CN102710212A

CN102710212A - Improved iterative learning control method and control system for permanent magnet linear synchronous motor

Info

Publication number: CN102710212A
Application number: CN2012101905219A
Authority: CN
Inventors: 党选举; 刘振丙; 姜辉; 朱骁; 杨青; 龙超
Original assignee: Guilin University of Electronic Technology
Current assignee: Guilin University of Electronic Technology
Priority date: 2012-06-11
Filing date: 2012-06-11
Publication date: 2012-10-03

Abstract

The present invention is an improved iterative learning control method and control system for a permanent magnet synchronous linear motor. The control method adopts the iterative control law algorithm of the superposition of the time axis and the iterative axis to obtain the control voltage of the permanent magnet synchronous linear motor stator; in the iterative control law An initial control variable u ₀ is introduced on the time axis of the algorithm, and an adaptive factor α is designed; the control voltage of the permanent magnet synchronous linear motor k+1 iterations of the iterative learning law

The control voltage also increases the amount of feedforward control. The control system includes an embedded controller, a power drive module connected to the PMLSM stator, and a mover displacement sensor installed on the PMLSM. The invention avoids the oscillation and oscillation of the initial iterative trajectory, accelerates the iterative convergence speed, and improves the control precision by 0.55%.

Description

Improved Iterative Learning Control Method and Control System for Permanent Magnet Synchronous Linear Motor

（一）技术领域 (1) Technical field

本发明涉及永磁同步直线电机的智能控制技术领域，具体为一种针对永磁同步直线电机往复运动中出现的周期性干扰，通过对历史数据迭代学习的一种永磁同步直线电机改进的迭代学习控制方法与控制系统。The invention relates to the technical field of intelligent control of permanent magnet synchronous linear motors, in particular to an improved iteration of permanent magnet synchronous linear motors through iterative learning of historical data for periodic interference occurring in the reciprocating motion of permanent magnet synchronous linear motors Learn control methods and control systems.

（二）背景技术 (2) Background technology

永磁同步直线电机(Permanent Magnet Linear Synchronous Motor，PMLSM)采用直接驱动，简化了齿轮、滚珠与螺杆等设备，所以其反应速度更快，灵敏度更高，随动性更好，可实现超高速运动。并在定位精度、效率等方面也比其它电机具有更多的优势。由于采用直接驱动，系统参数摄动、负载扰动以及外部非线性扰动等不确定因素将直接影响直线电机的静、动态特性，增加了控制的难度，特别是在高速运动的控制过程中，要保证高精度跟踪，更加不易。Permanent Magnet Linear Synchronous Motor (PMLSM) adopts direct drive, which simplifies equipment such as gears, balls and screws, so it has faster response speed, higher sensitivity, better follow-up performance, and can realize ultra-high-speed motion . And it also has more advantages than other motors in terms of positioning accuracy and efficiency. Due to the use of direct drive, uncertain factors such as system parameter perturbation, load disturbance and external nonlinear disturbance will directly affect the static and dynamic characteristics of the linear motor, which increases the difficulty of control, especially in the control process of high-speed motion. High-precision tracking is even more difficult.

迭代学习控制是智能控制的一个分支，它适用于具有重复运动性质的被控对象，其控制能够充分借助历史控制信息，构成当前控制输入且不依赖被控系统的准确模型，只根据系统的历史实际输出信号和期望输出信号的偏差，求取理想的控制信号，使得被控系统的实际输出轨迹在有限的时间区间上沿着期望输出的整个轨迹，实现零偏差的完全跟踪。Iterative learning control is a branch of intelligent control. It is suitable for controlled objects with repetitive motion properties. Its control can make full use of historical control information to form the current control input and does not depend on the accurate model of the controlled system. It is only based on the history of the system. The deviation between the actual output signal and the expected output signal is calculated to obtain the ideal control signal, so that the actual output trajectory of the controlled system follows the entire trajectory of the expected output in a limited time interval to achieve complete tracking with zero deviation.

永磁同步直线电机的迭代学习控制从迭代轴和时间轴两个方向同时进行，两轴之间互相影响，其中迭代轴上的过去时刻的电压控制量值影响着时间轴上的当前电压控制量。The iterative learning control of the permanent magnet synchronous linear motor is carried out simultaneously from the iteration axis and the time axis, and the two axes affect each other, in which the voltage control value at the past time on the iteration axis affects the current voltage control value on the time axis .

在永磁同步直线电机的一个控制周期内，根据迭代轴的前一次迭代，即k次迭代，计算当前次迭代的控制量，即k+1次迭代的控制电压，该控制电压是在一个周期内的系列（一组）值。In one control cycle of the permanent magnet synchronous linear motor, according to the previous iteration of the iteration axis, that is, k iterations, the control amount of the current iteration is calculated, that is, the control voltage of k+1 iterations, and the control voltage is in one cycle A series (group) of values within a .

传统的比例P型迭代算法在时间轴上仅表现为对偏差信息的比例控制，提供的信息相对较少。比例微分PD型迭代算法针对P型迭代算法信息量少，在时间轴上引入了动子目标位移与动子实际位移偏差的导数信息，但在迭代轴上未作改变，这样做只能在一定程度上加快系统的跟踪速度，如果要实现系统的高速跟踪，必须在迭代轴上也做出相应的改进。The traditional proportional P-type iterative algorithm only shows proportional control of deviation information on the time axis, and provides relatively little information. Proportional-differential PD-type iterative algorithm for P-type iterative algorithm with less information, introduces the derivative information of the deviation between the target displacement of the mover and the actual displacement of the mover on the time axis, but does not change the iterative axis, which can only be done in a certain To speed up the tracking speed of the system to a certain extent, if the high-speed tracking of the system is to be realized, corresponding improvements must be made on the iteration axis.

现有的迭代学习算法，在实际系统面临新的环境和控制任务时，系统必须重新进行学习。这主要体现在对一个新的控制对象初始控制时,初始控制值的选取不含有任何控制经验，故在初始控制阶段跟踪偏差有大幅的摆动，需要耗费相当一段时间，才能有效实现对永磁同步直线电机的跟踪控制。With existing iterative learning algorithms, when the actual system faces new environment and control tasks, the system must re-learn. This is mainly reflected in the initial control of a new control object, the selection of the initial control value does not contain any control experience, so the tracking deviation has a large swing in the initial control stage, and it takes a considerable period of time to effectively realize the permanent magnet synchronization. Tracking control of linear motors.

