Weight-factor-free predictive torque control method for permanent magnet synchronous motor

Abstract

The invention relates to the technical field of motor control, in particular to a weight-factor-free predictive torque control method for a permanent magnet synchronous motor, and the method comprises the following steps: building an OEW-PMSM mathematical model, carrying out the discretization of the OEW-PMSM mathematical model, respectively building cost functions for zero-sequence current, electromagnetic torque and flux linkage, determining the number of voltage vectors in a finite set of each layer, and carrying out the calculation of the weight factor-free predictive torque of the permanent magnet synchronous motor. An optimal voltage vector is screened out through a zero-sequence current cost function, a magnetic torque cost function and a flux linkage cost function, and the finally screened optimal voltage vector acts on an inverter and is input into the permanent magnet synchronous motor to control the torque and flux linkage of the permanent magnet synchronous motor. By introducing a cascade type predictive torque control method and utilizing the idea of hierarchical optimization, introduction of a weight factor is directly avoided, and by gradually increasing the number of voltage vectors of a zero-sequence current cost function until the OEW-PMSM system is out of control and optimizing the number of voltage vectors of an electromagnetic torque cost function, the zero-sequence current can be further suppressed, and the electromagnetic torque cost can be effectively improved. And the burden can be reduced.

Description

Technical Field

[0001] The invention relates to the technical field of motor control, and in particular to a method for predicting torque control of a permanent magnet synchronous motor without weighting factors. Background Art

[0002] The open-winding permanent magnet synchronous motor system has higher power density, greater low-speed output torque, higher high-speed output power, better weak magnetic characteristics and smaller phase current, and is favored by researchers. It has been rapidly promoted and applied in high-power occasions.

[0003] High-performance AC speed control methods generally include field-oriented control (FOC), direct torque control (DTC) and model predictive control (MPC). FOC can achieve good torque and flux control of the motor, but the internal current loop requires complex adjustment; DTC has a simple control structure and can achieve fast dynamic response, but the torque pulsation is large. MPC has the advantages of fast dynamic response and good control effect, but MPC requires simultaneous adjustment of the stator flux weight factor and the zero-sequence current weight factor, which is cumbersome and has a heavy calculation burden, which is not conducive to practical application.

[0004] In view of this, the applicant conducted an in-depth study on the above-mentioned issues, which led to the present case. Summary of the invention

[0005] The object of the present invention is to provide a permanent magnet synchronous motor weight-free predictive torque control method that helps to reduce the computational burden.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A method for predicting torque control of a permanent magnet synchronous motor without weighting factors comprises the following steps:

[0008] S1: Establish the mathematical model of OEW-PMSM;

[0009] S2: Discretize the OEW-PMSM mathematical model to obtain the stator flux ψ s , stator current i s and stator voltage u s The measured value of

[0010] S3: Predict the zero-sequence current i at time k+1 based on the measured value in step 2 s0 (k+1), electromagnetic torque T e (k+1) and stator flux ψ s (k+1), where k is the current time;

[0011] S4: for the zero-sequence current i s0 (k+1) Construct zero-sequence current cost function JΔis0 , for the electromagnetic torque T e (k+1) Construct electromagnetic torque cost function And construct the flux cost function J for the stator flux Δψ ;

[0012] S5: List the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ All hierarchical orders of the three cost functions, simulation analysis for each order, and selection of zero-sequence current cost function -Magnetic torque cost function - flux linkage cost function as the zero-sequence current cost function, the magnetic torque cost function And the magnetic link cost function J Δψ Hierarchical optimization order of the three cost functions;

[0013] S6: Determine the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each finite set is determined as follows:

[0014] S6-1: First determine the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each finite set;

[0015] S6-2: Gradually expand the zero-sequence current cost function The number of voltage vectors allocated by the optimization layer is increased until the entire OEW-PMSM system is out of control;

[0016] S6-3: Select the zero-sequence current cost function of the OEW-PMSM system before it loses control The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each finite set is used as the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each layer's finite set;

