Super-local model predictive control method and system and storage medium

Abstract

The invention relates to the technical field of motor control, in particular to a hyper-local model predictive control method and system and a storage medium, and the method comprises the following steps: sampling the three-phase current of a motor, constructing a hyper-local model based on Poincare mapping causality, discretizing the hyper-local model, constructing a gain recursive vector and a data recursive vector, and carrying out the prediction control of the hyper-local model. And updating the gain recursion vector and the data recursion vector, constructing a control equation, obtaining a modulation signal, converting the modulation signal into a pulse control signal through a modulation module, and controlling a motor driver. According to the method, the hyper-local model based on Poincare mapping causal is introduced, and the coupled current component of Poincare mapping and the electrical angular velocity additional factor are brought into the causal relationship, so that the hyper-local model can effectively process multiple input signals, time-varying physical parameters do not need to be identified in the control process, and the adaptability is good; meanwhile, the motion state of the motor system can be accurately represented, and the description precision is good.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control, and in particular to a super-local model predictive control method, system and storage medium. Background Art

[0002] Model Predictive Control (MPC) is a type of computer control algorithm that emerged in the field of industrial engineering control in the late 1970s and has been widely used in process control industries such as the chemical industry. As a control strategy that has emerged in recent years, MPC, compared to vector control (VOC), does not require a current inner loop or parameter tuning. It directly generates inverter drive signals without pulse modulation, making it easier to handle system constraints or add other control objectives. It has the advantages of simple structure, fast dynamic response, and easy scalability. Compared with direct torque control (DTC), MPC optimizes the selection of the optimal voltage vector by predicting the motor state. This makes vector selection more accurate and effective, and it can more easily account for various nonlinear constraints, including switching frequency reduction. It has the advantages of good steady-state performance and flexible control.

[0003] However, traditional model predictive control has a strong dependence on system model parameters. When the motor is subjected to external disturbances and motor parameters change, the predicted optimal voltage vector will deviate, thereby affecting the overall control performance of the motor. It has poor adaptability and cannot accurately describe the motion state of the motor system, resulting in low description accuracy.

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

[0005] The object of the present invention is to provide a super-local model predictive control method, system and storage medium with good adaptability and description accuracy.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A hyperlocal model predictive control method comprises the following steps: S1: Sample the three-phase current of the motor and transform the sampled three-phase current into the rotating coordinate system through Park to obtain the d-axis component of the stator current and the q-axis component of the stator current , based on the d-axis component of the stator current and the q-axis component of the stator current Constructing a hyperlocal model based on Poincare map causality, wherein the hyperlocal model is a multi-input single-output model; S2: Discretize the hyperlocal model in step S1 and swap the model causality, reorganize the coefficients in the hyperlocal model, the coefficients including the next cycle state gain, the current cycle state gain, the coupling state gain, the electrical angular velocity gain, and the integrated variable, to obtain a hyperlocal model based on the Poincare map causality discretized at time k+1, where k is the current time; S3: constructing the coefficients of the hyperlocal model based on the Poincare map causality discretized at time k+1 into a gain recursive vector, and constructing the state variables of the hyperlocal model based on the Poincare map causality discretized at time k+1 into a data recursive vector; S4: updating the gain recursive vector and the data recursive vector, so that the gain recursive vector at time k is updated to the gain recursive vector at time k+1, and the data recursive vector at time k is updated to the data recursive vector at time k+1; S5: Construct a control equation, substitute the gain recursive vector at time k+1 and the data recursive vector at time k+1 into the control equation to obtain a modulated signal; S6: The modulation signal is converted into a pulse control signal through a modulation module and controls the motor driver.

