Predictive Control Method for AC Induction Motor Based on Gradient Descent

Abstract

The present invention discloses a predictive control method for an AC induction motor based on gradient descent, comprising the following steps: establishing a mathematical model of the AC induction motor, adopting a sliding model reference adaptive (MRAS) observer, and a gradient descent iterative method. The gradient descent iterative method is adopted as an adaptive law, iterating in the opposite direction of the gradient of a quantization error to find the minimum value of the quantization error, performing model predictive torque control, and obtaining the switching state of an upper arm switch tube of each phase in an inverter in a drive system of the AC induction motor. The switching state that minimizes a cost function is selected and output to the inverter, so as to accurately and quickly minimize the error and achieve current tracking performance similar to that of a carrier.

Description

Technical Field

[0001] The present invention relates to the field of motor control, and in particular to an AC induction motor predictive control method based on gradient descent. Background Art

[0002] An AC induction motor is a device that converts alternating current into mechanical energy and is widely used in industries such as industry, home appliances, and transportation.

[0003] AC induction motors can be controlled using sensorless control methods, such as MRAS model control or sliding mode observer control. These methods work by estimating the speed / position through motor models and algorithms, eliminating the need for encoders and offering the advantages of low cost and high reliability. The core of MRAS model control is to dynamically adjust system parameters through the error between a reference model and an adjustable model, so that the actual system output approaches the ideal reference model. However, it is difficult to accurately minimize the error between MRAS models in existing MARS model control.

[0004] In view of this, the applicant conducted in-depth research on this basis, which led to the emergence of this case. Summary of the Invention

[0005] The object of the present invention is to provide an AC induction motor predictive control method based on gradient descent, which can quickly and accurately minimize the error between models.

[0006] To achieve the above object, the solution of the present invention is:

[0007] A predictive control method for an AC induction motor based on gradient descent, wherein the drive system of the AC induction motor includes an inverter, wherein the inverter has three-phase bridge arms, wherein the three-phase bridge arms correspond to phases a, b, and c, respectively, and each phase is provided with two switching tubes, wherein the two switching tubes in each phase bridge arm correspond to an upper bridge arm and a lower bridge arm, respectively; and the method comprises the following steps:

[0008] Step S1, establishing a mathematical model of the AC induction motor, mathematically modeling the stator and rotor flux, stator and rotor current, stator and rotor inductance, and rotor parameters of the AC induction motor to obtain the mathematical model of the AC induction motor;

[0009] Step S2: Using a sliding model reference adaptive MRAS observer, an adjustable model and a reference model in the MRAS observer are obtained according to the mathematical model in step S1, and the error between the adjustable model and the reference model is corrected by a sliding surface in the sliding model;

[0010] Step S3: gradient descent iteration method, using gradient descent iteration method as the adaptive law, iterating in the opposite direction of the gradient of the quantization error to find the minimum value of the quantization error;

[0011] Step S4, model predictive torque control, outputting a speed prediction value through the MRAS observer, calculating the error between the actual speed value output by the AC induction motor and the speed prediction value through a PI regulator to obtain an electromagnetic torque reference value, and calculating a cost function by predicting the tracking deviation between the stator flux and the torque and the reference value;

[0012] Step S5: Obtain the switching state of the upper bridge arm switch tube of each phase in the inverter, select the switching state that minimizes the cost function, and output it to the inverter.

[0013] In step S1, the mathematical model of the AC induction motor is established as follows:

[0014] (1)

[0015] (2)

[0016] (3)

[0017] (4),

[0018] Where, represents the stator voltage vector, 、 represent the stator flux and rotor flux respectively, 、 represent the stator current and rotor current respectively, and They are represented as stator resistance and rotor resistance respectively, Indicates the speed, 、 and They represent stator inductance, rotor inductance and mutual inductance respectively.

[0019] In step S2, the designed MRAS observer model is as follows,

[0020] (5),

[0021] (6),

[0022] (7),

[0023] (8),

[0024] Where, represents the predicted value of rotor voltage flux, represents the predicted value of rotor current flux, represents the output voltage of the inverter, is the predicted speed value, 、 、 、 Respectively and exist 、 The component on the axis, j represents the imaginary number, σ is the motor magnetic leakage coefficient, and s represents the error between the reference model and the adjustable model.