[0017] S7: The obtained finite concentrated voltage vector is passed through the delay compensation link in the OEW-PMSM mathematical model to obtain the zero-sequence current i at time k+2 s0 Electromagnetic torque T at time (k+2) and k+2e The electromagnetic torque ψ at time (k+2) and k+2 s (k+2), the electromagnetic torque i at time k+2 is predicted s0 Substitute (k+2) into the zero-sequence current cost function J Δis0 The electromagnetic torque T at time k+2 is predicted e Substitute (k+2) into the electromagnetic torque cost function In the prediction, the stator flux ψ at time k+2 is s Substitute (k+2) into the stator flux cost function J Δψ In the process, we first use the zero-sequence current cost function Filter out The smallest m voltage vectors are selected from the m voltage vectors. The smallest n voltage vectors are then selected from the n voltage vectors to make J Δψ The minimum voltage vector, where m and n are both positive integers, is selected so that J Δψ The smallest voltage vector is taken as the optimal voltage vector finally screened out;

[0018] S8: Applying the optimal voltage vector finally selected to the inverter and inputting it into the open-winding permanent magnet synchronous motor to control the torque and flux linkage of the open-winding permanent magnet synchronous motor.

[0019] Furthermore, in step S2, the stator flux ψ s , the stator current i s and the stator voltage u s Specific ways to obtain:

[0020] Using constant power conversion, the OEW-PMSM mathematical model is transformed into a rotating coordinate system through coordinate transformation. In the rotating coordinate system, the stator flux ψ s The expression is as follows:

[0021]

[0022] In the rotating coordinate system, the electromagnetic torque T of the motor e The expression is as follows:

[0023]

[0024] In the rotating coordinate system, the stator voltage balance equation is expressed as follows:

[0025]

[0026] Among them, ψ sd ,ψ sq ,ψ s0are the d-axis component, q-axis component and 0-axis component of the stator flux; u sd ,u sq ,u s0 are the d-axis component, q-axis component and 0-axis component of the stator voltage; i sd ,i sq ,i s0 are the d-axis component, q-axis component and 0-axis component of the stator current, p is the differential operator, ω r is the motor speed, L d is the d-axis inductance, L q is the q-axis inductance, L sσ1 is the motor leakage inductance, ψ f is the fundamental permanent magnet flux, ψ 3f is the third harmonic component of permanent magnet flux, θ r is the rotor position angle, R s is the stator resistance.

[0027] Furthermore, in step S3, the zero-sequence current i at time k+1 is s0 (k+1), electromagnetic torque T e (k+1) and stator flux ψ s The calculation formula for (k+1) is as follows:

[0028] The mechanical time constant is much larger than the electrical time constant. The motor speed ω is r and the rotor position angle θ r When regarded as unchanged, the first-order Euler discretization expression is as follows:

[0029]

[0030] Calculate the stator flux ψ at time k+1 s The (k+1) amplitude is:

[0031]

[0032] Calculate the d-axis component i of the stator point current at time k+1 sd (k+1), q-axis component i of stator flux sq (k+1) and the d-axis component i of the stator flux s0 (k+1):

[0033]

[0034] The electromagnetic torque T at time k+1 is calculated by formula (6): e The (k+1) amplitude is:

[0035]

[0036] Among them, ψ sd (k+1) is the d-axis component of the stator flux at time k+1, ψ sq (k+1) is the q-axis component of the stator flux at time k+1, ψ s0 (k+1) is the 0-axis component of the stator flux at time k+1, u sd (k) is the d-axis component of the stator voltage at time k+1, T s To control the cycle.

[0037] Furthermore, the d-axis component u of the stator voltage sd (k), the q-axis component u of the stator voltage sq (k) and the zero-axis component u of the stator voltage s0 The calculation formula of (k) is as follows:

[0038]

[0039] in, is the stator voltage of the current control cycle.

