Permanent magnet motor driving system switching point selection model prediction control method

Abstract

The invention relates to a motor control method, in particular to a permanent magnet motor driving system switching point selection model prediction control method, a permanent magnet motor is a synchronous motor, and a driving circuit of the permanent magnet motor comprises two level converters. Comprising the following steps of establishing a permanent magnet synchronous motor mathematical model under a dq-axis rotating coordinate system, performing forward Euler discretization on differential terms of the permanent magnet synchronous motor mathematical model, quantifying the differential terms of the permanent magnet synchronous motor, performing cost function error term disassembly and the like. According to the model predictive control method adopted by the invention, current predictive substitution is carried out by combining a current differential term with a switching point balance equation, the influence of external disturbance and load change is not easily caused, the control precision is relatively high, the response speed is relatively high, a target function can be quickly pre-selected, an error term can be decomposed, the algorithm complexity is reduced, and the efficiency is improved. And meanwhile, the differential terms of the permanent magnet synchronous motor are quantized, so that the number of current iterations is reduced, and the steady-state performance of the system is improved.

Description

Technical Field

[0001] The invention relates to a motor control method, in particular to a switch point selection model predictive control method for a permanent magnet motor drive system. Background Art

[0002] Satellite attitude control is usually completed using a permanent magnet motor drive system. Traditional permanent magnet motor drive systems usually use PI control methods, which are easily affected by external disturbances and load changes, and have insufficient control accuracy. At the same time, the dynamic response of this control method is affected by current, and the response speed is relatively slow.

[0003] Model predictive control can make up for the shortcomings of traditional control methods in some aspects. However, the algorithms of existing model predictive control methods are complex, the number of current iterations increases, and the computing power requirements are relatively high, which affects the steady-state performance of the system.

[0004] In view of this, the applicant conducted in-depth research on the model predictive control method for switch point selection of permanent magnet motor drive system, which led to the present case. Summary of the invention

[0005] The object of the present invention is to provide a model predictive control method for selecting a switch point of a permanent magnet motor drive system with relatively high control accuracy, relatively fast dynamic response speed, relatively small computing power requirements and relatively good system steady-state performance.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A model predictive control method for selecting a switch point of a permanent magnet motor drive system, wherein the permanent magnet motor is a synchronous motor and a drive circuit thereof includes a two-level converter, comprising the following steps: S1, establishing a mathematical model of a permanent magnet synchronous motor in a dq axis rotating coordinate system, and transforming the circuit topology of the two-level converter from an abc three-phase coordinate system to an αβ coordinate system; S2, performing forward Euler discretization on the differential term of the permanent magnet synchronous motor mathematical model, and simplifying it to obtain the current prediction value of the permanent magnet synchronous motor in the dq axis coordinate system, then introducing a vector switching point and updating the current prediction value, and then obtaining the current differential reference value and the permanent magnet synchronous motor differential term according to the updated current prediction value; S3, quantifying the differential term of the permanent magnet synchronous motor to obtain a set of all switching points that can make the single-axis current reach the current differential reference value in advance, and establishing a vector switching point balance equation according to the set; S4, decompose the cost function error term of all possible output results of the two-level converter to obtain the corresponding error term, and arrange the error terms in order of size, then introduce soft constraints, establish the initial boundary, and the boundary error adjustment range of the vector dynamic selection process is ±5%. The optimal vector is selected inside or at the edge of the boundary, and finally the corresponding binary number is used as the system input signal to obtain the optimal output.

[0007] As an improvement of the present invention, in step S1, a mathematical model of the permanent magnet synchronous motor in the dq axis rotating coordinate system is established as follows: (1)

[0008] (2) (3) in, u d and u q is the stator voltage in the dq axis coordinate system, i sd and i sq is the stator current in the dq axis coordinate system, is the stator resistance of the permanent magnet synchronous motor, and is the dq axis component of the stator inductance, and is the dq-axis component of the stator flux, is the permanent magnet flux, is the mechanical angular velocity of the permanent magnet synchronous motor rotor, T e is the electromagnetic torque, p is the number of magnetic pole pairs of permanent magnet synchronous motor.

[0009] As an improvement of the present invention, in step S1, the circuit topology of the two-level converter is transformed from the abc three-phase coordinate system to the αβ coordinate system to obtain the following formula: (4) Then transform formula (4) from the αβ coordinate system to the dq coordinate system to obtain the following formula: (5) in, θ is the motor rotor mechanical angle, , They are Clark and Park transformation matrices respectively.