[0025] In a fixed reference system, the voltage model in formula (5) is selected as the reference model, and the current model in formula (6) is selected as the adjustable model;

[0026] In formula (6), As the sliding surface, Represents a continuous function, the formula is (9), where α represents the sliding mode parameter, x represents the function independent variable, represents the coefficients of the constituent functions;

[0027] In formula (5), Indicates stator resistance compensation, the calculation formula is: (10), where K p , K i represent the proportional coefficient and the integral coefficient respectively, Represents the stator resistance error, and the calculation formula of the stator resistance error is: (11), where 、 They represent the components of the stator current on the α and β axes respectively.

[0028] In step S3, the gradient descent iteration method is as follows,

[0029] Step S3-1: Establish the objective function of gradient iteration as follows: J = (12), where s is obtained by formula (7);

[0030] Step S3-2: Obtaining the target function J Speed prediction value Gradient , the calculation formula is as follows,

[0031] (13),

[0032] (14),

[0033] (15),

[0034] Where, Indicates the predicted value of rotor current flux Speed prediction value The gradient, The error s between the reference model and the adjustable model is expressed as the speed prediction value gradient; and They represent the component gradients of the rotor flux in the α and β axes respectively;

[0035] Step S3-3, using The iterative formula to obtain the speed prediction value is: (16), where is the observer gain greater than 0; use formula (16) to iterate until The value of s Convergence, get the speed prediction value .

[0036] In step S4, the model predicts torque control as follows,

[0037] Step S4-1: The speed prediction value obtained in step S3-3 is The error between the actual speed value ω output by the AC induction motor is calculated by the PI regulator to obtain the electromagnetic torque reference value ;

[0038] Step S4-2, the calculation formulas for predicting stator flux, predicted current and predicted torque are as follows:

[0039] (17),

[0040] (18),

[0041] (19),

[0042] Where, represents the stator flux, represents the rotor flux, T represents the electromagnetic torque, represents the stator flux at time k+1, represents the predicted value of the stator flux at time k+1, v s represents the stator voltage, represents the stator voltage at time k, represents the stator current at time k, represents the predicted value of the stator current at time k+1, represents the predicted speed value at time k, kr represents the rotor coupling coefficient, represents the rotor time constant, represents the predicted value of rotor flux at time k;

[0043] represents the sampling period, p represents the number of pole pairs, represents the predicted value of electromagnetic torque, represents the predicted value of the electromagnetic torque at time k+1, represents the imaginary part of a complex number; where , R σ represents the equivalent resistance; , , stator transient time ;

[0044] Step S4-3: Establish the cost function as follows: (20), represents the cost function value in the case of switch state j; where, Indicates the stator flux reference value, which is directly input by the AC induction motor; is the weighting factor that modifies the torque and flux linkage values according to the operating conditions; represents the torque prediction value at time k+2, Represents the predicted value of stator flux at time k+2.

[0045] In step S5, a cost function is selected that minimizes the cost function , obtain the cost function The corresponding switching state j is transmitted to the inverter, and the inverter switches the switching state of the upper bridge arm switch tube of each phase according to the switching state.

[0046] Step S4-4, parameter optimization, k The torque prediction value and flux linkage prediction value in the +2 cycle are compared with the reference value in the cost function for optimization. The optimization formula is: (21), where represents the maximum current limit at time k+2, γ represents a value much greater than zero, i max Indicates the set maximum current value.

[0047] The voltage vector is obtained by Clarke transform , the formula is as follows,

[0048] (twenty two),

[0049] Where, is the DC voltage, Indicates the switch status, 、 and They correspond to the switching states of the upper bridge arm switches of phase a, phase b and phase c in the two-level three-phase inverter, =0, =0 and =0 respectively indicates that the upper bridge arm switches of phase a, phase b and phase c in the inverter are turned on, =1, =1 and =1 respectively indicates that the upper arm switches of phase a, phase b and phase c in the inverter are turned off.

[0050] According to the switching state of the upper bridge arm switch tube of the three phases in the inverter, a three-bit binary number is obtained. , using control input j as input variable, j represents the current state of the switch state.

[0051] After adopting the above method, the present invention has the following beneficial effects:

[0052] 1. The present invention uses the gradient descent iteration method to process the error between the reference model and the adjustable model in the MRAS observer. The appropriate MRAS observer gain is selected to make it descend iteratively in the opposite direction of the gradient, so as to accurately and quickly minimize the error, make the control process more accurate, and obtain the ideal parameter results.