（三）发明内容 (3) Contents of the invention

本发明的目的是提出一种永磁同步直线电机改进的迭代学习控制方法，对永磁同步直线电机采用时间轴与迭代轴叠加的迭代控制律算法求得电机的控制电压，在时间轴上引入一个初始控制量，并设计了一个自适应因子，初始控制量能有效地抑制迭代开始过程中动子跟踪偏差大幅摆动，实现永磁同步直线电机动子的快速跟踪控制。The purpose of the present invention is to propose an improved iterative learning control method for permanent magnet synchronous linear motors. For permanent magnet synchronous linear motors, the iterative control law algorithm of superimposing the time axis and the iteration axis is used to obtain the control voltage of the motor. An initial control quantity, and an adaptive factor is designed, the initial control quantity can effectively suppress the large swing of the mover tracking deviation in the beginning of the iteration, and realize the fast tracking control of the mover of the permanent magnet synchronous linear motor.

本发明的另一目的是提出实现上述永磁同步直线改进的迭代学习控制方法的永磁同步直线电机迭代学习控制系统。Another object of the present invention is to propose an iterative learning control system for permanent magnet synchronous linear motors that implements the above-mentioned improved iterative learning control method for permanent magnet synchronous linear motors.

考虑到干扰的存在，永磁同步直线电机的电压模型，即直线电机的输入电压及动力学方程表示如下：Considering the existence of interference, the voltage model of the permanent magnet synchronous linear motor, that is, the input voltage and dynamic equation of the linear motor are expressed as follows:

$U u ((t t)) = = {K K}_{e e} \overset{. .}{y the y} ((t t)) + + Ri Ri ((t t)) + + L L \frac{di di ((t t))}{dt dt},, - - - - - - ((11))$

f(t)=K_fi(t) （2）f(t)=K _f i(t) (2)

$f f ((t t)) = = M m \overset{. . . .}{y the y} ((t t)) + + {f f}_{r r} ((y the y)) + + {f f}_{f f} ((\overset{. .}{y the y})) + + {f f}_{Δ Δ},, - - - - - - ((33))$

式中：U(t)是永磁同步直线电机的输入电压，该电压控制电机的运动，也称为电机的控制电压；i(t)是永磁同步直线电机电枢电流；In the formula: U(t) is the input voltage of the permanent magnet synchronous linear motor, which controls the movement of the motor, also known as the control voltage of the motor; i(t) is the armature current of the permanent magnet synchronous linear motor;

分别为永磁同步直线电机动子的位移、速度和加速度；

are the displacement, velocity and acceleration of the permanent magnet synchronous linear motor mover, respectively;

k_e是永磁同步直线电机的反电动势常数；k _e is the counter electromotive force constant of the permanent magnet synchronous linear motor;

R是永磁同步直线电机定子绕组电阻；R is the stator winding resistance of the permanent magnet synchronous linear motor;

L是永磁同步直线电机的电感；L is the inductance of the permanent magnet synchronous linear motor;

k_f是永磁同步直线电机的电磁推力系数；k _f is the electromagnetic thrust coefficient of the permanent magnet synchronous linear motor;

M是永磁同步直线电机动子质量；M is the mass of the permanent magnet synchronous linear motor mover;

f_r(y)是永磁同步直线电机的推力波动；f _r (y) is the thrust fluctuation of the permanent magnet synchronous linear motor;

是永磁同步直线电机的摩擦力；

is the friction force of the permanent magnet synchronous linear motor;

f_Δ是永磁同步直线电机的不确定扰动，如模型误差、参数扰动和系统噪音等所造成的永磁同步直线电机的扰动干扰。f _Δ is the uncertain disturbance of permanent magnet synchronous linear motor, such as the disturbance interference of permanent magnet synchronous linear motor caused by model error, parameter disturbance and system noise.

由于电气时间常数要远小于机械时间常数，电气响应的延迟时间可以忽略不计，因此，从工程应用的角度，可以忽略式（1）的最后一项。根据式（1）、（2）及（3）式，得到永磁同步直线电机的动力学方程为：Since the electrical time constant is much smaller than the mechanical time constant, the delay time of the electrical response can be ignored. Therefore, from the perspective of engineering application, the last term of formula (1) can be ignored. According to the formulas (1), (2) and (3), the dynamic equation of the permanent magnet synchronous linear motor is obtained as:

$\overset{. . . .}{y the y} ((t t)) = = - - \frac{{K K}_{f f} {K K}_{e e}}{RM RM} \overset{. .}{y the y} - - \frac{11}{M m} {f f}_{r r} ((y the y)) - - \frac{11}{M m} {f f}_{f f} ((\overset{. .}{y the y})) - - \frac{11}{M m} {f f}_{Δ Δ} + + \frac{{K K}_{f f}}{RM RM} U u ((t t)),, - - - - - - ((44))$

式中，是电机动子加速度。In the formula, is the motor mover acceleration.

令： ${\overset{*}{f}}_{r} (y) = {Rf}_{r} (y) / K_{f},$ ${\overset{*}{f}}_{f} (\dot{y}) = {Rf}_{f} (\dot{y}) / K_{f},$ ${\overset{*}{f}}_{Δ} = {Rf}_{Δ} / K_{f},$ （4）式简化为：make: ${\overset{*}{f}}_{r} (the y) = {Rf}_{r} (the y) / K_{f},$ ${\overset{*}{f}}_{f} (\dot{the y}) = {Rf}_{f} (\dot{the y}) / K_{f},$ ${\overset{*}{f}}_{Δ} = {Rf}_{Δ} / K_{f},$ (4) can be simplified as:

$\overset{. . . .}{y the y} ((t t)) = = - - \frac{{K K}_{f f} {K K}_{e e}}{RM RM} \overset{. .}{y the y} - - \frac{{K K}_{f f}}{RM RM} {\overset{* *}{f f}}_{r r} ((y the y)) - - \frac{{K K}_{f f}}{RM RM} {\overset{* *}{f f}}_{f f} ((\overset{. .}{y the y})) - - \frac{{K K}_{f f}}{RM RM} {\overset{* *}{f f}}_{Δ Δ} + + \frac{{K K}_{f f}}{RM RM} U u ((t t)) - - - - - - ((55))$

令 $\overset{*}{u} (t) = U (t) - {\overset{*}{f}}_{r} (y) - {\overset{*}{f}}_{f} (\dot{y}) - {\overset{*}{f}}_{Δ},$ 则式（5）简化为make $\overset{*}{u} (t) = u (t) - {\overset{*}{f}}_{r} (the y) - {\overset{*}{f}}_{f} (\dot{the y}) - {\overset{*}{f}}_{Δ},$ Then formula (5) can be simplified as

$\overset{. . . .}{y the y} ((t t)) = = - - \frac{{K K}_{f f} {K K}_{e e}}{RM RM} \overset{. .}{y the y ((t t))} + + \frac{{K K}_{f f}}{RM RM} \overset{* *}{u u} ((t t)),, - - - - - - ((66))$

令a=K_fK_e/(RM)，b＝K_f/(RM)，则有 $\overset{. .}{y} (t) = - a \dot{y} (t) + b \overset{*}{u} (t), - - - (7)$ 对式（7）进行拉普拉斯变换后，得到Let a=K _f K _e /(RM), b=K _f /(RM), then we have $\overset{. .}{the y} (t) = - a \dot{the y} (t) + b \overset{*}{u} (t), - - - (7)$ After performing Laplace transform on formula (7), we get

${y the y ((s the s)) s the s}^{22} = = - - ay ay ((s the s)) s the s + + b b \overset{* *}{u u} ((s the s)),, - - - - - - ((88))$

式（8）中，s为拉普拉斯算子，y(s)与分别为y(t)与

的拉普拉斯变换结果。In formula (8), s is the Laplacian operator, y(s) and are y(t) and

The Laplace transform result of .