[0010] As an improvement of the present invention, in step S2, the differential term of formula (1) is forward Euler discretized and simplified to obtain the current prediction value of the permanent magnet synchronous motor in the dq axis coordinate system. i sd ( k + 1) and i sq ( k + 1), (6) (7) in, is the predicted value of the q-axis current in the k+1th sampling period d, represents the sampling value of the d and q axis currents in the kth period, is the time interval of the sampling period; Introduce vector switching points to update current prediction values i sd ( k + 1) and i sq ( k + 1), the process is as follows: (8) (9) (10) (11) Among them, m represents arrive The vector switching point position between are the current magnitudes corresponding to the vector switching time m in the kth sampling period, namely The d-axis and q-axis components of the current at the moment; Indicated in arrive The voltage vector pre-applied for the time period, are all intermediate variables. The time interval from the vector switching point position m to the end of the sampling period is the action time interval of the new vector after switching, that is, arrive Time period, and The sampling period ends The predicted d-axis and q-axis components of the current at time instant, Indicated in arrive Voltage vector updated in time period; The current differential reference values ​​of the d-axis and q-axis are: (12) (13) in, Indicates the d-axis current reference value, set to 0; q-axis current reference Defined according to the reference torque of the speed outer loop; Combining formula (1) and formula (2), the current differential term of the permanent magnet synchronous motor can be obtained as follows: (14) (15) in, and Proportional, Hecheng Proportional.

[0011] As an improvement of the present invention, in step S3, the vector switching point balance equation is established as follows: (16) (17) Reasonable range of vector switching point m (0, ), within which the switching point is set at the minimum value in the range.

[0012] As an improvement of the present invention, in step S4, the global optimization method is used to decompose the error term of the cost function as follows: (18) (19) (20) in, The cost function designed for traditional predictive current control includes two current error terms in the d and q axes. , Decompose the above two current error terms.

[0013] By adopting the above scheme, the present invention has the following beneficial effects: The model predictive control method adopted in the present invention utilizes the current differential term in combination with the switching point balance equation to perform current prediction instead of the traditional current inner loop. It is not easily affected by external disturbances and load changes, has relatively high control accuracy, and relatively fast response speed. It can quickly pre-select the objective function and decompose the error term without negatively affecting the control performance, thereby reducing the algorithm complexity and requiring relatively little computing power. At the same time, by quantifying the differential term of the permanent magnet synchronous motor, it can achieve fast switching point selection, reduce the number of current iterations, and improve the steady-state performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Schematic diagram of the circuit topology structure of the two-level converter of the permanent magnet motor in the embodiment; Figure 2 This is a block diagram of the model-predicted current control in the embodiment; Figure 3 A quantified schematic diagram of a set of all switching points in which the single-axis current reaches a reference value in advance in an embodiment; Figure 4 Schematic diagram of optimal vector selection in the embodiment. DETAILED DESCRIPTION

[0015] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0016] This embodiment provides a model predictive control method for selecting a switch point of a permanent magnet motor drive system. The permanent magnet motor is a synchronous motor, and its drive circuit includes a two-level converter. The two-level converter is a conventional converter that can be purchased directly from the market. The circuit topology is as follows: Figure 1 shown.

[0017] The model predictive control method includes the following steps: S1, establish the mathematical model of permanent magnet synchronous motor in dq axis rotating coordinate system as follows: (1)

[0018] (2) (3) in, u d and u q is the stator voltage in the dq axis coordinate system, i sd and i sq is the stator current in the dq axis coordinate system, is the stator resistance of the permanent magnet synchronous motor, and is the dq axis component of the stator inductance, and is the dq-axis component of the stator flux, is the permanent magnet flux, is the mechanical angular velocity of the permanent magnet synchronous motor rotor, T e is the electromagnetic torque, p is the number of magnetic pole pairs of permanent magnet synchronous motor.

[0019] like Figure 1 As shown in the figure, the permanent magnet synchronous motor is used as the load end of the inverter. The driving circuit topology includes a three-phase bridge arm composed of 6 switch tubes. The upper and lower switch tubes of each phase bridge arm form the upper and lower bridge arms. The binary numbers "1" and "0" represent the on and off states of the switch tubes respectively. During operation, the upper and lower switch tubes of each phase bridge arm cannot be turned on at the same time. Therefore, the switching state of the inverter can be expressed by the three-bit binary number Indicates, for example represent , The upper switch is turned on. The upper switch tube is cut off.

[0020] The three phases abc with a phase difference of 120° are converted into two orthogonal dq axis phases for analysis, and the coordinate transformation is as follows: First, Clark transformation is used to transform the circuit topology of the two-level converter from the abc three-phase coordinate system to the αβ coordinate system, and the following formula is obtained: (4) Then, Park transformation is used to transform formula (4) from the αβ coordinate system to the dq coordinate system, and the following formula is obtained: (5) in, θ is the motor rotor mechanical angle, , They are Clark and Park transformation matrices respectively.

