DPCC predictive control system and method for introducing sliding mode component into Luenberger observer

Abstract

The invention discloses a DPCC predictive control system and method for introducing a sliding mode component into a Luenberger observer. According to the system, a dead-beat current prediction controller generates predicted direct-axis voltage and quadrature-axis voltage according to given direct-axis current and quadrature-axis current and direct-axis current and quadrature-axis current fed back by a control system in combination with parameter disturbance; the three-state pulse width modulator generates six paths of pulse signals according to the predicted direct-axis voltage and quadrature-axis voltage; the inverter inverts the six paths of pulse signals to generate three-phase current; the control system controls the permanent magnet synchronous motor according to the three-phase current and obtains the actual output of the permanent magnet synchronous motor; and the Luenberger observer is used for generating parameter disturbance according to the actual output of the permanent magnet synchronous motor. According to the prediction control system, the tracking precision of the DPCC control system can be improved, the common-mode voltage can be reduced, the current deviation caused by parameter change can be inhibited, and the current deviation caused by motor parameter change can still be eliminated when the load is suddenly changed.

Description

Technical Field

[0001] The invention relates to the technical field of permanent magnet synchronous motor control, and in particular to a DPCC prediction control system and method in which a synovial component is introduced into a Romberg observer. Background Art

[0002] With the rapid development of the global economy, optimizing the current energy structure has become an urgent problem to be solved. Improving the performance of motor control algorithms is an important way to optimize the current energy structure. Permanent magnet synchronous motors (PMSMs) have been widely used due to their simple structure and high efficiency. With the development of electronic technology and control theory, more and more research has been conducted on improving the performance of motor control algorithms.

[0003] At present, PMSM is mainly vector controlled, in which PI controller is used for regulation. However, the working environment of the motor is complex and there are problems such as external disturbances. PI control can no longer meet the high performance requirements of the motor system. In order to improve the requirements of motor control performance, many advanced control strategies have been proposed, such as adaptive control, anti-disturbance control, sliding film control and predictive control. Among them, predictive control has the advantages of fast response speed and has become a current research hotspot. At present, the predictive control that is widely used in PMSM control systems is: continuous control set model predictive control, finite control set model predictive control and deadbeat current predictive control (DPCC). Among them, DPCC has the advantages of simple control and excellent performance, and can make the current error converge to zero in a short time, which has become a research hotspot for PMSM control strategies.

[0004] However, the existing DPCC-based control system has the following two main problems: 1) The key part of PMSM vector control is space vector pulse width modulation (SVPWM), but SVPWM uses zero voltage vector to synthesize reference voltage, which leads to a large common mode voltage in the DPCC control system. 2) Since the DPCC control system relies on motor parameters, it is prone to current tracking errors. Therefore, how to reduce the common mode voltage of the DPCC control system and suppress the current deviation caused by parameter changes is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In view of the deficiencies of the above-mentioned prior art, the technical problem to be solved by the present invention is: how to provide a DPCC predictive control method in which a synoviometric component is introduced into a Lumberg observer, by replacing the original SVPWM with a three-state pulse width modulator (TSPWM), so as to improve the tracking accuracy of the DPCC control system and reduce the common-mode voltage; at the same time, a Lumberg observer is introduced to observe the changes in motor parameters in real time to correct the predictive control and suppress the current deviation caused by the parameter changes; and a synoviometric component based on the constant velocity approach rate is introduced into the feedback loop of the Lumberg observer to improve the load-carrying capacity of the Lumberg observer, so that the current deviation caused by the change in motor parameters can still be eliminated when the load suddenly changes.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The DPCC predictive control system with the synovial component introduced into the Romberg observer includes:

[0008] A deadbeat current prediction controller, used for generating predicted direct-axis voltage and quadrature-axis voltage according to a given direct-axis current, quadrature-axis current and the direct-axis current and quadrature-axis current fed back by a control system in combination with parameter disturbance;

[0009] A three-state pulse width modulator is used for generating six pulse signals according to the predicted direct-axis voltage and quadrature-axis voltage;

[0010] An inverter is used to invert six pulse signals to generate three-phase current;

[0011] A control system is used to control the permanent magnet synchronous motor according to the three-phase current to obtain the actual output of the permanent magnet synchronous motor; and feed back the direct axis current and the quadrature axis current in the actual output to the deadbeat current prediction controller;

[0012] The Romberg observer is used to generate parameter disturbance according to the actual output of the permanent magnet synchronous motor; and the parameter disturbance is feed-forward compensated to the deadbeat current prediction controller.

[0013] Preferably, a synovial component is introduced into the feedback loop of the Romberg observer to obtain a synovial-Romberg observer; and a parameter disturbance is generated according to the actual output of the permanent magnet synchronous motor by the synovial-Romberg observer.

[0014] Preferably, it also includes:

[0015] A Park inverse converter is provided between the deadbeat current prediction controller and the three-state pulse width modulator, and is used to perform Park inverse conversion on the predicted direct-axis voltage and quadrature-axis voltage output by the deadbeat current prediction controller, and then input the predicted direct-axis voltage and quadrature-axis voltage to the three-state pulse width modulator;

[0016] The Clark converter and the Park converter are arranged between the inverter and the Romberg observer. The Clark converter is used to perform Clark transformation on the three-phase current and voltage output by the inverter and then input them to the Park converter. The Park converter is used to perform Park transformation on the output of the Clark converter and then input them to the Romberg observer.

[0017] A PI controller is arranged between the permanent magnet synchronous motor and the deadbeat current prediction controller, and is used to take the actual speed of the permanent magnet synchronous motor and the given speed as input to generate a given direct-axis current input to the deadbeat current prediction controller;

[0018] The sensor is arranged between the permanent magnet synchronous motor and the PI controller, and is used for collecting the actual output of the permanent magnet synchronous motor and inputting it into the Park converter, and at the same time inputting the actual speed in the actual output into the PI controller.

[0019] Preferably, the processing steps of the deadbeat current prediction controller are as follows:

[0020] S101: discretize the voltage equation of the permanent magnet synchronous motor by the forward Euler method to obtain a PMSM discrete current model;

[0021] S102: Calculate a given direct-axis voltage and a quadrature-axis voltage by combining a given direct-axis current and a direct-axis current fed back by the motor with a PMSM discrete current model;

[0022] The formula is:

[0023]

[0024] Where: u * d (k),u * q (k) respectively represent the given direct-axis voltage and quadrature-axis voltage; R s represents the stator resistance; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; ω e represents the electrical angular velocity; ψ f Represents the permanent magnet flux; L = L d =L q represents inductance;

[0025] S103: Calculating predicted direct-axis voltage and quadrature-axis voltage by combining given direct-axis voltage and quadrature-axis voltage with parameter disturbance;

[0026] The formula is:

[0027]

[0028] Where: u d ′、u q ′ represents the predicted direct-axis voltage and quadrature-axis voltage respectively; f d 、f q denote the parameter perturbations of the direct and quadrature axes respectively.