[0040] Furthermore, the stator voltage of the current control cycle The calculation formula is as follows:

[0041]

[0042] Among them, T A1 is the A-phase upper tube action time of inverter 1, T A2 is the A-phase upper tube action time of inverter 2, T B1 is the B-phase upper tube action time of inverter 1, T B2 is the B-phase upper tube action time of inverter 2, T C1 is the C phase upper tube action time of inverter 1, T C2 is the C phase upper tube action time of inverter 2.

[0043] Furthermore, in step S4, the electromagnetic torque evaluation function J ΔTe The expression is as follows:

[0044]

[0045] Magnetic linkage evaluation function J Δψ The expression is as follows:

[0046] J Δψ =||ψ s | * -|ψ s (k+2)|| 2 (11);

[0047] in, is the electromagnetic torque reference value, |ψ s | * is the stator flux reference value.

[0048] By adopting the above technical solution, the present invention has the following beneficial effects:

[0049] By introducing the cascade predictive torque control method and utilizing its hierarchical optimization idea, the introduction of weight factors is directly avoided. By gradually expanding the number of voltage vectors of the zero-sequence current cost function until the OEW-PMSM system is out of control, the number of voltage vectors of the electromagnetic torque cost function is optimized, which can further suppress the zero-sequence current and reduce the burden. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a block diagram of the predicted torque control of the OEW-PMSM system provided by the present invention. DETAILED DESCRIPTION

[0051] The present invention will be further described below in conjunction with specific embodiments:

[0052] This embodiment provides a method for predicting torque control of a permanent magnet synchronous motor without weighting factors, such as Figure 1 As shown, the following steps are included:

[0053] S1: Establish the OEW-PMSM mathematical model.

[0054] S2: Discretize the OEW-PMSM mathematical model to obtain the stator flux ψ s , stator current i s and stator voltage u s The measured value of

[0055] S3: Predict the zero-sequence current i at time k+1 based on the measured value in step 2 s0 (k+1) 、 Electromagnetic torque T e (k+1) and stator flux ψ s (k+1) ， Among them, k is the current moment.

[0056] S4: for the zero-sequence current i s0 (k+1) Construct zero-sequence current cost function The electromagnetic torque T e (k+1) Construct electromagnetic torque cost function And construct the flux cost function J for the stator flux Δψ .

[0057] S5: List the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ All hierarchical orders of the three cost functions, simulation analysis for each order, and selection of zero-sequence current cost function -Magnetic torque cost function - Flux cost function as zero-sequence current cost function, magnetic torque cost function J ΔTe and magnetic link cost function J Δψ Hierarchical optimization order of the three cost functions.

[0058] S6: Determine the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each finite set is determined as follows:

[0059] S6-1: First determine the zero-sequence current cost function Electromagnetic torque cost function and magnetic link cost function J Δψ The number of voltage vectors in each finite set. Taking the dual inverter as an example, there are 64 voltage vectors in the dual inverter. The zero-sequence current cost function The number of voltage vectors in the initial finite set is 64, and the electromagnetic torque cost function The number of voltage vectors in the initial finite set is 44, and the flux cost function J Δψ The number of voltage vectors in the initial finite set is 14.

[0060] S6-2: Gradually expand the number of voltage vectors allocated to the optimization layer of the zero-sequence current cost function until the entire OEW-PMSM system is out of control.

[0061] S6-3: Final determination of the zero-sequence current cost function of the OEW-PMSM system without losing control The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each finite set, in this embodiment, the zero-sequence current cost function The number of finite set voltage vectors is 64, and the electromagnetic torque cost function The number of finite set voltage vectors is 28, and the flux cost function J Δψ The number of finite set voltage vectors is 14.

