JP2010130852A - Motor controller and estimation control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a motor controller having high followability at a constant speed by reducing a feedforward amount in accordance with a machine to follow a command, simplifying control operation to reduce overshoot in acceleration and deceleration, and to provide an estimation control method. <P>SOLUTION: The motor controller includes: a position command difference buffer section (21) for storing position commands per sampling time; and an estimation operating section (22) for generating a speed command based on a position command of the position command difference buffer section and a position deviation of the position of the motor. The position command difference buffer section has two stages when using only speed feedforward, and four stages when using only torque feedforward. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a predictive control method and an electric motor control device including the predictive control method, and predicts a future position command from a current position command, and controls so that position feedback follows the position command.

The conventional predictive control method generates a target command value from the current time to M step ahead at the sampling time when movement starts, and generates only a target command value M steps ahead from the current time after the next sampling time. Foreseeing (synonymous with prediction) control calculation is performed (for example, see Patent Document 1).
In FIG. 8, 21 is a command generator, 22 is a memory for storing command increment values, 23 is a memory for storing preview control parameters, 24 is a memory for storing output increment values, and 25 is a memory for storing past control inputs. , 28 is an arithmetic unit, and a memory 22 for storing a command increment value stores a target command increment value from the current time to M steps ahead in the memory at the first sampling time for performing the preview control by a target command switching signal. After the next sampling time, the memory contents are shifted one sampling past from the current time, and a future command increment value is newly stored. The computing unit 28 calculates the control input v (i) by the equation (1). The calculated v (i) is stored in the memory 25 and output to the motor controller for control.
As described above, the conventional control method generates a target command value from the current time to M step ahead at the sampling time when movement starts, and only the target command value of M step ahead from the current time after the next sampling time. It is generated and the predictive control is calculated.
JP-A-8-123537 (page 4, FIG. 2)

The conventional method of controlling the position so as to follow the command requires a target command value of M steps ahead and a correction command value of N-1 steps in predictive (predictive) control, and the control calculation is complicated. The follow-up to the position is determined by the target command value M steps ahead and the correction command value before the N-1 step, and since there is no adjustment of the feedforward amount, there is a problem that the follow-up to the command cannot be matched with the machine.
The present invention has been made in view of such problems. The feedforward amount is adjusted in accordance with the machine so as to follow the command, the control calculation is simplified, the overshoot at the time of acceleration / deceleration is reduced, and constant. It is an object of the present invention to provide an electric motor control device and a predictive control method with improved followability during speed.

In order to solve the above problem, the present invention is configured as follows.
The invention according to claim 1 is a position command difference buffer unit that stores a position command for each sampling time, and a prediction calculation that generates a speed command based on the position command of the position command difference buffer unit and the position deviation of the motor position. A speed control unit that generates a torque command based on a speed deviation between the speed command and the motor speed, a motor drive unit that drives the motor based on the torque command, and a time difference between the motor positions to determine the motor speed. When using only the speed feedforward in an electric motor control device including an electric motor speed generation unit to be generated, the electric motor control device includes a position command difference buffer unit having two stages of the position command difference buffer.
In the invention according to claim 2, the position command difference buffer unit has four stages of position command difference buffers when using torque feed forward.

According to a third aspect of the present invention, the position command difference buffer unit uses the same position command difference without using the position command difference buffer when feedforward is not used.
According to a fourth aspect of the present invention, a position command difference buffer unit that stores a position command for each sampling time, and a speed command is generated based on the position command of the position command difference buffer unit and the position deviation of the motor position. A prediction calculation unit; a speed control unit that generates a torque command based on a speed deviation between the speed command and the motor speed; a motor drive unit that drives the motor based on the torque command; and a time difference between the motor positions. In a control method of an electric motor control device including an electric motor speed generation unit that generates a speed, when only speed feedforward is used, the position command difference buffer is stored in two stages and the position command difference is stored. .

According to a fifth aspect of the present invention, when using torque feed forward, the position command difference buffer is stored in four stages and the position command difference is stored.
According to the sixth aspect of the present invention, when feedforward is not used, the position command difference is not used but the prediction control is performed with the same position command difference.