[0029] Preferably, in step S101, the voltage equation of the permanent magnet synchronous motor is expressed by the following formula:

[0030]

[0031] Where: u d 、u q Respectively represent the direct-axis voltage and quadrature-axis voltage of the stator; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; R s Indicates stator resistance; L d , L q They represent the direct-axis inductance and quadrature-axis inductance respectively; ψ f represents the permanent magnet flux; ω e represents the electrical angular velocity; ψ d , q They represent the direct-axis flux and the quadrature-axis flux respectively;

[0032] The PMSM discrete current model is expressed by the following formula:

[0033]

[0034] Where: i d 、i q They represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor respectively; T s represents the sampling time; k represents the current time.

[0035] Preferably, in step S103, in the initial state, the parameter disturbances of the direct axis and the quadrature axis are calculated by the following formula:

[0036]

[0037] Where: f d 、f q They represent the parameter perturbations of the direct axis and quadrature axis respectively; ΔR s represents the deviation of stator resistance; ΔL represents the deviation of inductance; Δψ f Indicates the deviation of magnetic flux; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; ω e represents the electrical angular velocity; ψ fRepresents the flux linkage of a permanent magnet.

[0038] Preferably, the model formula of the Lumberg observer is expressed as:

[0039]

[0040] Where: represents the estimated value of the state quantity of the Lumberg observer; represents the output estimate of the Lumberg observer; K represents the gain matrix, k1 and k2 are gain coefficients; R s represents the stator resistance; ω e represents the electrical angular velocity; u d 、u q They represent the direct-axis voltage and quadrature-axis voltage output by the permanent magnet synchronous motor respectively; ψ f represents the permanent magnet flux; L represents inductance; i d 、i q They represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor respectively;

[0041] The model formula of the Lumberg observer is discretized by the forward Euler formula to obtain the discretized model formula of the Lumberg observer:

[0042]

[0043] Where: represents the estimated value of the direct axis and quadrature axis parameter disturbance; T s represents the sampling time; k represents the current time.

[0044] Preferably, the model formula of the synovial membrane-Lomborg observer is expressed as:

[0045]

[0046] Where: N represents the synovial gain matrix, n1 and n2 are the synovial gain coefficients respectively; sgn represents the sign function in synovial control;

[0047] The model formula of the synovial membrane-Lomborg observer is discretized by the forward Euler formula to obtain the discretized model formula of the synovial membrane-Lomborg observer:

[0048]

[0049] Where: represents the estimated values ​​of the direct and quadrature axis parameter perturbations.

[0050] Preferably, the stability analysis of the synovial-Lomberg observer is performed using the Lyapunov function, and the specific steps are as follows:

[0051] S201: construct a full-order observer with parameter disturbance and direct-axis current and quadrature-axis current as state variables;

[0052] The model formula of the full-order observer is expressed as:

[0053]

[0054]

[0055] Where: x represents the state of the full-order observer; y represents the output of the full-order observer; u represents the input of the full-order observer; A is the state matrix; B is the input matrix; C is the output matrix; R s represents the stator resistance; ω e represents the electrical angular velocity; ψ f represents the permanent magnet flux; u d 、u q Respectively represent the direct-axis voltage and quadrature-axis voltage output by the permanent magnet synchronous motor; i d , i q The direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor respectively;

[0056] S202: Subtract the model formula of the full-order observer from the discretized model formula of the synovial-Lomborg observer to obtain Formula 1;

[0057] Formula 1 is expressed as:

[0058]

[0059]

[0060] Where: k1, k2 are gain coefficients;

[0061] S203: define a Lyapunov function; obtain Formula 2 by analyzing the Lyapunov function in combination with the discretization model formula of the synovial-Lomborg observer; and obtain Formula 3 by updating the Lyapunov function through Formula 2;

[0062] The Lyapunov function is defined as:

[0063]

[0064] Where: V represents the defined scalar function, i.e., the Lyapunov function; S represents the function of the current;

[0065] According to the Lyapunov stability criterion, when the Lyapunov function is greater than, the derivative should be less than 0;

[0066] Formula 2 is expressed as:

[0067]

[0068] Formula 3 is expressed as:

[0069]

[0070] S204: According to Formula 3, it can be known that the Lyapunov function V is always greater than 0; the derivative of Formula 3 is analyzed, and Formula 4 is obtained by taking the derivative of Formula 3;

[0071] Formula 4 is expressed as:

[0072]

[0073] S205: converting formula 1 into formula 5;

[0074] Formula 5 is expressed as:

[0075]

[0076] S206: Obtain the derivative of the Lyapunov function according to Formula 4 and Formula 5 to obtain Formula 6; simplify Formula 6 into Formula 7;

[0077] Formula 6 is expressed as:

[0078]

[0079] Formula 7 is expressed as:

[0080]

[0081] S207: Derivation of Formula 8 satisfied by the synovial gain coefficient n1 according to Formula 7;

[0082] Formula eight is expressed as:

[0083]

[0084] S208: According to Formula 8, it can be ensured that the derivative of the obtained Lyapunov function is less than 0, that is, the Synovial-Lumberg observer is stable.

[0085] The present invention also discloses a DPCC predictive control method in which a synovial component is introduced into a Romberg observer. The DPCC predictive control system according to the present invention is implemented, and includes:

[0086] S1: Generate predicted direct-axis voltage and quadrature-axis voltage according to the given direct-axis current, quadrature-axis current and the direct-axis current and quadrature-axis current fed back by the control system combined with parameter disturbance;

[0087] S2: Generate six pulse signals according to the predicted direct-axis voltage and quadrature-axis voltage; invert the six pulse signals to generate three-phase current;

[0088] S3: Control the permanent magnet synchronous motor according to the three-phase current to obtain the actual output of the permanent magnet synchronous motor;

[0089] S4: Generate parameter disturbance according to the actual output of the permanent magnet synchronous motor;

[0090] S5: Feedback the direct-axis current and quadrature-axis current in the actual output in step S3 and the parameter disturbance generated in step S4 to step S1 to assist in generating the predicted direct-axis voltage and quadrature-axis voltage.

[0091] Compared with the prior art, the DPCC predictive control method of introducing the synovial component into the Romberg observer in the present invention has the following advantages:

[0092] Beneficial effects:

[0093] The present invention generates predicted direct-axis voltage and quadrature-axis voltage according to given direct-axis current, quadrature-axis current and direct-axis current and quadrature-axis current fed back by control system combined with parameter disturbance by deadbeat current prediction controller (DPCC). Firstly, DPCC can accurately control the direct-axis and quadrature-axis current of motor by real-time prediction of direct-axis and quadrature-axis voltage at future moment, so that it can quickly track given value, significantly improve the precision and dynamic response speed of current control, and reduce current fluctuation and steady-state error. Secondly, DPCC is an accurate mathematical model based on permanent magnet synchronous motor, which can effectively suppress the influence of external interference and internal parameter change on system performance through predictive control strategy, and enhance the robustness and stability of permanent magnet synchronous motor control system. Finally, DPCC can optimize the torque output and efficiency of motor and reduce unnecessary energy loss by accurately controlling current. At the same time, DPCC can reduce the calculation delay caused by waiting for feedback signal in traditional control method by predicting the control amount at next moment and implementing it at current moment, and improve the real-time and control efficiency of predictive control system.