[0021] S2, perform model prediction current control based on the vector switching point. Specifically, the model prediction current control block diagram used in this embodiment is as follows: Figure 2 As shown, firstly, the differential term of the permanent magnet synchronous motor mathematical model is forward Euler discretized, and simplified to obtain the current prediction value of the permanent magnet synchronous motor in the dq axis coordinate system, that is, the differential term of formula (1) is forward Euler discretized, and simplified to obtain the current prediction value of the permanent magnet synchronous motor in the dq axis coordinate system. i sd ( k + 1) and i sq ( k + 1), the process is as follows:

[0022] (6) (7) in, is the predicted value of the q-axis current in the k+1th sampling period d, represents the sampling value of the d and q axis currents in the kth period, is the time interval of the sampling period.

[0023] Then the vector switching point is introduced and the current prediction value is i sd ( k + 1) and i sq ( k + 1) To update, the process is as follows: (8) (9) (10) (11) Among them, m represents arrive The vector switching point position between are the current magnitudes corresponding to the vector switching time m in the kth sampling period, namely The d-axis and q-axis components of the current at the moment; Indicated in arrive The voltage vector pre-applied for the time period, are all intermediate variables. The time interval from the vector switching point position m to the end of the sampling period is the action time interval of the new vector after switching, that is, arrive Time period, and The sampling period ends The predicted d-axis and q-axis components of the current at time instant, Indicated in arrive The voltage vector is updated during the time period.

[0024] It should be noted that the above formulas (8) and (9) are the current prediction values ​​corresponding to the vector switching point positions within the sampling interval, and formula (10) and formula (11) are the corresponding current prediction values ​​at the end of the sampling period.

[0025] Then, the current differential reference value and the permanent magnet synchronous motor differential term are obtained according to the updated current prediction value. Specifically, the forward Euler equation is as follows:

[0026] Will Moment Definition Reach the reference value, , at this time the differential term is the reference value Similarly, the current differential reference values ​​of the d-axis and q-axis are: (12) (13) in, Indicates the d-axis current reference value, set to 0; q-axis current reference Defined according to the reference torque of the speed outer loop; Combining formula (1) and formula (2), the current differential term of the permanent magnet synchronous motor can be obtained as follows: (14) (15) Assume that within a continuous time interval, the speed Approximately unchanged, and ignoring the magnetic saturation effect, then and unchanged, from formula (14) and formula (15), we can know that and Proportional, Hecheng Proportional.

[0027] S3, switch point quantization and fast switch point selection predictive current control. Specifically, the permanent magnet synchronous motor differential term is quantized to obtain a set of all switching points that can make the single-axis current reach the current differential reference value in advance. The quantization schematic diagram is shown in Figure 3 As shown, assuming that the initial voltage vector of the kth sampling period is , corresponding to the d, q axis voltage components , , the slope size , ; According to the geometric principle, in order to achieve error-free tracking of the d and q axes at the end of the kth sampling period, the d and q axis reference currents of the two-dimensional plane correspond to the voltage vector , Reverse extension.

[0028] At the same time, the vector switching point equilibrium equation is established according to the set as follows: (16) (17) Reasonable range of vector switching point m (0, ), within this range, the switching point is set at the minimum value in the range, which meets the requirements of rapidity and performance.

[0029] S4, in order to balance the excitation component and torque component of the permanent magnet synchronous motor current, a global optimization method is used to decompose all possible output results of the two-level converter into the cost function error term to obtain the corresponding error term. The cost function error term is decomposed as follows: (18) (19) (20) in, The cost function designed for traditional predictive current control includes two current error terms in the d and q axes. , Decompose the above two current error terms.

[0030] According to the eight possible output results of the three-phase two-level voltage source inverter, the voltage value is substituted into formula (10) and formula (11) to predict The current at the moment is used to calculate the error term according to formula (19) and formula (20).

[0031] like Figure 4 As shown, the error terms are arranged in order of size, and then the error term is introduced , The limited range, i.e. soft constraint, combines the traditional model to predict the output value of the current control cost function, predefines the initial error boundary, ensures that the optimization range is not an empty set, establishes the initial boundary, and dynamically selects the vector: the initial boundary error adjustment range is ±5%, so that only 2-4 vectors are included in the optimization boundary, and the optimal vector is selected inside or on the edge. The selection principle is as follows: If , There is an intersection within the boundary, and the output intersection has a higher priority in controlling torque fluctuations The corresponding smaller voltage vector is taken as the optimal output; otherwise, when satisfying In the voltage vector of the soft constraint premise, the output The smaller corresponding voltage vector is finally used as the system input signal to obtain the optimal output.

