JP2005278225A - Controller and controlling method of rotator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To control rotation of a rotator easily to a target rotational state in feedback control. <P>SOLUTION: During feedforward control of a motor B, rotational position of motors A and B is detected and then the rotational speed is detected from the rotational position thus detected (72). Positional error Δθ<SB>p</SB>is detected from the rotational position of the motors A and B and speed error Δω is detected from the rotational speed of the motors A and B (74). The ratio δ of the positional error Δθ<SB>p</SB>and the speed error Δω is operated (76). A decision is made whether the δ falls within a predetermined range where a rotational state can be feedback controlled to a target state without inverting the sign of torque during feedforward control (78). If the δ falls within the predetermined range, a maximum torque is estimated (80) and switching is made to feedback control (84, 86) if the estimated maximum torque falls within an allowable range (82; Y). <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a rotating body control device and method, and in particular, a rotating body that performs feed-forward control, switches from the feed-forward control to feed-back control, and controls the rotating state of the rotating body to a target rotating state. The present invention relates to a control apparatus and method.

  In recent years, various studies have been made to achieve higher efficiency of hybrid vehicles (HV). For example, a clutch is generally used for connecting / disconnecting a power transmission element. However, a conventional clutch mechanism using a friction material or hydraulic pressure is not a good measure because it causes a reduction in fuel consumption.

  Therefore, it is conceivable to introduce a dog clutch (meshing clutch) as a clutch mechanism. Since the dog clutch meshes and connects the dog teeth, it does not require a pressing force due to hydraulic pressure and has no loss due to sliding. On the other hand, if the state of the input / output shafts is not synchronized well, there is a risk of impact torque being generated at the time of engagement, or the teeth of the dog being damaged.

  Therefore, for example, feedforward control is performed so that the rotation speed of one of the dog teeth to be engaged in the speed change mechanism approaches the rotation speed of the other dog tooth, and then the engagement position of the dog teeth is accurate. Attempt to feedback control to engage.

  FIG. 12 shows an example in which position control is performed after speed synchronization as a simple example. In this example, the speed of the reference motor (motor A) is 12 rad / s, and the motor (motor B) to be controlled is suddenly decelerated from the speed of 180 rad / s (1720 rpm) and synchronized as quickly as possible. Indicated.

  That is, as shown in FIG. 12 (C), reverse torque (−12 Nm) is applied to the motor B at a speed of 180 rad / s (1720 rpm), and the rotational speed of the motor B as shown in FIG. 12 (A). Is moved closer to the speed of motor A (12 rad / s), and when the speed of motor B matches the speed of motor A, the feedforward control is switched to the feedback control. Thus, since the control is switched to the feedback control, as shown in FIG. 12B, the position error converges smoothly after the start of the position control (when switching to the feedback control).

  However, even if the speed of motor B matches the speed of motor A, if there is a position error in the meshing position of the dog teeth, it is necessary to reduce this position error, so the speed of motor B must be increased once . That is, as shown in FIG. 12C, after starting position control (when switching to feedback control), the sign of the torque so far is reversed (from-to +) to increase the speed of motor B, The motor B is controlled to converge the position error.

  As described above, after the speeds of the motors A and B coincide with each other, it is a useless operation to re-accelerate the speed of the motor B. As a result, the synchronization time is extended. Further, a large torque fluctuation that reverses the sign of the torque becomes a cause of sound and vibration.

Therefore, a control system has been proposed that switches from feedforward control to feedback control when the position error reaches a predetermined amount before the speed matches (see Non-Patent Document 1).
Electrical Engineering D, Vol. 118, No. 7/8, 1998, “High-speed, high-accuracy positioning control system by mode switching control”

  However, in the control system, since it is not determined whether the rotational state of the motor B can accurately control the rotational position of the motor B to the rotational position of the motor A, the feedback control after switching is complicated. That is, it is necessary to prepare various parameters of the control model when switching from feedforward control to feedback control.

  The present invention has been made in view of the above-described facts, and an object of the present invention is to provide a rotating body control device and method capable of easily controlling the rotating state of a rotating body to a target rotating state in feedback control.

  In order to achieve the above object, the invention according to claim 1 is a first detection means for detecting the rotation state of the first rotating body and a second detection means for detecting the rotation state of the second rotating body. A predetermined torque is generated in the first rotating body to perform feed-forward control, and from the feed-forward control, the rotation state of the first rotating body detected by the first detecting means, the first Based on a target rotation state determined from the rotation state of the second rotating body detected by the two detection means, and a feedback control model determined in advance by the torque and rotation state generated in the first rotating body. And a control means for controlling the rotation state of the first rotating body to the target rotation state by switching to feedback control for changing the torque generated in the first rotating body. The The control means detects that the rotation state of the first rotating body detected by the first detection means during the feed-forward control does not reverse the sign of the torque during the feed-forward control. It is determined whether or not the state is within a predetermined range in which feedback control can be performed to the target rotational state, and when the rotational state is determined to be within the predetermined range, It is characterized by switching to feedback control.