[0094] The present invention introduces a three-state pulse width modulator (TSPWM) to generate six pulse signals based on the generation of predicted direct-axis voltage and quadrature-axis voltage by a DPCC predictive control system. First, the current loop of the present invention adopts DPCC instead of the traditional PI control, and the modulation strategy adopts TSPWM instead of SVPWM, so that the tracking accuracy of the control system can be improved and the common-mode voltage can be reduced. Compared with SVPWM, TSPWM uses two non-zero vectors and one zero vector to synthesize the voltage reference vector in the low modulation area, and three non-zero vectors to synthesize the voltage reference vector in the high modulation area, which can more effectively distribute the voltage vector, thereby improving the voltage utilization rate without increasing the switching frequency of the inverter, so that the permanent magnet synchronous motor can output a larger torque at the same voltage level. Secondly, TSPWM can reduce the harmonic content in the output voltage and current, improve the electromagnetic performance and operation stability of the permanent magnet synchronous motor, and reduce mechanical vibration and noise by optimizing the distribution of the voltage vector. Finally, TSPWM provides more control freedom, allowing the synthesis mode and timing of the voltage vector to be flexibly adjusted according to the specific application scenario and performance requirements to adapt to different control strategies and performance optimization goals.

[0095] The present invention aims at the problem that the DPCC predictive control system is prone to current deviation (i.e., current tracking error) due to parameter mismatch, and designs a Lumberg observer to observe the changes of motor parameters in real time to correct the prediction model and suppress the current deviation caused by parameter changes. At the same time, in order to further improve the load-carrying capacity of the Lumberg observer, so that it can still eliminate the current deviation caused by the change of motor parameters when the load changes suddenly, a synovial component based on the constant velocity approach rate is introduced into the Lumberg observer feedback loop, and the stability analysis is performed using the Lyapunov function. First, the synovial component is introduced on the basis of the traditional Lumberg observer, which can more effectively estimate and compensate for the parameter disturbance during the operation of the motor, and improve the accuracy and robustness of the Lumberg observer. Secondly, the synovial control has the characteristics of being insensitive to parameter changes and external disturbances. Combined with the state observation function of the Lumberg observer, it can significantly enhance the system's adaptability to unknown disturbances and uncertainties, so that the DPCC predictive control system can still maintain high-performance operation under complex working conditions. Finally, the synovial-Lumberg observer estimates and compensates for parameter disturbances in real time, further optimizes the control strategy, better adapts to the changes in motor operating conditions, and thus improves the overall performance and efficiency of the DPCC predictive control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] In order to make the purpose, technical solution and advantages of the invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which:

[0097] Figure 1 The overall logic block diagram of the DPCC predictive control system with the introduction of the Romberg observer for the synovial component;

[0098] Figure 2 It is the sector division of TSPWM;

[0099] Figure 3 This is the structural principle diagram of the synovial membrane-Romberg observer;

[0100] Figure 4 The common mode voltage when the reference speed is 2500r / min: (a) is SVPWM; (b) is TSPWM;

[0101] Figure 5 The common mode voltage when the reference speed is 3000r / min: (a) is SVPWM; (b) is TSPWM;

[0102] Figure 6 When a load of 10 N·m is applied at 0.4 s, the resistance is 0.3 times the rated resistance: (a) is the traditional DPCC; (b) is the DPCC improved by the Romberg observer;

[0103] Figure 7 When a load of 10 N·m is applied at 0.4 s, the inductance is 0.3 times the rated inductance: (a) is the traditional DPCC; (b) is the DPCC improved by the Romberg observer;

[0104] Figure 8 When a load of 10 N·m is applied at 0.4 s, the flux is twice the rated flux: (a) is the traditional DPCC; (b) is the DPCC improved by the Romberg observer;

[0105] Fig. 9 When the load suddenly increases to 25 N·m at 0.4 s, the resistance is 0.3 times the rated resistance: (a) DPCC improved by Lumberg observer; (b) DPCC improved by synovial-Lumberg observer;

[0106] Fig.10 When the load suddenly increases to 25 N·m at 0.4 s, the inductance is 0.3 times the rated inductance: (a) DPCC improved by Lumberg observer; (b) DPCC improved by sliding film-Lumberg observer;

[0107] Fig.11 When the load suddenly increases to 25 N·m at 0.4 s, the flux is twice the rated flux: (a) DPCC improved by Romberg observer; (b) DPCC improved by sliding film-Romberg observer. DETAILED DESCRIPTION

[0108] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but only represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0109] The following is a further detailed description through specific implementation methods:

[0110] Embodiment 1:

[0111] This embodiment discloses a DPCC predictive control system in which a synovial component is introduced into a Romberg observer.

[0112] like Figure 1 As shown in FIG. 1 , the DPCC predictive control system in which the synovial component is introduced into the Lumberg observer includes:

[0113] Permanent magnet synchronous motor (PMSM);

[0114] A deadbeat current prediction controller (DPCC) for generating predicted direct-axis voltage and quadrature-axis voltage according to a given direct-axis current, quadrature-axis current and the direct-axis current and quadrature-axis current fed back by a control system combined with parameter disturbance;

[0115] A three-state pulse width modulator (TSPWM) for generating six pulse signals according to the predicted direct-axis voltage and quadrature-axis voltage;

[0116] An inverter is used to invert six pulse signals to generate three-phase current;

[0117] In this embodiment, the inverter is arranged between the three-state pulse width modulator and the permanent magnet synchronous motor.

[0118] A control system is used to control the permanent magnet synchronous motor according to the three-phase current to obtain the actual output of the permanent magnet synchronous motor; feed back the direct axis current and the quadrature axis current in the actual output to the deadbeat current prediction controller to assist it in generating the predicted direct axis voltage and quadrature axis voltage;

[0119] The Romberg observer is used to generate parameter disturbance according to the actual output of the permanent magnet synchronous motor; the parameter disturbance is feed-forward compensated to assist the deadbeat current prediction controller to generate predicted direct-axis voltage and quadrature-axis voltage.

[0120] Also includes:

[0121] A Park inverse converter is provided between the deadbeat current prediction controller and the three-state pulse width modulator, and is used to perform Park inverse conversion on the predicted direct-axis voltage and quadrature-axis voltage output by the deadbeat current prediction controller, and then input the predicted direct-axis voltage and quadrature-axis voltage to the three-state pulse width modulator;

[0122] The Clark converter and the Park converter are arranged between the inverter and the Romberg observer. The Clark converter is used to perform Clark transformation on the three-phase current and voltage output by the inverter and then input them to the Park converter. The Park converter is used to perform Park transformation on the output of the Clark converter and then input them to the Romberg observer.

[0123] A PI controller is arranged between the permanent magnet synchronous motor and the deadbeat current prediction controller, and is used to take the actual speed of the permanent magnet synchronous motor and the given speed as input to generate a given direct-axis current input to the deadbeat current prediction controller;

[0124] The sensor is arranged between the permanent magnet synchronous motor and the PI controller, and is used for collecting the actual output of the permanent magnet synchronous motor and inputting it into the Park converter, and at the same time inputting the actual speed in the actual output into the PI controller.

