CN109995290B - Open-loop iterative learning control method and system based on fractional calculus - Google Patents

Abstract

The invention discloses a control method and system for open-loop iterative learning based on fractional calculus, wherein the method comprises the following steps: establishing a discrete fractional-order open-loop iterative learning controller; establishing a permanent magnet synchronous motor position servo system based on vector control; Equivalent transformation of fractional calculus iterative learning controller; convert i _q , _id and

The difference obtained by comparison is sent to the current regulator respectively, and the voltage control quantities _ud and u _q are obtained through the current regulator; _ud and u _q are converted to the voltage control quantities u _α and u in the αβ coordinate system through the inverse PARK transformation _β , and then generate a pulse-modulated PWM signal according to u _α and u _β , and control the three-phase inverter to generate a three-phase voltage signal through the SVPWM principle. The present invention has higher control precision, and takes both practicability and accuracy into consideration.

Description

A control method and system for open-loop iterative learning based on fractional calculus

technical field

The invention belongs to the technical field of motor control, and relates to an open-loop iterative learning control method and system based on fractional-order calculus.

Background technique

In actual industrial production, there are many systems with repetitive or periodic motion processes, such as robot welding, various industrial production lines such as mobile phones, and the textile industry. These devices have a common feature, that is, repetitive motion and production processes. Existing control algorithms generally use PID, sliding mode, and adaptive control algorithms. These algorithms generally use a single feedback or feedforward technology. Although satisfactory accuracy can be obtained for a single product, it is not a problem for product consistency. can be well resolved.

Iterative learning control is a control algorithm for a system with repetitive motion. It uses the data information in the previous control to find the appropriate control input through online iteration, and can obtain accurate control effects. Fractional calculus has good memory function and genetic characteristics. Combining fractional calculus with iterative learning control will be a feasible control algorithm. However, due to the large amount of calculation, fractional calculus is difficult to implement in industrial control using microprocessors such as DSP.

SUMMARY OF THE INVENTION

In order to solve the above problems, the purpose of the present invention is to provide a position servo control method of a permanent magnet synchronous motor (PMSM) based on fractional order open-loop iterative learning. The invention applies discrete fractional iterative learning to position control design, and designs fractional calculus engineering implementation, and the introduction of fractional order makes the system have higher control precision. The invention takes both practicability and accuracy into consideration, and has high application value.

In order to achieve the above object, the technical solution of the present invention is a control method based on fractional-order calculus open-loop iterative learning, comprising the following steps:

S10, establish a discrete fractional-order open-loop iterative learning controller, wherein the iterative learning control law adopts PD ^α type iterative learning control law for position control;

S20, establish a permanent magnet synchronous motor position servo system based on vector control, and improve the performance of the controller by combining fractional calculus, wherein the learning gain and fractional calculus factor are adjusted according to the dynamic performance and steady-state performance of the system;

S30, equivalently transform the fractional-order calculus iterative learning controller;

比较得到的差值，分别送入电流调节器，经过电流调节器得到电压控制量u_d和u_q；S40, the convergence proof is performed on the motor position control variable _uk (i), i _q , _id and

The difference obtained by comparison is sent to the current regulator respectively, and the voltage control quantities _ud and u _q are obtained through the current regulator;

S50, after obtaining the motor position control quantity, adopt the position loop + current loop control strategy, the control quantity is converted into the current control quantity in the q coordinate system through the position regulator, and the d-axis given reference is 0;

S60, _ud and u _q are converted to the voltage control variables u _α and u _β in the αβ coordinate system through PARK inverse transformation, and then the pulse modulation PWM signal is generated according to u _α and u _β , and the three-phase inverter is controlled by the SVPWM principle Generate three-phase voltage signals.

Preferably, the S10 includes the following steps:

In the S11, dq coordinate system, the discrete mechanical dynamics equation of the permanent magnet synchronous motor is:

B＝[0 1/J]^T，C＝[1 0]，θ(t)和ω(t)分别表示系统t时刻的位置和转速信号，T_e(t)为电磁转矩输入，k_T为转矩系数，i_q（t）为q轴电流，B_f为摩擦系数，J为转动惯量，d(t)为包括负载的干扰信息；where x(t)=[θ(t) ω(t)] ^T , u(t)=T _e (t)=k _T i _q (t),

