JP6059891B2 - Positioning device control device and electronic component mounting device - Google Patents

Description

  The present invention relates to a controller for a positioning device used in an electronic component mounting apparatus such as a chip mounter.

  Generally, a positioning device used in a chip mounter or the like has a configuration in which a ball screw is connected to a drive shaft of a rotary motor, or a configuration in which a timing belt is stretched around a pulley fixed to the drive shaft of the rotary motor. . In this positioning device, the load is positioned at a predetermined position by transmitting the output from the rotary motor to a load that is a mover via a ball screw or a timing belt. In this case, since the motor and the load are connected by an elastic body such as a ball screw or a timing belt, it can be regarded as a resonance system having two or more inertias. In such a resonance system with two or more inertias, for example, it has been proposed to perform resonance ratio control provided with a plurality of disturbance observers (see Patent Document 1 and Non-Patent Document 1).

JP 2007-038884 A

Yuki et al., “Vibration Suppression Control of Two-Inertia Resonance System by Resonance Ratio Control”, IEEJ Transactions D, 1993, Vol. 113, No. 10, pages 1162-1169.

  When positioning using a chip mounter or the like, high-speed and high acceleration / deceleration control of the motor drive shaft is required in order to increase the productivity of mounted components. If the motor drive shaft speed and acceleration / deceleration are increased, overshoot is likely to occur during positioning settling when the control gain is greatly increased in order to increase the positioning performance to the upper limit. Since overshoot adversely affects settling time and positioning accuracy, it is desirable to eliminate overshoot. However, in order to eliminate overshoot, the speed and acceleration / deceleration must be lowered or the control gain must be lowered.

  The present invention has been made in view of such circumstances, and provides a positioning device control device and an electronic component mounting device capable of suppressing overshoot and performing high-speed and high-precision positioning. For the purpose.

A control device for a positioning device according to the present invention is a control device for a positioning device that positions a load at a predetermined position by a motor, a control system that controls the motor based on a current command value for the motor, A disturbance observer that estimates disturbance from the command value and motor speed and feeds back to the current command value for the motor, and the disturbance observer includes a friction parameter included in a part of the disturbance as the motor speed increases. It is a function that reduces friction and is based on a disturbance model including the function .

  According to this configuration, overshoot can be suppressed by the friction parameter of the disturbance observer, and high-speed settling and high-accuracy positioning can be performed. Since it is only necessary to provide a friction parameter as a disturbance observer disturbance, the amount of calculation is small compared to state feedback control and H∞ control, and an expensive CPU or the like is unnecessary. It is also possible to make adjustments on site. Moreover, since overshoot is suppressed without increasing the rigidity of the apparatus, it is possible to reduce the weight of the entire apparatus and realize energy saving and cost reduction.

In the control device for a positioning apparatus according to the present invention, the disturbance observer is based on a disturbance model including a function of equation (1) as the friction parameter, where B is a friction coefficient, u is a motor speed, and a is a constant. ing.

In the control device of the positioning device of the present invention, the disturbance observer is based on a disturbance model including a function of equation (2) as the friction parameter, where B is a friction coefficient and u is a motor speed.

In the control device of the positioning device of the present invention, the disturbance observer is based on a disturbance model including a function of equation (3) as the friction parameter, where B is a friction coefficient, u is a motor speed, and a is a constant. ing.

  According to these configurations, since the influence of friction is reduced when the motor speed is increased, loss of motor output due to the friction parameter can be prevented. In addition, since the influence of friction increases when the motor speed is reduced, overshooting can be suppressed and the settling time can be shortened. Therefore, positioning can be performed with high speed and high accuracy.

In the control device of the positioning device of the present invention, the disturbance observer minimizes the friction coefficient when the motor is started. According to this configuration, it is possible to prevent the loss of the motor output due to the friction coefficient when the motor is started.

In the control device of the positioning device of the present invention, the disturbance observer minimizes the friction coefficient when driven by a motor speed equal to or lower than a predetermined speed. According to this configuration, at a motor speed that does not cause overshoot, loss of motor output due to the friction coefficient can be prevented.

  In the control device of the positioning device of the present invention, the disturbance observer has a plurality of friction coefficients with respect to the function as the friction parameter, and dynamically switches the friction coefficient according to the positioning position. According to this configuration, the optimum positioning can be performed by switching to the friction coefficient that matches the behavior of the overshoot that changes depending on the positioning position.

  In the control device for a positioning device according to the present invention, the disturbance observer dynamically adjusts the friction coefficient so that the amount of positional deviation is a predetermined value or less. According to this configuration, optimal positioning is possible by changing the friction coefficient in accordance with the amount of position deviation.

  According to another aspect of the invention, there is provided an electronic component mounting apparatus including the positioning device control apparatus. According to this configuration, the electronic component can be mounted on the substrate at high speed and with high accuracy.

