JP2008211904A - Voltage control method for actuator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voltage control method for an actuator capable of enhancing positioning accuracy by greatly improving acceleration disturbance removal performance in DC voltage-driving of a DC drive source. <P>SOLUTION: This voltage control method for controlling a voltage source for voltage-driving the DC motor of an actuator by a controller, includes: a step for expressing the actuator in a transfer function including the inductance, resistance and counter electromotive force of the coil of the DC motor to compensate for the gain of the inductance, resistance and counter electromotive force with the transfer function; and a step for controlling the voltage drive source of the DC motor by a control signal to which the compensated gain is applied. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to actuator control, and more specifically to an actuator voltage control method in which an actuator including a DC (direct current) drive source and a position or angle sensor is driven by a voltage source.

  Conventionally, a DC motor that is a DC drive source is generally driven by a current source, and a drive circuit using a current source is usually used for positioning control using the DC motor. The current and the output (torque) generated by the DC motor are approximately proportional, and the output and the position of the DC motor output movable part are in a simple second-order integral relationship, so it is easy to model and the control system algorithm is simple That's why.

On the other hand, a method of driving a DC motor with a voltage source has been used for a long time. When driving a DC motor with a voltage source, the point where the counter electromotive voltage proportional to the speed (rotational speed) and the applied voltage balance is the DC motor speed. Is also useful. However, in the conventional voltage source drive, the characteristics of the coil of the DC motor are not strictly modeled, that is, the optimization design including the inductance and resistance of the DC motor and the torque constant (back electromotive voltage constant) has not been made. Therefore, the gain of the DC motor control system is very low, and accuracy and disturbance suppression performance are limited. For example, a toy DC motor is driven by a battery.
Therefore, a control method by voltage source driving in consideration of coil characteristics has been proposed (Patent Document 1).
Japanese Patent Laid-Open No. 2002-237152

  However, since the main purpose of the voltage source drive control method described in Patent Document 1 is to simplify the circuit (the number of resistors for current detection is reduced by one), the characteristics of the coil that changes due to heat or the like can be obtained. There is no description or suggestion regarding compensation methods and means. In addition, there is no description regarding the addition of another integrator or a minor loop, and there is no description or suggestion regarding the optimization of the characteristics of the DC motor such as the inductance and the torque constant. For this reason, the characteristics and performance obtained by the control method of Patent Document 1 are not improved as compared with the driving method using a current source. Therefore, this control method cannot be applied to a positioning control system for a head actuator of a hard disk drive using a DC drive source that requires high-precision rotation control and position control, for example, a VCM (Voice Coil Motor).

  The present invention has been made to solve the above-described problems of the prior art, and provides a voltage control method for an actuator that improves positioning accuracy by greatly improving acceleration disturbance removal performance in DC voltage drive of a DC drive source. The purpose is to provide. Furthermore, the present invention provides a voltage control method for an actuator that improves zero deviation tracking with respect to a ramp target by means of second order integration inside the controller, reduces operating noise in the course, and avoids nonlinear saturation of the analog drive system. For the purpose.

  The voltage control method for an actuator of the present invention that solves such a problem is a voltage control method for controlling a voltage source that drives a voltage of a DC motor of an actuator by a controller, and the actuator includes an inductance of a coil of the DC motor, The transfer function includes a resistance and a counter electromotive force. The transfer function compensates the gain of the inductance, the resistance, and the counter electromotive force, and the voltage drive of the DC motor is performed by a control signal multiplied by the compensated gain. And the step of controlling the source.

  In the actuator voltage control method of the present invention, the inductance, resistance and back electromotive force (parameter) of the DC motor are optimized including the compensation of the gain.

Actually, the step of detecting the state of the DC motor, the step of second-order integration of the deviation between the target value set by the target setting means and the detected state signal, and the second-order integrated deviation, Adding the compensated gain.
The step of detecting the state of the DC motor includes a step of detecting a current flowing through the DC motor. The inductance and resistance of the coil are estimated based on the detected current value, and the gains are corrected and adjusted. It is preferable to control.
Furthermore, it is preferable that a minor feedback is applied from the detected current to change the apparent impedance of the coil.