[0062] S7: The obtained finite concentrated voltage vector is passed through the delay compensation link in the OEW-PMSM mathematical model to obtain the zero-sequence current i at time k+2 s0 Electromagnetic torque T at time (k+2) and k+2e The electromagnetic torque ψ at time (k+2) and k+2 s (k+2), the electromagnetic torque i at time k+2 is predicted s0 Substitute (k+2) into the zero-sequence current cost function The electromagnetic torque T at time k+2 is predicted e Substitute (k+2) into the electromagnetic torque cost function In the prediction, the stator flux ψ at time k+2 is s Substitute (k+2) into the stator flux cost function J Δψ In the process, we first use the zero-sequence current cost function Filter out The smallest m voltage vectors are selected from the m voltage vectors. The smallest n voltage vectors are then selected from the n voltage vectors to make J Δψ The minimum voltage vector, where m and n are both positive integers, is selected so that J Δψ The minimum voltage vector is taken as the optimal voltage vector finally screened out. It should be noted that the OEW-PMSM mathematical model has delay compensation, and the delay compensation is conventional delay compensation, which will not be described in detail here.

[0063] S8: The optimal voltage vector finally screened out sends out a pulse to act on the inverter and input it into the open-winding permanent magnet synchronous motor to control the torque and flux of the open-winding permanent magnet synchronous motor.

[0064] In step S2, the stator flux ψ s , stator current i s and stator voltage u s Specific ways to obtain:

[0065] Using constant power conversion, the OEW-PMSM mathematical model is transformed into a rotating coordinate system through coordinate transformation. In the rotating coordinate system (dq coordinate system), the stator flux ψ s The expression is as follows:

[0066]

[0067] In the rotating coordinate system (dq coordinate system), the electromagnetic torque T of the motor is e The expression is as follows:

[0068]

[0069] In the rotating coordinate system (dq coordinate system), the expression of the stator voltage balance equation is as follows:

[0070]

[0071] Among them, ψ sd ,ψ sq ,ψ s0 are the d-axis component, q-axis component and 0-axis component of the stator flux; u sd ,u sq ,u s0 are the d-axis component, q-axis component and 0-axis component of the stator voltage; i sd ,i sq ,i s0 are the d-axis component, q-axis component and 0-axis component of the stator current, p is the differential operator, ω r is the motor speed, L d is the d-axis inductance, L q is the q-axis inductance, L sσ1 is the motor leakage inductance, ψ f is the fundamental permanent magnet flux, ψ 3f is the third harmonic component of permanent magnet flux, θ r is the rotor position angle, R s is the stator resistance.

[0072] In step S3, the zero-sequence current i at time k+1 is s0 (k+1), electromagnetic torque T e (k+1) and stator flux ψ s The calculation formula for (k+1) is as follows:

[0073] The mechanical time constant is much larger than the electrical time constant. The motor speed ω is r and the rotor position angle θ r When regarded as unchanged, the first-order Euler discretization expression is as follows:

[0074]

[0075] Calculate the stator flux ψ at time k+1 s The (k+1) amplitude is:

[0076]

[0077] Calculate the d-axis component i of the stator point current at time k+1 sd (k+1), q-axis component i of stator flux sq (k+1) and the d-axis component i of the stator flux s0 (k+1):

[0078]

[0079] The electromagnetic torque T at time k+1 is calculated by formula (6): e The (k+1) amplitude is:

[0080]

[0081] Among them, ψ sd (k+1) is the d-axis component of the stator flux at time k+1, ψ sq (k+1) is the q-axis component of the stator flux at time k+1, ψ s0 (k+1) is the 0-axis component of the stator flux at time k+1, u sd (k) is the d-axis component of the stator voltage at time k+1, T s To control the cycle.

[0082] The d-axis component of the stator voltage u sd (k), q-axis component u of stator voltage sq (k) and the 0-axis component u of the stator voltage s0 The calculation formula of (k) is as follows:

[0083]

[0084] in, is the stator voltage of the current control cycle.