  Another positioning device control device of the present invention is a positioning device control device for positioning a load at a predetermined position by a motor, the control system for controlling the motor based on a current command value for the motor, and the motor An angular position detector for detecting the angular position of the output shaft, a shaft torsional reaction force corresponding to the angular position of the output shaft, a current command value for the motor, and a motor speed to estimate a disturbance, and a current to the motor A disturbance observer for feeding back to the command value is provided.

  According to this configuration, since the disturbance due to the shaft torsional reaction force is estimated and fed back to the current command value for the motor, the vibration can be converged early even if the rigidity between the motor and the load is small. it can. In addition, since overshoot during positioning is suppressed, the settling time can be shortened and positioning can be performed at high speed and with high accuracy. In addition, since the operation on the load side can be reflected in the control through the axial torsional reaction force, it is not affected by fluctuations in the nominal value of the load inertia. Further, compared with state feedback control, H∞ control, and resonance ratio control, the calculation amount is small and the amount of calculation is small, and an expensive CPU or the like is not necessary. Furthermore, since overshoot is suppressed without increasing the rigidity of the apparatus, it is possible to reduce the weight of the entire apparatus and realize energy saving and cost reduction.

  In the control device for the other positioning device according to the present invention, the disturbance observer is a system in which a load is connected to the output shaft of the motor with a spring property. A shaft torsional reaction force is calculated based on a transfer function that outputs force.

Also, in the control device for the other positioning device of the present invention, the disturbance observer has K _f as a spring constant, J _l as a load inertia, s as a Laplace operator, and θ _m as an angular position of the output shaft. The torsional reaction force is calculated using Equation (4) as the transfer function.

  According to these configurations, the axial torsional reaction force can be calculated without directly detecting the load moving position. Therefore, even in a semi-closed control system that does not have a moving position detector such as a linear encoder, the disturbance observer can be made to estimate a disturbance in consideration of the axial torsional reaction force.

  In the control device for another positioning device of the present invention, the disturbance observer corrects a spring constant used for the transfer function in accordance with a moving position of the load. According to this configuration, by correcting the spring constant according to the rigidity between the motor and the load that changes according to the moving position of the load, higher robustness can be maintained and vibration can be effectively suppressed.

  The control device for the other positioning device of the present invention further includes a moving position detector for detecting the moving position of the load, and the disturbance observer is detected by the angular position of the output shaft and the moving position detector. A shaft torsion reaction force is calculated based on the load moving position. According to this configuration, it is possible to directly detect the moving position of the load and calculate the shaft twist reaction force. Therefore, even in a full-closed control system in which the position of the load is directly detected by the linear encoder, it is possible to estimate the disturbance in consideration of the shaft torsion reaction force with respect to the disturbance observer.

  In the control device for the other positioning device of the present invention, the disturbance observer may calculate the shaft torsional reaction force by multiplying a difference between an angular position of the output shaft and a moving position of the load by a spring constant. Therefore, the spring constant is corrected according to the moving position of the load. According to this configuration, by correcting the spring constant according to the rigidity between the motor and the load that changes according to the moving position of the load, higher robustness can be maintained and vibration can be effectively suppressed.

  According to another aspect of the invention, there is provided an electronic component mounting apparatus including the control device for the other positioning device. According to this configuration, the electronic component can be mounted on the substrate at high speed and with high accuracy.

  According to the present invention, by designing a disturbance observer based on a disturbance model that includes a friction parameter as a part of the disturbance, it is possible to suppress overshoot and perform high-speed and high-accuracy positioning.

1 is an overall view of an electronic component mounting apparatus. It is a block diagram of a normal disturbance observer. It is a block diagram of a disturbance observer including a friction parameter. It is a block diagram of a disturbance observer including a friction parameter. It is a block diagram of the control apparatus of the positioning device to which the disturbance observer including a friction parameter is applied. It is explanatory drawing of the characteristic of the function shown to Formula (1). It is a simulation result of positioning settling when the function of Formula (1) is used. It is explanatory drawing of the characteristic of the function shown to Formula (2). It is explanatory drawing of the characteristic of the function shown to Formula (3). It is explanatory drawing of the model of a 2 inertia resonance system. It is a block diagram of the control apparatus of the positioning device to which a normal disturbance observer is applied. It is a block diagram of a control device of a positioning device to which a shaft torsion reaction force feedback type disturbance observer is applied in full closed control. It is a block diagram of a control device of a positioning device to which a shaft torsion reaction force feedback type disturbance observer is applied in semi-closed control. It is a simulation result of positioning settling in case a spring constant is 1250. It is a simulation result of positioning settling in case a spring constant is 120. It is a simulation result of positioning settling in case a spring constant is 10. It is a simulation result of positioning settling when adjusting an observer gain. It is a simulation result of positioning settling when the nominal value of load inertia changes. It is a simulation result of positioning settling when the nominal value of load inertia changes.