[0007] Furthermore, in step S1, the hyperlocal model incorporates the coupled current component of the Poincare map and the additional factor of the electrical angular velocity into the causal relationship. The hyperlocal model is shown in formula (1): (1); Where, is the d-axis component of the stator current in the rotating coordinate system, is the q-axis component of the stator current in the rotating coordinate system, is the d-axis component of the stator voltage in the rotating coordinate system, is the q-axis component of the stator voltage in the rotating coordinate system, and are the input gains before d-axis and q-axis reorganization, and are the state gains before d-axis and q-axis recombination, and are the coupling state gains before d-axis and q-axis recombination, and are the electrical angular velocity gains of the d-axis and q-axis before reorganization, and are the integrated variables before reorganization on the d-axis and q-axis, is the electrical angular velocity.

[0008] Furthermore, in step S2, the discretized hyperlocal model based on the Poincare map causality at time k+1 is as shown in formula (2): (2); Where, is the d-axis component of the stator voltage at time k+1, is the q-axis component of the stator voltage at time k+1, is the d-axis component of the stator current at time k+1, is the q-axis component of the stator current at time k+1, is the d-axis component of the stator current at time k, is the q-axis component of the stator current at time k, is the coupled current component of the d-axis component of the stator current at time k+1, is the coupled current component of the q-axis component of the stator current at time k+1, is the electrical angular velocity at time k+1, and are the periodic state gains on the d-axis and q-axis at time k, and are the state gains of the d-axis and q-axis in this cycle at time k, and are the d-axis and q-axis coupling state gains at time k, and are the d-axis and q-axis electrical angular velocity gains at time k, and are the d-axis and q-axis integrated variables at time k respectively; The reorganized next cycle state gain, current cycle state gain, coupling state gain, electrical angular velocity gain and integrated variable are derived through formula (1) and formula (2), as shown in formula (3): (3); in, For the control cycle.

[0009] Furthermore, the coefficients in formula (2) are constructed into the gain recursive vector, as shown in formula (4): (4); The state variables of formula (2) are constructed into the data recursive vector, and the state variables include stator voltage, stator current, coupling circuit components and electrical angular velocity, as shown in formula (5): (5); in, and are the d-axis and q-axis gain recursive vectors at time k, and is the recursive vector of d-axis and q-axis data at time k.

[0010] Furthermore, in step S4, the gain recursive vector and the data recursive vector are updated as follows: S4-1: Split the gain recursive vector of formula (4) and the data recursive vector of formula (5) into three groups, namely case 1 group, case 2 group and case 3 group: (6); (7); (8); Where, is the d-axis gain recursive vector at moment k in case1 group, is the q-axis gain recursive vector at moment k in case1 group, is the recursive vector of d-axis data at time k in case1 group, is the recursive vector of q-axis data at time k in case1 group, is the d-axis gain recursive vector at moment k in case2 group, is the q-axis gain recursive vector at moment k in case2 group, is the recursive vector of d-axis data at time k in case2 group, is the recursive vector of q-axis data at moment k in case2 group, is the d-axis gain recursive vector at time k in case3 group, is the q-axis gain recursive vector at time k in case3 group, is the recursive vector of d-axis data at time k in case3 group, is the recursive vector of q-axis data at time k in case3 group, 、 、 and Corresponding to case1 group, 、 、 and Corresponding to case2 group, 、 、 and Corresponding to case3 group, set the subscript Used to distinguish calculations between groups; S4-2: Construct a fast-slow recursive least squares algorithm, which is as follows: (9); (10); (11); (12); Where, and are the d-axis and q-axis stator voltage errors at time k, 、 and 、 are the d-axis and q-axis covariance matrices at time k and time k-1 respectively, and are the d-axis and q-axis recursive gain matrices at time k, For the forgetting factor, is the identity matrix, T is the matrix transpose; S4-3: within three consecutive sampling periods, sequentially executing the fast-slow recursive least squares algorithm on the case 1, the case 2, and the case 3, and repeating this cycle to update the gain recursive vector; S4-4: Substitute the result of formula (12) into formula (4) so ​​that the vector and Updated to and ; S4-5: Calculate the data recursive vector at time k+1 and , where the d-axis component of the stator current at time k+1 is and the q-axis component of the stator current at time k+1 Their reference signals and Instead, the reference signal and The reference signal is obtained by Lagrangian algorithm. and the reference signal The calculation formula is as follows: (13); 、 、 and 、 、 are the d-axis and q-axis stator current reference values ​​at time k, k-1, and k-2 respectively.

