阅读说明：本技术 一种永磁同步电机的模型预测电流控制方法 (Model prediction current control method of permanent magnet synchronous motor ) 是由 杨凯 杨帆 姜峰 柳岸明 辜成林 李健 熊飞 于 2020-05-21 设计创作，主要内容包括：本发明公开了一种永磁同步电机的模型预测电流控制方法,该方法包括如下步骤：通过获取d轴电流扰动项和q轴电流扰动项；获取永磁同步电机的三相相电流、三相线电压和转速信号,利用线性扩张状态观测器得到d轴电流观测值、q轴电流观测值、d轴电流扰动项观测值和q轴电流扰动项观测值；进而获取扰动控制器的d轴输出值和q轴输出值,利用电感提取算法获取电感预测值,并结合增量预测模型得到d轴电流预测值和q轴预测电流值；利用电感预测值、d轴电流预测值和q轴预测电流值对永磁同步电机的开关信号进行调整,以实现永磁同步电机的闭环控制,从而提高永磁同步电机模型预测电流控制算法的鲁棒性。(The invention discloses a model prediction current control method of a permanent magnet synchronous motor, which comprises the following steps: obtaining a d-axis current disturbance term and a q-axis current disturbance term; acquiring three-phase current, three-phase line voltage and rotating speed signals of the permanent magnet synchronous motor, and acquiring a d-axis current observation value, a q-axis current observation value, a d-axis current disturbance term observation value and a q-axis current disturbance term observation value by using a linear extended state observer; obtaining a d-axis output value and a q-axis output value of the disturbance controller, obtaining an inductance predicted value by using an inductance extraction algorithm, and obtaining a d-axis current predicted value and a q-axis predicted current value by combining an increment prediction model; and adjusting the switching signal of the permanent magnet synchronous motor by using the inductance predicted value, the d-axis current predicted value and the q-axis predicted current value to realize closed-loop control of the permanent magnet synchronous motor, so that the robustness of a permanent magnet synchronous motor model predicted current control algorithm is improved.)

1. A model prediction current control method of a permanent magnet synchronous motor is characterized by comprising the following steps:

s1, acquiring a voltage equation set of the permanent magnet synchronous motor in a rotating coordinate system, and acquiring a d-axis current disturbance term and a q-axis current disturbance term by using the voltage equation set;

s2, three-phase current, three-phase line voltage and rotating speed signals of the permanent magnet synchronous motor are obtained, and a d-axis current observation value, a q-axis current observation value, a d-axis current disturbance item observation value and a q-axis current disturbance item observation value are obtained by utilizing a linear extended state observer, a d-axis current disturbance item and a q-axis current disturbance item;

s3, obtaining a d-axis output value and a q-axis output value of the disturbance controller by using the d-axis current disturbance item observation value and the q-axis current disturbance item observation value, obtaining an inductance predicted value by using an inductance extraction algorithm, and obtaining a d-axis current predicted value and a q-axis current predicted value by combining an increment prediction model;

and S4, adjusting a switching signal of the permanent magnet synchronous motor by using the inductance predicted value, the d-axis current predicted value and the q-axis current predicted value so as to realize closed-loop control of the permanent magnet synchronous motor.

2. The model predictive current control method of a permanent magnet synchronous motor according to claim 1, wherein the voltage equation set specifically is:

wherein i_d、i_q、u_d、u_q、w_eAnd psi_fRespectively d-axis current, alternating current, d-axis voltage, q-axis voltage, electric angular velocity and permanent magnet flux linkage of the permanent magnet synchronous motor in a rotating coordinate system, R₀、L₀Respectively an initial value of armature resistance and an initial value of armature inductance of the permanent magnet synchronous motor, wherein Delta L is an error between a real value and the initial value of the armature inductance, Delta R is an error between the real value and the initial value of the armature resistance, and Delta psi_fIs the error between the true and initial values of the permanent magnet flux linkage.

3. The method of claim 2, wherein the voltage equation is used to control the model predictive current of the PMSMGroup acquisition d-axis current disturbance term f_dAnd q-axis current disturbance term f_qThe method specifically comprises the following steps:

4. the model predictive current control method of a permanent magnet synchronous machine according to claim 1, characterized in that the iterative expression of the inductance extraction algorithm is:

wherein E is_LdIs the output value of the d-axis disturbance controller, E_LqFor the output value of the q-axis perturbation controller, α is the weighting factor,predicted values of inductance, T, of the PMSM at the previous moment and the current moment respectively_sIs the sampling period.

