A single-loop deadbeat predictive anti-interference control method for permanent magnet synchronous motor

Abstract

The present invention discloses a single-loop deadbeat predictive anti-interference control method for a permanent magnet synchronous motor. The method includes establishing a generalized disturbed mathematical model of a surface-mounted permanent magnet synchronous motor in a dq synchronous rotating coordinate system; designing a discrete-time disturbance observer to online reconstruct the total disturbance in each channel; constructing a reference system and a state tracking error system based on a disturbance estimation signal, a reference speed, and a d-axis reference current; using the reference system model to obtain the d- and q-axis reference control input voltages; and calculating the d- and q-axis error control input voltages for the state tracking error system based on the deadbeat predictive control principle. The reference control input voltage and the error control input voltage are added to obtain the control input voltage required for motor drive. Based on the deadbeat predictive control concept, a single-loop deadbeat predictive anti-interference control method for a PMSM is proposed. To improve the robustness of the system, a set of discrete-time disturbance observers are designed to online correct the prediction model.

Description

Technical Field

[0001] The present invention belongs to the technical field of motor control algorithm design, and in particular relates to a single-loop deadbeat prediction anti-interference control method for a permanent magnet synchronous motor. Background Art

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in various industrial applications due to their compact size, high efficiency, low electrical losses, and low failure rate. However, high-precision control of PMSMs remains challenging because they are nonlinear systems with multiple coupled variables. In particular, when affected by internal parameter perturbations and external load torque, conventional linear control methods such as linear PID control may fail to achieve the desired control effect and may even fail to guarantee the stability of the closed-loop system.

[0003] To improve the control accuracy of PMSM systems, many advanced control methods have been proposed, including adaptive control, sliding mode control, internal model control, fuzzy logic control, and model predictive control (MPC). Among them, MPC is considered one of the most promising control methods for AC drive systems due to its simple implementation and ease of handling nonlinear and multivariable constraints. Based on the different implementation methods of the control action, MPC applied to PMSMs can be divided into two categories: finite control set MPC (FCS-MPC) and continuous control set MPC (CCS-MPC). FCS-MPC uses discrete models of the motor and inverter to predict future dynamics. By minimizing a cost function, it selects the optimal switching state from a limited set of switching states and directly applies it to the inverter. FCS-MPC is known for its fast dynamic response performance. However, as the prediction horizon increases, the online computational complexity of this method increases exponentially, which poses a challenge for real-time control. In addition, since the FCS-MPC system does not contain a modulator, the switching frequency of the inverter is not fixed, and the limited voltage vector makes the current harmonics of this method larger. The main idea of ​​CCS-MPC is to obtain a continuously variable voltage vector reference by solving an optimization problem, and then modulate the voltage vector reference using space vector pulse width modulation (SVPWM) technology. What is very different from FCS-MPC is that CCS-MPC can set the modulator to generate a fixed inverter switching frequency as needed, and the harmonic content of the current is smaller. However, CCS-MPC based on multi-step prediction also has the problem of heavy online computational burden, and the weight factor in the cost function increases the complexity of controller parameter tuning.

[0004] Deadbeat predictive control (DPC), a type of CCS-MPC, is considered a practical control technology for AC drive systems. Its design does not involve a cost function or require weighting factors, and it exhibits low current harmonics and fast response. Like other MPC methods, DPC is a model-based control technique. When applied to PMSM systems, its control accuracy is susceptible to motor parameter perturbations and external loads. Furthermore, existing MPC strategies for PMSMs, including DPC, are mostly based on the classic field-oriented control (FOC) framework. Within the FOC framework, a dual-loop cascade control structure is typically used, in which both speed and current are regulated as first-order subsystems by separate controllers. This dual-loop control structure facilitates controller design, but to ensure closed-loop stability, the inner and outer loops typically use different control periods. Specifically, the control period of the inner loop (current loop) must be significantly smaller than that of the outer loop (speed loop), which can hinder further improvements in the dynamic response performance of the PMSM system. In addition, designing different controllers for the inner and outer loops usually means that more parameters need to be tuned, which is not friendly to engineering applications. Summary of the Invention

[0005] In order to solve the technical problems existing in the background technology, the present invention aims to provide a single-loop zero-beat predictive anti-interference control method for a permanent magnet synchronous motor, which simplifies the traditional speed-current dual-loop cascade control structure in the permanent magnet synchronous motor system into a single-loop control structure. The closed-loop system adopts the same control period, and only the observer bandwidth parameter needs to be adjusted, which is very convenient for engineering applications. By establishing an uncertainty estimation and compensation mechanism based on a discrete-time disturbance observer, the closed-loop control system has strong robustness against both internal parameter perturbations and external load disturbances.