[0125] The present invention generates predicted direct-axis voltage and quadrature-axis voltage according to given direct-axis current, quadrature-axis current and direct-axis current and quadrature-axis current fed back by control system combined with parameter disturbance by deadbeat current prediction controller (DPCC). Firstly, DPCC can accurately control the direct-axis and quadrature-axis current of motor by real-time prediction of direct-axis and quadrature-axis voltage at future moment, so that it can quickly track given value, significantly improve the precision and dynamic response speed of current control, and reduce current fluctuation and steady-state error. Secondly, DPCC is an accurate mathematical model based on permanent magnet synchronous motor, which can effectively suppress the influence of external interference and internal parameter change on system performance through predictive control strategy, and enhance the robustness and stability of permanent magnet synchronous motor control system. Finally, DPCC can optimize the torque output and efficiency of motor and reduce unnecessary energy loss by accurately controlling current. At the same time, DPCC can reduce the calculation delay caused by waiting for feedback signal in traditional control method by predicting the control amount at next moment and implementing it at current moment, and improve the real-time and control efficiency of predictive control system.

[0126] The present invention introduces a three-state pulse width modulator (TSPWM) to generate six pulse signals based on the generation of predicted direct-axis voltage and quadrature-axis voltage by a DPCC predictive control system. First, the current loop of the present invention adopts DPCC instead of the traditional PI control, and the modulation strategy adopts TSPWM instead of SVPWM, so that the tracking accuracy of the control system can be improved and the common-mode voltage can be reduced. Compared with SVPWM, TSPWM uses two non-zero vectors and one zero vector to synthesize the voltage reference vector in the low modulation area, and three non-zero vectors to synthesize the voltage reference vector in the high modulation area, which can more effectively distribute the voltage vector, thereby improving the voltage utilization rate without increasing the switching frequency of the inverter, so that the permanent magnet synchronous motor can output a larger torque at the same voltage level. Secondly, TSPWM can reduce the harmonic content in the output voltage and current, improve the electromagnetic performance and operation stability of the permanent magnet synchronous motor, and reduce mechanical vibration and noise by optimizing the distribution of the voltage vector. Finally, TSPWM provides more control freedom, allowing the synthesis mode and timing of the voltage vector to be flexibly adjusted according to the specific application scenario and performance requirements to adapt to different control strategies and performance optimization goals.

[0127] The present invention aims at the problem that the DPCC predictive control system is prone to current deviation (i.e., current tracking error) due to parameter mismatch, and designs a Lumberg observer to observe the changes of motor parameters in real time to correct the prediction model and suppress the current deviation caused by parameter changes. At the same time, in order to further improve the load-carrying capacity of the Lumberg observer, so that it can still eliminate the current deviation caused by the change of motor parameters when the load changes suddenly, a synovial component based on the constant velocity approach rate is introduced into the Lumberg observer feedback loop, and the stability analysis is performed using the Lyapunov function. First, the synovial component is introduced on the basis of the traditional Lumberg observer, which can more effectively estimate and compensate for the parameter disturbance during the operation of the motor, and improve the accuracy and robustness of the Lumberg observer. Secondly, the synovial control has the characteristics of being insensitive to parameter changes and external disturbances. Combined with the state observation function of the Lumberg observer, it can significantly enhance the system's adaptability to unknown disturbances and uncertainties, so that the DPCC predictive control system can still maintain high-performance operation under complex working conditions. Finally, the synovial-Lumberg observer estimates and compensates for parameter disturbances in real time, further optimizes the control strategy, better adapts to the changes in motor operating conditions, and thus improves the overall performance and efficiency of the DPCC predictive control system.

[0128] In order to better introduce the technical solution of the present invention, this embodiment is described through the following parts.

[0129] 1. PMSM Model

[0130] In order to simplify the design, in this embodiment, when building the PMSM model of the permanent magnet synchronous motor, eddy current and hysteresis losses are ignored, and the damping effect of the permanent magnet and the winding is ignored. The magnetic field is sinusoidally distributed, and the voltage equation of the PMSM in the two-phase rotating coordinate is as follows:

[0131]

[0132] Where: u d 、u q They represent the direct-axis voltage and quadrature-axis voltage of the permanent magnet synchronous motor (stator in synchronous rotating coordinate system) respectively; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; R s Indicates stator resistance; L d , L q They represent the direct-axis inductance and quadrature-axis inductance respectively; ψ f represents the permanent magnet flux; ω e represents the electrical angular velocity; ψ d , q They represent the direct-axis flux and the quadrature-axis flux respectively;

[0133] The electromagnetic torque equation of the permanent magnet synchronous motor is expressed by the following formula:

[0134] T e =1.5P n [ψ f i q +(L d -L q )i d i q ];

[0135] Where: T e Represents electromagnetic torque; P n Represents the number of pole pairs.

[0136] 2. Deadbeat Predictive Current Control (DPCC)

[0137] In this embodiment, the processing steps of the deadbeat current prediction controller are as follows:

[0138] S101: discretize the voltage equation of the permanent magnet synchronous motor by the forward Euler method to obtain a PMSM discrete current model;

[0139] The PMSM discrete current model is expressed by the following formula:

[0140]

[0141] Where: i d 、i qThey represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor respectively; T s Indicates the sampling time.

[0142] S102: Calculate a given direct-axis voltage and a quadrature-axis voltage by combining a given direct-axis current and a direct-axis current fed back by the motor with a PMSM discrete current model;

[0143] The control goal of DPCC is to make the current value i(k+1) at the next moment accurately track the given value i * (k), we can get i d (k+1)=i * d (k), i q (k+1)=i * q (k), where i * d (k), i * q (k) are the given direct-axis and quadrature-axis currents of the motor, so the given direct-axis and quadrature-axis voltages u can be obtained. * d (k),u * q The calculation formula of (k) is:

[0144]

[0145]

[0146] The above formula is discretized by Euler method to obtain the following formula:

[0147]

[0148] Where: u * d (k),u * q (k) respectively represent the given direct-axis voltage and quadrature-axis voltage; i * d (k), i * q (k) respectively represent the given direct-axis current and quadrature-axis current; R s represents the stator resistance; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; ω e represents the electrical angular velocity; ψ f represents the permanent magnet flux; T s Indicates sampling time; L = L d =L q represents inductance;

[0149] S103: Calculating predicted direct-axis voltage and quadrature-axis voltage by combining given direct-axis voltage and quadrature-axis voltage with parameter disturbance;

[0150] Current predictive control is a model-based control that includes three parameters: stator resistance, inductance, and flux linkage. As the motor parameters change under the external conditions of motor operation, its control performance deteriorates.

[0151] The formula is:

[0152]

[0153] Where: u d ′、u q ′ represents the predicted direct-axis voltage and quadrature-axis voltage respectively; f d 、f q denote the parameter perturbations of the direct and quadrature axes respectively.