[0053] 2. The present invention adopts the sliding surface reference adaptive MRAS observer of the sliding model to correct it according to the error between the adjustable model and the reference model. Compared with the traditional MRAS observer, the MRAS designed in the present invention has stronger robustness and easier tuning.

[0054] 3. In the model prediction torque control of the present invention, k The torque prediction value and flux linkage prediction value in the +2 cycle are compared with the reference values in the cost function and optimized to compensate for the time lag in the model predictive torque control. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a circuit topology diagram of the AC induction motor and the two-level inverter in the present invention.

[0056] Figure 2 This is a control principle block diagram of the MRAS observer in the present invention.

[0057] Figure 3 This is a control principle block diagram of the model predictive torque control in the present invention. DETAILED DESCRIPTION

[0058] In order to further explain the technical solution of the present invention, the present invention is described in detail below through specific embodiments.

[0059] A gradient descent-based predictive control method for AC induction motors, which is based on AC induction motors or permanent magnet motors to control common drive systems, such as Figure 1-3 As shown, the drive system includes an inverter and a conventional two-level inverter topology. The inverter adopts a conventional two-level three-phase inverter. The inverter includes three-phase bridge arms, which correspond to the following phases a, b and c respectively, and each phase is provided with two switching tubes. The two switching tubes in each phase bridge arm correspond to the upper bridge arm and the lower bridge arm respectively.

[0060] In addition, the above-mentioned drive system also includes a flux estimation module, a torque and flux prediction module and a PI regulator, and each of its input quantities can be obtained through conventional means in this field or the following method, so it will not be described in detail.

[0061] In this embodiment, the predictive control method includes the following steps.

[0062] Step S1, establishing a mathematical model of the AC induction motor: mathematically modeling the stator and rotor fluxes, stator and rotor currents, stator and rotor inductances, and rotor parameters of the AC induction motor to obtain a mathematical model of the AC induction motor.

[0063] To elaborate, the established mathematical model of AC induction motor (i.e. IM model) is as follows:

[0064] (1)

[0065] (2)

[0066] (3)

[0067] (4),

[0068] Where, represents the stator voltage vector, 、 represent the stator flux and rotor flux respectively, 、 represent the stator current and rotor current respectively, and They are represented as stator resistance and rotor resistance respectively, Indicates the speed, 、 and They represent stator inductance, rotor inductance and mutual inductance respectively.

[0069] Step S2, using a sliding model (SM) to refer to an adaptive MRAS observer: According to the mathematical model in step S1, an adjustable model and a reference model in the MRAS observer are obtained, wherein the voltage model is used as the reference model and the current model is used as the adjustable model, and then the error between the adjustable model and the reference model is corrected by the sliding surface in the sliding model (SM).

[0070] To expand on this, Figure 2 As shown, the designed MRAS observer model is as follows:

[0071] (5),

[0072] (6),

[0073] (7),

[0074] (8),

[0075] Where, represents the predicted value of rotor voltage flux, represents the predicted value of rotor current flux, represents the output voltage of the inverter, is the predicted speed value, 、 Respectively exist 、 The component on the axis, 、 Respectively exist 、 The component on the axis, j represents the imaginary number, σ is the motor magnetic leakage coefficient, and s represents the error between the reference model and the adjustable model. That is, the error between the reference model and the adjustable model can be obtained by formula (7).

[0076] Furthermore, in this embodiment, in a fixed reference frame, the voltage model in formula (5) is selected as the reference model in the MRAS observer, and the current model in formula (6) is selected as the adjustable model in the MRAS observer.

[0077] Furthermore, in order to alleviate the problem of chatter, the above formula (6) is replaced by As the sliding surface in the sliding model, Represents a continuous function, and the calculation formula is (9), in formula (9), α represents the sliding mode parameter, x represents the function independent variable, Represents the coefficients of the constituent functions.

[0078] In the above formula (5), Represents stator resistance compensation, and its calculation formula is: (10), where K p , K i represent the proportional coefficient and the integral coefficient respectively; among them, Represents the stator resistance error, which is calculated using the following formula: (11), in formula (11), in 、 Represent the components of the stator current in the α and β axes respectively; it should be noted that the stator resistance error Stator resistance compensation is obtained through feedback from a conventional PI regulator .