  Hereinafter, a first embodiment of the present invention will be described in detail with reference to the accompanying drawings with reference to FIGS. FIG. 1 is an overall view of an electronic component mounting apparatus according to the present embodiment. In the following description, a configuration in which the control device for a positioning device of the present invention is applied to a modular mounting device will be described as an example, but the present invention is not limited to this. For example, the control device of the positioning device of the present invention can be applied to a positioning device used in a rotary type mounting device.

  The mounting apparatus 1 is configured to mount the electronic component supplied from the feeder 3 on the substrate W (base material) on the base 2 by the mounting head 4. A mounting head moving mechanism 5 that moves the mounting head 4 in the X-axis direction and the Y-axis direction is provided on the base 2. The mounting head moving mechanism 5 is supported by support columns 6 erected at the four corners of the base 2, and moves the mounting head 4 in the X-axis direction and the Y-axis direction at a predetermined height from the upper surface of the base 2. .

  The mounting head moving mechanism 5 moves the mounting head 4 (load) in the X-axis direction along the X-axis table 11 (load) by a motor (not shown). Further, the mounting head moving mechanism 5 moves the mounting head 4 together with the X-axis table 11 along the Y-axis table 12 and the slide guide 13 in the Y-axis direction by a motor (not shown). With such a configuration, the mounting head 4 is moved horizontally above the substrate W, and can transport the electronic component supplied from the feeder 3 to a desired position on the substrate W.

  On the base 2, a substrate transport unit 7 is provided below the mounting head 4 to position the substrate W (load). The board transport unit 7 takes in the board W before component mounting from one end side and transports the board W after component mounting from the other end side by a transport belt or the like extending in the X-axis direction. The control device for the positioning device according to the present embodiment is applied to the mounting head moving mechanism 5 and the substrate transport unit 7 that are motor-driven in the mounting device 1. And in the structure which connected the motor and load with elastic bodies, such as a molding screw and a timing belt, high-speed and highly accurate positioning is realized.

A normal disturbance observer will be described with reference to FIG. FIG. 2 is a block diagram of a normal disturbance observer. In FIG. 2, I _ref is a current command value, I _cmp is a current compensation value, K _t is a torque constant, K _tn is a nominal value of the torque constant, J _m is a motor inertia, J _mn is a nominal value of the motor inertia, s is the Laplace operator, T _dism is the disturbance torque, T ^{^} _dism is the estimated value of the disturbance torque, θ ^· _m is the motor speed, and G _dis is the observer gain. The motor speed θ ^· _m is measured based on information from an encoder provided on the output shaft of the motor, for example. In the following description, “·” attached to the symbol θ represents differentiation, and “^” attached to the symbol T represents an estimated value.

As shown in FIG. 2, the motor control system is configured to compensate the disturbance torque T _dism to the motor by the disturbance observer 10. The current command value I _ref is multiplied by the torque constant K _{t in the} calculation element e102, and is output to the subtraction element e103 as the torque command value. A deviation of the torque command value from the disturbance torque T _dism is calculated by a subtraction element e103. The torque deviation is divided by the motor inertia J _m by the calculation element E104, those that are further integrated the motor speed theta ^· _m. Here, the transfer function from the current command value I _ref to the motor speed θ ^· _m is expressed by the following equation (5).

電流指令値Ｉ_refとモータ速度θ^・ _mから外乱トルクを推定演算できることが分かる。外乱オブザーバ１０は、この原理を用いて外乱トルクを推定してモータ制御系にフィードバックすることで外乱を除去している。 When formula (5) is arranged with respect to disturbance torque T _dism , the following formula (6) is obtained.

It can be seen that the disturbance torque can be estimated and calculated from the current command value I _ref and the motor speed θ ^· _m . The disturbance observer 10 eliminates the disturbance by estimating the disturbance torque using this principle and feeding it back to the motor control system.

In this case, the disturbance observer 10 is inputted with the current command value I _ref and the motor speed θ ^· _m, and the estimated value T ^{^} _dism of the disturbance torque is used as a command value for removing the disturbance to the motor from these values. Calculated. In the disturbance observer 10, the current command value I _ref is multiplied by the nominal value K _tn of the torque constant in the calculation element e 105, and the nominal value J _mn of the motor inertia is obtained by differentiating the motor speed θ ^· _m in the calculation element e 106. Multiply to match the unit.

Then, the output from the calculation element e105 and the output from the calculation element e106 are compared by the subtraction element e107, and the low frequency band component is extracted by the low-pass filter e108, and this is used as the disturbance torque estimated value T ^{^} _dism. Is output. Estimate T ^{^} _dism of disturbance torque is divided by the nominal value K _tn torque constant by the computing element E109, it is converted to a compensation current value I _cmp for compensating for the disturbance. The compensation current value I _cmp is added to the current command value I _ref input to the motor by the addition element e101. In this way, the estimated value T ^{^} _dism of the disturbance torque is fed back to the motor control system.