[0154] In the initial state, the parameter disturbances of the direct axis and the quadrature axis are calculated by the following formula:

[0155]

[0156] Where: f d 、f q They represent the parameter perturbations of the direct axis and quadrature axis respectively; ΔR s represents the deviation of stator resistance; ΔL represents the deviation of inductance; Δψ f Indicates the deviation of magnetic flux; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; ω e represents the electrical angular velocity; ψ f Represents the flux linkage of a permanent magnet.

[0157] 3. Three-state pulse width modulator (TSPWM)

[0158] The key part of PMSM vector control is space vector pulse width modulation (SVPWM). SVPWM controls the output voltage amplitude by turning on and off the IGBT, making it output a sine wave as much as possible. However, the traditional SVPWM uses a zero voltage vector to synthesize the reference voltage, resulting in a large output common-mode voltage. This embodiment uses TSPWM to reduce the common-mode voltage. The difference from SVPWM is that TSPWM uses two non-zero vectors and one zero vector to synthesize in the low modulation area, and three non-zero vectors to synthesize the voltage reference vector in the high modulation area. Figure 2 This is the sector division of TSPWM. The area within the hexagonal dotted line is the low modulation area, and the area outside the dotted line and within the solid line is the high modulation area.

[0159] Taking sector I as an example, when the reference voltage v ref When it is in the low modulation area, v4, v5, and v6 cannot be used for synthesis, so a non-zero vector v4 is discarded and a zero vector v7 is added. According to the synthesis principle, we can get:

[0160]

[0161] Where: k r , k0, k l are the right, zero, and left voltage vector coefficients respectively.

[0162] When the reference voltage v ref When it is in the high modulation area, it can be directly synthesized using non-zero vectors v4, v5, and v6. According to the synthesis principle, it can be obtained:

[0163]

[0164] Where: k r , k c , k l are the right, center and left voltage vector coefficients respectively.

[0165] 4. Lumberg Observer

[0166] When the motor model deviates due to external interference, the u obtained by DPCC * d (k),u * q (k) There is a deviation from the real voltage vector of the system, and this deviation is f d 、f q Therefore, the Lumberg observer is designed to compensate the voltage vector in real time.

[0167] With system disturbance and d q The full-order observer is constructed with the shaft current as the state variable:

[0168]

[0169] According to the Lumberg observer theory, a Lumberg observer is constructed to obtain the state variables that are difficult to obtain by reconstructing the state quantity. In this embodiment, it refers to the system deviation f d 、f q So the Lumberg observer is as follows:

[0170]

[0171] Where: represents the estimated value of the state quantity of the Lumberg observer; represents the output estimate of the Lumberg observer; K represents the gain matrix, k1 and k2 are gain coefficients; Rs represents the stator resistance; ω e represents the electrical angular velocity; u d 、u q They represent the direct-axis voltage and quadrature-axis voltage output by the permanent magnet synchronous motor respectively; ψ f represents the permanent magnet flux; L represents inductance;

[0172] The model formula of the Lumberg observer is discretized by the forward Euler formula to obtain the discretized model formula of the Lumberg observer:

[0173]

[0174] Where: represents the estimated value of the direct axis and quadrature axis parameter disturbance; T s Indicates the sampling time.

[0175] When designing an observer, it is necessary to ensure that the designed observer is observable.

[0176]

[0177] If the observability matrix P is full rank, that is, rankP=4, that is, det(P)≠0, it is easy to obtain the rank of the observability matrix P from the A and C matrices to be 4, so the designed observer is observable.

[0178] In order to test the stability of the observer established above, equation (2) is subtracted from equation (3) to obtain the error equation as shown in equation (5). As long as the stability of the error equation is guaranteed, the stability of the observer established can be obtained.

[0179]

[0180] Where:

[0181] In order to ensure the stability of the error equation, the feedback gain matrix K must be reasonably designed so that the coefficient matrix (A n -KC) are all negative. So the characteristic equation of formula (5) is:

[0182] det[λE-(A n -KC)]=0 (7);

[0183] Assuming that the characteristic roots of the observer are p0, p1, and p2, according to the design criteria of the observer, the absolute value of the negative real part of the characteristic root cannot be too large. After sorting and calculating, we get p1=k1-2,p2=1. According to the stability criterion, we can get:

[0184]

[0185] According to the above theory, the gain coefficients of the feedback matrix are reasonably configured. After multiple simulations, the coefficients of the feedback matrix are k1=1.5 and k2=-0.5.

[0186] 5. Synovial-Romberg Observer

[0187] Introducing the synovial component into the feedback loop of the Romberg observer, we can obtain Figure 3 The synovial-Lumberg observer shown in FIG. is used to generate parameter disturbances according to the actual output of the permanent magnet synchronous motor. In order to further improve the robustness of the Lumberg observer and reduce the observation error when the load increases, the synovial component is introduced into the feedback loop of the Lumberg observer and the Lumberg observer is redesigned.

[0188] The sliding film observer is not easily affected by parameter disturbances, and has good dynamic response performance and strong robustness. However, if the traditional sliding mode control is used to redesign the observer, high-frequency jitter will be introduced into the control system, affecting the performance of the control system. Therefore, this embodiment introduces the sliding film component into the feedback loop of the Lumberg observer, which will not only not bring high-frequency jitter to the control system, but also further improve the robustness of the Lumberg observer when the load increases, and reduce the influence of parameter disturbance on the observer observation error.

[0189] The synovial film control is mainly composed of two motion stages, one is the approach motion outside the sliding surface, and the other is the motion near the sliding surface, so the motion mode of the synovial film variable structure is composed of the approach mode and the sliding mode. The general synovial film control only considers whether it can approach the sliding surface, but does not reflect how to approach the sliding surface, so it is particularly important to design the approach rate of the synovial film control, which can not only speed up the convergence speed of the synovial film controller, but also ensure the dynamic quality of the approach motion.

[0190] This paper adopts the constant approach rate function, which is expressed as:

[0191]

[0192] Based on formula (1), the output matrix y is the direct axis current i d and the quadrature axis current i q , the model formula of the synovial-Lomborg observer is expressed as:

[0193]

[0194] Where: N represents the synovial gain matrix, n1 and n2 are the synovial gain coefficients respectively;

[0195] The model formula of the synovial membrane-Lomborg observer is discretized by the forward Euler formula to obtain the discretized model formula of the synovial membrane-Lomborg observer:

[0196]

[0197] 6. Stability Analysis

[0198] The synovial approach rate selected in this embodiment is a constant velocity approach rate. Since the observer introduces a synovial component, which affects the stability of the control system, it is necessary to perform stability analysis on the observer that introduces the synovial component.