  That is, the present invention provides a first detection means for detecting the rotation state of the first rotating body, a second detection means for detecting the rotation state of the second rotating body, and a predetermined amount for the first rotating body. Torque is generated to perform feed-forward control, and from the feed-forward control, the rotation state of the first rotating body detected by the first detection means, and the first detection detected by the second detection means. Torque generated in the first rotating body based on a target rotation state determined from the rotating state of the second rotating body and a feedback control model determined in advance by the torque and rotating state generated in the first rotating body. And a control unit that controls the rotation state of the first rotator to the target rotation state by switching to the feedback control to be changed.

  In order to synchronize in the shortest time, it is desirable to set the predetermined torque to the maximum allowable torque of the first rotating body, but it is not limited to this.

  The first detecting means detects the rotational state of the first rotating body, and the second detecting means detects the rotating state of the second rotating body.

The control means generates a predetermined torque in the first rotating body to perform feed-forward control, and a control model determined in advance by the torque generated in the first rotating body and the rotation state from the feed-forward control, Based on the rotation state of the first rotating body detected by the first detection means and the target rotation state determined from the rotation state of the second rotating body detected by the second detection means. Switching to the feedback control that changes the torque generated in one rotating body, and controlling the rotation state of the first rotating body to the target rotation state. In this case, it is not possible to change from the feed forward control to the feedback control. If switched to, the rotation state of the first rotating body may not be controlled to the target rotation state. Further, if the sign of the torque generated in the first rotating body during the feed-forward control is switched to the feed-back control and reversed, sound and vibration may be generated in the first rotating body.

  Therefore, in the present invention, the control means, during the feed-forward control, indicates that the rotation state of the first rotating body detected by the first detection means indicates the sign of torque during the feed-forward control. It is determined whether or not the rotation state is within a predetermined range in which the rotation state can be feedback-controlled to the target rotation state without reversing, and when the rotation state is determined to be within the predetermined range, The feed-forward control is switched to the feed-back control.

  Thus, during the feed-forward control, it is determined whether the rotation state of the first rotating body is within a predetermined range that can be controlled to the target rotation state based on the control model, and the rotation state Is switched from the feed-forward control to the feed-back control when it is determined that it is within a predetermined range that can be controlled to the target rotation state. The rotating body may be controlled, and the rotation state of the first rotating body can be easily controlled to the target rotation state in the feedback control.

  Note that even if the feed-forward control is switched to the feed-back control, the torque during the feed-forward control is not largely reversed, so that the sound and vibration generated in the first rotating body due to the torque reversal are prevented. be able to.

  Here, as in claim 2, the first detection means detects the rotational position of the first rotating body, and the control means is determined based on the detected rotational position and the rotational position. Based on at least one of the rotation speeds and the target rotation state, it may be determined whether or not the rotation state is within a predetermined range that can be controlled to the target rotation state.

  That is, for example, the control means may have a relationship between the detected rotational position and a target rotational position determined based on the target rotational state, for example, a relationship between a position error, a rotational speed and a target rotational speed determined based on the target rotational state. For example, based on the speed error and the relationship between the position error and the speed error, for example, the ratio, it is determined whether or not the rotational state is within a predetermined range that can be controlled to the target rotational state. May be.

  The predetermined range that can be controlled to the target rotation state is determined based on the control parameter of the control model.

  By the way, after switching to the feedback control, the torque cannot be changed without limitation based on the control model, and needs to be within an allowable range determined from the first rotating body or the like.

  Therefore, as in claim 5, the control means estimates the maximum value of torque required after switching to the feedback control, and the estimated maximum value of the torque is within a predetermined allowable range. It is also possible to switch from the feed-forward control to the feed-back control when it is determined that the estimated maximum value of the torque is within the predetermined allowable range.

  In this case, when the load torque is applied to the first rotating body, if this is also ignored, the torque may exceed the allowable range.

  Therefore, as described in claim 6, when the load torque is applied to the first rotating body, the control means allows the predetermined sum of the maximum value of the estimated torque and the load torque to be determined in advance. It may be determined whether it is within a range, and when it is determined that the sum is within the predetermined allowable range, the feed / forward control may be switched to the feedback control.