In the voltage control method for an actuator according to the present invention, the transfer function includes u as a control input for controlling the voltage of the DC motor, x as an angle output of the DC motor, v as an angular velocity, i as a current flowing through the DC motor, and the voltage as described above. The gain of the state detector for detecting the state of the drive source and the DC motor is collectively expressed as g, and the transfer function G (s) from the control input u to the angle output x is expressed by the following equation (1).
G (s) = x (s) / u (s)
= gk / [s (k ² + JR s + JL s ² )] ・ ・ ・ Equation 1
(Where R is the resistance of the DC motor, L is the inductance, k is the torque constant, J is the moment of inertia, and w is the torque disturbance.)

The sensitivity function S (s) with respect to the torque disturbance w is expressed by the following equation 2.
S (s) = (R + L s) / [s (k ² + JR s + JL s ² )] Equation 2

Furthermore, in the voltage control method of the actuator of the present invention, the sensitivity function of the voltage source drive is (R + L s) / [s (k ² + JR s + JL s ² )], and the sensitivity function of the current source drive is 1 / (J s ² ), these ratios are expressed by the following formula 3, and when the formula 3 is transformed into the following formula 5 with the gain a, the zero point q, and the pole p1, p2, the pole p1, It is preferable to use a region where p2 is a complex pole.
J s (R + L s) / (k ² + JR s + JL s ² ) Equation 3
as (s + q) / [(s + p1) (s + p2)] ... Equation 5

The current function i (s) is expressed by the following formula 7, and R, L, g, and k are estimated from the history of x, u, and i.
i (s) = (gu (s) − kv (s)) / (R + s L) = (gu (s) − ksx (s)) / (R + s L) Equation 7

The angle output function x (s) is expressed by the following equation 8, and w and J are estimated from the history of x and i and the estimated value of k.
x (s) = (ki (s) + w (s)) / (J s ² ) Equation 8

When h = R, the sensitivity function S (s) is expressed by the following formula 12.
S (s) = 1 / (J s ² + k ² / L) Equation 12
When h = R + s L1, the sensitivity function S (s) is expressed by the following equation (13).
(However, L1 <L.)
When h = −R1, the sensitivity function S (s) is expressed by the following formula 15.
S (s) = ((R + R1) + L s) / [s (k 2 + J (R + R1) s + JL s 2)] ··· Equation 15
(However, L1 <L.)

According to the present invention, the acceleration disturbance removal performance is greatly improved and the positioning accuracy of the actuator is improved.
By integrating the output of the detection means for detecting the driving state of the DC motor second-order, it is possible to follow the zero deviation with respect to the ramp-like target, reduce the operating noise at high frequencies, and avoid nonlinear saturation of the analog drive system.

  The best mode to which the present invention is applied will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the hardware configuration of the voltage control apparatus of the present invention. This voltage control apparatus is a position (angle) detector for a control object including a DC motor 11 as a DC drive source and a position (angle) detector 13 as means for detecting the state of the DC drive source. A controller 15 for calculating a control input from the output of 13, a voltage driver (voltage source) 17 for driving the DC motor 11 based on the control input u, and a current detection as a current sensor for detecting a current flowing through the DC motor 11. A device (AD converter) 19 is provided.

  The controller 15 is composed of a digital computing unit 21 such as a DSP. The digital calculator 21 further includes an AD converter 23 that converts an analog signal from the position angle detector 13 into a digital signal, and a DA that converts the output from the digital calculator 21 into an analog signal to generate a control input. A converter 25 is attached.

  The current detector 19 is an A / D converter that AD converts the terminal voltage Vs of the low resistor Rs connected in series to the negative input of the DC motor 11 and outputs a current value. A Hall element or the like may be used instead of the low resistor Rs.

  The voltage control device further includes a target value setting circuit 27 for setting a target of output of the position angle detector 13. The digital calculator 21 calculates a voltage for driving the DC motor 11 so that the deviation of the output of the position angle detector 13 from the target value ref output from the target value setting means 27 is eliminated.

  When this voltage control device is applied to a hard disk drive, the DC motor 11 is applied to a VCM (Voice Coil Motor), the output of the position angle detector 13 is a PES (Position Error Signal) read by the magnetic head, or the magnetic head is applied. This corresponds to the encoder output that detects the rotation angle of the arm to be tracked. In this specification, a case where the present invention is applied to a rotary type DC motor (VCM) will be described for the sake of simplicity, but the configuration when applied to a linear type DC motor is also the same.