[0199] In this embodiment, the stability analysis of the synovial membrane-Lomberg observer is performed using the Lyapunov function, and the specific steps are as follows:

[0200] S201: construct a full-order observer with parameter disturbance and direct-axis current and quadrature-axis current as state variables;

[0201] The model formula of the full-order observer is expressed as:

[0202]

[0203] Where: x represents the state of the full-order observer; y represents the output of the full-order observer; u represents the input of the full-order observer; A is the state matrix; B is the input matrix; C is the output matrix; R s represents the stator resistance; ω e represents the electrical angular velocity; ψ f represents the permanent magnet flux; u d 、u q They represent the direct-axis voltage and quadrature-axis voltage of the stator in the synchronous rotating coordinate system respectively; i d 、i q They represent the direct-axis current and quadrature-axis current of the stator in the synchronous rotating coordinate system respectively;

[0204] S202: Subtract the model formula of the full-order observer from the discretized model formula of the synovial-Lomborg observer to obtain Formula 1;

[0205] Formula 1 is expressed as:

[0206]

[0207] Where: k1, k2 are gain coefficients;

[0208] S203: define a Lyapunov function; obtain Formula 2 by analyzing the Lyapunov function in combination with the discretization model formula of the synovial-Lomborg observer; and obtain Formula 3 by updating the Lyapunov function through Formula 2;

[0209] The Lyapunov function is defined as:

[0210]

[0211] Where: V represents the defined scalar function, namely the Lyapunov function; S represents the function of the current; this formula is equivalent to the theorem.

[0212] According to the Lyapunov stability criterion, when the Lyapunov function is greater than, the derivative should be less than 0;

[0213] Formula 2 is expressed as:

[0214]

[0215] Formula 3 is expressed as:

[0216]

[0217] S204: According to Formula 3, it can be known that the Lyapunov function V is always greater than 0; the derivative of Formula 3 is analyzed, and Formula 4 is obtained by taking the derivative of Formula 3;

[0218] Formula 4 is expressed as:

[0219]

[0220] S205: converting formula 1 into formula 5;

[0221] Formula 5 is expressed as:

[0222]

[0223] S206: Obtain the derivative of the Lyapunov function according to Formula 4 and Formula 5 to obtain Formula 6; simplify Formula 6 into Formula 7;

[0224] Formula 6 is expressed as:

[0225]

[0226] Formula 7 is expressed as:

[0227]

[0228] S207: Derivation of Formula 8 satisfied by the synovial gain coefficient n1 according to Formula 7;

[0229] Formula eight is expressed as:

[0230]

[0231] S208: According to Formula 8, it can be ensured that the derivative of the obtained Lyapunov function is less than 0, that is, the Synovial-Lumberg observer is stable.

[0232] VII. Experimental Description

[0233] In order to verify the effectiveness of the algorithm proposed in the present invention, this embodiment uses MATLAB / Simulink to build SVPWM, TSPWM modulation strategies and the Romberg observer and the Romberg observer DPCC control system with the synovial component for verification. The motor parameters in the simulation are shown in Table 1.

[0234] Table 1 Motor parameters

[0235]

[0236]

[0237] 1. Common mode voltage simulation

[0238] In order to verify the suppression effect of TSPWM modulation on common-mode voltage compared with SVPWM applied to DPCC motor control system, the reference speeds are given as 2500r / min and 3000r / min respectively, no-load start, sampling time is 5e-6s, and simulation time is 0.6s. Figure 4 and Figure 5 SVPWM and TSPWM are used for the common mode voltage of DPCC control system respectively. Figure 4 When the reference speed is 2500r / min, the SVPWM common mode voltage is U dc Although the TSPWM common-mode voltage has spikes, the density of the common-mode voltage begins to decrease significantly, and the common-mode voltage begins to gather near the neutral point. The minimum peak-to-peak value of the common-mode voltage varies between +58.3V and -58.3V. At this time, the TSPWM modulation strategy is in a transition zone.

[0239] Figure 5 When the reference speed is 3000r / min, the SVPWM common mode voltage is still U dc , the common-mode voltage peak-to-peak value varies between +175V and -175V, while the TSPWM common-mode voltage has no spikes in the area and the common-mode voltage peak-to-peak values ​​all vary between +58.3V and -58.3V. At this time, the TSPWM modulation strategy is in the high modulation area.

[0240] It can be seen that no matter the voltage vector is in the high modulation area or the low modulation area, the SVPWM modulation strategy will produce a peak-to-peak value of U dcThe TSPWM modulation strategy has better common-mode voltage suppression effect than SVPWM in both high and low modulation areas, especially in the high modulation area, the common-mode voltage is reduced by about 0.67U compared with the SVPWM common-mode voltage. dc .

[0241] 2. Current simulation

[0242] When the stator resistance, flux linkage and inductance of the DPCC prediction model deviate, the DPCC improved by the Romberg observer and the traditional DPCC under a load of 10 N·m are respectively changed from i d and i q The current tracking performance is simulated and compared. The motor is started without load, the given speed is 1000r / min, a load of 10N·m is added at 0.4s, and the simulation time is 1s.

[0243] When the resistance is deviated, a comparative simulation is performed, and the resistance is set to 0.3 times the rated resistance, and the flux linkage and inductance have no error. The results are as follows Figure 6 .

[0244] When a load of 10 N·m is applied at 0.4 s, Figure 6 Due to the mismatch of resistance parameters, the traditional DPCC has a tracking error in the cross-axis current, while the direct-axis current has no obvious tracking error, which shows that the resistance mismatch has little effect on the direct-axis current. The DPCC improved based on Romberg significantly improves the tracking ability of the cross-axis current and basically eliminates the current static error. Figure 7 It can be seen that the inductance mismatch affects both the direct-axis and quadrature-axis currents and has the greatest impact on the direct-axis current. It not only deviates significantly from the given value of the direct-axis current, but also has large fluctuations in the direct-axis current. The improved DPCC can not only significantly improve the problem of inaccurate current tracking, but also reduce the fluctuations in the direct-axis current. Figure 8 When the flux linkage is mismatched, the cross-axis current will have obvious errors, while the improved DPCC completely eliminates the cross-axis current static error phenomenon.

[0245] When the load suddenly increases to 25 N·m at 0.4s, the resistance, inductance and flux linkage deviate respectively, and the Lumberg observer has obvious tracking error, so the improved Lumberg observer of the present invention is simulated and compared. Fig. 9 , Fig.10 and Fig.11 These are the simulation results when resistance, inductance and flux deviation occur respectively.

[0246] Depend on Fig. 9 , Fig.10 and Fig.11It can be clearly seen that when the load increases further, the quadrature-axis current based on the Lumberg DPCC predictive control system has errors to varying degrees, the direct-axis current has no obvious deviation, the quadrature-axis feedback current cannot accurately track the given current, and the robustness and anti-disturbance capabilities are significantly reduced. The DPCC predictive control system based on the improved Lumberg observer significantly improves the anti-disturbance capability when the control system parameters change, and improves the robustness of the system.

[0247] Embodiment 2:

[0248] This embodiment discloses a DPCC predictive control method in which a synovial component is introduced into a Romberg observer, which is implemented based on the deadbeat current predictive control system described in the first embodiment.