5. The model predictive current control method of the permanent magnet synchronous motor according to claim 1, wherein the d-axis current observed value and the d-axis current disturbance term observed value of the linear extended state observer are specifically:

wherein the content of the first and second substances,u_d(k) sampled values for the d-axis voltage at the present time, L₀Is the initial value of the armature inductance of the permanent magnet synchronous motor,is d-axis electricity of the current timeThe observed value of the flow is,

6. The model predictive current control method of a PMSM according to claim 5, characterized in that d-axis current prediction value i in the incremental prediction model^p _dThe expression (k +1) is:

wherein R is_sIs the actual value, L, of the armature resistance of the PMSM_sIs the actual value of the armature inductance, u, of the permanent magnet synchronous machine_d(k-1) is the sampled value of the d-axis voltage at the previous moment, i_d(k-1) is the sampled value of the d-axis current at the previous moment, i_q(k-1) and i_q(k) Sampled values, w, of the q-axis current at the previous and present time, respectively_eIs the electrical angular velocity of the permanent magnet synchronous motor.

7. The model predictive current control method of the permanent magnet synchronous motor according to claim 1, wherein the q-axis current observed value and the q-axis current disturbance term observed value of the linear extended state observer are specifically:

wherein the content of the first and second substances,u_q(k) is a sampled value of the q-axis voltage at the present moment, L₀Is the initial value of the armature inductance, w, of the permanent magnet synchronous motor_eIs the electrical angular velocity, psi, of a permanent magnet synchronous machine_fIs a permanent magnet flux linkage of a permanent magnet synchronous motor, delta psi_fAs the error between the true value and the initial value of the permanent magnet flux linkage,as an observation of the q-axis current at the present time,

8. The model predictive current control method of a permanent magnet synchronous motor according to claim 7, characterized in that a q-axis current predicted value i in the incremental prediction model^p _q(k +1) is represented by：

Wherein R is_sIs the actual value, L, of the armature resistance of the PMSM_sIs the actual value of the armature inductance, u, of the permanent magnet synchronous machine_d(k-1) is the sampled value of the d-axis voltage at the previous moment, i_d(k-1) and i_d(k) Sampled values of d-axis current, i, at the previous and present time, respectively_qAnd (k-1) is a sampled value of the q-axis current at the previous moment.

9. A terminal device, comprising at least one processing unit and at least one memory unit, wherein the memory unit stores a computer program which, when executed by the processing unit, causes the processing unit to carry out the steps of the method according to any one of claims 1 to 8.

10. A computer-readable medium, in which a computer program is stored which is executable by a terminal device, and which, when run on the terminal device, causes the terminal device to carry out the steps of the method of any one of claims 1 to 8.

Technical Field

The invention belongs to the field of permanent magnet synchronous motors, and particularly relates to a model prediction current control method of a permanent magnet synchronous motor.

Background

In recent years, with continuous development and application of rare earth permanent magnet materials in China, performances of a permanent magnet synchronous motor in power density, efficiency and the like exceed those of a conventional induction motor and a conventional direct current motor. Therefore, the permanent magnet synchronous motor is widely applied to industrial fields such as high-performance numerical control machine tools, industrial robots, electric vehicles and the like. The large number of applications in the industrial field put higher demands on the control performance of permanent magnet synchronous machines. The industrial control system of the permanent magnet synchronous motor is required to have good steady-state precision and excellent dynamic performance so as to adapt to more complex occasions with worse environment.

A commonly used control system for a permanent magnet synchronous motor consists of an outer ring of rotational speed and an inner ring of current. Wherein the bandwidth of the inner loop is a key factor influencing the dynamic performance of the control system. The existing permanent magnet synchronous motor control system mostly adopts magnetic field directional control, and d-axis and q-axis currents of the motor are respectively controlled by a PI regulator after three-phase stator currents are converted into a synchronous rotating coordinate system, so that decoupling of an excitation current component and a torque current component of the motor is realized, a control process is simplified, and control precision is improved. But due to the inherent delays of the digital control system, such as hold, quantization, dead zone, filtering, etc., the bandwidth of the current loop is severely affected. In digital control systems, therefore, the run time of the control algorithm is to be made as short as possible in order to improve the control performance. The model prediction current control algorithm is an algorithm capable of effectively improving the dynamic performance of a current loop, and is widely researched in industry.