A block diagram of a transfer function of the entire control target including the voltage control device shown in FIG. 1 is shown in FIG. Here, the control input is u, the angle output is x, the angular velocity is v, the current flowing through the DC motor 11 is i, and the gains of the voltage driver 17 and the position angle detector 13 are collectively g. Furthermore, the resistance of the DC motor 11 is R, the inductance is L, the torque constant is k, the moment of inertia is J, and the torque disturbance is w. At this time, the transfer function G (s) from the control input u to the angle output x is expressed by the following equation 1.
G (s) = x (s) / u (s)
= gk / [s (k ² + JR s + JL s ² )] ・ ・ ・ Equation 1
The transfer function from the torque disturbance w to the angle output x, that is, the sensitivity function S (s) for the disturbance is expressed by the following equation (2).
S (s) = (R + L s) / [s (k ² + JR s + JL s ² )] Equation 2

  The transfer function shown in FIG. 2 will be described in detail. A constant gain g is applied to the control input u, which is applied to the DC motor 11 as a voltage. The current i flows by the applied voltage and the impedance 1 / (R + s L) of the DC motor 11, and the current i multiplied by the torque constant k is the torque (i k). Dividing the torque (i k) by the moment of inertia J and first-order integration gives the angular velocity v, and further angular integration of the angular velocity v gives the angular output x. The voltage generated by multiplying the angular velocity v by the counter electromotive voltage constant (= torque constant) k is the counter electromotive voltage, and works in the opposite direction to the voltage from the input.

From the above, assuming that torque disturbance w = 0,
v (s) = [gu (s) − kv (s)] [1 / (R + s L)] k [1 / (s J)]
x (s) = v (s) / s
Equation 1 is obtained by solving.
Also, if the control input u = 0,
v (s) = [w (s) − k ² / (R + s L) v (s)] [1 / (s J)]
x (s) = v (s) / s
Equation (2) is obtained by solving.

Next, from Equation 1 and Equation 2, the disturbance rejection ratio in an actual control system to which the present invention is applied will be described in comparison with the disturbance rejection ratio in a conventional current source drive. In the current source drive, the transfer function (sensitivity function) from the torque disturbance w to the angle output x is 1 / (J s ² ).

Voltage source drive sensitivity function (R + L s) / [s (k ² + JR s + JL s ² )] and current source drive sensitivity function 1 / (J s ² ) If the value is smaller than 1, the voltage source drive is less sensitive to torque disturbance w and is suitable for control purposes. When calculating the ratio of the above sensitivity functions,
J s (R + L s) / (k ² + JR s + JL s ² ) Equation 3
It becomes.
If we organize Equation 3 by 1 / s
[R / (L s ² ) + 1] / [k ² / (JL s ² ) + R / (L s) + 1]
Therefore, the gain is 1 when s is infinite.

Equation 3 is expressed as follows: gain is a, zero is q, pole is p1, p2
as (s + q) / [(s + p1) (s + p2)] ... Equation 5
When the frequency characteristic is expressed by a log-log plot, it is roughly divided into patterns (A), (B), and (C) in FIG. 3 according to the values of the poles p1 and p2. In each graph of FIG. 3, the vertical axis represents gain (dB) and the horizontal axis represents frequency.

In each graph of FIG. 3, the hatched area indicates that the voltage control according to the present embodiment is less sensitive to disturbance, that is, is more resistant to disturbance. The region indicated by the horizontal line indicates the opposite.
The case of FIG. 3A is when the poles p1 and p2 are real numbers and p1 ≠ p2. In this case, the ratio of the sensitivity function is 0 dB in almost the entire region, that is, there is almost no difference.
The case of FIG. 3 (B) is when the pole p1 = p2. In this case, it can be seen that the voltage control of the present embodiment is preferable in the frequency range above the middle range.
The case of FIG. 3C is when the poles p1 and p2 are complex poles. In this case, a large sensitivity reduction is achieved by voltage control, and it can be seen that this is the most preferable state in the present embodiment. The voltage control method in the embodiment of the present invention mainly uses this mode, that is, a case where (k ² + JR s + JL s ² ) = 0 has a complex root with respect to s. However, in the vicinity of the frequencies of the poles p1 and p2, there is a region P where the sensitivity function is high, so this region P is set to be excluded and this region P is set not to be used.