  According to the present invention, it is possible to provide an electric motor control device and a predictive control method in which the position command difference buffer is reduced to simplify the calculation, the overshoot at the time of acceleration / deceleration is reduced, and the followability at a constant speed is improved.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of an electric motor control device of the present invention. In the figure, 1 is an electric motor control device, 2 is an electric motor, 3 is a position detector, 11 is a prediction control unit, 12 is a speed control unit, 13 is a current control unit, 14 is a power conversion unit, 15 is a speed generation unit, 16 Is an encoder signal input / output unit.
The prediction control unit 11 generates a speed command based on the position command and the position deviation of the motor position, and the speed control unit 12 performs proportional integral control on the speed command and the speed deviation of the motor speed to generate a torque command. The current control unit 13 converts the torque command into a current command, and generates a voltage command based on a current deviation between the current command and the motor current. The power conversion unit 14 converts the voltage command into a PWM signal, drives the power converter, and flows a current according to the current command to the motor. The position detector 3 is coupled to the electric motor 2 and detects the position of the electric motor. The speed generation unit 15 generates a motor speed by taking the time difference of the motor position.
A characteristic part of the present invention is that the position command difference buffer of the prediction control unit 11 is set to two stages when only speed feedforward is used or four stages when torque feedforward is used.

FIG. 2 is a block diagram illustrating a configuration of the prediction control unit 11. In the figure, 21 is a command buffer unit, 22 is a prediction calculation unit, and 23 is a feedforward calculation unit.
Here, the predictive control of the present invention will be described.
In the predictive control of the present invention, at the current time i · Ts (time i, sampling period Ts), the target command increment value Δr (i + M) of M sampling future and the position increment value y ( Using i) as an input, the speed command v _ref as a control input is given as follows. When only speed feedforward is used, equation (1) is used, and when torque feedforward is used, equation (2) is obtained.

Here, Δr (i) is a position command difference, Δy (i) is a position difference, e (i) is a position deviation, v _FF (i) is a speed feedforward, and v _FF = ffv × Δr (i + 1), t _FF (i) is torque feed forward and tFF (i) = ffa × Δr (i + 4). ffv and ffa are feed forward gains, respectively, and vm, E, p ₀ , g ₁ , x _n and t _n are constants calculated in advance.
When only the speed feedforward is used, the command buffer 21 performs the two-stage buffer processing of the position command difference in order to calculate the equation (3).

When torque feed forward is used, a 4-step buffer process for position command difference is performed in order to calculate equation (4).

Moreover, the prediction calculation part 22 performs the calculation process of Formula (1). The feedforward calculation unit 23 performs a feedforward calculation of speed feedforward v _FF (i) and torque feedforward t _FF (i).

FIG. 3 is a block diagram showing a detailed configuration for the prediction control unit 11 to further execute the calculation of the expression (1) or (2). In the figure, 21 is a command buffer unit, 22 is a prediction calculation unit, and 23 is a feedforward calculation unit. The processing is the same as described above.
Predictive control is based on the control input so that the future deviation predicted value obtained using the transfer function model from the speed command that is the control target input to the output position, the deviation, and the evaluation function related to the control input are minimized. Is a speed command.
Using the position command increment Δr (i), the future deviation predicted value e ^* (i + M) is given by equation (5), and the evaluation function is given by equation (6).

Here, M is a prediction function, e ^* (i + M) is a future prediction deviation M samples ahead of the control period i, e is a deviation, and C, C _d , and α are adjustment coefficients.
(6) The control input v _ref (i) that minimizes the evaluation function of the equation is obtained and controlled.
The speed command v _ref is given by a predictive control expression of Expression (7).

FIG. 6 is a block diagram showing a detailed configuration for the prediction control unit 11 to further execute the calculation of Equation (7). In the figure, 21 is a command buffer unit, 22 is a prediction calculation unit, and 23 is a feedforward calculation unit.
The difference from the present invention is that the command buffer 21 performs M-stage buffer processing of the position command difference in order to perform the calculation of equation (8), and the prediction calculation unit 22 performs the calculation processing of equation (7). The feedforward calculation unit 23 performs the same process as described above.

The command buffer 21 has an M-stage buffer for position command difference, but if the number of stages of the buffer is large, the speed command becomes gradual at the time of transition.

FIG. 4 is a diagram showing the relationship between the position command difference buffer size and the speed command depending on the control cycle. FIG. 4A shows a speed command in the case of a position command differential buffer of about 8 ms and a control cycle of about 8 ms. It can be seen that the position command is delayed from the position command at a constant speed after acceleration. A large feedforward is required to make this follow the position command, but a large feedforward leads to an overshoot. FIG. 4B shows a speed command in the case of a position command difference buffer of about 0.5 ms as a control cycle and about two stages, and it can be seen that the position command follows from the position command even at a constant speed from the acceleration. For further tracking, if the control cycle is shortened to 0.25 ms as shown in FIG. 4C, the speed command further follows the position command. This is the weight of the command

This is because the future is larger as v ₁ <v ₂ <v ₃ . That weight v part multiplied by _m and the position command difference command becomes (9), in order to approach the position command difference shorter prediction interval.
Further, when the control period is shortened, in terms of the position command difference is constant speed is a portion obtained by multiplying the position command difference between the weight v _m command (10) is larger because the position command difference becomes closer to the predetermined speed position This is to approach the command difference.