  In this embodiment, by adding a friction parameter (friction term) in the disturbance observer, overshoot can be suppressed even in positioning at high speed and high acceleration / deceleration. Hereinafter, a motor control system to which the disturbance observer according to the present embodiment is applied will be described with reference to FIGS.

3 and 4 are block diagrams of a disturbance observer including friction parameters. FIG. 5 is a block diagram of the controller of the positioning device to which the disturbance observer including the friction parameter is applied. 3 to FIG. 5, I _ref , I _cmp , K _t , K _tn , J _m , J _mn , s, T _dism , T ^{^} _dism , θ ^· _m , and G _dis are the same as those in FIG. K _p is the position gain, K _v is the speed gain, K _f is the spring constant (shaft stiffness), J _l is the load inertia, T _disl is the load side disturbance torque, T _reac is the shaft twist reaction force, and θ _cmd is The target position, θ _m is the motor position, θ _l is the load position, θ _t is the amount of shaft twist, θ ^· _cmd is the speed command value, and B is the friction.

As shown in FIG. 3, in the motor control system according to the present embodiment, the motor speed θ ^· _m is calculated from the current command value I _ref including the friction B as part of the disturbance. In this case, the transfer function from the current command value I _ref to the motor speed θ ^· _m is expressed by the following equation (7).

図３に示すような外乱オブザーバ２０を構成することができる。図３に示す外乱オブザーバ２０を実用的に変形すると図４のような構成にすることができる。 When formula (7) is arranged for disturbance torque T _dism , the following formula (8) is obtained.

A disturbance observer 20 as shown in FIG. 3 can be configured. If the disturbance observer 20 shown in FIG. 3 is practically modified, the configuration shown in FIG. 4 can be obtained.

Next, a description will be given of a controller for the positioning device to which the disturbance observer is applied. As shown in FIG. 5, the motor position theta _m fed back from the position command value theta _cmd and motor in subtraction element e201 target is subtracted, the position deviation is calculated. Position deviation is multiplied position gain K _p is at computing element E202, is output to the subtraction element e203 as a speed command value theta ^· _cmd. In the subtraction element e203, the speed command value θ ^· _cmd and the motor speed θ ^· _m fed back from the motor are subtracted to calculate a speed deviation. Speed deviation is multiplied speed gain K _v by computing elements e204 is a unit are output to the summing element e206 as converted by the current command value I _ref by the calculating element E205.

The current command value I _ref is added to the current compensation value I _cmp by the addition element e206 and output to the calculation element e207 and the disturbance observer 20. The current command value I _ref is multiplied by the torque constant Kt in the calculation element e207, and is output to the subtraction element e208 as a torque command value. The torque command value is calculated as a deviation from the disturbance torque T _dism by the subtraction element e208, and is further calculated as a deviation from the shaft torsional reaction force T _reac by the subtraction element e209. The torque deviation is divided by the motor inertia J _m by the calculation element E211, those that are further integrated the motor speed theta ^· _m. The motor speed θ ^· _m is output to the calculation element e212 and the disturbance observer 20, and is fed back to the subtraction element e203 in order to obtain a speed deviation.

Motor speed theta ^· _m is is integrated and converted into motor position theta _m by the calculation element E212. The motor position θ _m is an angular position of the output shaft of the motor, and is measured based on information of a rotary encoder as an angular position detector provided on the output shaft, for example. The angular position detector may have any configuration as long as it can detect the angular position of the output shaft. The motor position θ _m is output to the subtraction element e213 and is fed back to the subtraction element e201 in order to obtain a position deviation. The motor position θ _m is subtracted from the load position θ _l fed back from the load side by the subtraction element e 213, and the shaft twist amount θ _t is calculated. The load position θ _l is a moving position (angular position) of the load moved by driving the motor.

Torsional amount theta _t is multiplied spring constant K _f is in operation element E214, is fed back to the subtraction element e209 is output to the subtraction element e215 as torsional reaction force T _REAC. The shaft torsional reaction force T _reac is subtracted from the load-side disturbance torque T _disl by the subtraction element e215 and output to the calculation element e216. The shaft torsion reaction force T _reac is divided by the load inertia J _l at the calculation element e 216 and integrated at the calculation element e 217 to calculate the load position θ _l . Load position theta _l is fed back to the subtraction element e213 to determine the torsional amount theta _t.

The disturbance observer 20 receives the current command value I _ref and the motor speed θ ^· _m . The current command value I _ref is multiplied by the nominal value K _{tn of the} torque constant at the calculation element e218 and output to the addition element e221 on the upstream side of the low-pass filter e222. The motor speed θ ^· _m is converted into a part of the disturbance by the calculation elements e219 and e223 in consideration of the friction B, and is output to the addition element e221 on the front stage side and the subtraction element e224 on the rear stage side of the low-pass filter e222. . A low frequency band component is extracted from the output signal from the addition element e221 by the low-pass filter e222, and the output signal from the calculation element e223 is subtracted by the subtraction element e224, whereby the estimated value T ^{^} _dism of the disturbance torque is obtained. Calculated.