将式（8）写成传递函数形式，即得永磁同步直线电机的简化传递函数模型：Write the formula (8) in the form of transfer function, that is, the simplified transfer function model of the permanent magnet synchronous linear motor:

$\frac{y the y ((s the s))}{\overset{* *}{u u} ((s the s))} = = \frac{b b}{{s the s}^{22} + + as as} - - - - - - ((99))$

（9）式可理解为在不受f_r(y)推力波动、

电机摩擦力及f_Δ不确定扰动影响时，永磁同步直线电机的简化传递函数模型。Equation (9) can be interpreted as being affected by f _r (y) thrust fluctuations,

The simplified transfer function model of the permanent magnet synchronous linear motor when the motor friction force and f _Δ are uncertainly affected by the disturbance.

从此可知，永磁同步直线电机可等价为：一个受摩擦力、推力波动和模型误差等因素影响的一阶惯性加积分环节的复杂系统。It can be seen from this that the permanent magnet synchronous linear motor can be equivalent to: a complex system of first-order inertia plus integral link affected by factors such as friction, thrust fluctuation and model error.

本发明提出的永磁同步直线电机改进的迭代学习控制方法，对永磁同步直线电机采用时间轴与迭代轴的叠加的迭代控制律算法求得电机的控制电压，在时间轴上引入一个初始控制量u₀，并设计了一个自适应因子α。永磁同步直线电机k+1次迭代（即当前迭代）的控制电压由下式确定的u_k+1(t)：The improved iterative learning control method of the permanent magnet synchronous linear motor proposed by the present invention adopts the iterative control law algorithm of the superposition of the time axis and the iteration axis for the permanent magnet synchronous linear motor to obtain the control voltage of the motor, and introduces an initial control on the time axis Quantity u ₀ , and designed an adaptive factor α. The control voltage of the permanent magnet synchronous linear motor k+1 iterations (that is, the current iteration) is determined by u _k+1 (t) as follows:

${u u}_{k k + + 11} ((t t)) = = ((11 - - α α)) {u u}_{00} ((t t)) + + {αu αu}_{k k} ((t t)) + + {Γe Γ e}_{k k} ((t t)) + + {Φ Φ \overset{. .}{e e}}_{k k} ((t t)) - - - - - - ((1010))$

$α α = = {e e}^{{- - ξE ξE}_{M m}} - - - - - - ((1111))$

${E E.}_{M m} = = \frac{11}{22} {(({r r}_{k k + + 11} ((t t)) - - {y the y}_{k k + + 11} ((t t))))}^{22} - - - - - - ((1212))$

式中α为自适应因子，其随当前迭代，即k+1次迭代的动子目标位移r_k+1(t)与动子实际位移y_k+1(t)偏差的函数E_M,自适应调整。In the formula, α is an adaptive factor, which follows the current iteration, that is, the function E _M of the deviation between the target displacement r _k+1 (t) of the mover and the actual displacement y _k+1 (t) of the mover in the k+1 iteration. Adapt to adjust.

u_k+1(t)是k+1次迭代的控制电压量，根据（10）式得到在永磁同步直线电机的一个控制周期内的系列（一组）控制电压值。u _k+1 (t) is the control voltage value of k+1 iterations, and a series (group) of control voltage values in one control cycle of the permanent magnet synchronous linear motor can be obtained according to formula (10).

ξ是一个固定系数，其值的大小影响α变化的幅度,其取值范围：0.001～1，通过实验调试的方式得到。ξ is a fixed coefficient, and its value affects the range of α change. Its value range: 0.001～1, which is obtained through experimental debugging.

r_k+1(t)为永磁同步直线电机跟踪定位k+1次迭代的动子目标位移，y_k+1(t)为动子的实际位移。r _k+1 (t) is the target displacement of the mover for k+1 iterations of permanent magnet synchronous linear motor tracking and positioning, and y _k+1 (t) is the actual displacement of the mover.

e_k+1(t)＝r_k+1(t)-y_k+1(t)为永磁同步直线电机跟踪定位k+1次迭代的跟踪偏差。e _k+1 (t)=r _k+1 (t)-y _k+1 (t) is the tracking deviation of the permanent magnet synchronous linear motor tracking and positioning k+1 iterations.

为永磁同步直线电机跟踪定位k+1次迭代的跟踪偏差e_k+1(t)的导数。

is the derivative of the tracking deviation e k+ _{1 (t) of the permanent magnet synchronous linear motor tracking and positioning for k} +1 iterations.

e_k(t)=r_k(t)-y_k(t)为永磁同步直线电机跟踪定位k次迭代的跟踪偏差；

为永磁同步直线电机跟踪定位k次迭代的跟踪偏差e_k(t)的导数。e _k (t)=r _k (t)-y _k (t) is the tracking deviation of the permanent magnet synchronous linear motor tracking and positioning for k iterations;

is the derivative of the tracking deviation e _k (t) for k iterations of permanent magnet synchronous linear motor tracking and positioning.

Γ,Φ是迭代学习律参数，Γ的取值范围为0～0.5，Φ的取值范围为0～0.003，通过实验调试的方法取得其值。Γ and Φ are the parameters of the iterative learning law. The value range of Γ is 0-0.5, and the value range of Φ is 0-0.003. The values are obtained through experimental debugging.

对于周期性目标值，在一个完整周期内产生对应的偏差e_k+1(t)=r_k+1(t)-y_k+1(t)和其导数将其按时间次序存放入迭代轴数据。在当前时刻，在时间轴上对应取出迭代轴上前一次周期内对应时刻的偏差值及其导数的数据，分别记为e_k(t)和

For a periodic target value, the corresponding deviation e _k+1 (t)=r _k+1 (t)-y _k+1 (t) and its derivative are generated in a complete cycle Store it in chronological order into the iteration axis data. At the current moment, on the time axis, the deviation value and its derivative data at the corresponding time in the previous cycle on the iteration axis are correspondingly taken out, which are recorded as e _k (t) and

α值的大小与实际输出值和期望值输出密切相关，在迭代初期，二者差值较大，即E_M较大，

α的值趋于零，那么计算控制电压u_k+1(t)的学习律式（10）中第二项的αu_k(t)趋于零，即第一项(1-α)u₀(t)起主要作用；The value of α is closely related to the actual output value and the expected value output. In the early stage of iteration, the difference between the two is relatively large, that is, E _M is relatively large.