  According to a seventh aspect of the present invention, in the feed-forward control, the control means brings the rotation speed of the first rotating body close to a target rotation speed determined based on the target rotation state, and the feedback / feedback control. In the control, the rotational position of the first rotating body is brought close to a target rotational position determined based on the target rotational state.

  According to an eighth aspect of the present invention, there is provided a rotating body control method according to an eighth aspect of the present invention, wherein a predetermined torque is generated in the first rotating body to perform feed forward control, and the torque generated in the first rotating body is controlled from the feed forward control. A rotating body control method that switches to a feedback control to be changed based on a model, and controls the rotation state of the first rotating body to a target rotation state determined from the rotation state of the second rotating body, During the feed-forward control, the rotation state of the first rotating body can be controlled in advance to the target rotation state based on the control model without inverting the sign of the torque during the feed-forward control. It is determined whether the rotation range is within a predetermined range, and when it is determined that the rotation state is within a predetermined range that can be controlled to the target rotation state, the feed-forward control is performed. It is intended to switch on the feed-back control from.

  Note that when the first rotation state is controlled to a predetermined target rotation state without considering the rotation state of the second rotation body, that is, the rotation body 1 is quickly stopped and fixed at a predetermined rotation position. In such a case, the rotational speed of the rotating body 2 can be regarded as equivalent to a zero state, and the control can be performed only with the rotation information of the rotating body 1 without considering the state of the rotating body 2. The following invention according to claim 9 is proposed.

  That is, a detecting means for detecting a rotating state of the rotating body, a feed-forward control by generating a predetermined torque in the rotating body, and a rotating state of the rotating body detected by the detecting means from the feed-forward control Switching to feedback control for changing the torque generated in the rotating body based on a predetermined target rotation state and a feedback control model determined in advance by the torque and rotating state generated in the rotating body, Control means for controlling the rotational state of the rotary body to the target rotational state, wherein the control means detects the rotation detected by the detection means during the feed-forward control. The rotation state of the body does not significantly reverse the torque during the feed-forward control, and the rotation state is fed to the target rotation state. It is determined whether or not the rotation range is within a predetermined range, and when the rotation state is determined to be within the predetermined range, switching from the feed-forward control to the feed-back control is performed. The rotating body control device is characterized.

  In the present invention, it is desirable that the predetermined torque is the maximum allowable torque of the rotating body in order to synchronize in the shortest time as in the above invention, but the present invention is not limited to this.

  The detection means detects the rotation state of the rotating body. The control means generates a predetermined torque in the rotating body and performs feed-forward control. From the feed-forward control, the rotating state of the rotating body detected by the detecting means, a predetermined target rotating state, and the By switching to feedback control that changes the torque generated in the rotating body based on a feedback control model that is determined in advance by the torque generated in the rotating body and the rotation state, the rotation state of the rotating body is changed to the target rotation state. To control.

  Then, the control means of the present invention is configured so that the rotation state of the rotating body detected by the detection means during the feed-forward control can be changed to the rotation state without greatly reversing the torque during the feed-forward control. It is determined whether or not the target rotation state is within a predetermined range in which the feedback control can be performed, and when the rotation state is determined to be within the predetermined range, the feedback from the feed-forward control is performed. Switch to control.

  Thus, during feed-forward control, it is determined whether or not the rotation state of the rotating body is within a predetermined range that can be controlled to the target rotation state based on the control model, and the rotation state is determined as the target state. When it is determined that the rotation state is within a predetermined range, the feed / forward control is switched to the feed-back control. In the feedback control, the rotation state of the rotating body can be easily controlled to the target rotation state.

  Note that even when switching from the feed-forward control to the feed-back control, the torque during the feed-forward control is not significantly reversed, so that it is possible to prevent noise and vibration generated in the rotating body due to the torque reversal. .

  As in claims 10 to 15, the inventions of claims 2 to 7 can also be applied to the present invention.

  In the present application, the following rotating body control method is also proposed. That is, a predetermined torque is generated in the rotating body to perform feed-forward control, and the feed-forward control is switched from the feed-forward control to the feedback control that changes the torque generated in the rotating body based on the feedback control model, A rotating body control method for controlling a rotating state of the rotating body to a target rotating state, wherein during the feed-forward control, the rotating state of the rotating body does not significantly reverse the torque during the feed-forward control. It is determined whether or not the rotation state is within a predetermined range that can be controlled to the target rotation state based on the feedback control model, and the predetermined range in which the rotation state can be controlled to the target rotation state. Switching from the feed-forward control to the feed-back control when it is determined that A rotator control method.