[0249] The DPCC predictive control method with the synovial component introduced into the Lumberg observer includes:

[0250] S1: Generate predicted direct-axis voltage and quadrature-axis voltage according to given direct-axis current, quadrature-axis current and direct-axis current, quadrature-axis current fed back by control system combined with parameter disturbance through deadbeat current prediction controller (DPCC);

[0251] S2: Generate six pulse signals according to the predicted direct-axis voltage and quadrature-axis voltage through a three-state pulse width modulator; invert the six pulse signals through an inverter to generate three-phase current;

[0252] S3: Control the permanent magnet synchronous motor according to the three-phase current through the control system to obtain the actual output of the permanent magnet synchronous motor;

[0253] S4: Generate parameter disturbance according to the actual output of the permanent magnet synchronous motor through the Romberg observer;

[0254] S5: Feedback the direct-axis current and quadrature-axis current in the actual output in step S3 and the parameter disturbance generated in step S4 to step S1 to assist in generating the predicted direct-axis voltage and quadrature-axis voltage.

[0255] In step S4, a synovial component is introduced into the feedback loop of the Romberg observer to obtain a synovial-Romberg observer; and a parameter disturbance is generated according to the actual output of the permanent magnet synchronous motor through the synovial-Romberg observer.

[0256] The present invention generates predicted direct-axis voltage and quadrature-axis voltage according to given direct-axis current, quadrature-axis current and direct-axis current and quadrature-axis current fed back by control system combined with parameter disturbance by deadbeat current prediction controller (DPCC). Firstly, DPCC can accurately control the direct-axis and quadrature-axis current of motor by real-time prediction of direct-axis and quadrature-axis voltage at future moment, so that it can quickly track given value, significantly improve the precision and dynamic response speed of current control, and reduce current fluctuation and steady-state error. Secondly, DPCC is an accurate mathematical model based on permanent magnet synchronous motor, which can effectively suppress the influence of external interference and internal parameter change on system performance through predictive control strategy, and enhance the robustness and stability of permanent magnet synchronous motor control system. Finally, DPCC can optimize the torque output and efficiency of motor and reduce unnecessary energy loss by accurately controlling current. At the same time, DPCC can reduce the calculation delay caused by waiting for feedback signal in traditional control method by predicting the control amount at next moment and implementing it at current moment, and improve the real-time and control efficiency of predictive control system.

[0257] The present invention introduces a three-state pulse width modulator (TSPWM) to generate six pulse signals based on the generation of predicted direct-axis voltage and quadrature-axis voltage by a DPCC predictive control system. First, the current loop of the present invention adopts DPCC instead of the traditional PI control, and the modulation strategy adopts TSPWM instead of SVPWM, so that the tracking accuracy of the control system can be improved and the common-mode voltage can be reduced. Compared with SVPWM, TSPWM uses two non-zero vectors and one zero vector to synthesize the voltage reference vector in the low modulation area, and three non-zero vectors to synthesize the voltage reference vector in the high modulation area, which can more effectively distribute the voltage vector, thereby improving the voltage utilization rate without increasing the switching frequency of the inverter, so that the permanent magnet synchronous motor can output a larger torque at the same voltage level. Secondly, TSPWM can reduce the harmonic content in the output voltage and current, improve the electromagnetic performance and operation stability of the permanent magnet synchronous motor, and reduce mechanical vibration and noise by optimizing the distribution of the voltage vector. Finally, TSPWM provides more control freedom, allowing the synthesis mode and timing of the voltage vector to be flexibly adjusted according to the specific application scenario and performance requirements to adapt to different control strategies and performance optimization goals.

[0258] The present invention aims at the problem that the DPCC predictive control system is prone to current deviation (i.e., current tracking error) due to parameter mismatch, and designs a Lumberg observer to observe the changes of motor parameters in real time to correct the prediction model and suppress the current deviation caused by parameter changes. At the same time, in order to further improve the load-carrying capacity of the Lumberg observer, so that it can still eliminate the current deviation caused by the change of motor parameters when the load changes suddenly, a synovial component based on the constant velocity approach rate is introduced into the Lumberg observer feedback loop, and the stability analysis is performed using the Lyapunov function. First, the synovial component is introduced on the basis of the traditional Lumberg observer, which can more effectively estimate and compensate for the parameter disturbance during the operation of the motor, and improve the accuracy and robustness of the Lumberg observer. Secondly, the synovial control has the characteristics of being insensitive to parameter changes and external disturbances. Combined with the state observation function of the Lumberg observer, it can significantly enhance the system's adaptability to unknown disturbances and uncertainties, so that the DPCC predictive control system can still maintain high-performance operation under complex working conditions. Finally, the synovial-Lumberg observer estimates and compensates for parameter disturbances in real time, further optimizes the control strategy, better adapts to the changes in motor operating conditions, and thus improves the overall performance and efficiency of the DPCC predictive control system.

[0259] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit the technical solution. Those skilled in the art should understand that those modifications or equivalent substitutions of the technical solution of the present invention that do not depart from the purpose and scope of the technical solution should be included in the scope of the claims of the present invention.

Claims

1. The DPCC predictive control system with the synovial component introduced into the Lumberg observer is characterized by: include: A deadbeat current prediction controller, used for generating predicted direct-axis voltage and quadrature-axis voltage according to a given direct-axis current, quadrature-axis current and the direct-axis current and quadrature-axis current fed back by a control system in combination with parameter disturbance; A three-state pulse width modulator is used for generating six pulse signals according to the predicted direct-axis voltage and quadrature-axis voltage; An inverter is used to invert six pulse signals to generate three-phase current; A control system is used to control the permanent magnet synchronous motor according to the three-phase current to obtain the actual output of the permanent magnet synchronous motor; and feed back the direct axis current and the quadrature axis current in the actual output to the deadbeat current prediction controller; Romberg observer, used to generate parameter disturbances based on the actual output of the permanent magnet synchronous motor; The parameter disturbance is feed-forward compensated to the deadbeat current predictive controller.

2. The DPCC predictive control system with the synovial component introduced into the Romberg observer as claimed in claim 1, characterized in that: The synovial component is introduced into the feedback loop of the Romberg observer to obtain the synovial-Romberg observer; the parameter disturbance is generated according to the actual output of the permanent magnet synchronous motor through the synovial-Romberg observer.

3. The DPCC predictive control system with the synovial component introduced into the Romberg observer as claimed in claim 1, characterized in that: Also includes: A Park inverse converter is provided between the deadbeat current prediction controller and the three-state pulse width modulator, and is used to perform Park inverse conversion on the predicted direct-axis voltage and quadrature-axis voltage output by the deadbeat current prediction controller, and then input the predicted direct-axis voltage and quadrature-axis voltage to the three-state pulse width modulator; The Clark converter and the Park converter are arranged between the inverter and the Romberg observer. The Clark converter is used to perform Clark transformation on the three-phase current and voltage output by the inverter and then input them to the Park converter. The Park converter is used to perform Park transformation on the output of the Clark converter and then input them to the Romberg observer. A PI controller is arranged between the permanent magnet synchronous motor and the deadbeat current prediction controller, and is used to take the actual speed of the permanent magnet synchronous motor and the given speed as input to generate a given direct-axis current input to the deadbeat current prediction controller; The sensor is arranged between the permanent magnet synchronous motor and the PI controller, and is used for collecting the actual output of the permanent magnet synchronous motor and inputting it into the Park converter, and at the same time inputting the actual speed in the actual output into the PI controller.