Estimate T ^{^} _dism of disturbance torque is divided by the nominal value K _tn torque constant by the computing element E 225, it is converted to a compensation current value I _cmp for compensating for the disturbance. The compensation current value I _cmp is fed back to the addition element e206 and added to the command current value _Iref . As described above, the disturbance observer 20 according to the present embodiment feeds back the compensation current value I _cmp considering the friction to the command current value I _ref . Therefore, by adjusting the value of the friction parameter B, it is possible to obtain the behavior of the motor when arbitrary friction is applied to the motor control system.

By the way, it is possible to reduce the overshoot by applying friction to the disturbance observer 20, but the friction acts even during high-speed driving and the motor output is lost. For this reason, it is desirable that friction does not affect during high-speed driving, and that friction acts only during positioning settling. Thus, the friction B can be used as a function instead of a constant parameter. For example, a function represented by the following equation (1) can be used as the friction. In equation (1), B represents a friction coefficient, u represents an input, and f (u) represents an output.

  Hereinafter, the characteristic of the function shown in Formula (1) as friction will be described. FIG. 6 is an explanatory diagram of the characteristics of the function shown in Expression (1). FIG. 6 shows a case where the friction coefficient B is set to 0.3.

  As shown in FIG. 6, equation (1) shows that the input f is 0 and the output f (u) has the maximum value (B = 0.3), and the output f (u) increases as the input u increases. It has a characteristic of approaching zero. By using this function, the influence of friction can be eliminated during high-speed driving, that is, when the input u is large. Therefore, loss of motor output can be prevented. On the other hand, at the time of positioning settling, that is, when the input u approaches 0, the friction increases as the speed decreases, and the overshoot can be reduced to shorten the settling time.

When positioning settling was performed using the function of the equation (1), a simulation result as shown in FIG. 7 was obtained. Here, as shown in FIG. 7A, the parameters of the motor control system are set, and the simulation is performed for each of the case where friction is not applied (B = 0) and the case where friction is applied (B = 5.0 × 10 ⁻⁴ ). Carried out. The settling time is the time from the end of the position command until the response signal converges within ± 0.01% of the target positioning position. Moreover, the simulation results on the right side of FIGS. 7B and 7C are partially enlarged views of the simulation results on the left side of the drawings.

  As shown in FIG. 7B, when friction is not applied, when a step-like position command is input, the overshoot of the response signal at the time of positioning settling is large. For this reason, the settling time until the response signal converges from the end of the position command becomes 104 ms and becomes long. On the other hand, as shown in FIG. 7C, when friction is applied, if a step-like position command is input, the overshoot of the response signal during positioning settling is small. For this reason, the settling time until the response signal converges from the end of the position command is 57 ms, which is significantly shortened compared to the case where no friction is applied.

Note that the function is not limited to the function of Expression (1). Any function that reduces the friction as the motor speed increases can be used. For example, the functions shown in the following expressions (2) and (3) can be used instead of the function of the expression (1). In Expressions (2) and (3), B indicates a friction coefficient, u indicates an input, and f (u) indicates an output.

  The characteristics of the functions shown in equations (2) and (3) as friction will be described. FIG. 8 is an explanatory diagram of the characteristics of the function shown in Expression (2). FIG. 9 is an explanatory diagram of the characteristics of the function shown in Expression (3). FIG. 8 shows a case where the friction coefficient B in the equation (2) is set to 0.3. FIG. 9 shows a case where the friction coefficient B in the equation (3) is set to 0.3 and a is set to 1.

  As shown in FIG. 8, the equation (2) indicates that the input u (velocity) is 0 and the output f (u) is the maximum value (B = 0.3), and the output f ( u) has the property of approaching zero. By using this function, the effect of friction can be eliminated during speed driving as in the case where equation (1) is used as a function, and the settling time can be reduced by reducing the overshoot during positioning settling. Can be shortened.

  As shown in FIG. 9, Equation (3) shows that the input u (velocity) is 0 and the output f (u) is the maximum value (B = 0.3), and the output f (u ) Approaches 0 and the output f (u) becomes 0 when the input u becomes a or more. By using this function, the effect of friction can be eliminated during speed driving as in the case where equation (1) is used as a function, and the settling time can be reduced by reducing the overshoot during positioning settling. Can be shortened. It should be noted that any function can be applied to the functions shown in the equations (1) to (3) according to the purpose.

  As described above, according to the control device of the positioning device according to the present embodiment, overshoot can be suppressed by the friction parameter of the disturbance observer 20, and high-speed settling and high-accuracy positioning can be performed. Since it is only necessary to provide a friction parameter as the disturbance of the disturbance observer 20, the calculation amount is small with a simple configuration compared to the state feedback control and the H∞ control, and an expensive CPU or the like is unnecessary. It is also easy to make adjustments on site. Moreover, since overshoot is suppressed without increasing the rigidity of the apparatus, it is possible to reduce the weight of the entire apparatus and realize energy saving and cost reduction.