  As described above, according to the present invention, whether or not the rotation state of the first rotating body is within a predetermined range that can be controlled to the target rotation state based on the control model during feed-forward control. When the rotation state is determined to be within a predetermined range that can be controlled to the target rotation state, the feed-forward control is switched to the feed-back control. The first rotating body may be controlled based on the model, and the effect that the rotation state of the first rotating body can be easily controlled to the target rotation state in the feedback control is obtained.

  Further, according to the present invention, during the feed-forward control, it is determined whether the rotation state of the rotating body is within a predetermined range that can be controlled to the target rotation state based on the control model, and the rotation When it is determined that the state is within a predetermined range that can be controlled to the target rotational state, the feed / forward control is switched to the feedback control. Thus, it is possible to easily control the rotation state of the rotating body to the target rotation state in the feedback control.

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

  As shown in FIG. 1, the rotating body control device according to the present embodiment includes an encoder 20 as a first detection unit that detects a rotational position as a rotation state of a motor B as a first rotating body, And an encoder 18 as second detection means for detecting a rotational position as a rotational state of the motor A as the second rotating body. The encoders 18 and 19 can be configured by a resolver or the like.

  Further, the rotating body control device generates a predetermined torque in the motor B and performs feed-forward control, and switches from the feed-forward control to the feed-back control so that the rotational position of the motor B is changed to the rotational state of the motor A. A current control circuit 22 and a calculator 24 are provided as control means for controlling the target rotational position as a target rotational state determined from Note that the feedback control includes a feedback control model (details will be described later) determined in advance by the torque generated in the motor B and the rotation state, the rotational position of the motor B detected by the encoder 20, and the rotation of the motor A. Based on the position, the torque generated in the motor B is changed so that the rotational position of the motor B matches the target rotational position of the motor A.

  A gear 12 is attached to the motor A, and a gear 16 is attached to the motor B via an inertial load 14. As shown in FIG. 2, the gears 12 and 16 are provided with dog teeth 26. That the rotational position of the motor B matches the target rotational position of the motor A means that the dog teeth 26 of the gear 12 of the motor A and the dog teeth 26 of the gear 16 of the motor B can be engaged with each other. .

As shown in FIG. 3, the current control circuit 22 is detected by the encoder 18 and the tooth position calculator 30 that calculates the tooth position of the dog tooth 26 of the motor B from the rotational position of the motor B detected by the encoder 20. The tooth position calculator 31 for calculating the tooth position of the dog tooth 26 of the motor A from the rotational position of the motor A, the tooth position of the dog tooth 26 of the motor B calculated by the tooth position calculator 30, and the tooth position calculation A subtractor 32 for obtaining a position error Δθ _p between the tooth position of the dog tooth 26 of the motor A calculated by the device 31 is provided.

  In addition, the current control circuit 22 calculates the rotation speed of the motor A from the rotation position of the motor A detected by the encoder 18 and the rotation position of the motor B detected by the encoder 20 from the rotation position of the motor B. Subtraction for calculating the speed error Δω between the rotational speed of the motor A calculated by the speed calculator and the rotational speed of the motor B calculated by the speed calculator and the rotational speed of the motor B calculated by the speed calculator. A container 40 is provided.

The subtractor 32 is a position controller, and is connected to a P controller 34 that multiplies the position error Δθ _p by a predetermined transfer function (K _p ) of the position controller. The P controller 34 is connected to a subtractor 35 that subtracts the speed error Δω from the value (transfer function (K _p ) × position error Δθ _p ) calculated by the P controller 34. A PI controller 42 that is a speed controller that calculates a torque command value from the value W _err (= K _p × Δθ _p −Δω) calculated by the subtractor 35 is connected to the subtractor 35.

  Further, the current control circuit 22 includes a disturbance observer 48 for calculating a disturbance torque applied to the motor B from the rotation speed of the motor B calculated by the speed calculator 38 and a torque command value to be described later, and a PI controller. An adder 44 that adds the torque command value calculated by 42 and the disturbance torque calculated by the disturbance observer 48 to obtain the torque command value is provided.

  By the way, in this embodiment, as will be described in detail later, in order to control in the shortest time, the maximum torque is generated and feed-forward control is performed, and the feed-forward control is switched to the feed-back control that changes the torque. Then, the motor B is controlled. The current control circuit 22 includes a switch 45 that switches between feed-forward control and feed-back control based on a command from the calculator 24. The switch 45 is connected to a motor B having a current controller (PWM) (not shown) that controls the current to the motor B based on the torque command.

  The switch 45 is switched according to the control signal from the calculator 24 so that the maximum torque command from the calculator 24 is output to the current controller of the motor B in the feed-forward control. In the case of the back control, switching is performed so that the torque command from the adder 44 is output to the current controller of the motor B.