[0085] Stator voltage of the current control cycle The calculation formula is as follows:

[0086]

[0087] Among them, T A1 is the A-phase upper tube action time of inverter 1, T A2 is the A-phase upper tube action time of inverter 2, T B1 is the B-phase upper tube action time of inverter 1, T B2 is the B-phase upper tube action time of inverter 2, T C1 is the C phase upper tube action time of inverter 1, T C2 is the C phase upper tube action time of inverter 2.

[0088] Electromagnetic torque evaluation function in step S4 The expression is as follows:

[0089]

[0090] Magnetic linkage evaluation function J Δψ The expression is as follows:

[0091]

[0092] Zero-sequence current cost function The expression is as follows:

[0093]

[0094] in, is the electromagnetic torque reference value, ψ s | * is the stator flux reference value, is the zero-sequence current reference value.

[0095] In summary, by introducing the cascade predictive torque control method and utilizing its hierarchical optimization idea, the introduction of weight factors is directly avoided. By gradually expanding the number of voltage vectors of the zero-sequence current cost function until the OEW-PMSM system is out of control, the number of voltage vectors of the electromagnetic torque cost function is optimized. This can not only further suppress the zero-sequence current, but also reduce the burden. At the same time, the three-phase current ripple and zero-sequence current amplitude are reduced, the response speed of the electromagnetic torque becomes faster, and the control performance of the entire system is better.

[0096] The present invention is described in detail above with reference to the accompanying drawings, but the embodiments of the present invention are not limited to the above embodiments. Those skilled in the art can make various modifications to the present invention based on the prior art, which all belong to the protection scope of the present invention.

Claims

1. A method for predicting torque control of a permanent magnet synchronous motor without weighting factors, characterized in that: The steps include: S1: Establish the mathematical model of OEW-PMSM; S2: Discretize the OEW-PMSM mathematical model to obtain the stator flux ψ s , stator current i s and stator voltage u s The measured value of S3: Predict the zero-sequence current i at time k+1 based on the measured value in step 2 s0 (k+1), electromagnetic torque T e (k+1) and stator flux ψ s (k+1), where k is the current time; S4: for the zero-sequence current i s0 (k+1) Construct zero-sequence current cost function The electromagnetic torque T e (k+1) Construct electromagnetic torque cost function And construct the flux cost function J for the stator flux Δψ ; S5: List the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ All hierarchical orders of the three cost functions, simulation analysis for each order, and selection of zero-sequence current cost function -Magnetic torque cost function - flux linkage cost function as the zero-sequence current cost function, the magnetic torque cost function And the magnetic link cost function J Δψ Hierarchical optimization order of the three cost functions; S6: Determine the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each finite set is determined as follows: S6-1: First determine the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each finite set; S6-2: Gradually expand the zero-sequence current cost function The number of voltage vectors allocated by the optimization layer is increased until the entire OEW-PMSM system is out of control; S6-3: Select the zero-sequence current cost function of the OEW-PMSM system before it loses control The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each finite set is used as the zero-sequence current cost function The electromagnetic torque cost function And the magnetic link cost function J Δψ The number of voltage vectors in each layer's finite set; S7: The obtained finite concentrated voltage vector is passed through the delay compensation link in the OEW-PMSM mathematical model to obtain the zero-sequence current i at time k+2 s0 Electromagnetic torque T at time (k+2) and k+2 e The electromagnetic torque ψ at time (k+2) and k+2 s (k+2), the electromagnetic torque i at time k+2 is predicted s0 Substitute (k+2) into the zero-sequence current cost function The electromagnetic torque T at time k+2 is predicted e Substitute (k+2) into the electromagnetic torque cost function In the prediction, the stator flux ψ at time k+2 is s Substitute (k+2) into the stator flux cost function J Δψ In the process, we first use the zero-sequence current cost function Filter out The smallest m voltage vectors are selected from the m voltage vectors. The smallest n voltage vectors are then selected from the n voltage vectors to make J Δψ The minimum voltage vector, where m and n are both positive integers, is selected so that J Δψ The smallest voltage vector is taken as the optimal voltage vector finally screened out; S8: Applying the optimal voltage vector finally selected to the inverter and inputting it into the open-winding permanent magnet synchronous motor to control the torque and flux linkage of the open-winding permanent magnet synchronous motor.