[0006] In order to solve the technical problem, the technical solution of the present invention is:

[0007] Based on the deadbeat predictive control concept, the present invention proposes a single-loop deadbeat predictive disturbance rejection control (SDPDRC) method for PMSM. To improve the robustness of the system, a set of discrete-time disturbance observers (DDOBs) are designed to correct the prediction model online.

[0008] A single-loop deadbeat predictive anti-interference control method for a permanent magnet synchronous motor, the method comprising:

[0009] S1: In the dq synchronous rotating coordinate system, a generalized disturbed mathematical model of the surface-mounted permanent magnet synchronous motor is established. The uncertainty of the parameters in the motor system is considered, the dynamic behavior of the system is described by the state space model, and an unknown total disturbance is introduced.

[0010] S2: Design a discrete-time disturbance observer to estimate the total disturbance online. By assuming that the variation of the disturbance is bounded, the observer accurately estimates the total disturbance in each channel. The observer gain is adjusted according to the system bandwidth parameter.

[0011] S3: Based on the estimated total disturbance, reference speed, and d-axis reference current, the reference system and state tracking error system models are constructed.

[0012] S4: deriving the d-axis and q-axis reference control input voltages using the reference system model;

[0013] S5: for the state tracking error system model, according to the deadbeat predictive control principle, obtain the d-axis and q-axis error control input voltages;

[0014] S6: Add the reference control input voltage obtained in step S4 and the error control input voltage obtained in step S5 to obtain the control input voltage required for the motor drive in the current control cycle.

[0015] Furthermore, the step S1 includes:

[0016] S101: In the dq synchronous rotating coordinate system, the mathematical model of the surface-mount PMSM is:

[0017]

[0018] Where R and L are the stator resistance and inductance respectively, p is the number of motor pole pairs, J is the moment of inertia of the rotor, and B is the v is the viscous damping coefficient, ψ is the magnetic flux of the rotor permanent magnet, i d and i q are d-axis and q-axis currents, respectively, and u d and u q are the d and q axis voltages, ω, ω e =pω are the mechanical angular velocity and electrical angular velocity of the rotor respectively, T l is the load torque;

[0019] S102: Further consider the following parameter uncertainties in the PMSM system:

[0020]

[0021] Where, for *=R,L,ψ,J, “*0” represents the known nominal value of parameter “*”, and “δ *” represents the unknown perturbation of parameter “*”;

[0022] S103: Based on sub-steps S101 and S102, the generalized disturbed state space model of the PMSM is established as:

[0023]

[0024] In the formula, the state variable

[0025] Control Input

[0026] The nonlinear coupling vector is known:

[0027]

[0028] Measurable output

[0029] Total disturbance vector where d d (t), d q (t), d ω (t) represents the unknown “total disturbance” in each subsystem, and its expressions are:

[0030]

[0031] Coefficient matrix:

[0032]

[0033] Where I3 represents the 3×3 order unit matrix;

[0034] S104: T c To control the period, the forward Euler method is used to discretize the generalized disturbed state space model of the PMSM in step S103 to obtain a discrete controlled model:

[0035]

[0036] Further, the step S2 includes:

[0037] S201: Determine the disturbance as well as The changes in the amount are bounded, that is, there is a positive constant μ i (i=d,q,ω) and μ ω2 So that:

[0038]

[0039] S202: Design the following discrete-time disturbance observer-1 online estimation and

[0040]

[0041] Where, subscript i = d,q; is an auxiliary variable; nonlinear coupling term for estimates; is the observer gain, which can be selected as:

[0042]

[0043] in, represents the adjustable bandwidth of the observer;

[0044] S203: Definition Design the following discrete-time disturbance observer-2 online estimation

[0045]

[0046] Where, is an auxiliary variable; and They are and Estimates of ; coefficient matrix:

[0047]

[0048] is the observer gain matrix, which is selected according to the following rules:

[0049]

[0050] Furthermore, the construction of the reference system and state tracking error system models in step S3 includes:

[0051] S301: Assume that the discrete-time disturbance observer in step S2 accurately estimates the total disturbance, that is,

[0052]

[0053] in, For disturbance estimated value of;

[0054] S302: Based on the output regulation theory, the reference model of the discrete controlled PMSM system in sub-step S104 is introduced:

[0055]

[0056] Where, is the reference state vector; is the reference control input voltage;