  As shown in FIG. 4, the disturbance observer 48 includes a multiplier 50 that multiplies the speed of the motor B by a predetermined value (J / τ−D), and an addition that adds the torque command and a value calculated by the multiplier 50. 52, a low-pass filter 54 that multiplies the value calculated by the adder 52 by a predetermined value (1 / (τs + 1)), a multiplier 56 that multiplies the speed of the motor B by a predetermined value (J / τ), and a low-pass A subtractor 58) for subtracting the value calculated by the multiplier 56 from the output from the filter 54 to obtain a disturbance torque (estimated disturbance) is provided. As will be described later, J is the moment of inertia on the rotor side of the motor B, τ is the filter time constant, and D is the friction coefficient.

As shown in FIG. 5, the PI controller 42 is connected to the subtractor 35 and the initial value I ₀ is set by the calculator 24. Thereafter, the integrator 60 integrates the value W _err calculated by the subtractor 35. And a multiplier 62 that multiplies the value calculated by the integrator 60 by a predetermined parameter (1 / Tω). The PI controller 42 also adds an adder 64 that adds the value W _err calculated by the subtractor 35 and the value calculated by the multiplier 62, and multiplies the value added by the adder 64 by Kω. And a multiplier 66 for calculating a torque command value.

  Next, the operation of the present embodiment will be described.

  In the present embodiment, the rotational state of the motor B is controlled to the target rotational state of the motor A, and the speed between the rotational speed of the motor B and the target rotational speed of the motor A as shown in FIG. When the difference is large, feed-forward control (see the maximum torque section T1) that accelerates (or decelerates) at the maximum allowable torque of the motor B is performed. That is, the calculator 24 controls the switch 45 so that the allowable maximum torque from the calculator 24 is commanded to the current controller of the motor B.

  As described above, during the feed-forward control, the rotation state of the motor B can be fed back to the target rotation state without largely reversing the sign of the torque during the feed-forward control. When it is determined that the position is within the predetermined range, the position is switched to position control by feedback control (see position control section T2) using the feedback control model.

  Next, the feedback control model will be described.

  The speed controller is a PI control 42, the position controller is a P control 34, and the transfer function of each controller is designed as follows.

At this time, assuming T _w >> j / Kω (generally valid assumption), the frequency response of the velocity loop is

And the frequency response of the position loop is

It becomes.

For further simplicity, if K _p = γ ² h / τ _r , the frequency response characteristic of the position loop is

It becomes. Here, h is the pitch angle of the gear, and h = 2π / N (N: number of teeth (for example, 60)). In addition, γ is a parameter that determines this characteristic.

FIG. 7 shows the position error (unit: pitch) of the step response when γ ² is variously set. Here, the step input value (initial value of the position error) is set to 1. The values of γ ² are Binominal: 0.25, Butterworth, O.5, and CDM: O. 4. The response oscillates as γ ² increases. Binominal is called critical braking when γ = 0.25, CDM (Coefficient Diagram Method) means coefficient diagram, and Binominal, Butterworth, and CDM are feedback control models. It is.

  Looking at the settling time α of these waveforms, as shown in FIG. 7, the normalized time is 5.O for Binominak, 3.3 for Butterworth, and 4.1 for CDM. Binominal's settling time is the longest, but the others are oscillatory, so there is some overshoot. The response waveform to be used is determined in light of the actual system requirements.

  Generally, in the step response of the position control, the initial torque fluctuation is the largest, and then the torque tends to decrease as the error converges. Therefore, the constraint condition of the control parameter is obtained from the maximum value of the initial torque. The initial generated torque is given by the following equation since the value of the integrator 60 of the PI controller 42 is zero.

  Here, when switching to position control is considered after the speed error has converged to some extent, if the second term on the right side of the above equation is ignored, a relational expression with the maximum torque is derived as follows.

  Organize the above formula

It becomes. This is an intuitive relational expression that if there is a torque limitation, the gain for position control must be increased and the gain for speed control must be decreased.

  At the time of design, first, γ is determined from a desired response waveform, and then Kω is determined based on equation (9). Then, τγ is obtained (see equation (3)).

  In addition, the maximum estimate of the synchronization time at that time is

It becomes. ω _diff is the speed difference at the start of maximum torque control.

 In the present embodiment, the control system is employed in the Bimmia1 model in consideration of the ease of analysis and the non-vibrating response waveform.

  Next, a method for determining a switching point for switching from feed-forward control to feedback control will be described.

  During maximum torque control, there is no causal relationship between speed error and position error, so even if switching from feed-forward control to feedback control from the middle, the consistency with the internal state of the controller is not achieved, and the error does not converge. Sometimes.

  Therefore, it is considered to search for a point where the relationship between the speed and position error that changes every moment during the maximum torque control and the time response of the controller coincide.

