 respectively satisfy U α21 = U α22 = And based on the α-axis alternative voltage U α2k (k=1,2) Predict the α-axis current i at the next time step. α2k Based on the predicted two α-axis currents i α2k Calculate the value of the second cost function g 22k ; compared to the values ​​of the two second cost functions g obtained 22k To obtain the value g of the second cost function 22k The candidate voltage for the α-axis that corresponds to the minimum value is the optimal α-axis voltage U. αopt .

Images