B=[0 1/J] ^T , C=[1 0], θ(t) and ω(t) represent the position and rotational speed signals of the system at time t, respectively, _Te (t) is the electromagnetic torque input, and k _T is the torque coefficient, i _q (t) is the q-axis current, B _f is the friction coefficient, J is the moment of inertia, and d(t) is the disturbance information including the load;

S12, the discrete fractional open-loop iterative learning control is adopted, and the difference between the given position θ ^* and the position θ fed back by the position sensor is sent to the discrete fractional open-loop iterative learning controller, and the output of the discrete fractional open-loop iterative learning controller is the torque or current command signal, and the discrete fractional-order open-loop iterative learning controller is

u _k+1 (t)=u _k (t)+K _p e _k (t)+K _D Δ ^α e _k (t) (2)

Among them, _uk (t) is the control variable of the k-th iteration at time t, uk ₊₁ (t) is the control variable of the k+1-th iteration at time t, and _ek (t) is the error of the k-th iteration at time t, namely e _k (t)=y _d (t)-y _k (t), K _p is the proportional adjustment coefficient, K _D is the differential adjustment coefficient, Δ represents the discrete differential operator, Δ ^α is the α-order differential, α∈(0, 1);

The discrete fractional calculus is defined as follows,

Among them, h is the sampling time, and m is the discrete time.

P is defined as follows,

Preferably, the S30 includes the following steps:

根据式(3)和(4)，可得t时刻第k次误差的α阶微分为S31, let the sampling time h→0, then

According to equations (3) and (4), the α-order differential of the kth error at time t can be obtained as

将式(5)变换为：S32, order

Transform equation (5) into:

S33, according to formula (6), the discrete fractional-order open-loop iterative learning controller in S12 is transformed into

in,

Preferably, the S40 includes the following steps:

S41, obtained from formula (1),

Let Y _k = [y _k (1)y _k (2),...y _k (N)] ^T (10)

x _k = [x _k (0)x _k (1),...x _k (N-1)] ^T (11)

U _k = [δu _k (0)δu _k (1),...δu _k (N-1)] ^T (12)

δu _k (i)=u _d (i)-u _k (i) (13)

Among them, _uk (i) is the motor control amount, _ud (i) is the expectation, and _ek (t) is the k-th iteration error at time t.

S42, according to equation (13), subtract u _d (i) from both sides of equation (7) at the same time, then invert, and then bring t=0 into the discrete fractional-order open-loop iterative learning controller as:

in

According to formula (14), formula (15) is simplified, and (14) is brought into

=δu _k (0)-k _p CBδu _k (0)

Extract the common factor to get

δu _k+1 (0)=(Ik _p CB)δu _k (0)

δu _k+1 (0) is the difference between the expectation of the k+1 iteration and the control amount at the time of t=0;

S43, according to equations (13) and (14), subtract u _d (i) from both sides of equation (7) at the same time, then invert, and then bring t=1 into the discrete fractional-order open-loop iterative learning controller as:

δu _k+1 (1)=δu _k (1)-k _p CABδu _k (0)-k _p CBδu _k (1)+k _p c ₁ e _k (0) (16)

Wherein, k _p c ₁ e _k (0)=0;

Simplify equation (16) and extract the common factor to get,

δu _k+1 (1)=(Ik _p cB)δu _k (1)-k _p CABδu _k (0) (17)

S44, according to equations (13) and (14), subtract u _d (i) from both sides of equation (7) at the same time, then invert, and then bring t=2 into the discrete fractional-order open-loop iterative learning controller as,

According to formula (14), formula (18) is simplified, and the common factor is extracted to obtain,

δu _k+1 (2)=(Ik _p cB)δu _k (2)+Δ (19)

Wherein, Δ=-k _p e _k (c ₁ -c ₃ );

S45, after finishing,

That is, δU _k+1 =QδU _k

in,

Corresponding to the above method, the present invention also provides a system of a fractional-order calculus-based open-loop iterative learning control method, including a discrete fractional open-loop iterative learning controller, a current regulator, a DSP microprocessor, an inverse variable circuit and position sensor, where,

和

The position sensor collects the position information of the permanent magnet synchronous motor, and inputs the difference between the given position θ ^* and the position θ fed back by the position sensor into the discrete fractional open-loop iterative learning controller, and the discrete fractional open-loop iterative learning The output of the controller is the torque or current command signal

and

比较得到的差值，分别输入q轴电流调节器和d轴电流调节器，输出电压控制量u_d和u_q；The current regulator includes a q-axis current regulator and a _d -axis current regulator, which combine i _q , id and