  The present invention is not limited to the first embodiment, and can be implemented with various modifications. In the above-described embodiment, the size, shape, and the like illustrated in the accompanying drawings are not limited to this, and can be appropriately changed within a range in which the effects of the present invention are exhibited. In addition, various modifications can be made without departing from the scope of the object of the present invention.

  For example, in the block diagram of the control device according to the first embodiment described above, P control is used for speed calculation, but PI control, PD control, and PID control are used according to speed deviation and speed response. May be. Further, the control device for the positioning device according to the present embodiment can be applied not only to a rotary motor but also to a linear motor.

  For example, in the above formulas (1) to (3), the friction coefficient B may be minimized when the motor is started so that strong friction is applied only during positioning settling. Thereby, it is possible to distinguish between a low speed state at the time of starting the motor and a low speed state at the time of positioning settling, and to prevent loss of motor output due to friction at the time of starting the motor. The minimum value of the friction coefficient B may be 0, or may be a very small value near 0.

  In the above formulas (1) to (3), when driven at a low speed below a predetermined speed, the friction coefficient B is set to the minimum value so that strong friction is applied only at the time of positioning settling. Also good. As a result, loss of motor output due to friction can be prevented when driving continues at a speed that does not cause overshoot. The minimum value of the friction coefficient B may be 0, or may be a very small value near 0.

  Further, in the above formulas (1) to (3), a plurality of friction coefficients B may be provided, and the friction coefficient B may be dynamically switched according to the positioning position. As a result, the friction coefficient B can be switched in accordance with the behavior of the overshoot that changes depending on the load positioning position, and optimal positioning becomes possible. In this case, the correspondence relationship between the positioning position and the friction coefficient B is obtained in advance.

  Further, in the above formulas (1) to (3), even if the friction coefficient B is dynamically adjusted so that the amount of load side positional deviation or the amount of motor side positional deviation is less than a certain value. Good. Thereby, optimal positioning is attained by changing a friction coefficient according to a position deviation. In this case, the correspondence between the position deviation and the friction coefficient B is obtained in advance.

  Next, with reference to FIGS. 10 to 19, a second embodiment of the present invention will be described in detail with reference to the accompanying drawings. Note that the electronic component mounting apparatus according to the second embodiment is the same as the electronic component mounting apparatus according to the first embodiment, and a description thereof will be omitted here. In a positioning device such as a mounting head moving mechanism 5 used in an electronic component mounting device, it can be regarded as a two-inertia resonance system in which a motor and a load (mounting head 4 or the like) are coupled by a shaft having a spring property (see FIG. 10).

  When the mechanical rigidity between the motor and the load is low, vibration is generated during positioning, which deteriorates positioning accuracy and positioning time. Although the vibration can be suppressed by increasing the positioning rigidity, there is a limit to the vibration suppression due to the rigidity because of an increase in weight and cost. For such resonance suppression and disturbance suppression control, methods such as state feedback control, H∞ control, disturbance observer control, and resonance ratio control have been proposed.

  State feedback control and H∞ control are difficult to apply to actual machines because the control system is complicated and the amount of calculation becomes enormous. On the other hand, disturbance observer control and resonance ratio control are relatively simple control systems that are easy to adjust and highly practical. However, the disturbance observer is intended to compensate for disturbance torque and parameter fluctuations in one inertial system, and when applied to a two-inertia system with low rigidity, it may induce large vibrations due to load inertia and springiness. . In addition, the resonance ratio control requires an observer for estimating a shaft torsional reaction force in addition to a normal disturbance observer, and the amount of calculation increases as compared with a configuration in which a single observer is provided.

  Therefore, in the second embodiment, the shaft torsional reaction force is calculated from the angular position of the output shaft and the angular position (moving position) of the load, and the current command value for the motor, the motor speed, and the shaft torsional reaction force are calculated. A torsional reaction force feedback type disturbance observer that estimates the disturbance from the above is used. By using a shaft torsion reaction force feedback type disturbance observer, vibration suppression control of a two-inertia resonance load in which a motor and a load are coupled by a low-rigid elastic shaft can be easily realized.

Here, with reference to FIG. 11, a control device for a positioning device to which a normal disturbance observer is applied will be described. FIG. 11 is a block diagram of a controller for a positioning device to which a normal disturbance observer is applied. In FIG. 11, I _ref , I _cmp , K _p , K _v , K _t , K _tn , K _f , J _m , J _mn , J _l , s, T _dism , T ^{^} _dism , T _disl , T _reac , θ _cmd , θ _m , θ _l , θ _t , θ ^· _cmd , θ ^· _m , and G _dis are the same as those in FIG.