[0079] Thus, the above-mentioned MRAS observer is designed as an adaptive mechanism, and the error term of the MRAS observer is used as a sliding surface to improve the accuracy of flux linkage prediction.

[0080] Step S3, using the gradient descent iteration method: In order to minimize the prediction deviation, the gradient descent iteration method is used to replace the traditional PI regulator as the adaptive law, and iterate in the opposite direction of the gradient of the quantization error to find the minimum value of the quantization error. The specific gradient descent iteration method is described as follows.

[0081] like Figure 2 As shown, step S3-1, establish the objective function of gradient iteration, the objective function is as follows: J = (12), where s is the error between the reference model and the adjustable model of the MRAS observer in step S2. The error s can be obtained by formula (7), that is, the error s is incorporated into the objective function J of the gradient iteration.

[0082] Step S-2: Obtaining the target function J Speed prediction value Gradient : Predicted speed As parameters, the rotor current flux prediction values are calculated respectively Speed prediction value Gradient And the error s between the reference model and the adjustable model in the MRAS observer is proportional to the speed prediction value Gradient , and thus calculate the gradient , the specific calculation formula is as follows.

[0083] (13),

[0084] (14),

[0085] (15),

[0086] Where, Indicates the predicted value of rotor current flux Speed prediction value The gradient, The error s between the reference model and the adjustable model is expressed as the speed prediction value The gradient, and They represent the component gradients of the rotor flux on the α and β axes respectively.

[0087] Step S3-3, using The iterative formula to obtain the speed prediction value is: (16), where is the observer gain greater than 0; use formula (16) to iterate until The value of the error s Convergence, get the speed prediction value .

[0088] To illustrate, first determine an initial value, then continuously calculate according to formula (16). After each calculation, the newly calculated value replaces the original value. When the newly calculated value is equal to the original value, the calculation stops. In formula (16), when the gradient is equal to 0, it is the final result of the iteration.

[0089] It should be noted that the observer gain Too large may cause oscillation, too small may cause slow convergence, so The value of needs to be selected accurately, that is, The selection needs to be tested according to the actual situation of the experimental platform.

[0090] Step S4, model predictive torque control (PTC): torque and stator flux are selected as control objects, a speed prediction value is output through the MRAS observer, the error between the actual speed value output by the AC induction motor and the speed prediction value is calculated through the PI regulator to obtain an electromagnetic torque reference value, and a cost function is calculated by predicting the tracking deviation between the stator flux and the torque and the reference value. The specific control is described as follows.

[0091] like Figure 3 As shown, step S4-1, step S3-3 obtains the speed prediction value The error between the actual speed value ω output by the AC induction motor is calculated by the PI regulator to obtain the electromagnetic torque reference value .

[0092] Step S4-2, the calculation formulas for predicting stator flux, predicted current and predicted torque are as follows.

[0093] (17),

[0094] (18),

[0095] (19),

[0096] Where, represents the stator flux, represents the rotor flux, T represents the electromagnetic torque, represents the stator flux at time k+1, represents the predicted value of the stator flux at time k+1, represents the predicted value of the rotor flux at time k, v s represents the stator voltage, represents the stator voltage at time k, represents the stator current at time k, represents the predicted value of the stator current at time k+1, represents the predicted speed value at time k; k r It represents the rotor coupling coefficient, that is, it represents the ratio of mutual inductance to rotor inductance; represents the sampling period, p represents the number of pole pairs, represents the predicted value of electromagnetic torque, represents the predicted value of the electromagnetic torque at time k+1, Represents the imaginary part of a complex number; represents the rotor time constant.

[0097] Furthermore, in formula (18), R σ Represents the equivalent resistance, which is usually related to the resistance of the stator and rotor, and its calculation formula is .

[0098] Furthermore, in formula (18), , , stator transient time .

[0099] Step S4-3: Establish a cost function. The cost function is calculated by using the predicted tracking deviation between the stator flux and the torque, as well as the stator flux reference value and the torque reference value. The cost function is calculated as follows: (20), where represents the cost function value in the case of switch state j; Indicates the stator flux reference value, which is directly input by the AC induction motor; is the weighting factor that modifies the torque and flux linkage values according to the operating conditions; represents the torque prediction value at time k+2, Represents the predicted value of stator flux at time k+2.