[0011] Furthermore, in step S5, the control equation is as follows: (14).

[0012] A hyperlocal model predictive control system based on Poincare map causality comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any one of the above methods.

[0013] A computer-readable storage medium stores a computer program, wherein the computer program implements the steps of any of the above methods when executed by a processor.

[0014] By adopting the above technical solution, the present invention has the following beneficial effects: 1. The present invention introduces a hyperlocal model based on the causal relationship of the Poincare map, incorporating the coupled current component of the Poincare map and the additional factor of the electrical angular velocity into the causal relationship, so that the hyperlocal model can effectively process multiple input signals. The control process does not need to identify time-varying physical parameters, and has good adaptability. At the same time, it can accurately characterize the motion state of the motor system with good description accuracy.

[0015] 2. The present invention also introduces a fast-slow recursive least squares algorithm to estimate coefficients and maintain high fitting performance. There is no excessive computational burden during the update and prediction process, which simplifies the estimation process, avoids causing a processor resource crisis, and does not require excessive processor resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the structure of a super-local model predictive control method of the present invention; Figure 2 This is a flow chart of a super-local model predictive control method of the present invention; Figure 3 Schematic diagram of the principle of fast-slow recursive least squares algorithm. DETAILED DESCRIPTION

[0017] The invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1-Figure 2 As shown, this embodiment provides a hyperlocal model predictive control method, comprising the following steps: S1: Sample the three-phase current of the motor and transform the sampled three-phase current into the rotating coordinate system through Park to obtain the d-axis component of the stator current and the q-axis component of the stator current , based on the d-axis component of the stator current and the q-axis component of the stator current Construct a hyperlocal model based on Poincare mapping causality, which is a multi-input single-output model; S2: Since this embodiment is implemented under a continuous control set, it is necessary to discretize the hyperlocal model in step S1 and swap the model causality. The coefficients in the hyperlocal model are reorganized. The coefficients include the next cycle state gain, the current cycle state gain, the coupling state gain, the electrical angular velocity gain, and the integrated variable. The hyperlocal model based on the Poincare map causality discretized at time k+1 is obtained, where k is the current time. S3: constructing the coefficients of the hyperlocal model based on the Poincare map causality discretized at time k+1 into a gain recursive vector, and constructing the state variables of the hyperlocal model based on the Poincare map causality discretized at time k+1 into a data recursive vector; S4: updating the gain recursive vector and the data recursive vector, so that the gain recursive vector at time k is updated to the gain recursive vector at time k+1, and the data recursive vector at time k is updated to the data recursive vector at time k+1; S5: Construct a control equation using a time-shifting concept, substitute the gain recursive vector at time k+1 and the data recursive vector at time k+1 into the control equation to obtain a modulated signal; S6: The modulation signal is converted into a pulse control signal through the modulation module and controls the motor driver.

[0019] Figure 1 middle, is the electrical angular velocity reference value, and S is the pulse control signal.

[0020] In step S1, the hyperlocal model incorporates the coupled current component of the Poincare map and the additional factor of the electrical angular velocity into the causal relationship. In the rotating coordinate system, the d-axis component of the stator current The coupling current component is the q-axis component of the stator current , the q-axis component of the stator current The coupling current component is the d-axis component of the stator current , the hyperlocal model is shown in formula (1): (1); Where, is the d-axis component of the stator current in the rotating coordinate system, is the q-axis component of the stator current in the rotating coordinate system, is the d-axis component of the stator voltage in the rotating coordinate system, is the q-axis component of the stator voltage in the rotating coordinate system, and are the input gains before d-axis and q-axis reorganization, and are the state gains before d-axis and q-axis recombination, and are the coupling state gains before d-axis and q-axis recombination, and are the electrical angular velocity gains of the d-axis and q-axis before reorganization, and are the integrated variables before reorganization on the d-axis and q-axis, is the electrical angular velocity.