Thus, since a larger feed forward is not required than in the case of many position command differential buffers, overshoot and undershoot are difficult to follow even if the position command is followed.

  FIG. 5 is a flowchart showing the control method of the present invention. In the figure, in step S1, the command buffer of equation (11) is calculated when only speed feedforward is used, and the command buffer of equation (12) is calculated when torque feedforward is used.

Next, in step S2, the position deviation e (i) is calculated from e (i) = Δr (i) −Δy (i). In step S3, speed feed forward is calculated from equation (13).

In step S4, when using torque feed forward, torque feed forward is calculated using equation (14).

In step S5, an equation (1) or equation (2) speed command is calculated from the calculation results of steps S1 to S4. In step S 6, the position command difference Δr (i) is stored in the command buffer 21. The calculation of predictive control is performed every sampling period, that is, every control period.

  In this way, in the predictive control using the position command difference buffer as two or four stages, overshoot and undershoot can be prevented during positioning even if the position command is followed.

FIG. 7 is a block diagram showing in detail a configuration for the prediction control unit 11 of the second embodiment to execute the calculation of the following equation (15). In the figure, 22 is a prediction calculation unit. The prediction calculation unit 22 executes the calculation of equation (15). A different part from Example 1 is a part which does not have an instruction | command buffer part and a feedforward calculating part. The speed command v _ref is given by the predictive control expression of Expression (15).

Since the expression (16) in the first term of the expression (15) is a constant that is calculated in advance, there is no position command difference buffer and the calculation can be performed easily.

  In this way, with predictive control that does not use the position command difference buffer, overshoot and undershoot can be prevented during positioning even if the position command is followed.

The block diagram which shows the structure of the motor control apparatus of this invention The block diagram which shows the structure of the prediction control part of this invention. The block diagram which shows the detailed structure of the prediction control part of this invention Position command difference and speed command timing diagram due to difference in buffer size and control cycle of position command difference The flowchart which shows the prediction control method of the electric motor control apparatus of this invention Block diagram showing the detailed configuration of a prediction control unit having multiple stages of position command difference buffers The block diagram which shows the detailed structure of the prediction control part of 2nd Example. Block diagram for explaining a conventional control device

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Electric motor control apparatus 2 Electric motor 3 Position detector 11 Prediction control part 12 Speed control part 13 Current control part 14 Power conversion part 15 Motor speed generation part 16 Encoder signal input / output part 21 Position command buffer part 22 Prediction calculation part 23 Feedforward calculation Part

Claims (6)

A position command difference buffer unit that stores a position command every sampling time; a prediction calculation unit that generates a speed command based on a position command of the position command difference buffer unit and a positional deviation of the motor position; and the speed command and the motor speed A speed control unit that generates a torque command based on the speed deviation of the motor, a motor drive unit that drives the motor based on the torque command, and a motor speed generation unit that generates the motor speed based on a time difference between the motor positions. In the provided motor control device,
An electric motor control device comprising a position command difference buffer unit having two stages of the position command difference buffer when only speed feedforward is used.   The motor control device according to claim 1, wherein the position command difference buffer unit has four stages of position command difference buffers when torque feedforward is used.   The motor control device according to claim 1, wherein the position command difference buffer unit uses the same position command difference without using the position command difference buffer when feedforward is not used. A position command difference buffer unit that stores a position command every sampling time; a prediction calculation unit that generates a speed command based on a position command of the position command difference buffer unit and a positional deviation of the motor position; and the speed command and the motor speed A speed control unit that generates a torque command based on the speed deviation of the motor, a motor drive unit that drives the motor based on the torque command, and a motor speed generation unit that generates the motor speed based on a time difference between the motor positions. In the predictive control method of the provided motor control device,
A step of storing a position command difference by using two stages of the position command difference buffer when only speed feedforward is used, and a predictive control method for an electric motor control device.   5. The predictive control method for an electric motor control device according to claim 4, further comprising a step of storing the position command difference using four stages of the position command difference buffer when torque feed forward is used.   5. The predictive control method for an electric motor control device according to claim 4, further comprising a step of performing predictive control with the same position command difference without using a position command difference buffer when feedforward is not used.