4. The DPCC predictive control system with the synovial component introduced into the Romberg observer as claimed in claim 1, characterized in that: The processing steps of the deadbeat current prediction controller are as follows: S101: discretize the voltage equation of the permanent magnet synchronous motor by the forward Euler method to obtain a PMSM discrete current model; S102: Calculate a given direct-axis voltage and a quadrature-axis voltage by combining a given direct-axis current and a direct-axis current fed back by the motor with a PMSM discrete current model; The formula is: Where: Represent the given direct-axis voltage and quadrature-axis voltage respectively; R s represents the stator resistance; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; ω e represents the electrical angular velocity; ψ f Represents the permanent magnet flux; L = L d =L q represents inductance; S103: Calculating predicted direct-axis voltage and quadrature-axis voltage by combining given direct-axis voltage and quadrature-axis voltage with parameter disturbance; The formula is: Where: u d ′、u q ′ represents the predicted direct-axis voltage and quadrature-axis voltage respectively; f d 、f q denote the parameter perturbations of the direct and quadrature axes respectively.

5. The DPCC predictive control system with the synovial component introduced into the Romberg observer as claimed in claim 4, characterized in that: In step S101, the voltage equation of the permanent magnet synchronous motor is expressed by the following formula: Where: u d 、u q Respectively represent the direct-axis voltage and quadrature-axis voltage of the stator; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; R s Indicates stator resistance; L d , L q They represent the direct-axis inductance and quadrature-axis inductance respectively; ψ f represents the permanent magnet flux; ω e represents the electrical angular velocity; ψ d , q They represent the direct-axis flux and the quadrature-axis flux respectively; The PMSM discrete current model is expressed by the following formula: Where: i d 、i q They represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor respectively; T s represents the sampling time; k represents the current time.

6. The DPCC predictive control system with the synovial component introduced into the Romberg observer as claimed in claim 4, characterized in that: In step S103, in the initial state, the parameter disturbances of the direct axis and the quadrature axis are calculated by the following formula: Where: f d 、f q They represent the parameter perturbations of the direct axis and quadrature axis respectively; ΔR s Indicates the deviation of stator resistance; ΔL represents the deviation of inductance; Δψ f Indicates the deviation of magnetic flux; i d 、i q Respectively represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor; ω e represents the electrical angular velocity; ψ f Represents the flux linkage of a permanent magnet.

7. The DPCC predictive control system with the synovial component introduced into the Romberg observer as claimed in claim 2, characterized in that: The model formula of the Lumberg observer is expressed as: Where: represents the estimated value of the state quantity of the Lumberg observer; represents the output estimate of the Lumberg observer; K represents the gain matrix, k1 and k2 are gain coefficients; R s represents the stator resistance; ω e represents the electrical angular velocity; u d 、u q They represent the direct-axis voltage and quadrature-axis voltage output by the permanent magnet synchronous motor respectively; ψ f represents the permanent magnet flux; L represents inductance; i d 、i q They represent the direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor respectively; The model formula of the Lumberg observer is discretized by the forward Euler formula to obtain the discretized model formula of the Lumberg observer: Where: represents the estimated value of the direct axis and quadrature axis parameter disturbance; T s represents the sampling time; k represents the current time.

8. The DPCC predictive control system with the synovial component introduced into the Romberg observer as claimed in claim 7, characterized in that: The model formula of the synovial-Lomborg observer is expressed as: Where: N represents the synovial gain matrix, n1 and n2 are the synovial gain coefficients respectively; sgn represents the sign function in synovial control; The model formula of the synovial membrane-Lomborg observer is discretized by the forward Euler formula to obtain the discretized model formula of the synovial membrane-Lomborg observer: Where: represents the estimated values ​​of the direct and quadrature axis parameter perturbations.

9. The DPCC predictive control system with the synovial component introduced into the Romberg observer as claimed in claim 8, characterized in that: The stability analysis of the synovial-Lomborg observer is carried out using the Lyapunov function. The specific steps are as follows: S201: construct a full-order observer with parameter disturbance and direct-axis current and quadrature-axis current as state variables; The model formula of the full-order observer is expressed as: Where: x represents the state of the full-order observer; y represents the output of the full-order observer; u represents the input of the full-order observer; A is the state matrix; B is the input matrix; C is the output matrix; R s represents the stator resistance; ω e represents the electrical angular velocity; ψ f represents the permanent magnet flux; u d 、u q Respectively represent the direct-axis voltage and quadrature-axis voltage output by the permanent magnet synchronous motor; i d 、i q The direct-axis current and quadrature-axis current output by the permanent magnet synchronous motor respectively; S202: Subtract the model formula of the full-order observer from the discretized model formula of the synovial-Lomborg observer to obtain Formula 1; Formula 1 is expressed as: Where: k1, k2 are gain coefficients; S203: define a Lyapunov function; obtain Formula 2 by analyzing the Lyapunov function in combination with the discretization model formula of the synovial-Lomborg observer; and obtain Formula 3 by updating the Lyapunov function through Formula 2; The Lyapunov function is defined as: Where: V represents the defined scalar function, i.e., the Lyapunov function; S represents the function of the current; According to the Lyapunov stability criterion, when the Lyapunov function is greater than, the derivative should be less than 0; Formula 2 is expressed as: Formula 3 is expressed as: S204: According to Formula 3, it can be known that the Lyapunov function V is always greater than 0; the derivative of Formula 3 is analyzed, and Formula 4 is obtained by taking the derivative of Formula 3; Formula 4 is expressed as: S205: converting formula 1 into formula 5; Formula 5 is expressed as: S206: Obtain the derivative of the Lyapunov function according to Formula 4 and Formula 5 to obtain Formula 6; simplify Formula 6 into Formula 7; Formula 6 is expressed as: Formula 7 is expressed as: S207: Derivation of Formula 8 satisfied by the synovial gain coefficient n1 according to Formula 7; Formula eight is expressed as: S208: According to Formula 8, it can be ensured that the derivative of the obtained Lyapunov function is less than 0, that is, the Synovial-Lumberg observer is stable.

10. The DPCC predictive control method with the synovial component introduced into the Romberg observer is characterized by: The DPCC prediction control system according to claim 1 is implemented, comprising: S1: Generate predicted direct-axis voltage and quadrature-axis voltage according to the given direct-axis current, quadrature-axis current and the direct-axis current and quadrature-axis current fed back by the control system combined with parameter disturbance; S2: Generate six pulse signals according to the predicted direct-axis voltage and quadrature-axis voltage; invert the six pulse signals to generate three-phase current; S3: Control the permanent magnet synchronous motor according to the three-phase current to obtain the actual output of the permanent magnet synchronous motor; S4: Generate parameter disturbance according to the actual output of the permanent magnet synchronous motor; S5: Feedback the direct-axis current and quadrature-axis current in the actual output in step S3 and the parameter disturbance generated in step S4 to step S1 to assist in generating the predicted direct-axis voltage and quadrature-axis voltage.

Images