The value of α tends to zero, then the αu _k (t) of the second item in the learning law (10) for calculating the control voltage u _k+1 (t) tends to zero, that is, the first item (1-α)u ₀ (t) plays a major role;

当迭代控制系统运行一段时间后，系统趋于稳定，实际输出和期望输出接近，二者差值缩小，即E_M趋于零，α的值趋近于1，此时计算控制电压u_k+1(t)的学习律式（10）中第一项的(1-α)u₀(t)趋于零，第二项αu_k(t)起主要作用。When the iterative control system runs for a period of time, the system tends to be stable, the actual output is close to the expected output, and the difference between the two decreases, that is, E _M tends to zero, and the value of α tends to 1. At this time, the control voltage u _k+ is calculated In the learning law (10) of ₁ (t), the first item (1-α)u ₀ (t) tends to zero, and the second item αu _k (t) plays a major role.

u₀(t)为初始修正值，它避免迭代轨迹的大幅摆动，减少震荡过度过程，加快迭代收敛速度。u₀(t)的选择不是一个定值，是随着实际位置与期望位置差值而变化的一个动态值。u ₀ (t) is the initial correction value, which avoids the large swing of the iterative trajectory, reduces the excessive oscillation process, and speeds up the iterative convergence speed. The choice of u ₀ (t) is not a fixed value, but a dynamic value that changes with the difference between the actual position and the expected position.

本发明中，u₀(t)的值由传统的比例微分PD（Proportional and Derivative）计算得到，u₀(t)计算式如下：In the present invention, the value of u ₀ (t) is calculated by traditional proportional differential PD (Proportional and Derivative), and the calculation formula of u ₀ (t) is as follows:

${u u}_{00} ((t t)) = = {K K}_{P P} {e e}_{k k + + 11} ((t t)) {K K}_{D D.} {\overset{. .}{e e}}_{k k + + 11} ((t t)),, - - - - - - ((1313))$

其中，K_P，K_D分别是比例和微分系数，取值范围K_P=0.1～1，K_D=0.001～0.003。此式同时考虑了k+1次迭代控制的偏差e_k+1(t)及该偏差的变化率（导数），故选择比例微分计算得到的初始修正值u₀(t)提高了迭代控制的抗干扰能力。Among them, K _P and K _D are proportional and differential coefficients respectively, and the range of values is K _P =0.1～1, and K _D =0.001～0.003. This formula takes into account the deviation e _{k+1 (t) of k} +1 times of iterative control and the rate of change (derivative) of the deviation at the same time, so the initial correction value u ₀ (t) obtained by proportional differential calculation improves the performance of iterative control Anti-interference ability.

在式（10）中，永磁同步直线电机k+1次迭代的控制电压由三个电机控制电压分量的叠加组成：In Equation (10), the control voltage of the permanent magnet synchronous linear motor k+1 iterations consists of the superposition of three motor control voltage components:

第一部分为时间轴上初始修正值的电机控制电压分量u₁:The first part is the motor control voltage component u ₁ of the initial correction value on the time axis:

u₁＝(1-α)u₀(t)，（14）u ₁ =(1-α)u ₀ (t), (14)

第二部分为前一迭代（即k次迭代）的电机控制电压分量u₂:The second part is the motor control voltage component u ₂ of the previous iteration (that is, k iterations):

u₂＝αu_k(t)，（15）u ₂ =αu _k (t), (15)

式中u_k(t)为k次迭代的控制电压。每完成一次迭代，迭代控制中一系列的控制量按照时间次序存储，在下一次迭代时，将对应的相关值取出进行计算。where u _k (t) is the control voltage for k iterations. Every time an iteration is completed, a series of control quantities in the iterative control are stored in time order, and in the next iteration, the corresponding related values are taken out for calculation.

第三部分为经典的比例微分开环迭代控制学习算法所对应的电机控制电压分量u₃:The third part is the motor control voltage component u ₃ corresponding to the classic proportional differential split-loop iterative control learning algorithm:

${u u}_{33} = = {Γe Γe}_{k k} ((t t)) + + {Φ Φ \overset{. .}{e e}}_{k k} ((t t)),, - - - - - - ((1616))$

三个控制电压分量叠加，即k+1次迭代的永磁同步直线电机控制电压：The three control voltage components are superimposed, that is, the permanent magnet synchronous linear motor control voltage of k+1 iterations:

u_k+1(t)=u₁+u₂+u₃，（17）u _k+1 (t)=u ₁ +u ₂ +u ₃ , (17)

为了进一步提高控制精度，式（10）所得的k+1次迭代的永磁同步直线电机控制电压u_k+1(t)的基础上增加前馈控制u_f（t），采用U(t)=u_f(t)+u_k+1(t)作为电机控制电压。In order to further improve the control accuracy, the control voltage u k+ ₁ (t) of the permanent magnet synchronous linear motor of k+1 iterations obtained by formula (10) is added on the basis of feedforward control u _f (t), using U(t) =u _f (t)+u _k+1 (t) is used as the motor control voltage.

根据直线电机简化的传递函数模型式（9）:According to the simplified transfer function model formula (9) of the linear motor:

$\frac{y the y ((s the s))}{\overset{* *}{u u} ((s the s))} = = \frac{b b}{{s the s}^{22} + + as as},,$

得到永磁同步直线电机简化的传递函数的逆模型：The inverse model of the simplified transfer function of the permanent magnet synchronous linear motor is obtained:

$\frac{{s the s}^{22} + + as as}{b b},, - - - - - - ((1818))$

根据经典的前馈控制方法，对应传递函数为：According to the classic feedforward control method, the corresponding transfer function is:

$\frac{{u u}_{f f} ((s the s))}{{r r}_{k k + + 11} ((s the s))} = = \frac{{s the s}^{22} + + as as}{b b},, - - - - - - ((1919))$

根据（19）式，对应时域的u_f（t）计算式为：According to formula (19), the calculation formula of u _f (t) corresponding to the time domain is:

${u u}_{f f} ((t t)) = = \frac{11}{b b} {\overset{. . . .}{r r}}_{k k + + 11} ((t t)) + + \frac{a a}{b b} {\overset{. .}{r r}}_{k k + + 11} ((t t)),, - - - - - - ((2020))$

其中

是电机动子的运动速度目标，

是电机动子的运动加速度目标。in

is the movement speed target of the motor mover,

is the motion acceleration target of the motor mover.