However, model predictive current control algorithms are too dependent on the parameters of the model. In the operation process of the permanent magnet synchronous motor, along with the changes of the load and the environmental temperature of the permanent magnet synchronous motor, the electrical parameters of the armature resistance, the armature inductance and the like of the permanent magnet synchronous motor can change along with the changes of the working condition of the permanent magnet synchronous motor, so that the improvement of the robustness of the model predictive current control algorithm of the permanent magnet synchronous motor is a premise for being widely applied to industrial production.

In the existing method for improving the robustness of the permanent magnet synchronous motor model prediction current control algorithm, the traditional observer method (such as a Sliding Mode Observer (SMO), a Luenberger observer and the like) and the online parameter identification method are widely applied. However, in the method for improving the algorithm robustness by the traditional observer method, the robustness of the Lonberg observer method is poor, and the buffeting of the synovial observer has influence on the dynamic performance of the system, so that the running condition of the system sliding mode can be damaged, and the system has the problems of excessive overshoot, growth of the transition process and even an unstable state; in the existing online parameter identification method, only two parameters can be identified due to the lack of the identification matrix, so that all resistance and inductance parameters cannot be identified for the salient pole motor. The invention patent 201711252132.34 provides an algorithm for improving model prediction current control by adopting an extended sliding mode observer, however, in the algorithm, the sliding mode surface approach rate is selected from a simple constant speed approach rate, and the buffeting problem of the sliding mode is not considered; the invention patent 201910499568.5 adopts an extended state observer as a disturbance observer to be combined with dead-beat current prediction control, however, in the algorithm, a disturbance term observed by the observer is directly compensated into a voltage equation, and in doing so, when a parameter error is too large, convergence may not be achieved.

Disclosure of Invention

Aiming at the defects or improvement requirements in the prior art, the invention provides a model prediction current control method for a permanent magnet synchronous motor, aiming at solving the technical problem that the robustness of a model prediction current control algorithm of the permanent magnet synchronous motor is too low because electrical parameters such as armature resistance and armature inductance of the permanent magnet synchronous motor change along with the change of the load and the ambient temperature of the permanent magnet synchronous motor and the change of the working condition of the permanent magnet synchronous motor.

To achieve the above object, according to one aspect of the present invention, there is provided a model predictive current control method of a permanent magnet synchronous motor, the method including the steps of:

s1, acquiring a voltage equation set of the permanent magnet synchronous motor in a rotating coordinate system, and acquiring a d-axis current disturbance term and a q-axis current disturbance term by using the voltage equation set;

s2, three-phase current, three-phase line voltage and rotating speed signals of the permanent magnet synchronous motor are obtained, and a d-axis current observation value, a q-axis current observation value, a d-axis current disturbance item observation value and a q-axis current disturbance item observation value are obtained by utilizing a linear extended state observer, a d-axis current disturbance item and a q-axis current disturbance item;

s3, obtaining a d-axis output value and a q-axis output value of the disturbance controller by using the d-axis current disturbance item observation value and the q-axis current disturbance item observation value, obtaining an inductance predicted value by using an inductance extraction algorithm, and obtaining a d-axis current predicted value and a q-axis current predicted value by combining an increment prediction model;

and S4, adjusting a switching signal of the permanent magnet synchronous motor by using the inductance predicted value, the d-axis current predicted value and the q-axis current predicted value so as to realize closed-loop control of the permanent magnet synchronous motor.

As a further improvement of the invention, the voltage equation set is specifically as follows:

wherein i_d、i_q、u_d、u_q、w_eAnd psi_fRespectively d-axis current, alternating current, d-axis voltage, q-axis voltage, electric angular velocity and permanent magnet flux linkage of the permanent magnet synchronous motor in a rotating coordinate system, R₀、L₀Respectively an initial value of armature resistance and an initial value of armature inductance of the permanent magnet synchronous motor, wherein Delta L is an error between a real value and the initial value of the armature inductance, and Delta R is the armatureError between true and initial values of resistance, delta psi_fIs the error between the true and initial values of the permanent magnet flux linkage.