As an example of parameters of a voltage control device using an actual DC motor, a head actuator of a hard disk drive and a VCM as a DC motor for driving the head actuator will be described. In the actuator position control of this hard disk drive, VCM coil resistance = 8Ω, inductance = 5 mH, torque constant = 1 N / A, moment of inertia = 10 ⁻³ kg, etc. are common. This is a general case in which a ferrite magnet is applied to a VCM permanent magnet. Furthermore, the torque constant can be increased several times by using a strong permanent magnet, such as a neodymium magnet. For example, if the torque constant is 10,
Zero point = R / L = 1600 (rad / sec)
Pole = 800 ± I 4500 (rad / sec)
Thus, the zero point is obtained at about 254 Hz and the pole is obtained at about 728 Hz, and the characteristics shown in FIG. 3C are obtained.

  On the other hand, paying attention to the low-frequency range of the transfer function of Equation 1, the transfer function in the case of voltage source driving is equivalent to the first-order integral characteristic. Note that the input / output transfer function in current source driving is equivalent to the second-order integral characteristic. In other words, in the low frequency range, the transfer function is -6 dB / oct for voltage source drive and -12 dB / oct for current source drive, and the phases are -90 degrees and -180 degrees, respectively (Fig. 4). In the servo system whose purpose is to follow the target value with zero steady-state deviation, one or more integrators are included in the controller. With this integrator, in the case of current source driving, the phase is further rotated 90 degrees to -270 degrees. From here, two zeros are inserted at an appropriate frequency, and the closed loop is stabilized by restoring the phase at the frequency at which the open loop gain becomes 1 to about -150 degrees (phase margin 30 degrees). In the voltage control method of the present invention, the phase in the low frequency region is −90 degrees, so that even if the same phase margin as in the current control is obtained, there is a margin to put another integrator in the controller. Therefore, in the present invention, a control system including two integrators in the controller is configured. Thereby, not only a normal step-like target value input but also a ramp-like target value can be followed with zero deviation.

  According to the voltage control method and the voltage control apparatus of the present invention, the gain in the low frequency range is high even when there is a nonlinear behavior of the bearing, and the deviation from the ramp-shaped target value can be made zero. This greatly improves the follow-up performance to spiral tracks used in, for example, DVD and CD drives. Furthermore, it has been shown that a hard disk drive can realize a control system that follows a spiral track, as well as a conventional track configuration arranged as concentric circles. It shows that the access performance to can be greatly improved.

  5A and 5B show plot examples of the open loop transfer function of the control system. The configuration of the control system differs depending on whether the above complex pole is at a frequency higher than the zero cross frequency (FIG. 5A) or at a lower frequency (FIG. 5B). In other words, when the complex pole is at a frequency lower than the zero cross frequency, the complex pole may be included in the loop as a simple peak filter, but when it is at a high frequency, it is stabilized by means such as using a notch filter or phase compensation. It is necessary to apply. In this compensation, since the dependence of each parameter of the DC motor 11 is large, optimization of each parameter is important in order to make the control target easy to control.

  The voltage control method and voltage control apparatus of the present invention include the coil inductance and the torque constant as parameters, which vary with heat, changes with time, and the like. Therefore, a method of identifying the change and adaptively controlling it is useful. Therefore, in the embodiment of the present invention, it is preferable to provide a sensor for detecting the current flowing through the VCM as means for detecting the operating state of the VCM in the voltage source for driving the VCM. This sensor may be current detection by a Hall sensor, or may be configured to measure a voltage generated in a minute resistor connected in series with the VCM (see FIG. 1). In principle, the present invention can identify each parameter without the current detection sensor. However, if there is a current detection sensor, the disturbance torque can be separated and the load torque can be measured, so that the identification accuracy can be improved. .

Examples of parameter identification methods will be described below.
(1) When there is no current sensor From Equation 1 and Equation 2, the angle output function x (s) is
x (s) = [a / (s (s ² + bs + c))] u (s) + [(R + L s) / (s (s ² + bs + c))] w (s) ..Formula 6
It becomes. Assuming that x (s), u (s), and w (s) are sufficiently rich (including all frequency components required for control), R and L are obtained from the history of angle output x and control input u. , A, b, c, w are estimated and set.
Where a = gk / (JL), b = R / L, c = k ² / (JL)
It is. If there is no current sensor, g, k, and J cannot be set individually by estimation, but it is sufficient for control if a, b, and c can be set by estimation.

  In the case of this method, since all parameters must be set in a situation where the torque disturbance w is unknown and very large, the accuracy is low.