[0057] is the reference nonlinear coupling vector; according to the discrete controlled system model and MTPA principle:

[0058]

[0059] S303: Subtract the reference model in sub-step S302 from the discrete controlled model in step S104 to obtain the state tracking error system:

[0060]

[0061] The error variable is defined as follows:

[0062]

[0063] Furthermore, in step S4, the d-axis and q-axis reference control input voltages of the current control cycle are:

[0064]

[0065] Where, Represents the matrix T c Pseudo-inverse matrix of B; state vector in:

[0066]

[0067] Furthermore, in step S5, the d-axis and q-axis error control input voltages of the current control cycle are designed as follows:

[0068] According to the deadbeat predictive control theory, the state tracking error state of the system at the next moment in step S302 is:

[0069]

[0070] Among them, 03 represents a 3×3 order zero matrix; the error control input vector of the current control cycle is calculated as:

[0071]

[0072] Furthermore, in step S6, the control input voltage required for the motor drive in the current control cycle is:

[0073]

[0074] Compared with the prior art, the advantages of the present invention are:

[0075] The controller proposed in the present invention has a single-loop control structure, the current loop in the traditional cascade structure is eliminated, the system structure is simplified, and the closed-loop system has the same time constant;

[0076] The present invention adopts the PMSM single-loop deadbeat predictive control strategy for the first time. The controller has only one adjustable parameter, the observer bandwidth, which greatly reduces the complexity of parameter setting and the actual application cost.

[0077] By establishing a full-loop disturbance estimation and compensation mechanism based on DDOB, the closed-loop system is highly robust to both internal parameter perturbations and external load torque. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 , a structural diagram of a PMSM control system based on the method of the present invention disclosed in an embodiment of the present invention;

[0079] Figure 2 , the structural diagram of the SDPDRC controller disclosed in the present invention;

[0080] Figure 3 , DDOB-1 structure diagram;

[0081] Figure 4 , DDOB-2 structure diagram;

[0082] Figure 5 , set the reference speed ω r = Experimental results of the method of the present invention at 2000 rpm;

[0083] Figure 6 , set the reference speed ω r =2000rpm when the comparison method (double closed-loop PI control) experimental results. DETAILED DESCRIPTION

[0084] The specific implementation of the present invention is described below in conjunction with examples:

[0085] It should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the conditions under which the present invention can be implemented. Any structural modification, change in proportional relationship or adjustment of size should still fall within the scope of the technical content disclosed in the present invention without affecting the efficacy and purpose that can be achieved by the present invention.

[0086] At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" quoted in this specification are only for the convenience of description and are not used to limit the scope of implementation of the present invention. Changes or adjustments to their relative relationships should be regarded as the scope of implementation of the present invention without substantially changing the technical content.

[0087] Example 1:

[0088] Figure 1 The structure diagram of the PMSM control system using the method disclosed in the present invention, wherein the SDPDRC controller structure is as follows Figure 2 As shown, the structure of DDOB-1 is as follows Figure 3 As shown, the structure of DDOB-2 is as follows Figure 4 shown.

[0089] exist Figure 1-4 The method shown includes the following steps:

[0090] S1. In the dq synchronous rotating coordinate system, the mathematical model of the surface-mount PMSM is:

[0091]

[0092] Where R and L are the stator resistance and inductance respectively, p is the number of motor pole pairs, J is the moment of inertia of the rotor, and B is the v is the viscous damping coefficient, ψ is the magnetic flux of the rotor permanent magnet, i d and i q are d-axis and q-axis currents, respectively, and u d and u q are the d and q axis voltages, ω, ω e =pω are the mechanical angular velocity and electrical angular velocity of the rotor respectively, T l is the load torque.

[0093] Further consider the following parameter uncertainties in the PMSM system:

[0094]

[0095] Where, for *=R,L,ψ,J, “*0” represents the known nominal value of parameter “*”, and “δ * ” represents the unknown perturbation of parameter “*”.

[0096] Based on the above considerations, the generalized disturbed state space model of PMSM can be established as:

[0097]

[0098] In the formula, the state variable

[0099] Control Input

[0100] The nonlinear coupling vector is known:

[0101]

[0102] Measurable output

[0103] Total disturbance vector where d d (t), d q (t), d ω (t) represents the unknown “total disturbance” in each subsystem, and its expressions are:

[0104]

[0105] Coefficient matrix:

[0106]

[0107] Where I3 represents a 3×3 unit matrix.