  The time response of position error and velocity error in Binominal model is as follows.

Here, Δθ ₀ is a position error (initial error) when t = 0.

  Torque is calculated by multiplying acceleration by moment of inertia.

Given in.

  In addition, the ratio of speed error to position error

From the formulas (11) and (12)

This relationship is derived. It can be seen that T _n is a normalization time and δ takes a range of 0 ≦ δ <γ.

These relationships are shown in FIG. In FIG. 8, the initial error Δθo is set to 1 for convenience. It can be seen that this waveform has the same response as the experimental result of FIG. That is, performing position control after speed synchronization is equivalent to starting control from time T _n = 0 in FIG.

If the control is switched from the normalized time T _n = 1, the speed converges quickly without reacceleration, and the torque changes smoothly from zero. From this idea, it can be seen that the switching time can be determined so that the control starts from the region of T _{n =} = 1 (region on the right side of the green line) in FIG. In the relevant region, δ is

Take a range of. Therefore, if the position control is started when δ enters this range during the maximum torque control, the control can be performed without greatly reversing the sign of the torque.

Next, the initial value I ₀ given to the integrator 60 of the PI controller 42 will be described. In the present embodiment, the integrator 60 of the PI controller 42 is not operated from the beginning. As described above, the integrator 60 of the PI controller 42 is switched from the point of time when the feed-forward control is switched to the feedback control. , The initial value I _{0 at} this time is required.

As described above, if the value of δ taken when δ enters the range defined by equation (16) during the maximum torque control is δ _o and the normalized time T _nq at that point, the normalized time T _nq is obtained by the following equation:

  Furthermore, the initial error (virtual value) is

Is required. That is, the relationship between the speed error and the position error is as if the initial error Δθ. _Will be associated as a value when the time T _{n0 has} passed.

On the other hand, it is necessary to consider the torque limit. The torque after time _Tn0 is expressed by the following equation.

  After this time, the maximum torque that can be taken is

Given in.

  When the load torque is applied to the motor B, the estimated value of the disturbance torque estimated by the disturbance observer 48 may be added to the right side of the above equation.

  From the above examination, when the conditional expression (16) of δ is satisfied, the maximum torque is estimated based on the above expression, and when this is within the allowable range of the motor, it is sufficient to switch to the position control.

Torque T _q0 required at the time of switching is _calculated from equations (13) and (18)

Given in. Therefore, the torque may be calculated by the above equation using the position error Δθ _p at the time of switching, and the initial value T ₀ of the integrator of the PI controller may be calculated by the following equation.

  FIG. 9 shows a flowchart showing a switching control program for switching from feed-forward control to feed-back control executed by the calculator 24.

  In order to mesh the motor A and the motor B, the rotational speed of the motor B is reduced. At this time, in order to generate the maximum torque in the motor B, the switch 45 is switched to the feedforward side, and the maximum torque is commanded from the calculator 24 to the current controller of the motor B via the switch 45. As a result, the rotational speed of the motor B gradually decreases and approaches the rotational speed of the motor A.

  This switching control program starts during such feed-forward control. In step 72, the rotational positions of the motors A and B are detected from the signals of the encoders 18 and 20, and the rotational speed is detected from the detected rotational positions. To do.

In step 74, the position error Δθ _p is detected from the rotational positions of the motors A and B, and the speed error Δω is detected from the rotational speeds of the motors A and B.

In step 76, the relationship between the position error Δθ _p and the speed error Δω, for example, the ratio δ (see equation (14)) is calculated.

  In step 78, δ is within a predetermined range in which the rotational state can be feedback-controlled to the target rotational state without inverting the sign of the torque during the feed-forward control, that is, for example, Expression (16) It is determined whether it is within a predetermined range determined by. When it is determined that δ is within the predetermined range determined by the equation (16), the maximum torque is estimated using the equations (20) and (21) in step 80, and the estimated maximum torque is determined in step 82. Is determined to be within the allowable range.

  When it is determined that the estimated maximum torque is within the allowable range, the initial value (determined from the equation (14)) is set in the integration circuit 60 in step 84, and the switch 45 is controlled in step 86 to perform feedback control. Switch to.

In the feedback control, the subtractor 32 calculates the tooth position of the dog tooth 26 of the motor B calculated by the tooth position calculator 30 and the tooth position of the dog tooth 26 of the motor A calculated by the tooth position calculator 31. A position error Δθ _p is obtained. The P controller 34 multiplies the position error Δθ _p by a predetermined transfer function (K _p ) of the position controller.

  The subtractor 40 calculates a speed error Δω between the rotation speed of the motor A calculated by the speed calculator 36 and the rotation speed of the motor B calculated by the speed calculator 38.