例如：一维永磁同步直线电机的各电机参数如下：For example: the motor parameters of a one-dimensional permanent magnet synchronous linear motor are as follows:

R=18.7Ω,k_f=11.71N/A,k_e=9.6V/m/s,M=0.3kg。根据a=K_fK_e/(RM)，b=K_f/(RM)，可计算出a、b参数值。R=18.7Ω, k _f =11.71N/A, k _e =9.6V/m/s, M=0.3kg. According to a=K _f K _e /(RM), b=K _f /(RM), the parameter values of a and b can be calculated.

本发明设计的实现上述永磁同步直线电机改进的迭代学习控制方法的永磁同步直线电机迭代学习控制系统，包括与永磁同步直线电机（PMLSM）相连接的嵌入式控制器、功率驱动模块、动子位移传感器，所述嵌入式控制器包括中央处理单元CPU、程序存储模块(ROM)、数据存储模块(RAM)、脉宽调制模块(PWM)和数据输入输出模块（I/O）。在程序存储模块中存储有比例微分计算器、计算自适应因子

初始修正值

u_{0} (t) = K_{P} e_{k + 1} (t) + K_{D} {\dot{e}}_{k + 1} (t),

控制电压

U_{k + 1} (t) = (1 - α) u_{0} (t) + {αu}_{k} (t) + {Γe}_{k} (t) + Φ {\dot{e}}_{k} (t)

的程序。所述数据存储模块存储永磁同步直线电机的基本参数、动子位移传感器所测得的动子实际位移、中心处理单元计算所得的动子目标位移、动子实际位移和目标位移的偏差e_k+1(t)及迭代轴上的历史数据e_k(t)、u_k(t)。每完成一次迭代，中央处理单元将迭代控制中一系列的控制量，按照时间次序存入数据存储器中，在下一次迭代时，中央处理单元从数据存储器中将对应的相关值取出进行计算。中心处理单元连接程序存储模块(ROM)和数据存储模块(RAM)，中心处理单元经数据输入输出模块连接脉宽调制模块(PWM)，脉宽调制模块的调制信号接入功率驱动模块中，功率驱动模块连接永磁同步直线电机（PMLSM）定子向其提供控制电压。动子位移传感器的输出信号经数据输入输出模块接入中心处理器，存储于数据存储器。动子位移传感器安装于永磁同步电机。The permanent magnet synchronous linear motor iterative learning control system designed by the present invention to realize the improved iterative learning control method of the above permanent magnet synchronous linear motor includes an embedded controller connected to a permanent magnet synchronous linear motor (PMLSM), a power drive module, The mover displacement sensor, the embedded controller includes a central processing unit CPU, a program storage module (ROM), a data storage module (RAM), a pulse width modulation module (PWM) and a data input and output module (I/O). Stored in the program storage module is a proportional differential calculator and a calculation adaptive factor

initial correction value

u_{0} (t) = K_{P} e_{k + 1} (t) + K_{D.} {\dot{e}}_{k + 1} (t),

control voltage

u_{k + 1} (t) = (1 - α) u_{0} (t) + {αu}_{k} (t) + {Γ e}_{k} (t) + Φ {\dot{e}}_{k} (t)

program of. The data storage module stores the basic parameters of the permanent magnet synchronous linear motor, the actual displacement of the mover measured by the mover displacement sensor, the target displacement of the mover calculated by the central processing unit, and the deviation between the actual displacement of the mover and the target displacement e _{k +1} (t) and historical data e _k (t) and u _k (t) on the iteration axis. Every time an iteration is completed, the central processing unit stores a series of control quantities in the iterative control into the data memory in chronological order, and at the next iteration, the central processing unit takes out the corresponding relevant values from the data memory for calculation. The central processing unit is connected to the program storage module (ROM) and the data storage module (RAM). The central processing unit is connected to the pulse width modulation module (PWM) through the data input and output module. The modulation signal of the pulse width modulation module is connected to the power drive module. The drive module is connected to the permanent magnet synchronous linear motor (PMLSM) stator to provide it with control voltage. The output signal of the mover displacement sensor is connected to the central processor through the data input and output module, and stored in the data memory. The mover displacement sensor is installed on the permanent magnet synchronous motor.

所述嵌入式控制器可采用32位嵌入式控制芯片。The embedded controller can adopt a 32-bit embedded control chip.

所述动子位移传感器可采用光栅位移传感器。光栅位移传感器的两部分分别安装于永磁同步直线电机动子和定子上，精确测定动子位移。The mover displacement sensor can be a grating displacement sensor. The two parts of the grating displacement sensor are respectively installed on the mover and stator of the permanent magnet synchronous linear motor to accurately measure the displacement of the mover.

所述功率驱动模块根据脉宽调制模块输出的控制电压得到对应的驱动信号，实现对永磁同步直线电机的驱动电压的调整，达到控制动子位移的目的。The power drive module obtains a corresponding drive signal according to the control voltage output by the pulse width modulation module, so as to realize the adjustment of the drive voltage of the permanent magnet synchronous linear motor and achieve the purpose of controlling the displacement of the mover.

所述嵌入式控制器的中心处理器还经数据输入输出模块接有显示器，可对本控制系统对永磁同步直线电机的控制效果进行实时监控。The central processor of the embedded controller is also connected with a display through the data input and output module, which can monitor the control effect of the permanent magnet synchronous linear motor by the control system in real time.

在程序存储模块中还存储有前馈控制量的计算程序

增加前馈控制后的电机控制电压的计算程序U(t)=u_f(t)+u_k+1(t)。The calculation program of the feedforward control quantity is also stored in the program storage module

Add the calculation procedure of the motor control voltage U(t)=u _f (t)+u _k+1 (t) after feed-forward control.

本发明永磁同步直线电机改进的迭代学习控制方法与控制系统的优点为：1、是对现有的比例P型迭代算法和比例微分PD型迭代算法的改进，由于引入了初始修正值和自适应因子，避免初期迭代轨迹的大幅摆动，减少过度震荡的过程，加快迭代收敛速度；2、增加了前馈控制，进一步提高了控制精度，控制偏差最大仅为0.11mm，控制精度可达0.55%，而传统迭代学习控制方法最佳情况的最大控制偏差为0.84mm、控制精度4.2%，相比之下本发明的控制精度显著地提高；3、本控制系统可采用通用硬件实现，便于推广应用。The advantages of the improved iterative learning control method and control system of the permanent magnet synchronous linear motor of the present invention are: 1. It is an improvement to the existing proportional P-type iterative algorithm and proportional differential PD-type iterative algorithm. The adaptation factor avoids the large swing of the initial iteration trajectory, reduces the process of excessive oscillation, and speeds up the iteration convergence speed; 2. The feedforward control is added to further improve the control accuracy. The maximum control deviation is only 0.11mm, and the control accuracy can reach 0.55%. , while the maximum control deviation in the best case of the traditional iterative learning control method is 0.84mm, and the control accuracy is 4.2%. In contrast, the control accuracy of the present invention is significantly improved; 3. This control system can be realized by general-purpose hardware, which is convenient for popularization and application .