[0108] T c In order to control the period, the forward Euler method is used to discretize the generalized disturbed state space model of PMSM to obtain the discrete controlled model:

[0109]

[0110] S2, assuming the disturbance in step S1 as well as The changes in the amount are bounded, that is, there is a positive constant μ i (i=d,q,ω) and μ ω2 Make

[0111]

[0112] Design the following discrete-time disturbance observer-1 online estimation and

[0113]

[0114] Where, subscript i = d,q; is an auxiliary variable; nonlinear coupling term for estimates; is the observer gain, which can be selected as:

[0115]

[0116] in, represents the adjustable bandwidth of the observer.

[0117] definition Design the following discrete-time disturbance observer-2 online estimation

[0118]

[0119] Where, is an auxiliary variable; and They are and Estimates of ; coefficient matrix:

[0120]

[0121] is the observer gain matrix, which can be selected according to the following rules:

[0122]

[0123] S3. Assume that the discrete-time disturbance observer in step S2 accurately estimates the total disturbance, that is,

[0124]

[0125] in, For disturbance estimated value.

[0126] According to the output regulation theory, the reference model of the discrete controlled PMSM system in step S1 is introduced:

[0127]

[0128] Where, is the reference state vector; is the reference control input voltage; is the reference nonlinear coupling vector. According to the discrete controlled system model and MTPA principle:

[0129]

[0130] The discrete controlled model in step S1 minus the reference model in step S3 gives the state tracking error system:

[0131]

[0132] The error variable is defined as follows:

[0133]

[0134] S4. The d-axis and q-axis reference control input voltages of the current control cycle are:

[0135]

[0136] Where, Represents the matrix T c Pseudo-inverse matrix of B; state vector in

[0137]

[0138] S5. According to the deadbeat predictive control concept, the state in step S3 is set to track the error state of the system at the next moment:

[0139]

[0140] Among them, 03 represents a 3×3 order zero matrix. The error control input vector of the current control cycle is calculated as:

[0141]

[0142] S6. The control input voltage required for the motor drive in the current control cycle is:

[0143]

[0144] Where, and Defined in step S4 and step S5 respectively.

[0145] Example 2:

[0146] To test the control performance of a single-loop deadbeat predictive anti-interference control method for a permanent magnet synchronous motor (PMSM) disclosed in this invention, the proposed method was applied to a coaxial dual-PMSM test platform. This platform contained two identical PMSMs, with motor-1 serving as the measured motor and motor-2 as the load motor. The nominal parameters of the measured motors are shown in Table 1. The motor control algorithm was developed in MATLAB / Simulink, compiled, and generated into a C-language control program, which was downloaded to the CPU (F28379D) of the DSP control board and executed.

[0147] Table 1

[0148]

[0149] In the controller provided by the present invention, the inverter switching frequency is set to 10kHz, and the system control period T s =2×10 -4 s, observer bandwidth The traditional double closed-loop PI is selected as the comparison method, where the inverter switching frequency is set to 10kHz and the current loop control period is set to T s =2×10 -4 s, the speed loop control period is set to T s =1×10 -3 s; the proportional gain of the speed loop PI controller is selected as 4, and the integral gain is selected as 20; the proportional gain of the d-axis current loop PI controller is selected as 5, and the integral gain is selected as 5000; the proportional gain of the q-axis current loop PI controller is selected as 0.27, and the integral gain is selected as 8. In the experimental verification of the proposed method and the comparative method, the reference speed ω is set. r =2000rpm, the output torque of the load motor changes according to the following rules: When t<1s, T l =0Nm; 2s≤t≤4s, T l =0.188Nm; 7s≤t, T l =0.15+0.05sin[7π(t-7)]Nm.

[0150] Figure 5 (a)-(c) are the experimental results of the method proposed by the present invention under three load conditions. Figure 6 (a)-(c) are the experimental results of the double closed-loop PI method under three load conditions. Figure 5 and Figure 6 It can be seen that the output speed of the PMSM control system under the method proposed in the present invention can track the given reference speed quickly and without overshoot. Under sudden loads and fast time-varying loads, the speed deviates from the given value to a small extent, showing that the method has better dynamic response performance and robustness.

[0151] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0152] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0153] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0155] The preferred embodiments of the present invention are described in detail above, but the present invention is not limited to the above embodiments. Various changes can be made within the knowledge of ordinary technicians in this field without departing from the scope of the present invention.

[0156] Many other changes and modifications can be made without departing from the spirit and scope of the present invention. It should be understood that the present invention is not limited to the specific embodiments, and the scope of the present invention is defined by the appended claims.