[0021] In step S2, the discretized hyperlocal model based on the Poincare map causality at time k+1 is as shown in formula (2): (2); Where, is the d-axis component of the stator voltage at time k+1, is the q-axis component of the stator voltage at time k+1, is the d-axis component of the stator current at time k+1, is the q-axis component of the stator current at time k+1, is the d-axis component of the stator current at time k, is the q-axis component of the stator current at time k, is the coupled current component of the d-axis component of the stator current at time k+1, is the coupled current component of the q-axis component of the stator current at time k+1, is the electrical angular velocity at time k+1, and are the periodic state gains on the d-axis and q-axis at time k, and are the state gains of the d-axis and q-axis in this cycle at time k, and are the d-axis and q-axis coupling state gains at time k, and are the d-axis and q-axis electrical angular velocity gains at time k, and are the d-axis and q-axis integrated variables at time k respectively; The reorganized next cycle state gain, current cycle state gain, coupling state gain, electrical angular velocity gain and integrated variable are derived through formula (1) and formula (2), as shown in formula (3): (3); in, For the control cycle.

[0022] In step S3, the coefficients in formula (2) are used to construct a gain recursive vector, as shown in formula (4): (4); The state variables of formula (2) are constructed into a data recursive vector. The state variables include stator voltage, stator current, coupling circuit components and electrical angular velocity, as shown in formula (5): (5); in, and are the d-axis and q-axis gain recursive vectors at time k, and is the recursive vector of d-axis and q-axis data at time k.

[0023] Furthermore, in step S4, the gain recursive vector and the data recursive vector are updated as follows: S4-1: Split the gain recursive vector of formula (4) and the data recursive vector of formula (5) into three groups, namely case 1 group, case 2 group and case 3 group: (6); (7); (8); Where, is the d-axis gain recursive vector at moment k in case1 group, is the q-axis gain recursive vector at moment k in case1 group, is the recursive vector of d-axis data at time k in case1 group, is the recursive vector of q-axis data at time k in case1 group, is the d-axis gain recursive vector at moment k in case2 group, is the q-axis gain recursive vector at moment k in case2 group, is the recursive vector of d-axis data at time k in case2 group, is the recursive vector of q-axis data at moment k in case2 group, is the d-axis gain recursive vector at time k in case3 group, is the q-axis gain recursive vector at time k in case3 group, is the recursive vector of d-axis data at time k in case3 group, is the recursive vector of q-axis data at time k in case3 group, 、 、 and Corresponding to case1 group, 、 、 and Corresponding to case2 group, 、 、 and Corresponding to case3 group, set the subscript Used to distinguish calculations between groups; S4-2: If Figure 3 As shown, a fast-slow recursive least squares algorithm is constructed. The fast-slow recursive least squares algorithm is as follows: (9); (10); (11); (12); Where, and are the d-axis and q-axis stator voltage errors at time k, 、 and 、 are the d-axis and q-axis covariance matrices at time k and time k-1 respectively, and are the d-axis and q-axis recursive gain matrices at time k, For the forgetting factor, is the identity matrix, T is the matrix transpose; S4-3: In three consecutive sampling periods, the fast-slow recursive least squares algorithm is executed for case 1, case 2, and case 3 in sequence, and the gain recursive vector is estimated and updated in this cycle. S4-4: Substitute the result of formula (12) into formula (4) so ​​that the vector and Updated to and ; S4-5: Calculate the data recursive vector at time k+1 and , where the d-axis component of the stator current at time k+1 is and the q-axis component of the stator current at time k+1 Their reference signals and Instead, the reference signal and The reference signal is obtained by Lagrangian algorithm. and reference signal The calculation formula is as follows: (13); 、 、 and 、 、 are the d-axis and q-axis stator current reference values ​​at time k, k-1, and k-2 respectively.