Compare the difference obtained, input the q-axis current regulator and the d-axis current regulator respectively, and output the voltage control variables _ud and u _q ;

The DSP microprocessor includes a PARK inverse converter, a PARK converter, a CLARK converter and a SVPWM generator, a q-axis current regulator and a d-axis current regulator, and the output voltage control quantities _ud and u _q are input to the PARK inverse converter , converted to the voltage control quantities u _α and u _β in the αβ coordinate system, and then the SVPWM generator generates pulse-modulated PWM signals according to u _α and _{u β} , and controls the inverter circuit to generate three-phase voltage signals Va, V _b , V _c control the permanent magnet synchronous motor; convert V _a and V _b into current signals i _a , i _b through the CLARK converter to obtain the currents i _α and i _β in the αβ coordinate system, and then obtain i through the PARK converter _q , _id .

The beneficial effects of the present invention are as follows: the introduction of the fractional differential operator in the position control of the present invention increases the adjustable factor of the position controller, and also ensures that the control rate of the position controller is monotonically converged for the time-varying nonlinearity that occurs in the system , so that the position controller has better stability and adaptability. The present invention effectively utilizes the unique advantage of fractional order D ^α type learning law compared with traditional iterative learning in regulating the monotonous convergence of tracking learning, and combines the P type learning law and the increased adjustable parameter fractional order to improve the tracking performance and the convergence. speed.

Description of drawings

Fig. 1 is the flow chart of the steps of the control method based on fractional calculus based open-loop iterative learning of the method embodiment 1 of the present invention;

2 is a system block diagram of a control method based on fractional-order calculus-based open-loop iterative learning of a system embodiment of the present invention;

FIG. 3 is a schematic diagram of the control logic of the open-loop iterative learning based on fractional calculus according to an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

On the contrary, the present invention covers any alternatives, modifications, equivalents and arrangements within the spirit and scope of the present invention as defined by the appended claims. Further, in order to give the public a better understanding of the present invention, some specific details are described in detail in the following detailed description of the present invention. The present invention can be fully understood by those skilled in the art without the description of these detailed parts.

Method Example 1

Referring to FIG. 1, the technical solution of the present invention is a flowchart of steps of a control method based on fractional calculus based open-loop iterative learning, including the following steps:

S10, establish a discrete fractional-order open-loop iterative learning controller, wherein the iterative learning control law adopts PD ^α type iterative learning control law for position control;

S20, establish a permanent magnet synchronous motor position servo system based on vector control, and improve the performance of the controller by combining fractional calculus, wherein the learning gain and fractional calculus factor are adjusted according to the dynamic performance and steady-state performance of the system;

S30, equivalently transform the fractional-order calculus iterative learning controller;

比较得到的差值，分别送入电流调节器，经过电流调节器得到电压控制量u_d和u_q；S40, the convergence proof is performed on the motor position control variable _uk (i), i _q , _id and

The difference obtained by comparison is sent to the current regulator respectively, and the voltage control quantities _ud and u _q are obtained through the current regulator;

S50, after obtaining the motor position control quantity, adopt the position loop + current loop control strategy, the control quantity is converted into the current control quantity under the q coordinate system through the position regulator, and the d-axis given reference is 0;

S60, _ud and u _q are converted to the voltage control variables u _α and u _β in the αβ coordinate system through PARK inverse transformation, and then the pulse modulation PWM signal is generated according to u _α and u _β , and the three-phase inverter is controlled by the SVPWM principle Generate three-phase voltage signals.

In the above method, the iterative learning control law adopts PD ^α type iterative learning control law for position control. The reason is that the introduction of the fractional differential operator in the position control increases the adjustable factor of the position controller and ensures the The control rate of the position controller is monotonically convergent for the time-varying nonlinearity of the system, so that the position controller has better stability and adaptability.