The subtracter 35 subtracts the speed error Δω obtained by the subtractor 40 from the value calculated by the P controller 34 (transfer function (K _p ) × position error Δθ _p ).

By the way, when switching to the feedback control, the calculator 24 calculates the initial value I ₀ from the above equation (24) and supplies it to the integrator 60 of the PI controller 42. The PI controller 42 obtains a torque command value using the initial value I ₀ and outputs it to the adder 44.

  The adder 44 adds the torque command value calculated by the PI controller 42 and the disturbance torque calculated by the disturbance observer 48 to obtain a final torque command value, and obtains a current controller for the motor B. To control the motor B.

  Thereafter, the PI controller 42 adds (integrates) the speed error Δω obtained by the subtractor 40 to obtain a torque command value and outputs it to the adder 44. Thereafter, the same processing as described above is performed.

  Thus, by feedback control, the rotational speed of motor B approaches the rotational speed of motor A as shown in FIG. 10A, and the position error of the dog teeth of motors A and B as shown in FIG. Becomes smaller.

  By the way, as described above, when it is determined that δ is within the predetermined range determined by the equation (16), the torque is still a minus sign, and as shown in FIG. Even when switching to, the sign of torque is not largely reversed, and generation of sound and vibration can be prevented.

  In addition, after the speeds of the motors A and B match, the speed of the motor B is not reaccelerated, but is switched to feedback control before the speeds of the motors A and B match, so as shown in FIG. In addition, the control time can be shortened. That is, in the conventional case where the speed of the motor B is reaccelerated after the speeds of the motors A and B coincide with each other, the control time which required 194 (ms) is shortened to 178 (ms) in this method. We were able to.

  In the embodiment described above, the range of δ is

However, in order to shift more smoothly from the maximum torque, it becomes a region of Tn ≧ 2.

It is desirable to limit to the range. At this time, the formula for estimating the maximum torque is

Thus, the calculation of the exponential function becomes unnecessary.

  Further, a predetermined range in which the rotation state of the motor B can be fed back and controlled to the target rotation state without greatly reversing the sign of the torque during the feed-forward control using the value of δ. However, the present invention is not limited to this, and the following values may be used.

  That is, equations (14) to (16) can be expressed as follows.

  When this is developed, for example, it may be determined whether the speed error is within the following predetermined range.

  Further, it may be determined whether the predetermined range of the position error is within the following predetermined range.

  Note that the feedback control model is not limited to Binominal, and switching can be performed in the same way using Butterworth or CDM.

  Note that the rotational state of the motor B may be controlled to a predetermined target rotational state without considering the rotational state of the motor A.

It is a block diagram of the rotary body control apparatus concerning this Embodiment. It is a figure which shows the mode of engagement of the dog teeth of motors A and B. It is a block diagram which mainly shows a current controller. It is a block diagram of a disturbance observer. It is a block diagram of a PI controller. It is a figure which shows the relationship between the time at the time of switching from feed forward control to feed back control, and the speed of the motor B. FIG. It is a figure which shows three control models in feedback control by the relationship between normalization time and a position error. (A)-(C) are figures which show a Binominal control model. It is a flowchart which shows the switching control program which switches from feed forward control to feed back control. (A)-(C) are figures which show the control result in this Embodiment. It is a figure which shows the comparison result of each control time of the past and this Embodiment. This is a result when the conventional feed-forward control is switched to the feed-back control.

Explanation of symbols

18 Encoder (second detection means)
20 Encoder (first detection means)
22 Current controller (control means)
24 Computer (control means)

Claims (15)