(2) When there is a current sensor From the transfer function in Fig. 2, the current function i (s) is
i (s) = (gu (s) − kv (s)) / (R + s L) = (gu (s) − ksx (s)) / (R + s L) Equation 7
Therefore, R, L, g, and k are estimated and set from the history of x, u, and i. When there is a current sensor, the disturbance term (torque disturbance w) is not included, and therefore setting by high-precision estimation is possible.

Similarly, the angle output function x (s) from the transfer function in FIG.
x (s) = (ki (s) + w (s)) / (J s ² ) Equation 8
Therefore, w and J can be set by estimation from the history of x and i and the estimated value of k. Although this equation 8 includes a disturbance term (w (s)), since the parameters to be obtained are only the torque disturbance w and the moment of inertia J, the influence on the (estimation) accuracy can be ignored.

  In any of the above methods, a known estimation algorithm such as a least square method or a gradient method can be used. The gain of the control system is determined and set using the parameters thus estimated. For this gain determination method, a known algorithm such as LQ method or pole placement method can be used. Although the above is parameter estimation, it is also possible to perform adaptive control collectively with a policy that a closed loop system including a control system is always kept at the same transfer function.

Further, a minor feedback h (s) returning from the current i flowing through the DC motor 11 to the control input u is added to the transfer function system shown by the block in FIG. 2 to constitute the transfer function system (control system) shown in FIG. You can also. As a result, the apparent coil impedance of the DC motor 11 is
1 / (R + s L − h (s)) Equation 9
And can be controlled by minor feedback h (s). At this time, the transfer function G (s) of the entire control target and the sensitivity function S (s) with respect to the torque disturbance w are expressed by Expression 10 and Expression 11, respectively.
G (s) = x (s) / u (s) = gk / [s (k ² + J s (R + L s −h (s)))] Equation 10
S (s) = (R + L s −h (s)) / [s (k ² + J s (R + L s −h (s)))] Equation 11

The minor feedback h (s) can be arbitrarily set as long as the minor loop is stable (the pole of Equation 9 is stable). The voltage control method of the present invention corresponding to the method for obtaining the minor feedback h (s) will be described below. To do.
When h = R, the coil impedance of the DC motor 11 is only the inductance s and L, so the sensitivity function S (s) is
S (s) = 1 / (J s ² + k ² / L) Equation 12
Thus, compared with the current source drive, it is uniformly 12 dB / oct lower in the sub-frequency band. However, when h> R, the system becomes unstable, so h is set slightly smaller than R when actually installed. The same applies hereinafter.

h = R + s L1
By including inductance L1 in minor feedback h (s) in addition to R, the sensitivity function S (s) is
S (s) = 1 / [J s ² + k ² / (L−L1)] Equation 13
It can be. As a result, the pole can be freely controlled.
However, L1 <L.

When h = -R1, Equation 9 has an infinite Q value. Even in such a system, the closed loop can be stabilized by feedback. However, when there is a need to lower the Q value for some reason, such as when there is a mechanical resonance with a large fluctuation nearby, Q can be reduced by negatively feeding back large R1 as h. The value can be lowered. That is, such a control system can be expressed by a transfer function G (s) of the following formula 14.
G (s) = x (s ) / u (s) = gk / [s (k 2 + J (R + R1) s + JL s 2)] ··· Equation 14

The sensitivity function S (s) in this case is as shown in Equation 15. As for the sensitivity function S (s), the frequency of the zero point also increases, so that the torque disturbance suppression performance deteriorates as R1 increases. However, it still has an advantage of 6 dB / oct over current source driving.
S (s) = ((R + R1) + L s) / [s (k 2 + J (R + R1) s + JL s 2)] ··· Equation 15

When h = -R1 + L1 s When the purpose is to avoid deterioration of the sensitivity function as much as possible, to lower the Q value as much as possible, and to increase the pole frequency, the inductance is also controlled. For example, there is a case where it becomes one design guideline to match the zero point and the pole frequency in Expression 14. The sensitivity function S (s) in this case is as shown in Equation 16.
S (s) = [(R + R1) + (L-L1) s] / [s (k ² + J (R + R1) s + J (L-L1) s ² )] Equation 16
However, L1 <L.