In a specific embodiment, S10 includes the following steps:

In the S11, dq coordinate system, the discrete mechanical dynamics equation of the permanent magnet synchronous motor is:

B＝[0 1/J]^TC＝[1 0]，θ(t)和ω(t)分别表示系统t时刻的位置和转速信号，T_e(t)为电磁转矩输入，k_T为转矩系数，i_q(t)为q轴电流，B_f为摩擦系数，J为转动惯量，d(t)为包括负载的干扰信息；where x(t)=[θ(t) ω(t)] ^T , u(t)=T _e (t)=k _T i _q (t),

B=[0 1/J] ^T C=[1 0], θ(t) and ω(t) represent the position and rotational speed signals of the system at time t, respectively, _Te (t) is the electromagnetic torque input, and k _T is Torque coefficient, i _q (t) is the q-axis current, B _f is the friction coefficient, J is the moment of inertia, and d(t) is the disturbance information including the load;

S12, the discrete fractional open-loop iterative learning control is adopted, and the difference between the given position θ ^* and the position θ fed back by the position sensor is sent to the discrete fractional open-loop iterative learning controller, and the output of the discrete fractional open-loop iterative learning controller is the torque or current command signal, and the discrete fractional-order open-loop iterative learning controller is

u _k+1 (t)=u _k (t)+K _p e _k (t)+K _D Δ ^α e _k (t) (2)

Among them, _uk (t) is the control variable of the k-th iteration at time t, uk ₊₁ (t) is the control variable of the k+1-th iteration at time t, and _ek (t) is the error of the k-th iteration at time t, namely e _k (t)=y _d (t)-y _k (t), K _p is the proportional adjustment coefficient, K _D is the differential adjustment coefficient, Δ represents the discrete differential operator, Δ ^α is the α-order differential, α∈(0, 1);

The discrete fractional calculus is defined as follows,

Among them, h is the sampling time, and m is the discrete time.

P is defined as follows,

S30 includes the following steps:

根据式(3)和(4)，可得t时刻第k次误差的α阶微分为S31, let the sampling time h→0, then

According to equations (3) and (4), the α-order differential of the kth error at time t can be obtained as

将式(5)变换为：S32, order

Transform equation (5) into:

S33, according to formula (6), the discrete fractional-order open-loop iterative learning controller in S12 is transformed into

in,

S40 includes the following steps:

S41, obtained from formula (1),

Let Y _k = [y _k (1)y _k (2),...y _k (N)] ^T (10)

x _k = [x _k (0)x _k (1),...x _k (N-1)] ^T (11)

U _k = [δu _k (0)δu _k (1),...δu _k (N-1)] ^T (12)

δu _k (i)=u _d (i)-u _k (i) (13)

Among them, u _k (i) is the motor control amount, u _d (i) is the expectation, and _ek (t) is the k-th iteration error at time t;

S42, according to equation (13), subtract u _d (i) from both sides of equation (7) at the same time, then invert, and then bring t=0 into the discrete fractional-order open-loop iterative learning controller as:

in

According to formula (14), formula (15) is simplified, and (14) is brought into

=δu _k (0)-k _p CBδu _k (0)

Extract the common factor to get

δu _k+1 (0)=(Ik _p CB)δu _k (0)

δu _k+1 (0) is the difference between the expectation of the k+1 iteration and the control amount at the time of t=0;

S43, according to equations (13) and (14), subtract u _d (i) from both sides of equation (7) at the same time, then invert, and then bring t=1 into the discrete fractional-order open-loop iterative learning controller as:

δu _k+1 (1)=δu _k (1)-k _p CABδu _k (0)-k _p CBδu _k (1)+k _p c ₁ e _k (0) (16)

Wherein, k _p c ₁ e _k (0)=0;

Simplify equation (16) and extract the common factor to get,

δu _k+1 (1)=(Ik _p cB)δu _k (1)-k _p CABδu _k (0) (17)

S44, according to equations (13) and (14), subtract u _d (i) from both sides of equation (7) at the same time, then invert, and then bring t=2 into the discrete fractional-order open-loop iterative learning controller as,

According to formula (14), formula (18) is simplified, and the common factor is extracted to obtain,

δu _k+1 (2)=(Ik _p cB)δu _k (2)+Δ (19)

Wherein, Δ=-k _p e _k (c ₁ -c ₃ );

S45, after finishing,

That is, δU _k+1 =QδU _k

in,

System embodiment

Referring to FIG. 2, the present invention also provides a system of a fractional-order calculus-based open-loop iterative learning control method, including a discrete fractional-order open-loop iterative learning controller 10, a current regulator, a DSP microprocessor, an inverter circuit 40 and position sensor 50, wherein,

和

The position sensor 50 collects the position information of the permanent magnet synchronous motor 60, and inputs the difference between the given position θ ^* and the position θ fed back by the position sensor 50 into the discrete fractional open-loop iterative learning controller 10, which is a discrete fractional open-loop iterative learning controller. The output of 10 is the torque or current command signal

and

比较得到的差值，分别输入q轴电流调节器21和d轴电流调节器22，输出电压控制量u_d和u_q；The current regulator includes a q-axis current regulator 21 and a _d -axis current regulator 22, which convert i _q , id and