As a further improvement of the invention, the d-axis current disturbance term f is obtained by utilizing the voltage equation set_dAnd q-axis current disturbance term f_qThe method specifically comprises the following steps:

as a further improvement of the present invention, the iterative expression of the inductance extraction algorithm is:

wherein E is_LdIs the output value of the d-axis disturbance controller, E_LqFor the output value of the q-axis perturbation controller, α is the weighting factor,

predicted values of inductance, T, of the PMSM at the previous moment and the current moment respectively_sIs the sampling period.

As a further improvement of the present invention, the d-axis current observed value and the d-axis current disturbance term observed value of the linear extended state observer are specifically:

wherein the content of the first and second substances,u_d(k) sampled values for the d-axis voltage at the present time, L₀Is the initial value of the armature inductance of the permanent magnet synchronous motor,

as an observed value of the d-axis current at the present time,

is the observed value of the d-axis current at the next moment, i_d(k) Sampled values of d-axis current at the present moment, e_d(k) As an error between the observed value and the sampled value of the d-axis current at the present time,

for the d-axis current disturbance term observed value at the current moment,

is observed value of d-axis current disturbance term at next moment, T_sFor the sampling period, β₁、β₂A first gain and a second gain of the linear extended state observer, respectively.

As a further improvement of the invention, the d-axis current predicted value i in the incremental prediction model^p _dThe expression (k +1) is:

wherein R is_sIs the actual value, L, of the armature resistance of the PMSM_sIs the actual value of the armature inductance, u, of the permanent magnet synchronous machine_d(k-1) is the sampled value of the d-axis voltage at the previous moment, i_d(k-1) is the sampled value of the d-axis current at the previous moment, i_q(k-1) and i_q(k) Sampled values, w, of the q-axis current at the previous and present time, respectively_eIs the electrical angular velocity of the permanent magnet synchronous motor.

As a further improvement of the present invention, the q-axis current observed value and the q-axis current disturbance term observed value of the linear extended state observer are specifically:

wherein the content of the first and second substances,

u_q(k) is the current timeSampled values of the q-axis voltage of the moment, L₀Is the initial value of the armature inductance, w, of the permanent magnet synchronous motor_eIs the electrical angular velocity, psi, of a permanent magnet synchronous machine_fIs a permanent magnet flux linkage of a permanent magnet synchronous motor, delta psi_fAs the error between the true value and the initial value of the permanent magnet flux linkage,as an observation of the q-axis current at the present time,

is the observed value of the q-axis current at the next moment, i_q(k) Sampled value of q-axis current at present time, e_q(k) As an error between the observed value and the sampled value of the q-axis current at the present time,

for the observed value of the q-axis current disturbance term at the current moment,is observed value of q-axis current disturbance term at the next moment, T_sSampling period for a linear extended state observer, β₁、β₂A first gain and a second gain of the linear extended state observer, respectively.

As a further improvement of the invention, the q-axis current predicted value i in the incremental prediction model^p _qThe expression (k +1) is:

wherein R is_sIs the actual value, L, of the armature resistance of the PMSM_sIs the actual value of the armature inductance, u, of the permanent magnet synchronous machine_d(k-1) is the sampled value of the d-axis voltage at the previous moment, i_d(k-1) and i_d(k) Sampled values of d-axis current, i, at the previous and present time, respectively_qAnd (k-1) is a sampled value of the q-axis current at the previous moment.

To achieve the above object, according to another aspect of the present invention, there is provided a terminal device comprising at least one processing unit, and at least one memory unit, wherein the memory unit stores a computer program which, when executed by the processing unit, causes the processing unit to perform the steps of the above method.

To achieve the above object, according to another aspect of the present invention, there is provided a computer readable medium storing a computer program executable by a terminal device, the program, when executed on the terminal device, causing the terminal device to perform the steps of the above method.