  The control system including the minor feedback h (s) can be configured by detecting the voltage Vs generated in the low resistance Rs shown in FIG. 1 and feeding it back as h (s) Vs / Rs. This may be constituted by an analog circuit such as an analog operational amplifier, or after the voltage Vs is taken in using the AD converter 19a as shown in FIG. 1, the minor feedback h (s) is applied as a digital filter to control the result. It can also be configured by adding to the input u.

  FIG. 7 shows an example of the arithmetic block 210 inside the digital arithmetic unit 21 shown in FIG. 1 when the hardware circuit for detecting the current flowing through the coil of the DC motor 11 and the driving by the voltage source is provided. This calculation block 210 includes an integration circuit 211 that performs second-order integration of the deviation between the detected position and the target value, a differentiation circuit 212 that differentiates the difference, and a control gain f1 that is appropriate for the control input u up to one or several samples before. A control input determining circuit 213 for determining a control input by multiplying by ~ f5, and adaptively identifying a parameter to be controlled from the deviation, current value, and past control input history, and controlling gains f1 to f5 Is composed of an adaptive identification / adaptive controller 214 for tuning and a coil impedance conversion filter 215 for apparently changing the coil impedance of the DC motor 11.

  The control gains f1 to f5 are obtained by integrating the gain f3 applied to the deviation between the position signal and the target value ref, the gain f2 applied to the first-order integral value obtained by integrating the deviation with the first integrator, and the first-order integral value integrated with the integrator. The gain f1 applied to the second order integral value, the gain f4 applied to the differential value obtained by differentiating the deviation by the differentiator, and the gain f5 applied to the past input value delayed by the delay circuit by the output multiplied by the gains f1 to f5. The sum of the values multiplied by these gains f1 to f5 is output as the control input value u.

  The adaptive identification / adaptive controller 214 tunes the gains f1 to f5 based on the control input value u, the current value from the AD converter 19 and the position signal from the position detector 13.

  The coil impedance conversion filter 215 includes gains f6, f7, and f8. The current value from the AD converter 19 is multiplied by a gain f6, the current value is multiplied by a delay (DELAY) after the calculation, and the gain f7 is further multiplied, and the current value multiplied by the gains f6 and f7 is further obtained by the delay circuit. Delayed (DELAY) is multiplied by a gain f8, and the current value multiplied by these gains f6, f7, f8 is added to the control input value u.

  The sum of the added current value and control input value u is output to the DA converter 25, converted to a new analog control input value u, and output to the voltage driver 17.

  In FIG. 7, the control gains f1 to f5 can be set by a known control system algorithm. For example, there are various methods such as a transfer function method, a pole placement method, an LQ method, and an H∞ method, and a state estimator may be inserted in the middle. The present invention may use any control method and is not limited to the control method. Further, since the coil impedance conversion filter can include a filter having a higher order, the configuration of the filter is not limited as long as it is stable as a system.

  The present invention is particularly useful for controlling a positioning system used for a DC drive source such as a VCM such as a hard disk drive.

It is a figure which shows the embodiment of the hardware constitutions of the voltage control apparatus of this invention with a block. It is a figure which shows the transfer function of the voltage control apparatus shown in FIG. FIG. 3 is a diagram illustrating a graph in which frequency characteristics are plotted based on the transfer function of FIG. 2, and (A), (B), and (C) are diagrams illustrating cases in which the values of the poles p1 and p2 are different from each other. is there. It is a figure which shows the relationship between the frequency of the transfer function between input-output of the voltage drive of this invention, and the conventional current drive, and a gain with a graph. (A), (B) is a figure which shows the example of a plot of the open loop transfer function of a control system, respectively in the voltage control apparatus of this invention. It is a figure which shows the embodiment of the transfer function system which added minor feedback h (s) to the transfer function system shown in FIG. 2 with a block. It is a figure which shows embodiment of the calculation block inside the digital arithmetic unit shown in FIG.