The difference obtained by comparison is input to the q-axis current regulator 21 and the d-axis current regulator 22 respectively, and the output voltage control quantities _ud and u _q are output;

The DSP microprocessor includes a PARK inverse converter 31, a PARK converter 32, a CLARK converter 33 and a SVPWM generator 34, a q-axis current regulator 21 and a d-axis current regulator 22, and the output voltage control quantities _ud and u _q are input The PARK inverter 31 is converted to the voltage control variables u _α and u _β in the αβ coordinate system, and then the SVPWM generator 34 generates pulse-modulated PWM signals according to u _α and u _β , and controls the inverter circuit 40 to generate three-phase voltage signals V _a , V _b , V _c control the permanent magnet synchronous motor 60 ; convert V _a and V _b into current signals i _a , i _b through the CLARK converter 33 to obtain currents i _α and i _β in the αβ coordinate system, and then pass PARK transformer 32 obtains i _q , _id .

进行离线运算，得到结果表，在运算过程中通过查表快速得到c_j的值。In a specific embodiment, referring to FIG. 3 , in order to reduce the computational burden of the DSP microprocessor for multiplication, the

Perform off-line operation to obtain a result table, and quickly obtain the value of c _j by looking up the table during the operation.

The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (2)

1. the control method based on the open-loop iterative learning of fractional calculus, is characterized in that, comprises the following steps: S10, a discrete fractional open-loop iterative learning controller is established, wherein the iterative learning control law adopts PD ^α type iterative learning control law for position control, and the PD ^α type fractional open-loop iterative learning controller is in the form of u _k+1 (t )=u _k (t)+k _p e _k (t)+k _D Δ ^α e _k (t), where u _k (t) is the k-th iteration control amount at time t, u _k+1 (t) is the control variable of the k+1 iteration at time t, e _k (t) the error of the k th iteration at time t, that is, e _k (t)=y _d (t)-y _k (t), k _p is the proportional adjustment coefficient , k _D differential adjustment coefficient, Δ represents the discrete differential operator, Δ ^α is the α-order differential, α∈(0,1); S20, establish a permanent magnet synchronous motor position servo system based on vector control, and improve the performance of the controller by combining fractional calculus, wherein the learning gain and fractional calculus factor are adjusted according to the dynamic performance and steady-state performance of the system; S30, equivalently transform the discrete fractional-order open-loop iterative learning controller; S40, a convergence proof is performed on the motor position control algorithm, and i _q , _id and

The difference obtained by comparison is sent to the current regulator respectively, and the voltage control quantities _ud and u _q are obtained through the current regulator; S50, after the motor position control quantity is obtained, the position loop + current loop control strategy is adopted, the control quantity is converted into the current control quantity in the q coordinate system through the position regulator, and the d-axis reference current control quantity is 0; S60, _ud and u _q are converted to the voltage control variables u _α and u _β in the αβ coordinate system through PARK inverse transformation, and then the pulse modulation PWM signal is generated according to u _α and u _β , and the three-phase inverter is controlled by the SVPWM principle Generate three-phase voltage signal; The S10 includes the following steps: In the S11, dq coordinate system, the discrete mechanical dynamics equation of the permanent magnet synchronous motor is:

where x(t)=[θ(t) ω(t)] ^T , u(t)=T _e (t)=k _T i _q (t),

B=[0 1/J] ^T , C=[1 0], θ(t) and ω(t) represent the position and rotational speed signals of the system at time t, respectively, _Te (t) is the electromagnetic torque input, and k _T is the torque coefficient, i _q (t) is the q-axis current, B _f is the friction coefficient, J is the moment of inertia, and d(t) is the disturbance information including the load; S12, the discrete fractional open-loop iterative learning control is adopted, and the difference between the given position θ ^* and the position θ fed back by the position sensor is sent to the discrete fractional open-loop iterative learning controller, and the output of the discrete fractional open-loop iterative learning controller is the torque or current command signal, and the discrete fractional-order open-loop iterative learning controller is u _k+1 (t)=u _k (t)+k _p e _k (t)+k _D Δ ^α e _k (t) (2) Among them, _uk (t) is the control variable of the k-th iteration at time t, uk ₊₁ (t) is the control variable of the k+1-th iteration at time t, and _ek (t) is the error of the k-th iteration at time t, namely e _k (t)=y _d (t)-y _k (t), k _p is the proportional adjustment coefficient, k _D is the differential adjustment coefficient, Δ represents the discrete differential operator, Δ ^α is the α-order differential, α∈(0, 1); The discrete fractional calculus is defined as follows,