（四）附图说明 (4) Description of drawings

图1为永磁同步直线电机改进的迭代学习控制方法实施例中永磁同步直线电机模型结构示意图；Fig. 1 is a schematic structural diagram of a permanent magnet synchronous linear motor model in an embodiment of an improved iterative learning control method for a permanent magnet synchronous linear motor;

图2为永磁同步直线电机改进的迭代学习控制系统实施例硬件结构框图；Fig. 2 is the block diagram of the hardware structure of the embodiment of the iterative learning control system improved by the permanent magnet synchronous linear motor;

图3为永磁同步直线电机改进的迭代学习控制系统实施例各控制电压分量组合示意图；Fig. 3 is a schematic diagram of combinations of control voltage components of an embodiment of an improved iterative learning control system for a permanent magnet synchronous linear motor;

图4为永磁同步直线电机改进的迭代学习控制系统实施例电机动子往复式运动控制效果图；Fig. 4 is an effect diagram of the reciprocating motion control of the motor mover in an embodiment of an improved iterative learning control system for a permanent magnet synchronous linear motor;

图5为永磁同步直线电机改进的迭代学习控制系统实施例电机动子往复式运动控制偏差曲线图；Fig. 5 is a graph showing the deviation curve of the motor mover reciprocating motion control embodiment of the improved iterative learning control system of the permanent magnet synchronous linear motor;

图6为永磁同步直线电机传统迭代学习控制系统对比例电机动子往复式运动控制效果图；Fig. 6 is the effect diagram of the reciprocating motion control of the mover of the permanent magnet synchronous linear motor compared with the traditional iterative learning control system of the proportional motor;

图7为永磁同步直线电机传统迭代学习控制系统对比例电机动子往复式运动控制偏差曲线图。Fig. 7 is a deviation curve of the traditional iterative learning control system of the permanent magnet synchronous linear motor compared with the reciprocating motion control of the proportional motor mover.

（五）具体实施方式 (5) Specific implementation methods

永磁同步直线电机迭代学习控制方法实施例Embodiment of Iterative Learning Control Method for Permanent Magnet Synchronous Linear Motor

本例永磁同步直线电机的简化传递函数模型为：The simplified transfer function model of the permanent magnet synchronous linear motor in this example is:

$\frac{y the y ((s the s))}{\overset{* *}{u u} ((s the s))} = = \frac{b b}{{s the s}^{22} + + as as}$

该模型用图1表示， $\overset{*}{u} (t) = U (t) - {\overset{*}{f}}_{r} (y) - {\overset{*}{f}}_{f} (\dot{y}) - {\overset{*}{f}}_{Δ},$ 本例控制方法的控制电压U(t)=u₁+u₂+u₃+u_f（t），为4个电压分量的和，其计算如下：The model is shown in Figure 1, $\overset{*}{u} (t) = u (t) - {\overset{*}{f}}_{r} (the y) - {\overset{*}{f}}_{f} (\dot{the y}) - {\overset{*}{f}}_{Δ},$ The control voltage U(t)=u ₁ +u ₂ +u ₃ +u _f (t) of the control method in this example is the sum of 4 voltage components, and its calculation is as follows:

u₁=(1-α)u₀(t)，u ₁ =(1-α)u ₀ (t),

自适应因子 $α = e^{{- ξE}_{M}},$ adaptive factor $α = e^{{- ξE}_{m}},$

${E E.}_{M m} = = \frac{11}{22} {(({r r}_{k k + + 11} ((t t)) - - {y the y}_{k k + + 11} ((t t))))}^{22},,$

本例取ζ=0.01，ζ的取值是通过实验调试的方式得到的。In this example, ζ=0.01, the value of ζ is obtained through experimental debugging.

则 $u_{0} (t) = K_{P} e_{k + 1} (t) + K_{D} {\dot{e}}_{k + 1} (k + 1),$ but $u_{0} (t) = K_{P} e_{k + 1} (t) + K_{D.} {\dot{e}}_{k + 1} (k + 1),$

式中，K_P，K_D为PD算法中比例和微分系数。本例通过实验调试K_P=0.2，K_D=0.0025。e_k+1(t)是永磁同步直线电机跟踪定位的动子目标位移r_k+1(t)与输出为动子的实际位移y_k+1(t)的偏差值。In the formula, K _P and K _D are the proportional and differential coefficients in the PD algorithm. In this example, K _P =0.2 and K _D =0.0025 are adjusted through experiments. e _k+1 (t) is the deviation value between the target displacement r _k+1 (t) of the mover for tracking and positioning of the permanent magnet synchronous linear motor and the output is the actual displacement y _k+1 (t) of the mover.

u₂=αu_k(t)，α的计算方法同上，u_k(t)为迭代轴上按时间次序存放的k次迭代对应的控制电压。u ₂ =αu _k (t), the calculation method of α is the same as above, u _k (t) is the control voltage corresponding to k iterations stored in time order on the iteration axis.

${u u}_{33} = = {Γe Γe}_{k k} ((t t)) + + {Φ Φ \overset{. .}{e e}}_{k k} ((t t)),,$

式中Γ，Φ为学习律参数，本例经实验调试分别取Γ=0.05，Φ=0.0015。e_k(t)，

分别为迭代轴上k次迭代对应的位置偏差信息和其导数，通过迭代轴的数据存储器得到它们的值。In the formula, Γ and Φ are the parameters of the learning law. In this example, Γ=0.05 and Φ=0.0015 are respectively selected through experimental debugging. e _k (t),

are respectively the position deviation information corresponding to k iterations on the iteration axis and its derivative, and their values are obtained through the data memory of the iteration axis.