Explanation of symbols

11 (DC drive source) DC motor 13 Position (angle) detector 15 Controller 17 Voltage driver (voltage source)
19 Current detector (current sensor)
21 Digital operation unit 23 AD converter 27 Target value setting circuit
u Control input
x Angle output
v Angular velocity
i Current flowing in the DC motor
g Gain of voltage driver and position angle detector
ref target value
R DC motor resistance
L DC motor inductance
K DC motor torque constant
J Moment of inertia of DC motor
W Torque disturbance
G (s) transfer function (control input u to angle output x)
Sensitivity function for S (s) disturbance (torque disturbance w to angle output x)

Claims (14)

A voltage control method for controlling a voltage source for voltage driving a DC motor of an actuator by a controller,
The actuator is represented by a transfer function including the inductance, resistance, and counter electromotive force of the coil of the DC motor, and the transfer function compensates the gain of the inductance, resistance, and counter electromotive force,
And a step of controlling a voltage driving source of the DC motor by a control signal multiplied by the compensated gain. 2. The actuator voltage control method according to claim 1, wherein the inductance, resistance and back electromotive force (parameter) of the DC motor are optimized including compensation of the gain. 3. The voltage control method for an actuator according to claim 1, wherein the step of detecting the state of the DC motor and the step of second-order integrating the deviation between the target value set by the target setting means and the detected state signal. The voltage control method for an actuator includes a step of adding the compensated gain to the second-order integrated deviation. 4. The voltage control method for an actuator according to claim 3, wherein the step of detecting the state of the DC motor includes a step of detecting a current flowing through the DC motor, and the inductance and resistance of the coil based on the detected current value. A voltage control method for an actuator that performs adaptive control while estimating each gain and correcting each gain. 5. The voltage control method for an actuator according to claim 4, further comprising a step of changing an apparent impedance of the coil by applying minor feedback from the detected current. 2. The voltage control method for an actuator according to claim 1, wherein the transfer function includes u as a control input for controlling the voltage of the DC motor, x as an angular output of the DC motor, v as an angular velocity, and i as a current flowing through the DC motor. The gains of the voltage drive source and the state detector for detecting the state of the DC motor are collectively expressed as g, and the transfer function G (s) from the control input u to the angle output x is expressed by the following equation 1. Actuator voltage control method.
G (s) = x (s) / u (s)
= gk / [s (k ² + JR s + JL s ² )] ・ ・ ・ Equation 1
(Where R is the resistance of the DC motor, L is the inductance, k is the torque constant, J is the moment of inertia, and w is the torque disturbance.) 7. The actuator voltage control method according to claim 6, wherein the sensitivity function S (s) with respect to torque disturbance w is expressed by the following equation (2).
S (s) = (R + L s) / [s (k ² + JR s + JL s ² )] Equation 2 In the voltage control method of an actuator according to claim 7, the sensitivity function of the voltage source driving (R + L s) / [ s (k 2 + JR s + JL s 2)], the sensitivity function of the current source driving 1 When expressed as / (J s ² ), these ratios are expressed by the following formula 3,
When equation 3 is transformed into equation 5 below with gain a, zero point q, poles p1 and p2,
Actuator voltage control method using a region where the poles p1 and p2 are complex poles.
J s (R + L s) / (k ² + JR s + JL s ² ) Equation 3
as (s + q) / [(s + p1) (s + p2)] ... Equation 5 8. The voltage control method for an actuator according to claim 7, wherein the current function i (s) is expressed by the following formula 7, and R, L, g, k are estimated from the history of x, u, i.
i (s) = (gu (s) − kv (s)) / (R + s L) = (gu (s) − ksx (s)) / (R + s L) Equation 7 8. The voltage control method for an actuator according to claim 7, wherein the angle output function x (s) is expressed by the following formula 8, and w and J are estimated from the history of x and i and the estimated value of k. Method.
x (s) = (ki (s) + w (s)) / (J s ² ) Equation 8 In the transfer function G (s) of the above equation 1 in the voltage control method for an actuator according to claim 7, a minor feedback h (s) returning from the current i flowing through the DC motor to the control input u is added to the control function. The coil impedance of the motor 11 is expressed by the following formula 9
Actuator voltage control method set by.
1 / (R + s L − h (s)) Equation 9 12. The actuator voltage control method according to claim 11, wherein when h = R, the sensitivity function S (s) is expressed by the following equation (12).
S (s) = 1 / (J s ² + k ² / L) Equation 12 12. The actuator voltage control method according to claim 11, wherein when h = R + s L1, the sensitivity function S (s) is expressed by the following equation (13).
S (s) = 1 / [J s ² + k ² / (L−L1)] Equation 13
(However, L1 <L.) 12. The actuator voltage control method according to claim 11, wherein when h = −R1, the sensitivity function S (s) is expressed by the following equation 15.
S (s) = ((R + R1) + L s) / [s (k 2 + J (R + R1) s + JL s 2)] ··· Equation 15
(However, L1 <L.)

Images