Among them, h is the sampling time, m is the discrete time;

Defined as follows,

The S30 includes the following steps: S31, let the sampling time h→0, then

According to equations (3) and (4), the α-order differential of the kth error at time t can be obtained as

S32, order

Transform equation (5) into:

S33, according to formula (6), the discrete fractional-order open-loop iterative learning controller in S12 is transformed into

in,

The S40 includes the following steps: S41, obtained from formula (1),

Let Y _k = [y _k (1), y _k (2), ... y _k (N)] ^T (10) X _k = [x _k (0), x _k (1), ... x _k (N-1)] ^T (11) U _k = [δu _k (0), δu _k (1), ...δu _k (N-1)] ^T (12) δu _k (t)=u _d (t)-u _k (t) (13)

Among them, _uk (i) is the motor control variable, _ud (f) is the expectation, and _ek (t) is the k-th iteration error at time t; S42, according to equation (13), subtract u _d (t) from both sides of equation (7) at the same time, then invert, and then bring t=0 into the discrete fractional-order open-loop iterative learning controller as:

in

According to formula (14), formula (15) is simplified, and (14) is brought into δu _k+1 (0)=δu _k (0)-k _p CBδu _k (0) Extract the common factor to get δu _k+1 (0)=(Ik _p CB)δu _k (0) δu _k+1 (0) is the difference between the expectation of the k+1 iteration and the control amount at the time of t=0; S43, according to equations (13) and (14), subtract u _d (i) from both sides of equation (7) at the same time, then invert, and then bring t=1 into the discrete fractional-order open-loop iterative learning controller as: δu _k+1 (1)=δu _k (1)-k _p CABδu _k (0)-k _p CBδu _k (l)+k _p c ₁ e _k (0) (16) where k _p c ₁ e _k (0)=0, and c ₁ is when j=1

the value of ; Simplify equation (16) and extract the common factor to get, δu _k+1 (1)=(Ik _p CB)δu _k (1)-k _p CABδu _k (0) (17) S44, according to equations (13) and (14), subtract u _d (i) from both sides of equation (7) at the same time, then invert, and then bring t=2 into the discrete fractional-order open-loop iterative learning controller as,

According to formula (14), formula (18) is simplified, and the common factor is extracted to obtain, δu _k+1 (2)=(Ik _p CB)δu _k (2)+Δ (19) Among them, Δ=-k _p e _k (c ₁ -c ₃ ), c ₁ and c ₃ points are when j=1, 3

the value of ; S45, when t=0, 1, 2 is considered, formulas (15), (16) and (19) are sorted to obtain,

definition

Then equation (20) can be written as, U _k+1 =QU _k . 2. the system that adopts the control method of the open-loop iterative learning based on fractional calculus of claim 1, is characterized in that, comprises discrete fractional open-loop iterative learning controller, current regulator, DSP microprocessor, inverter circuit and position sensor, where, The position sensor collects the position information of the permanent magnet synchronous motor, and inputs the difference between the given position θ ^* and the position θ fed back by the position sensor into the discrete fractional open-loop iterative learning controller, and the discrete fractional open-loop iterative learning The output of the controller is the torque or current command signal

The reference value of 0; The current regulator includes a q-axis current regulator and a _d -axis current regulator, which combine i _q , id and

Compare the difference obtained, input the q-axis current regulator and the d-axis current regulator respectively, and output the voltage control variables _ud and u _q ; The DSP microprocessor includes a PARK inverse converter, a PARK converter, a CLARK converter and a SVPWM generator, a q-axis current regulator and a d-axis current regulator, and the output voltage control quantities _ud and u _q are input to the PARK inverse converter , converted to the voltage control quantities u _α and u _β in the αβ coordinate system, and then the SVPWM generator generates pulse-modulated PWM signals according to u _α and _{u β} , and controls the inverter circuit to generate three-phase voltage signals Va, V _b , V _c control the permanent magnet synchronous motor; convert V _a and V _b into current signals i _a , i _b through the CLARK converter to obtain the currents i _α and i _β in the αβ coordinate system, and then obtain i through the PARK converter _q , _id .

Images