Claims

1. A single-loop deadbeat predictive anti-interference control method for a permanent magnet synchronous motor, characterized in that: The method comprises: S1: In the dq synchronous rotating coordinate system, a generalized disturbed mathematical model of the surface-mounted permanent magnet synchronous motor is established. The uncertainty of the parameters in the motor system is considered, the dynamic behavior of the system is described by the state space model, and an unknown total disturbance is introduced. S2: Design a discrete-time disturbance observer to estimate the total disturbance online. By assuming that the variation of the disturbance is bounded, the observer accurately estimates the total disturbance in each channel. The observer gain is adjusted according to the system bandwidth parameter. S3: Based on the estimated total disturbance, reference speed, and d-axis reference current, the reference system and state tracking error system models are constructed. S4: deriving the d-axis and q-axis reference control input voltages using the reference system model; S5: for the state tracking error system model, according to the deadbeat predictive control principle, obtain the d-axis and q-axis error control input voltages; S6: Adding the reference control input voltage obtained in step S4 and the error control input voltage obtained in step S5 to obtain the control input voltage required for the motor drive in the current control period; The step S1 comprises: S101: In the dq synchronous rotating coordinate system, the mathematical model of the surface-mount PMSM is: Where R and L are the stator resistance and inductance respectively, p is the number of motor pole pairs, J is the moment of inertia of the rotor, and B is the v is the viscous damping coefficient, ψ is the magnetic flux of the rotor permanent magnet, i d and i q are d-axis and q-axis currents, respectively, and u d and u q are the d and q axis voltages, ω, ω e =pω are the mechanical angular velocity and electrical angular velocity of the rotor respectively, T l is the load torque; S102: Further consider the following parameter uncertainties in the PMSM system: Where, for *=R,L,ψ,J, "*0" represents the known nominal value of the parameter "*", "δ * ” represents the unknown perturbation of parameter “*”; S103: Based on sub-steps S101 and S102, the generalized disturbed state space model of the PMSM is established as: In the formula, the state variable Control Input The nonlinear coupling vector is known: Measurable output Total disturbance vector where d d (t), d q (t), d ω (t) represents the unknown "total disturbance" in each subsystem, and its expressions are: Coefficient matrix: Where I3 represents the 3×3 order unit matrix; S104: T c To control the period, the forward Euler method is used to discretize the generalized disturbed state space model of the PMSM in step S103 to obtain a discrete controlled model: The step S2 comprises: S201: Determine the disturbance as well as The changes in the amount are bounded, that is, there is a positive constant μ i (i=d,q,ω) and μ ω2 So that: S202: Design the following discrete-time disturbance observer-1 online estimation and Where, subscript i = d,q; is an auxiliary variable; nonlinear coupling term for estimates; is the observer gain, which can be selected as: in, represents the adjustable bandwidth of the observer; S203: Definition Design the following discrete-time disturbance observer-2 online estimation Where, is an auxiliary variable; and They are and Estimates of ; coefficient matrix: is the observer gain matrix, which is selected according to the following rules: The construction of the reference system and state tracking error system models in step S3 includes: S301: Assume that the discrete-time disturbance observer in step S2 accurately estimates the total disturbance, that is, in, For disturbance estimated value of; S302: Based on the output regulation theory, the reference model of the discrete controlled PMSM system in sub-step S104 is introduced: Where, is the reference state vector; is the reference control input voltage; is the reference nonlinear coupling vector; according to the discrete controlled system model and MTPA principle: S303: Subtract the reference model in sub-step S302 from the discrete controlled model in step S104 to obtain the state tracking error system: The error variable is defined as follows:

2. A permanent magnet synchronous motor single-loop deadbeat predictive anti-interference control method according to claim 1, characterized in that: In step S4, the d-axis and q-axis reference control input voltages of the current control cycle are: Where, Represents the matrix T c Pseudo-inverse matrix of B; state vector in:

3. The method for single-loop deadbeat predictive anti-interference control of a permanent magnet synchronous motor according to claim 1, characterized in that: In step S5, the d-axis and q-axis error control input voltages of the current control cycle are designed as follows: According to the deadbeat predictive control theory, the state tracking error state of the system at the next moment in step S302 is: Among them, 03 represents a 3×3 order zero matrix; the error control input vector of the current control cycle is calculated as:

4. A permanent magnet synchronous motor single-loop deadbeat prediction anti-interference control method according to claim 1, characterized in that: In step S6, the control input voltage required for the motor drive in the current control period is:

Images