Generally, compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects:

the invention discloses a model prediction current control method of a permanent magnet synchronous motor, which obtains parameters under a corresponding rotating coordinate system by measuring the parameters under the load condition of the permanent magnet synchronous motor; observing d and q axis prediction current disturbance items of the permanent magnet synchronous motor by using an extended state observer; and adjusting by using the disturbance term obtained by observation to obtain an inductance error value, obtaining an inductance estimation value by using an inductance extraction algorithm, substituting the inductance estimation value into a permanent magnet synchronous motor model prediction current control algorithm and a d-axis and q-axis extended state observer, and continuously correcting the inductance estimation value to enable the disturbance term generated due to parameter mismatching to be zero, so that the permanent magnet synchronous motor can obtain better dynamic and steady-state performances.

The invention relates to a model prediction current control method of a permanent magnet synchronous motor, which designs an extended state observer according to the idea of 'total disturbance' in an active disturbance rejection control technology, wherein a common observer only observes the state of a system, and the extended state observer also estimates the state of an external disturbance and an unknown model, so that the extended state observer does not depend on a mathematical model too much and has strong robustness, meanwhile, a linear extended state observer is adopted to estimate the total disturbance, two linear extended state observers are respectively used to form disturbance observers of d-axis prediction current and q-axis prediction current, the total disturbance caused by mismatching of electrical parameters in a model prediction current control algorithm is observed in real time, and the parameter design is convenient and easy to realize according to a bandwidth method.

According to the model prediction current control method of the permanent magnet synchronous motor, the sensitivity of electrical parameters in the traditional model prediction control is reduced by adopting the incremental prediction model, the extended state observers of the d axis and the q axis are established according to the idea of 'total disturbance' in active disturbance rejection control, and the disturbance caused by unmatched electrical parameters is observed. Observed disturbance and current are substituted into the disturbance controller, the output of the disturbance controller is substituted into an inductance extraction algorithm, an estimated value of the inductance is obtained in real time and is substituted into a prediction model, errors of inductance parameters are corrected on line, the steady state and dynamic performance of a control system are effectively improved, disturbance caused by mismatching of electrical parameters of the permanent magnet synchronous motor is effectively eliminated, and the robustness of the algorithm is improved.

Drawings

Fig. 1 is a schematic diagram of a model predictive current control method for a permanent magnet synchronous motor according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of an embodiment of the present invention;

fig. 3 is a schematic diagram of an inverter according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other. The present invention will be described in further detail with reference to specific embodiments.

Fig. 1 is a schematic diagram of a model predictive current control method for a permanent magnet synchronous motor according to an embodiment of the present invention. A model prediction current control method of a permanent magnet synchronous motor comprises the following steps:

s1, acquiring a voltage equation set of the permanent magnet synchronous motor in a rotating coordinate system, and acquiring a d-axis current disturbance term and a q-axis current disturbance term by using the voltage equation set;

s2, three-phase current, three-phase line voltage and rotating speed signals of the permanent magnet synchronous motor are obtained, and a d-axis current observation value, a q-axis current observation value, a d-axis current disturbance item observation value and a q-axis current disturbance item observation value are obtained by utilizing a linear extended state observer, a d-axis current disturbance item and a q-axis current disturbance item;

s3, obtaining a d-axis output value and a q-axis output value of the disturbance controller by using the d-axis current disturbance item observation value and the q-axis current disturbance item observation value, obtaining an inductance predicted value by using an inductance extraction algorithm, and obtaining a d-axis current predicted value and a q-axis current predicted value by combining an increment prediction model;

and S4, adjusting a switching signal of the permanent magnet synchronous motor by using the inductance predicted value, the d-axis current predicted value and the q-axis current predicted value so as to realize closed-loop control of the permanent magnet synchronous motor.

Fig. 2 is a schematic diagram of an embodiment of the present invention. As shown in fig. 2, the sampling part is used for sampling the rotating speed, three-phase current and three-phase line voltage of the permanent magnet synchronous motor; the coordinate change part is used for converting the three-phase voltage current into d and q axis voltage current; a speed controller section for generating a command value of the q-axis current; the disturbance controller part is used for controlling the size of a disturbance item and eliminating disturbance generated due to mismatching of inductance parameters; the inductance extraction algorithm part is used for obtaining an inductance estimation value in real time and on line and is used in a prediction model; the prediction model part is used for predicting q-axis and d-axis currents; the linear extended state observer part is used for observing the disturbance terms of q-axis and d-axis currents of the permanent magnet synchronous motor, and the cost function minimization part is used for finding an optimal voltage vector through rolling optimization and outputting the optimal voltage vector to the inverter