As shown in FIG. 11, the motor control system is configured so that the shaft torsional reaction force T _reac is subtracted from the torque command value together with the disturbance torque _dism between the calculation elements e307 to e311 and input to the motor. . The current command value I _ref is multiplied by a torque constant K _t and converted into a torque command value. The motor speed θ ^· _m is calculated from the torque deviation between the torque command value, the disturbance torque T _dism and the shaft torsional reaction force T _reac . Here, the transfer function from the current command value I _ref to the motor speed θ ^· _m (from the calculation element e307 to the calculation element e311) is expressed by the following equation (9).

In a normal disturbance observer, the torsional reaction force T _reac is handled as being included in the disturbance torque T _dism , but here, a case in which each is used separately will be considered. The shaft torsional reaction force T _reac is obtained by multiplying the amount of shaft torsion, which is the difference between the motor position θ _m that is the angular position of the output shaft of the motor and the load position θ _l that is the angular position of the load, by the spring constant K _f . It is shown by the following formula (10).

になり、電流指令値Ｉ_refとモータ速度θ^・ _mと軸ねじれ量（θ_m−θ_l）から外乱トルクを推定演算できることが分かる。フルクローズド制御系の軸ねじれ反力フィードバック型の外乱オブザーバ３０は、軸ねじれ量（θ_m−θ_l）とバネ定数のノミナル値Ｋ_fnとに基づいて算出される軸ねじれ反力の推定値Ｔ＾_reacがフィードバックされる構成となっている。 When formula (9) is arranged for disturbance T _dism and formula (10) is substituted,

Thus, it is understood that the disturbance torque can be estimated and calculated from the current command value I _ref , the motor speed θ ^· _m, and the shaft twist amount (θ _m −θ _l ). The shaft torsional reaction force feedback type disturbance observer 30 of the fully closed control system has an estimated value T of the shaft torsional reaction force calculated based on the shaft torsion amount (θ _m −θ _l ) and the nominal value K _{fn of the} spring constant. ^ _Reac is fed back.

A control device for a positioning device to which a shaft torsion reaction force feedback type disturbance observer is applied will be described. First, full closed control capable of directly detecting the load position will be described. FIG. 12 is a block diagram of a controller for a positioning device to which a shaft torsion reaction force feedback type disturbance observer is applied in full closed control. 12, I _ref , I _cmp , K _p , K _v , K _t , K _tn , K _f , J _m , J _mn , J _l , s, T _dism , T ^{^} _dism , T _disl , T _reac , θ _cmd , θ _m , θ _l , θ _t , θ ^· _cmd , θ ^· _m , and G _dis are the same as those in FIG. K _fn represents a nominal value of the spring constant.

As shown in FIG. 12, a motor position theta _m fed back from the position command value theta _cmd and motor in subtraction element e401 target is subtracted, the position deviation is calculated. Position deviation is multiplied position gain K _p is at computing element E402, is output to the subtraction element e403 as a speed command value theta ^· _cmd. In the subtraction element e403, the speed command value θ ^· _cmd and the motor speed θ ^· _m fed back from the motor are subtracted to calculate a speed deviation. Speed deviation is multiplied speed gain K _v by computing elements e404 is a unit are output to the summing element e406 as converted by the current command value I _ref by the calculating element E405.

The current command value I _ref is added with the current compensation value I _cmp by the addition element e 406 and is output to the calculation element e 407 and the disturbance observer 40. The current command value I _ref is multiplied by the torque constant Kt in the calculation element e407 and output to the subtraction element e408 as a torque command value. The torque command value is calculated as a deviation from the disturbance torque T _dism by a subtraction element e408, and is further calculated as a deviation from a shaft twist reaction force T _reac by a subtraction element e409. The torque deviation is divided by the motor inertia J _m by the calculation element E411, those that are further integrated the motor speed theta ^· _m. The motor speed θ ^· _m is output to the calculation element e412 and the disturbance observer 40, and is fed back to the subtraction element e403 in order to obtain a speed deviation.

Motor speed theta ^· _m is is integrated and converted into motor position theta _m by the calculation element E412. The motor position θ _m is an angular position of the output shaft of the motor, and is measured based on information of a rotary encoder as an angular position detector provided on the output shaft, for example. The motor position θ _m is output to the subtraction element e413 and is fed back to the subtraction element e401 to obtain a position deviation. The motor position θ _m is subtracted from the load position θ _l fed back from the load side by the subtraction element e 413, and the shaft twist amount θ _t is calculated. The load position θ _l is a movement position (angular position) of the load moved by driving the motor, and is measured by, for example, a linear encoder as a movement position detector provided on the load side.