${u u}_{f f} ((t t)) = = \frac{11}{b b} {\overset{. . . .}{r r}}_{k k + + 11} ((t t)) + + \frac{a a}{b b} {\overset{. .}{r r}}_{k k + + 11} ((t t)),,$

本例电机的各参数分别为K_f＝11.71N/A，K_e=9.6V/m/s，R=18.7Ω，The parameters of the motor in this example are K _f =11.71N/A, K _e =9.6V/m/s, R=18.7Ω,

M=0.3kg，M=0.3kg,

$a a = = \frac{{K K}_{f f} {K K}_{e e}}{RM RM} = = 2020,,$ $b b = = \frac{{K K}_{f f}}{RM RM} = = 2.08 2.08 . .$

永磁同步直线电机迭代学习控制系统实施例Embodiment of iterative learning control system for permanent magnet synchronous linear motor

永磁同步直线电机迭代学习控制系统实施例的硬件结构如图2所示，包括与永磁同步直线电机（PMLSM）相连接的嵌入式控制器、功率驱动模块、动子位移传感器和显示器，所述嵌入式控制器包括中央处理单元CPU、程序存储模块(ROM)、数据存储模块(RAM)、脉宽调制模块(PWM)和数据输入输出模块（I/O）。在程序存储模块中存储有比例微分计算器和计算自适应因子

初始修正值控制电压

u_{k + 1} (t) = (1 - α) u_{0} (t) + {αu}_{k} (t) + {Γe}_{k} (t) + {Φ \dot{e}}_{k} (t)

的程序。所述数据存储模块存储永磁同步直线电机（PMLSM）的基本参数、动子位移传感器所测得的动子实际位移、中心处理单元计算所得的动子目标位移、动子实际位移和目标位移的偏差e_k+1(t)及迭代轴上的历史数据e_k(t)。每完成一次迭代，中央处理单元将迭代控制中一系列的控制量，按照时间次序存入数据存储器中，在下一次迭代时，中央处理单元从数据存储器中将对应的相关值取出进行计算。中心处理单元连接程序存储模块(ROM)和数据存储模块(RAM)，中心处理单元经数据输入输出模块连接脉宽调制模块(PWM)，PWM的调制信号接入功率驱动模块，功率驱动模块连接永磁同步直线电机定子向其提供控制电压。本例的动子位移传感器为光栅位移传感器，光栅位移传感器的两部分分别安装于永磁同步直线电机动子和定子上。动子位移传感器的输出信号经数据输入输出模块接入中心处理器，存储于数据存储器中。The hardware structure of the embodiment of the permanent magnet synchronous linear motor iterative learning control system is shown in Figure 2, including an embedded controller connected to the permanent magnet synchronous linear motor (PMLSM), a power drive module, a mover displacement sensor and a display. The embedded controller includes a central processing unit CPU, a program storage module (ROM), a data storage module (RAM), a pulse width modulation module (PWM) and a data input and output module (I/O). The proportional differential calculator and the calculation adaptive factor are stored in the program storage module

initial correction value control voltage

u_{k + 1} (t) = (1 - α) u_{0} (t) + {αu}_{k} (t) + {Γ e}_{k} (t) + {Φ \dot{e}}_{k} (t)

program of. The data storage module stores the basic parameters of the permanent magnet synchronous linear motor (PMLSM), the actual displacement of the mover measured by the mover displacement sensor, the target displacement of the mover calculated by the central processing unit, the actual displacement of the mover and the target displacement Deviation e _k+1 (t) and historical data e _k (t) on the iteration axis. Every time an iteration is completed, the central processing unit stores a series of control quantities in the iterative control into the data memory in chronological order, and at the next iteration, the central processing unit takes out the corresponding relevant values from the data memory for calculation. The central processing unit is connected to the program storage module (ROM) and the data storage module (RAM). The central processing unit is connected to the pulse width modulation module (PWM) through the data input and output module. The PWM modulation signal is connected to the power drive module, and the power drive module is connected to the permanent The stator of the magnetic synchronous linear motor provides control voltage to it. The mover displacement sensor in this example is a grating displacement sensor, and the two parts of the grating displacement sensor are respectively installed on the mover and stator of the permanent magnet synchronous linear motor. The output signal of the mover displacement sensor is connected to the central processor through the data input and output module, and stored in the data memory.

所述嵌入式控制器的中心处理器还经数据输入输出模块接有显示器，本例的显示器为LCD显示器。The central processing unit of the embedded controller is also connected with a display through the data input and output module, and the display of this example is an LCD display.

本例的嵌入式控制器采用32位嵌入式控制芯片。The embedded controller in this example uses a 32-bit embedded control chip.

本例的各控制电压分量的叠加原理如图3所示，嵌入式控制器所得的控制电压U(t)送入功率驱动模块，功率驱动模块内有绝缘栅双极型功率管IGBT（Insulated Gate Bipolar Transistor)，通过IGBT的开关状态，控制电压U(t)直接加在永磁同步直线电机的定子上，达到克服干扰，控制动子位移的目的。The superimposition principle of each control voltage component in this example is shown in Figure 3. The control voltage U(t) obtained by the embedded controller is sent to the power drive module, which has an insulated gate bipolar power transistor IGBT (Insulated Gate Bipolar Transistor), through the switching state of the IGBT, the control voltage U(t) is directly added to the stator of the permanent magnet synchronous linear motor to overcome interference and control the displacement of the mover.

通过动子位移传感器采集到电机动子的位置信息并反馈回32位嵌入式控制器的CPU，通过CPU计算出动子位置偏差并且再次计算各控制电压分量，通过对控制电压量不断的修正，实现对永磁同步直线电机位移的精确控制，反复学习过程体现了迭代学习的优化过程。The position information of the motor mover is collected by the mover displacement sensor and fed back to the CPU of the 32-bit embedded controller. The position deviation of the mover is calculated by the CPU and each control voltage component is calculated again. Through continuous correction of the control voltage, the realization For the precise control of the displacement of the permanent magnet synchronous linear motor, the repeated learning process embodies the optimization process of iterative learning.

本例电机往复式运动行程20mm，本发明增加了前馈控制的永磁同步直线电机改进的迭代学习控制系统实施例的动子往复运动的控制效果如图4所示，其横坐标为时间、单位为秒，纵坐标为动子位置、单位为毫米，此图内实线表示期望的动力位置，虚线表示实际动子位置，二者基本重合，图4无法分辨两条曲线。图5所示为本实施例的动子往复运动位置的偏差，图中可见其最大偏差仅0.11mm，跟踪过程中动子速度为零时偏差最大，此时摩擦力为最大的静摩擦力，故体现为位置偏差的峰值；图5中可见本实施例动子位置控制精度达0.55%。相同的永磁同步直线电机采用传统迭代学习控制系统进行控制的对比例效果如图6、图7所示，图6中可见表示期望的动力位置的实线和表示实际动子位置的虚线在控制开始0至2秒内差异较大，由图7可见在控制开始0至1秒内，位置偏差极大、达到20mm，虽在2秒后逐渐减小，但其最大偏差仍达0.84mm，精度仅为4.2%。本发明实施例的效果显著。This example motor reciprocating motion stroke 20mm, the present invention has increased the permanent magnet synchronous linear motor improved iterative learning control system embodiment of feedforward control The control effect of the mover reciprocating motion is as shown in Figure 4, and its abscissa is time, The unit is seconds, the ordinate is the position of the mover, and the unit is mm. The solid line in this figure indicates the expected power position, and the dotted line indicates the actual position of the mover. The two basically coincide, and the two curves cannot be distinguished in Figure 4. Figure 5 shows the deviation of the reciprocating motion position of the mover in this embodiment. It can be seen from the figure that the maximum deviation is only 0.11 mm. The deviation is the largest when the speed of the mover is zero during the tracking process. At this time, the friction force is the largest static friction force, so It is reflected as the peak value of the position deviation; it can be seen from Fig. 5 that the position control accuracy of the mover in this embodiment reaches 0.55%. The comparative effect of the same permanent magnet synchronous linear motor controlled by traditional iterative learning control system is shown in Fig. The difference is relatively large in the first 0 to 2 seconds. It can be seen from Figure 7 that the position deviation is extremely large within 0 to 1 second of the control, reaching 20mm. Although it gradually decreases after 2 seconds, the maximum deviation still reaches 0.84mm. Only 4.2%. The effect of the embodiment of the present invention is remarkable.