2. A method for predicting torque control of a permanent magnet synchronous motor without weighting factors according to claim 1, characterized in that: In step S2, the stator flux ψ s , the stator current i s and the stator voltage u s Specific ways to obtain: Using constant power conversion, the OEW-PMSM mathematical model is transformed into a rotating coordinate system through coordinate transformation. In the rotating coordinate system, the stator flux ψ s The expression is as follows: In the rotating coordinate system, the electromagnetic torque T of the motor e The expression is as follows: In the rotating coordinate system, the stator voltage balance equation is expressed as follows: Among them, ψ sd ,ψ sq ,ψ s0 are the d-axis component, q-axis component and 0-axis component of the stator flux; u sd ,u sq ,u s0 are the d-axis component, q-axis component and 0-axis component of the stator voltage; i sd ,i sq ,i s0 are the d-axis component, q-axis component and 0-axis component of the stator current, p is the differential operator, ω r is the motor speed, L d is the d-axis inductance, L q is the q-axis inductance, L sσ1 is the motor leakage inductance, ψ f is the fundamental permanent magnet flux, ψ 3f is the third harmonic component of permanent magnet flux, θ r is the rotor position angle, R s is the stator resistance.

3. A method for predicting torque control of a permanent magnet synchronous motor without weighting factor according to claim 2, characterized in that: In step S3, the zero-sequence current i at time k+1 is s0 (k+1), electromagnetic torque T e (k+1) and stator flux ψ s The calculation formula for (k+1) is as follows: In two adjacent control cycles, the motor speed ω r and the rotor position angle θ r When regarded as unchanged, the first-order Euler discretization expression is as follows: Calculate the stator flux ψ at time k+1 s The (k+1) amplitude is: Calculate the d-axis component i of the stator point current at time k+1 sd (k+1), q-axis component i of stator flux sq (k+1) and the d-axis component i of the stator flux s0 (k+1): The electromagnetic torque T at time k+1 is calculated by formula (6): e The (k+1) amplitude is: Among them, ψ sd (k+1) is the d-axis component of the stator flux at time k+1, ψ sq (k+1) is the q-axis component of the stator flux at time k+1, ψ s0 (k+1) is the 0-axis component of the stator flux at time k+1, u sd (k) is the d-axis component of the stator voltage at time k+1, T s To control the cycle.

4. A method for predicting torque control of a permanent magnet synchronous motor without weighting factors according to claim 3, characterized in that: The d-axis component u of the stator voltage sd (k), the q-axis component u of the stator voltage sq (k) and the zero-axis component u of the stator voltage s0 The calculation formula of (k) is as follows: in, is the stator voltage of the current control cycle.

5. A method for predicting torque control of a permanent magnet synchronous motor without weighting factors according to claim 4, characterized in that: The stator voltage of the current control cycle The calculation formula is as follows: Among them, T A1 is the A-phase upper tube action time of inverter 1, T A2 is the A-phase upper tube action time of inverter 2, T B1 is the B-phase upper tube action time of inverter 1, T B2 is the B-phase upper tube action time of inverter 2, T C1 is the C phase upper tube action time of inverter 1, T C2 is the C phase upper tube action time of inverter 2.

6. A method for predicting torque control of a permanent magnet synchronous motor without weighting factors according to claim 1, characterized in that: Electromagnetic torque evaluation function in step S4 The expression is as follows: Magnetic Linkage Evaluation Function J Δψ The expression is as follows: J Δψ =||ψ s | * -|ψ s (k+2)|| 2 (11); Zero-sequence current cost function The expression is as follows: Among them, T e * is the electromagnetic torque reference value, ψ s | * is the stator flux reference value, is the zero-sequence current reference value.

Images