[0024] Furthermore, in step S5, the control equation is as follows: (14).

[0025] By using an online updated hyperlocal model based on the Poincare map causality, the control equation is established to generate the modulation signal, which can accurately predict and control the modulation signal of the permanent magnet synchronous motor drive system.

[0026] This embodiment also provides a hyperlocal model predictive control system based on Poincare mapping causality.

[0027] A hyperlocal model predictive control system based on Poincare map causality includes a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps of any one of the above methods.

[0028] This embodiment also provides a computer-readable storage medium.

[0029] A computer-readable storage medium stores a computer program, which implements the steps of any of the above methods when executed by a processor.

[0030] In summary, the present invention introduces a hyperlocal model based on the causal relationship of the Poincare map, and incorporates the coupled current component of the Poincare map and the additional factor of the electrical angular velocity into the causal relationship, so that the hyperlocal model can effectively process multiple input signals, and the control process does not need to identify time-varying physical parameters, and has good adaptability. At the same time, it can accurately characterize the motion state of the motor system with good description accuracy. In addition, the present invention also introduces a fast-slow recursive least squares algorithm to estimate coefficients and maintain high fitting performance. There is no excessive computational burden in the update and prediction process, which simplifies the estimation process, avoids causing a processor resource crisis, and does not require occupying too many processor resources.

[0031] The present invention has been described in detail above with reference to the accompanying drawings. However, 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 existing technology, and all of these modifications fall within the scope of protection of the present invention.

Claims

1. A hyperlocal model predictive control method, characterized in that: The steps include: S1: Sample the three-phase current of the motor and transform the sampled three-phase current into the rotating coordinate system through Park to obtain the d-axis component of the stator current and the q-axis component of the stator current , based on the d-axis component of the stator current and the q-axis component of the stator current Constructing a hyperlocal model based on Poincare map causality, wherein the hyperlocal model is a multi-input single-output model; S2: Discretize the hyperlocal model in step S1 and swap the model causality, reorganize the coefficients in the hyperlocal model, the coefficients including the next cycle state gain, the current cycle state gain, the coupling state gain, the electrical angular velocity gain, and the integrated variable, to obtain a hyperlocal model based on the Poincare map causality discretized at time k+1, where k is the current time; S3: constructing the coefficients of the hyperlocal model based on the Poincare map causality discretized at time k+1 into a gain recursive vector, and constructing the state variables of the hyperlocal model based on the Poincare map causality discretized at time k+1 into a data recursive vector; S4: updating the gain recursive vector and the data recursive vector, so that the gain recursive vector at time k is updated to the gain recursive vector at time k+1, and the data recursive vector at time k is updated to the data recursive vector at time k+1; S5: Construct a control equation, substitute the gain recursive vector at time k+1 and the data recursive vector at time k+1 into the control equation to obtain a modulated signal; S6: The modulation signal is converted into a pulse control signal through a modulation module and controls the motor driver.

2. The method of ultra-local model predictive control according to claim 1, characterized in that: In step S1, the hyperlocal model incorporates the coupled current component of the Poincare map and the additional factor of the electrical angular velocity into the causal relationship. The hyperlocal model is shown in formula (1): （1）； Where, is the d-axis component of the stator current in the rotating coordinate system, is the q-axis component of the stator current in the rotating coordinate system, is the d-axis component of the stator voltage in the rotating coordinate system, is the q-axis component of the stator voltage in the rotating coordinate system, and are the input gains before d-axis and q-axis reorganization, and are the state gains before d-axis and q-axis recombination, and are the coupling state gains before d-axis and q-axis recombination, and are the electrical angular velocity gains of the d-axis and q-axis before reorganization, and are the integrated variables before reorganization on the d-axis and q-axis, is the electrical angular velocity.