上述实施例，仅为对本发明的目的、技术方案和有益效果进一步详细说明的具体个例，本发明并非限定于此。凡在本发明的公开的范围之内所做的任何修改、等同替换、改进等，均包含在本发明的保护范围之内。The above-mentioned embodiments are only specific examples for further specifying the purpose, technical solutions and beneficial effects of the present invention, and the present invention is not limited thereto. Any modifications, equivalent replacements, improvements, etc. made within the disclosed scope of the present invention are included in the protection scope of the present invention.

Claims

1. the improved iterative learning control method of permanent magnetic linear synchronous motor, the input voltage of said permanent magnetic linear synchronous motor and kinetics equation are:

In the formula: U (t) is the input voltage of permanent magnetic linear synchronous motor;

I (t) is the permanent magnetic linear synchronous motor armature supply;

is respectively displacement and the speed that permanent magnetic linear synchronous motor has moved;

k _eIt is the back electromotive force constant of permanent magnetic linear synchronous motor;

R is that permanent magnetic linear synchronous motor has been decided winding resistance;

L is the inductance of permanent magnetic linear synchronous motor;

k _fIt is the electromagnetic push coefficient of permanent magnetic linear synchronous motor;

M is that permanent magnetic linear synchronous motor has moved quality;

f _r(y) be the force oscillation of permanent magnetic linear synchronous motor;

is the frictional force of permanent magnetic linear synchronous motor;

f _ΔIt is the uncertain disturbance of permanent magnetic linear synchronous motor;

is that motor has moved acceleration;

Order:

α=K _fK _e/ (RM), b=K _f/ (RM), obtain

After carrying out Laplace transform; Obtain

s and let it pass for Laplce, y (s) is respectively the Laplace transform result of y (t) with

with

; The simplification transfer function model of permanent magnetic linear synchronous motor does

This method is tried to achieve the control voltage of permanent magnetic linear synchronous motor stator for the iteration control rule algorithm of the stack of employing time shaft and iteration axle; It is characterized in that:

On the time shaft of said iteration control rule algorithm, introduce an initial controlled quentity controlled variable u ₀, and designed the self adaptation factor-alpha; The control voltage of the permanent magnetic linear synchronous motor k+1 iteration of iterative learning control law is the u that following formula is confirmed _K+1(t):

ξ is a fixed coefficient, span: 0.001～1,

r _K+1(t) be the mover displacement of targets of permanent magnetic linear synchronous motor track and localization k+1 iteration,

y _K+1(t) be the actual displacement of mover,

e _K+1(t)=r _K+1(t)-y _K+1(t) be the tracing deviation of permanent magnetic linear synchronous motor track and localization k+1 iteration,

Tracing deviation for permanent magnetic linear synchronous motor track and localization k+1 iteration

e _K+1(t)=r _K+1(t)-y _K+1(t) derivative,

Γ, Φ are the iterative learning control law parameters, are normal value.

2. the improved iterative learning control method of permanent magnetic linear synchronous motor according to claim 1 is characterized in that:

Said initial controlled quentity controlled variable u ₀Try to achieve by following formula:

K in the formula _P, K _DBe respectively ratio and differential coefficient.

3. the improved iterative learning control method of permanent magnetic linear synchronous motor according to claim 1 and 2 is characterized in that:

Said iterative learning control law parameter Γ=0～0.5, Φ=0～0.003.

4. the improved iterative learning control method of permanent magnetic linear synchronous motor according to claim 2 is characterized in that:

Said proportionality coefficient K _P=0.1～1, differential coefficient K _D=0.001～0.003.

5. the improved iterative learning control method of permanent magnetic linear synchronous motor according to claim 2 is characterized in that:

Said control voltage also increases the feedfoward control amount, and promptly Electric Machine Control voltage does

U(t)=u _f(t)+u _k+1(t)，

Wherein

is the movement velocity target of electric mover, and

is the acceleration of motion target of electric mover.

6. the improved iterative learning control systems of permanent magnetic linear synchronous motor of the improved iterative learning control method design of permanent magnetic linear synchronous motor according to claim 2 is characterized in that:

Comprise the embedded controller, power driver module, the mover displacement transducer that are connected with permanent magnetic linear synchronous motor, said embedded controller comprises CPU, program storage block, data memory module, pulse width modulation module and data input/output module; In program storage block, store the proportion differential calculator, calculate the self adaptation factor

The initial correction value

Control voltage

Program; The measured mover actual displacement of basic parameter, the mover displacement transducer of said data memory module storage permanent magnetic linear synchronous motor, the deviation e that center processing unit calculates mover displacement of targets, mover actual displacement and the displacement of targets of gained _K+1(t) and the historical data e on the iteration axle _k(t); Iteration of every completion, CPU deposits a series of controlled quentity controlled variable in the iteration control in the data storage in according to chronological order, and when next iteration, CPU is calculated the correlation taking-up of correspondence from data storage; Center processing unit linker memory module and data memory module; Center processing unit connects pulse width modulation module through the data input/output module; Pulse width modulation module is connected to power driver module, and power driver module connects the permanent magnetic linear synchronous motor stator provides control voltage to it; The output digital signal of mover displacement transducer inserts center processor through the data input/output module, and the mover displacement transducer is installed on permanent magnetic linear synchronous motor.

7. permanent magnetic linear synchronous motor iterative learning control systems according to claim 3 is characterized in that:

Said mover displacement transducer is a grating displacement sensor, and two parts of grating displacement sensor are installed on respectively on permanent magnetic linear synchronous motor mover and the stator.

8. permanent magnetic linear synchronous motor iterative learning control systems according to claim 3 is characterized in that:

Said embedded controller is 32 embedded control chip.

9. permanent magnetic linear synchronous motor iterative learning control systems according to claim 3 is characterized in that:

Said embedded controller also is connected to display.

10. permanent magnetic linear synchronous motor iterative learning control systems according to claim 3 is characterized in that:

Also store the feedfoward control amount in the said program storage block

Calculation procedure, increase Electric Machine Control voltage U (the t)=u after the feedfoward control _f(t)+u _K+1(t).