Obtaining d-axis current, alternating current, phase d-axis voltage, phase q-axis voltage and electrical angular velocity under a rotating coordinate system through Park transformation by utilizing three-phase current, three-phase line voltage and rotating speed signals of the permanent magnet synchronous motor, and further obtaining a voltage equation set under the rotating coordinate system of the permanent magnet synchronous motor; the voltage equation set specifically includes:

wherein i_d、i_q、u_d、u_q、w_eAnd psi_fRespectively d-axis current, alternating current, d-axis voltage, q-axis voltage, electric angular velocity and permanent magnet flux linkage of the permanent magnet synchronous motor in a rotating coordinate system, R₀、L₀Respectively an initial value of armature resistance and an initial value of armature inductance of the permanent magnet synchronous motor, wherein Delta L is an error between a real value and the initial value of the armature inductance, Delta R is an error between the real value and the initial value of the armature resistance, and Delta psi_fIs the error between the true and initial values of the permanent magnet flux linkage. Of course, the above expression of the voltage equation set is only an example, and the corresponding adjustment may be performed according to the selection of the coordinate system.

Extracting terms related to inductance resistance parameters, and distributing the differential of d-axis and q-axis currents on the left side of an equation, and further obtaining a voltage equation set under a permanent magnet synchronous motor rotating coordinate system containing a disturbance term as follows:

thereby obtaining d-axis current disturbance term f by using voltage equation system_dAnd q-axis current disturbance term f_qThe method specifically comprises the following steps:

as a preferred embodiment, a first-order linear extended state observer may be used to obtain a d-axis current observed value and a d-axis current disturbance observed value, where the d-axis current observed value and the d-axis current disturbance observed value of the linear extended state observer are specifically:

wherein the content of the first and second substances,u_d(k) sampled values for the d-axis voltage at the present time, L₀Is the initial value of the armature inductance of the permanent magnet synchronous motor,

as an observed value of the d-axis current at the present time,

is the observed value of the d-axis current at the next moment, i_d(k) Sampled values of d-axis current at the present moment, e_d(k) As an error between the observed value and the sampled value of the d-axis current at the present time,for the d-axis current disturbance term observed value at the current moment,

is observed value of d-axis current disturbance term at next moment, T_sFor the sampling period, β₁、β₂β, the first and second gains, respectively, of the linear extended state observer₁And β₂Can be in accordance with the bandwidth omega of the observer₀To determine:of course, the above expression of the first-order linear extended state observer is only an example, and different observers can be selected for observation according to needs, so that the expression of the observer can be adjusted accordingly.

As a preferred embodiment, a first-order linear extended state observer may be used to obtain a q-axis current observed value and a q-axis current disturbance observed value, where the q-axis current observed value and the q-axis current disturbance observed value of the linear extended state observer are specifically:

wherein the content of the first and second substances,

u_q(k) is a sampled value of the q-axis voltage at the present moment, L₀Is the initial value of the armature inductance, w, of the permanent magnet synchronous motor_eIs the electrical angular velocity, psi, of a permanent magnet synchronous machine_fIs a permanent magnet flux linkage of a permanent magnet synchronous motor, delta psi_fAs the error between the true value and the initial value of the permanent magnet flux linkage,as an observation of the q-axis current at the present time,is the observed value of the q-axis current at the next moment, i_q(k) Sampled value of q-axis current at present time, e_q(k) As an error between the observed value and the sampled value of the q-axis current at the present time,for the observed value of the q-axis current disturbance term at the current moment,is observed value of q-axis current disturbance term at the next moment, T_sSampling period for a linear extended state observer, β₁、β₂β, the first and second gains, respectively, of the linear extended state observer₁And β₂Can be in accordance with the bandwidth omega of the observer₀To determine:of course, the above expression of the first-order linear extended state observer is only an example, and different observers can be selected for observation according to needs, so that the expression of the observer can be adjusted accordingly.

Therefore, the d-axis and q-axis current voltage is input into the linear extended state observer part, and the disturbance term estimation values of the d-axis and q-axis prediction currents can be obtained.