[0100] As a preferred method, delay compensation is added after step S4-3. Specifically, the torque prediction value in the k+2 cycle is compared with the torque reference value of the cost function, and the flux prediction value in the k+2 cycle is compared with the flux reference value of the cost function, and the switching state j at this time is output. Then, the voltage vector is calculated based on the output of the inverter at this switching state, that is, the following formula (22); wherein, the optimization formula is (21), where represents the maximum current limit at time k+2, γ represents a value much greater than zero, i max Indicates the set maximum current value.

[0101] Step S5, obtaining the switching state of the upper arm switch tube of each phase in the inverter: selecting the switching state j that minimizes the cost function in step S4 and inputting it into the inverter, and the inverter controls the conduction or shutdown of the upper arm switch tube.

[0102] The calculation of the above voltage vector is as follows. The voltage vector is obtained by Clarke transformation. .

[0103] (twenty two),

[0104] Where, is the DC voltage, Indicates the switch status, 、 and They correspond to the switching states of the upper bridge arm switches of phase a, phase b and phase c in the two-level three-phase inverter, =0, =0 and =0 respectively indicates that the upper bridge arm switches of phase a, phase b and phase c in the inverter are turned on, =1, =1 and =1 respectively indicates that the upper arm switches of phase a, phase b and phase c in the inverter are turned off.

[0105] Furthermore, according to the switching state of the upper bridge arm switch tubes of the three phases in the above inverter, a three-bit binary number is obtained. , using control input j as input variable, j is j in the figure represents the current state of the switch state, that is, the jth switch state; in this embodiment, there are 8 switch states, j = 1, 2, 3, 4, 5, 6, 7 or 8, S abc The value of corresponds to j at this time, for example, j=4, then S abc =[0,1,1]; where Represents the value of the cost function in the case of switch state j, which means choosing the one with the smallest value , the corresponding j is the optimal switching state. Similarly, the voltage vector v is calculated by formula (22), which is also a quantity that changes with j.

[0106] The above description is only a preferred embodiment of this embodiment, and all equivalent changes and modifications made within the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A predictive control method for an AC induction motor based on gradient descent, wherein the drive system of the AC induction motor includes an inverter, wherein the inverter has three-phase bridge arms, wherein the three-phase bridge arms correspond to phases a, b, and c, respectively, and each phase is provided with two switching tubes, wherein the two switching tubes in each phase bridge arm correspond to the upper bridge arm and the lower bridge arm, respectively; characterized in that: The steps include: Step S1, establishing a mathematical model of the AC induction motor, mathematically modeling the stator and rotor flux, stator and rotor current, stator and rotor inductance, and rotor parameters of the AC induction motor to obtain the mathematical model of the AC induction motor; Step S2: Using a sliding model reference adaptive MRAS observer, an adjustable model and a reference model in the MRAS observer are obtained according to the mathematical model in step S1, and the error between the adjustable model and the reference model is corrected by a sliding surface in the sliding model; The designed MRAS observer model is as follows, （5）， （6）， （7）， （8）， Where, represents the predicted value of rotor voltage flux, represents the predicted value of rotor current flux, represents the output voltage of the inverter, is the predicted speed value, 、 、 、 Respectively and exist 、 The component on the axis, j represents the imaginary number, σ is the motor magnetic leakage coefficient, s represents the error between the reference model and the adjustable model, represents the stator voltage vector, represents the stator current, Expressed as the predicted value of stator resistance, Expressed as the rotor resistance, Indicates the speed, 、 and They represent stator inductance, rotor inductance and mutual inductance respectively; Step S3: Gradient descent iteration method, using the gradient descent iteration method as the adaptive law, iterating in the opposite direction of the gradient of the quantization error to find the minimum value of the quantization error; The gradient descent iterative method is as follows, Step S3-1: Establish the objective function of gradient iteration as follows: J = s 2 (12), where s is obtained by formula (7); Step S3-2: Obtaining the target function J Speed prediction value Gradient , the calculation formula is as follows, （13）， （14）， （15）， Where, Indicates the predicted value of rotor current flux Speed prediction value The gradient, The error s between the reference model and the adjustable model is expressed as the speed prediction value gradient; and They represent the component gradients of the rotor flux in the α and β axes respectively; Step S3-3, using The iterative formula to obtain the speed prediction value is: (16), where is the observer gain greater than 0; use formula (16) to iterate until The value of s Convergence, get the speed prediction value ; Step S4, model predictive torque control, outputting a speed prediction value through the MRAS observer, calculating the error between the actual speed value output by the AC induction motor and the speed prediction value through a PI regulator to obtain an electromagnetic torque reference value, and calculating a cost function by predicting the tracking deviation between the stator flux and the torque and the reference value; Step S5: Obtain the switching state of the upper bridge arm switch tube of each phase in the inverter, select the switching state that minimizes the cost function, and output it to the inverter.