[0032] 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 model predictive control method for selecting switch points of a permanent magnet motor drive system, wherein the permanent magnet motor is a synchronous motor, and its drive circuit includes a two-level converter, characterized in that: The following steps are involved: S1, establishing a mathematical model of a permanent magnet synchronous motor in a dq axis rotating coordinate system, and transforming the circuit topology of the two-level converter from an abc three-phase coordinate system to an αβ coordinate system; S2, performing forward Euler discretization on the differential term of the permanent magnet synchronous motor mathematical model, and simplifying it to obtain the current prediction value of the permanent magnet synchronous motor in the dq axis coordinate system, then introducing a vector switching point and updating the current prediction value, and then obtaining the current differential reference value and the permanent magnet synchronous motor differential term according to the updated current prediction value; S3, quantifying the differential term of the permanent magnet synchronous motor to obtain a set of all switching points that can make the single-axis current reach the current differential reference value in advance, and establishing a vector switching point balance equation according to the set; S4, decompose the cost function error term of all possible output results of the two-level converter to obtain the corresponding error term, and arrange the error terms in order of size, then introduce soft constraints, establish the initial boundary, and the boundary error adjustment range of the vector dynamic selection process is ±5%. The optimal vector is selected inside or at the edge of the boundary, and finally the corresponding binary number is used as the system input signal to obtain the optimal output.

2. The switch point selection model predictive control method for a permanent magnet motor drive system according to claim 1, characterized in that: In step S1, the mathematical model of the permanent magnet synchronous motor in the dq axis rotating coordinate system is established as follows: （1） （2） （3） in, u d and u q is the stator voltage in the dq axis coordinate system, i sd and i sq is the stator current in the dq axis coordinate system, is the stator resistance of the permanent magnet synchronous motor, and is the dq axis component of the stator inductance, and is the dq-axis component of the stator flux, is the permanent magnet flux, is the mechanical angular velocity of the permanent magnet synchronous motor rotor, T e is the electromagnetic torque, p is the number of magnetic pole pairs of permanent magnet synchronous motor.

3. The switch point selection model predictive control method for a permanent magnet motor drive system according to claim 2, characterized in that: In step S1, the circuit topology of the two-level converter is transformed from the abc three-phase coordinate system to the αβ coordinate system to obtain the following formula: （4） Then transform formula (4) from the αβ coordinate system to the dq coordinate system to obtain the following formula: （5） in, θ is the motor rotor mechanical angle, , They are Clark and Park transformation matrices respectively.

4. The switch point selection model predictive control method for a permanent magnet motor drive system according to claim 3, characterized in that: In step S2, the differential term of formula (1) is forward Euler discretized and simplified to obtain the current prediction value of the permanent magnet synchronous motor in the dq axis coordinate system: i sd ( k + 1) and i sq ( k + 1), （6） （7） in, is the predicted value of the q-axis current in the k+1th sampling period d, represents the sampling value of the d and q axis currents in the kth period, is the time interval of the sampling period; Introduce vector switching points to update current prediction values i sd ( k + 1) and i sq ( k + 1), the process is as follows: （8） （9） （10） （11） Among them, m represents arrive The vector switching point position between are the current magnitudes corresponding to the vector switching time m in the kth sampling period, namely The d-axis and q-axis components of the current at the moment; Indicated in arrive The voltage vector pre-applied for the time period, are all intermediate variables. The time interval from the vector switching point position m to the end of the sampling period is the action time interval of the new vector after switching, that is, arrive Time period, and The sampling period ends The predicted d-axis and q-axis components of the current at time instant, Indicated in arrive Voltage vector updated in time period; The current differential reference values ​​of the d-axis and q-axis are: （12） （13） in, Indicates the d-axis current reference value, set to 0; q-axis current reference Defined according to the reference torque of the speed outer loop; Combining formula (1) and formula (2), the current differential term of the permanent magnet synchronous motor can be obtained as follows: （14） （15） in, and Proportional, Hecheng Proportional.

5. The switch point selection model predictive control method for a permanent magnet motor drive system according to claim 4, characterized in that: In step S3, the vector switching point balance equation is established as follows: （16） （17） Reasonable range of vector switching point m (0, ), within which the switching point is set at the minimum value in the range.

6. The switch point selection model predictive control method for a permanent magnet motor drive system according to claim 5, characterized in that: In step S4, the global optimization method is used to decompose the error term of the cost function as follows: （18） （19） （20） in, The cost function designed for traditional predictive current control includes two current error terms in the d and q axes. , Decompose the above two current error terms.

Images