As a preferred embodiment, the disturbance controller is designed as a PI controller, and when there is no deviation in the inductance parameter, i.e. Δ L is 0, the d-axis disturbance controller results in w, as can be seen from the expression of the disturbance term_ei_qThe result of the q-axis disturbance controller is-w_ei_d. And the observed value is adopted to replace the sampling value, so the given values of the d-axis disturbance controller and the q-axis disturbance controller are respectively

Thus, the iterative expression of the inductance extraction algorithm can be:

wherein E is_LdIs the output value of the d-axis disturbance controller, E_LqFor the output value of the q-axis perturbation controller, α is the weighting factor,predicted values of inductance, T, of the PMSM at the previous moment and the current moment respectively_sIs the sampling period.

Due to the inherent delays of the digital signal processor and the controller, the predicted current is obtained at time k, but is not applied until time k + 1. Therefore, the voltage vector selected at the k-th instant may not be the best choice. The effect of this delay is particularly severe when the sampling frequency is low, so to further improve robustness, it is necessary to use two-step prediction to compensate one-step delay, and then obtain the predicted value i of d-axis current in the incremental prediction model^p _dThe expression (k +1) is:

wherein R is_sIs the actual value, L, of the armature resistance of the PMSM_sOf armature inductance of permanent-magnet synchronous machinesTrue value, u_d(k-1) is the sampled value of the d-axis voltage at the previous moment, i_d(k-1) is the sampled value of the d-axis current at the previous moment, i_q(k-1) and i_q(k) Sampled values, w, of the q-axis current at the previous and present time, respectively_eIs the electrical angular velocity of the permanent magnet synchronous motor.

Predicted value i of q-axis current in incremental prediction model^p _qThe expression (k +1) is:

wherein R is_sIs the actual value, L, of the armature resistance of the PMSM_sIs the actual value of the armature inductance, u, of the permanent magnet synchronous machine_d(k-1) is the sampled value of the d-axis voltage at the previous moment, i_d(k-1) and i_d(k) Sampled values of d-axis current, i, at the previous and present time, respectively_qAnd (k-1) is a sampled value of the q-axis current at the previous moment.

The last-time incremental prediction model is lower in parameter sensitivity than a traditional prediction model and is derived through the traditional prediction models at two different moments.

The expression of the traditional prediction model is as follows:

wherein i (k) represents the current sampling value at the current moment, i (k +1) represents the predicted value at the k +1 moment, and R_sIs the true value of armature resistance, L_sThe actual value of the armature inductance.

The prediction error of d-axis and q-axis currents of the traditional prediction model is as follows:

from the above formula, it can be seen that the conventional prediction model contains three parameters (resistance, inductance, flux linkage), so when the model parameters do not match, the steady-state accuracy and the dynamic response performance of the system are seriously affected.

Replacing k with k-1 to change the formula:

subtracting the two formulas to obtain an incremental prediction model expression:

as can be seen from the above equation, the incremental prediction model eliminates the influence caused by the mismatch of the permanent magnet flux linkage parameters.

In the incremental prediction model, the prediction error between the current predictions in the absence of parameter errors and in the presence of parameter errors is:

wherein D is_d、D_qPredicting errors occurring in the current for the d-axis and q-axis of the incremental prediction model, R_sIs the true value of armature resistance, L_sThe actual value of the armature inductance. And deltaL and deltaR are errors between the actual armature inductance and armature resistance and the armature inductance and armature resistance used by the algorithm respectively.

When the motor is in a steady state, i is satisfied_d(k)＝i_d(k-1) and i_q(k)＝i_q(k-1), whereby the prediction error is reduced to:

therefore, after the incremental prediction model is adopted, only errors caused by mismatching of inductance parameters need to be considered, and the influence of resistance parameters can be ignored. In order to obtain accurate inductance parameters, d-axis and q-axis disturbance observers are established. And observing a disturbance item through a disturbance observer, substituting the disturbance observation value and the current observation value into the disturbance controller, outputting an error value serving as an inductance parameter through a PI (proportional-integral) regulator, substituting the error value into an inductance extraction algorithm, outputting an inductance estimation value, substituting the inductance estimation value into the incremental prediction model, and gradually correcting the inductance parameter until the predicted current value is equal to the given current value. Meanwhile, the estimated value of the inductance parameter is also fed into the extended state observer, so that the size of the disturbance term observed by the observer is equal to the size of the given disturbance term.