First detecting means for detecting a rotation state of the first rotating body;
Second detection means for detecting a rotation state of the second rotating body;
A predetermined torque is generated in the first rotating body to perform feed-forward control. From the feed-forward control, the rotation state of the first rotating body detected by the first detecting means, the second Based on the target rotation state determined from the rotation state of the second rotating body detected by the detecting means, and the feedback control model determined in advance by the torque and rotation state generated in the first rotating body. Control means for controlling the rotation state of the first rotation body to the target rotation state by switching to feedback control for changing the torque generated in the rotation body of
A rotating body control device comprising:
The control means includes
During the feed-forward control, the rotation state of the first rotating body detected by the first detection means is the target rotation without inverting the sign of the torque during the feed-forward control. Judge whether it is within a predetermined range that can be feedback controlled to the state,
When the rotation state is determined to be within the predetermined range, the feed / forward control is switched to the feedback control.
A rotating body control device characterized by the above. The first detecting means detects a rotational position of the first rotating body;
The control means determines in advance that the rotational state can be controlled to the target rotational state based on at least one of the detected rotational position and a rotational speed determined based on the rotational position and the target rotational state. To determine whether it is within the specified range,
The rotating body control device according to claim 1.   The control means has a relationship between a positional error between the detected rotational position and a target rotational position determined based on the target rotational state, and a speed error between the rotational speed and the target rotational speed determined based on the target rotational state. The rotating body control device according to claim 2, wherein it is determined whether or not the rotational state is within a predetermined range that can be controlled to the target rotational state.   The rotator control apparatus according to any one of claims 1 to 3, wherein a predetermined range that can be controlled to the target rotation state is determined based on a control parameter of the feedback control model. . The control means includes
Estimate the maximum value of torque required after switching to the feedback control,
Determining whether the estimated maximum value of the torque is within a predetermined allowable range;
When it is determined that the estimated maximum value of the torque is within the predetermined allowable range, the feed-forward control is switched to the feedback control.
The rotating body control device according to any one of claims 1 to 4. The control means includes
When load torque is applied to the first rotating body, it is determined whether or not the sum of the estimated maximum value of the torque and the load torque is within a predetermined allowable range;
When it is determined that the sum is within the predetermined allowable range, the feed-forward control is switched to the feedback control.
The rotator control device according to claim 5. The control means includes
In the feed forward control, the rotational speed of the first rotating body is brought close to a target rotational speed determined based on the target rotational state,
In the feedback control, the rotational position of the first rotating body is brought close to a target rotational position determined based on the target rotational state.
The rotating body control device according to any one of claims 1 to 6, wherein Feed-back control in which a predetermined torque is generated in the first rotating body to perform feed-forward control, and the torque generated in the first rotating body is changed based on the feed-back control model from the feed-forward control. And the rotation state of the first rotation body is controlled to a target rotation state determined from the rotation state of the second rotation body,
During the feed-forward control, the rotation state of the first rotating body is changed to the target rotation state based on the feedback control model without inverting the sign of the torque during the feed-forward control. To determine whether it is within a predetermined range that can be controlled,
When it is determined that the rotation state is within a predetermined range that can be controlled to the target rotation state, the feed / forward control is switched to the feedback control.
A rotating body control method characterized by the above. Detection means for detecting the rotation state of the rotating body;
A predetermined torque is generated in the rotating body to perform feed-forward control, and from the feed-forward control, the rotation state of the rotating body detected by the detection means, a predetermined target rotation state, and the rotation body are generated. Control for switching the rotation state of the rotating body to the target rotation state by switching to the feedback control that changes the torque generated in the rotating body based on a feedback control model determined in advance by the torque to be rotated and the rotation state Means,
A rotating body control device comprising:
The control means is configured such that during the feed-forward control, the rotation state of the rotating body detected by the detection means is changed to the target rotation state without greatly reversing the torque during the feed-forward control. Judge whether it is within a predetermined range that can be feedback controlled,
When the rotational state is determined to be within the predetermined range, the feed control is switched from the feed forward control to the feed back control. The detecting means detects a rotational position of the rotating body;
The control means determines in advance that the rotational state can be controlled to the target rotational state based on at least one of the detected rotational position and a rotational speed determined based on the rotational position and the target rotational state. To determine whether it is within the specified range,
The rotating body control device according to claim 9.   The control means has a relationship between a positional error between the detected rotational position and a target rotational position determined based on the target rotational state, and a speed error between the rotational speed and the target rotational speed determined based on the target rotational state. 11. The rotating body control device according to claim 10, wherein it is determined whether or not the rotational state is within a predetermined range that can be controlled to the target rotational state.   12. The rotating body control device according to claim 9, wherein a predetermined range that can be controlled to the target rotation state is determined based on a control parameter of the feedback control model. . The control means includes
Estimate the maximum value of torque required after switching to the feedback control,
Determining whether the estimated maximum value of the torque is within a predetermined allowable range;
When it is determined that the estimated maximum value of the torque is within the predetermined allowable range, the feed-forward control is switched to the feedback control.
The rotating body control device according to any one of claims 9 to 12. The control means includes
When a load torque is applied to the rotating body, it is determined whether or not a sum of the estimated maximum value of the torque and the load torque is within a predetermined allowable range;
When it is determined that the sum is within the predetermined allowable range, the feed-forward control is switched to the feedback control.
The rotating body control device according to claim 13. The control means includes
In the feed-forward control, the rotational speed of the rotating body is brought close to a target rotational speed determined based on the target rotational state,
In the feedback control, the rotational position of the rotating body is brought close to a target rotational position determined based on the target rotational state.
The rotator control apparatus according to claim 9, wherein the rotator control apparatus is configured as described above.

Images