2. The AC induction motor predictive control method based on gradient descent according to claim 1, characterized in that: In step S1, the mathematical model of the AC induction motor is established as follows: （1） （2） （3） （4）， Where, 、 represent the stator flux and rotor flux respectively, represents the rotor current, Expressed as stator resistance, Indicates the rotation speed.

3. The AC induction motor predictive control method based on gradient descent according to claim 2, characterized in that: In a fixed reference system, the voltage model in formula (5) is selected as the reference model, and the current model in formula (6) is selected as the adjustable model; In formula (6), As the sliding surface, Represents a continuous function, the formula is (9), where α represents the sliding mode parameter, x represents the function independent variable, represents the coefficients of the constituent functions; In formula (5), Indicates stator resistance compensation, the calculation formula is: (10), where K p , K i represent the proportional coefficient and the integral coefficient respectively, Represents the stator resistance error, and the calculation formula of the stator resistance error is: (11), where 、 They represent the components of the stator current on the α and β axes respectively.

4. The AC induction motor predictive control method based on gradient descent according to claim 1, characterized in that: In step S4, the model predicts torque control as follows, Step S4-1: The speed prediction value obtained in step S3-3 is The error between the actual speed value ω output by the AC induction motor is calculated by the PI regulator to obtain the electromagnetic torque reference value ; Step S4-2, the calculation formulas for predicting stator flux, predicted current and predicted torque are as follows: （17）， （18）， （19）， Where, represents the stator flux, represents the rotor flux, T represents the electromagnetic torque, represents the stator flux at time k+1, represents the predicted value of the stator flux at time k+1, v s represents the stator voltage, represents the stator voltage at time k, represents the stator current at time k, represents the predicted value of the stator current at time k+1, represents the predicted speed value at time k, k r represents the rotor coupling coefficient, represents the rotor time constant, represents the predicted value of rotor flux at time k; represents the sampling period, p represents the number of pole pairs, represents the predicted value of electromagnetic torque, represents the predicted value of the electromagnetic torque at time k+1, represents the imaginary part of a complex number; where , R σ represents the equivalent resistance; , , stator transient time ; Step S4-3: Establish the cost function as follows: (20), represents the cost function value in the case of switch state j; where, Indicates the stator flux reference value, which is directly input by the AC induction motor; is the weighting factor that modifies the torque and flux linkage values according to the operating conditions; represents the torque prediction value at time k+2, Represents the predicted value of stator flux at time k+2.

5. The AC induction motor predictive control method based on gradient descent according to claim 4, characterized in that: In step S5, a cost function is selected that minimizes the cost function , obtain the cost function The corresponding switching state j is transmitted to the inverter, and the inverter switches the switching state of the upper bridge arm switch tube of each phase according to the switching state.

6. The AC induction motor predictive control method based on gradient descent according to claim 4, characterized in that: Step S4-4, parameter optimization, k The torque prediction value and flux linkage prediction value in the +2 cycle are compared with the reference value in the cost function for optimization. The optimization formula is: (21), where represents the maximum current limit at time k+2, γ represents a value much greater than zero, i max Indicates the set maximum current value.

7. The AC induction motor predictive control method based on gradient descent according to claim 5, characterized in that: The voltage vector is obtained by Clarke transform , the formula is as follows, （22）， Where, is the DC voltage, Indicates the switch status, 、 and They correspond to the switching states of the upper bridge arm switches of phase a, phase b and phase c in the two-level three-phase inverter, =0, =0 and =0 respectively indicates that the upper bridge arm switches of phase a, phase b and phase c in the inverter are turned on, =1, =1 and =1 respectively indicates that the upper arm switches of phase a, phase b and phase c in the inverter are turned off.

8. The AC induction motor predictive control method based on gradient descent according to claim 7, characterized in that: According to the switching state of the upper bridge arm switch tube of the three phases in the inverter, a three-bit binary number is obtained. , using control input j as input variable, j represents the current state of the switch state.

Images