Fig. 3 is a schematic diagram of an inverter according to an embodiment of the present invention. As shown in fig. 3, the basic voltage vectors of the inverter include 6 effective vectors and two zero vectors, and specifically, each voltage vector may be substituted into the prediction model to perform rolling optimization, and an optimal voltage vector is found and output to the inverter to implement control. Wherein the cost function can be expressed as:

g＝(i_d ^p(k+2))²+(i_q ^*-i_q ^p(k+2))²+f(i_d ^p(k+2),i_q ^p(k+2))

wherein the first term is reactive power minimization and comprises torque/ampere ratio optimization; the second term is tracking torque generating current; the last term is a non-linear function that limits the stator current. If the predicted current magnitude generated by the specified voltage vector is greater than the limit value, the cost function g is ∞, and therefore the voltage vector is not selected. On the other hand, if the predicted stator current is below the limit value, the cost function contains nearly the first two terms and will select a voltage vector that minimizes the current error.

Table 1 shows the basic parameters of the permanent magnet synchronous machine according to the embodiment of the present invention. As shown in table 1, taking a permanent magnet synchronous motor as an example, corresponding simulation verification is performed in MATLAB/SIMULINK software, and the basic relevant parameters of the motor are shown in table 1, wherein 5Nm load is suddenly added at the time of 2s and 10Nm load is suddenly added at the time of 4s during simulation.

Table 1 basic relevant parameters of a permanent magnet synchronous machine of an embodiment of the invention

Parameters of the electric machine	Value of parameter
		Armature resistance Rs	1.65Ω
d-axis reactance Ld	7.7mH
		q-axis reactance Lq	7.7mH
Permanent magnet flux linkage psif	0.208Wb

When inductance parameters are 0.2 times of actual inductance and resistance and flux linkage parameters are 5 times of actual parameters, the given value and the feedback value of the q-axis current are predicted by adopting the permanent magnet synchronous motor model of the embodiment of the invention; when the inductance parameter is 0.2 times of the actual inductance and the resistance and flux linkage parameters are 5 times of the actual parameters, the given value and the feedback value of the q-axis current in the traditional permanent magnet synchronous motor model prediction current control method are adopted; when the inductance parameter is 10 times of the actual inductance and the resistance and flux linkage parameters are 5 times of the actual parameters, the given value and the feedback value of the q-axis current are predicted by adopting the permanent magnet synchronous motor model of the embodiment of the invention; when the inductance parameter is 10 times of the actual inductance and the resistance and flux linkage parameters are 5 times of the actual parameters, the given value and the feedback value of the q-axis current in the traditional permanent magnet synchronous motor model prediction current control method are adopted; from the simulation results, under the condition that the electrical parameters are not matched, in the traditional model prediction current control method, a larger deviation occurs between the q-axis current feedback value and a given value, and the fluctuation of the q-axis current is larger. By adopting the method of the embodiment of the invention, the system is not influenced by the errors of the resistance and the flux linkage parameters, and the influence of the errors of the inductance parameters on the system can be effectively eliminated. The method of the embodiment of the invention obviously improves the robustness of the permanent magnet synchronous motor model prediction current control. Therefore, according to the permanent magnet synchronous motor system model prediction current control method provided by the embodiment of the invention, the incremental prediction model is adopted to replace the traditional prediction model, so that the parameter sensitivity of model prediction current control is reduced. According to the idea of active disturbance rejection control, d-axis and q-axis extended state observers are established, disturbance caused by parameter mismatching is observed on line in real time, a disturbance controller is established, the disturbance controller obtains the error of inductance by carrying out PI adjustment on a disturbance item, an estimated inductance value is obtained through an inductance extraction algorithm, the estimated inductance value is substituted into an increment prediction model and the d-axis and q-axis disturbance observers, the inductance value used by the prediction model is continuously corrected, the disturbance item observed by the observers is equal to the given value of the disturbance item, and finally the disturbance item generated due to parameter mismatching converges to zero.

A terminal device comprising at least one processing unit and at least one memory unit, wherein the memory unit stores a computer program which, when executed by the processing unit, causes the processing unit to carry out the steps of the above-mentioned method.

A computer-readable medium, in which a computer program is stored which is executable by a terminal device, the program, when run on the terminal device, causing the terminal device to perform the steps of the method described above.

It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.