3. The hyperlocal model predictive control method according to claim 2, wherein: In step S2, the discretized hyperlocal model based on the Poincare map causality at time k+1 is as shown in formula (2): （2）； Where, is the d-axis component of the stator voltage at time k+1, is the q-axis component of the stator voltage at time k+1, is the d-axis component of the stator current at time k+1, is the q-axis component of the stator current at time k+1, is the d-axis component of the stator current at time k, is the q-axis component of the stator current at time k, is the coupled current component of the d-axis component of the stator current at time k+1, is the coupled current component of the q-axis component of the stator current at time k+1, is the electrical angular velocity at time k+1, and are the periodic state gains on the d-axis and q-axis at time k, and are the state gains of the d-axis and q-axis in this cycle at time k, and are the d-axis and q-axis coupling state gains at time k, and are the d-axis and q-axis electrical angular velocity gains at time k, and are the d-axis and q-axis integrated variables at time k respectively; The reorganized next cycle state gain, current cycle state gain, coupling state gain, electrical angular velocity gain and integrated variable are derived through formula (1) and formula (2), as shown in formula (3): （3）； in, For the control cycle.

4. The method of ultra-local model predictive control according to claim 3, wherein: In step S3, the coefficients in formula (2) are constructed into the gain recursive vector, as shown in formula (4): （4）； The state variables of formula (2) are constructed into the data recursive vector, and the state variables include stator voltage, stator current, coupling circuit components and electrical angular velocity, as shown in formula (5): （5）； in, and are the d-axis and q-axis gain recursive vectors at time k, and is the recursive vector of d-axis and q-axis data at time k.

5. The method of ultra-local model predictive control according to claim 4, characterized in that: In step S4, the process of updating the gain recursive vector and the data recursive vector is as follows: S4-1: Split the gain recursive vector of formula (4) and the data recursive vector of formula (5) into three groups, namely case 1 group, case 2 group and case 3 group: （6）； （7）； （8）； Where, is the d-axis gain recursive vector at moment k in case1 group, is the q-axis gain recursive vector at moment k in case1 group, is the recursive vector of d-axis data at time k in case1 group, is the recursive vector of q-axis data at time k in case1 group, is the d-axis gain recursive vector at moment k in case2 group, is the q-axis gain recursive vector at moment k in case2 group, is the recursive vector of d-axis data at time k in case2 group, is the recursive vector of q-axis data at moment k in case2 group, is the d-axis gain recursive vector at time k in case3 group, is the q-axis gain recursive vector at time k in case3 group, is the recursive vector of d-axis data at time k in case3 group, is the recursive vector of q-axis data at time k in case3 group, 、 、 and Corresponding to case1 group, 、 、 and Corresponding to case2 group, 、 、 and Corresponding to case3 group, set the subscript Used to distinguish calculations between groups; S4-2: Construct a fast-slow recursive least squares algorithm, which is as follows: （9）； （10）； （11）； （12）； Where, and are the d-axis and q-axis stator voltage errors at time k, 、 and 、 are the d-axis and q-axis covariance matrices at time k and time k-1 respectively, and are the d-axis and q-axis recursive gain matrices at time k, For the forgetting factor, is the identity matrix, T is the matrix transpose; S4-3: within three consecutive sampling periods, sequentially executing the fast-slow recursive least squares algorithm on the case 1, the case 2, and the case 3, and repeating this cycle to update the gain recursive vector; S4-4: Substitute the result of formula (12) into formula (4) so ​​that the vector and Updated to and ; S4-5: Calculate the data recursive vector at time k+1 and , where the d-axis component of the stator current at time k+1 is and the q-axis component of the stator current at time k+1 Their reference signals and Instead, the reference signal and The reference signal is obtained by Lagrangian algorithm. and the reference signal The calculation formula is as follows: （13）； 、 、 and 、 、 are the d-axis and q-axis stator current reference values ​​at time k, k-1, and k-2 respectively.

6. The method of ultra-local model predictive control according to claim 5, characterized in that: In step S5, the control equation is as follows: （14）。 7. A hyperlocal model predictive control system based on Poincare map causality, comprising a memory, a processor, and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, wherein: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images