Torsional amount theta _t is multiplied spring constant K _f is in operation element E414, is fed back to the subtraction element e409 is output to the subtraction element e415 as torsional reaction force T _REAC. The shaft torsional reaction force T _reac is subtracted from the load-side disturbance torque T _disl by the subtraction element e415 and output to the calculation element e416. The shaft torsional reaction force T _reac is divided by the load inertia J _l at the calculation element e 416 and integrated at the calculation element e 417 to calculate the load position θ _l . Load position theta _l is fed back to the subtraction element e413 to determine the torsional amount theta _t.

The disturbance observer 40 receives a current command value I _ref , a motor speed θ ^· _m , and a shaft twist amount θ _t . The current command value I _ref is multiplied by the nominal value K _{tn of the} torque constant at the calculation element e418 and output to the subtraction element e422. The motor speed θ ^· _m is differentiated by the calculation element e419, multiplied by the nominal value J _{mn of the} motor inertia, and output to the subtraction element e422. Also, the shaft twist amount θ _t is multiplied by the nominal value K _{fn of the} spring constant in the calculation element e421, and is output to the subtraction element e422 as the estimated value T ^ _reac of the shaft twist reaction force.

The subtracting element e422 subtracts the outputs of the computing elements e419 and e421 from the output of the computing element e418, and a low-frequency band component is extracted by the low-pass filter e423, and this is output as an estimated value T ^{^} _dism of the disturbance torque. . Estimate T ^{^} _dism of disturbance torque is divided by the nominal value K _tn torque constant by the computing element E424, it is converted to a compensation current value I _cmp for compensating for the disturbance. The compensation current value I _cmp is fed back to the addition element e406 and added to the command current value _Iref . Thus, in the disturbance observer 40 according to the present embodiment, the compensation current value I _cmp considering the shaft torsional reaction force T _reac is fed back to the command current value I _ref .

Therefore, in the fully closed control in which the load position θ _l is directly detected by a linear encoder or the like, the disturbance observer 40 can compensate for the vibration due to the shaft torsional reaction force T _reac to reduce the overshoot at the time of positioning. Yes. Further, even when the spring constant (rigidity) between the motor and the load is low, the observer gain G _dis can be increased. Furthermore, since the operation on the load side can be reflected in the control through the axial torsional reaction force _Treac , it is not affected by the fluctuation of the nominal value of the load inertia.

In the above-mentioned full-closed control, the motor position θ _m is directly detected by the angular position detector provided on the motor, the load position θ _l is directly detected by the moving position detector provided on the load side, and the shaft twist amount θ _t Seeking. However, in semi-closed control without a movement position detector such as a linear encoder, generally to positioning control based on the angular position detector provided on the output shaft of the motor, it is impossible to detect the load position theta _l directly.

Hereinafter, a control device for a positioning device to which a shaft torsion reaction force feedback type disturbance observer is applied in semi-closed control in which the load position cannot be directly detected will be described. In FIG. 11, the shaft torsion reaction force T _reac is _expressed by the equation (10) by multiplying the shaft torsion amount (θ _m −θ _l ), which is the difference between the motor position θ _m and the load position θ _l , by the spring constant K _f. . The transfer function from the torsional reaction force T _reac to the load position θ _l (from the subtraction element e315 to the calculation element e317) is expressed by the following equation (12).

になり、この式（１３）と式（１０）とによって、

を得ることができ、モータ位置θ_mから軸ねじれ反力Ｔ_reacを推定演算できることがわかる。セミクローズド制御系の軸ねじれ反力フィードバック型の外乱オブザーバは、モータ位置θ_mから推定される軸ねじれ反力Ｔ_reacの推定値Ｔ＾_reacがフィードバックされる構成となっている。 Here, considering that the disturbance torque T _disl on the load side includes the axial torsional reaction force T _reac , T _disl = 0 and the equation (12) is

From this equation (13) and equation (10),

It can be seen that the torsional reaction force T _reac can be estimated and calculated from the motor position θ _m . Disturbance observer of torsional reaction force feedback type semi-closed control system is configured to estimate T ^ _REAC of torsional reaction force T _REAC estimated from motor position theta _m is fed back.

FIG. 13 is a block diagram of a control device of a positioning device to which a shaft torsion reaction force feedback type disturbance observer is applied in semi-closed control. In FIG. 13, I _ref , I _cmp , K _p , K _v , K _t , K _tn , K _f , J _m , J _mn , J _l , s, T _dism , T ^{^} _dism , T _disl , T _reac , θ _cmd , θ _m , θ _l , θ _t , θ ^· _cmd , θ ^· _m , and g are the same as those in FIG. K _fn represents a nominal value of the spring constant. Further, in the block diagram of the semi-closed control shown in FIG. 13, the description of the same configuration as the block diagram of the fully closed control shown in FIG. 12 is omitted, and only the differences will be described.

The disturbance observer 50 receives a current command value I _ref , a motor speed θ ^· _m , and a motor position θ _m . The current command value I _ref is multiplied by the nominal value K _{tn of the} torque constant at the calculation element e518 and output to the subtraction element e522. The motor speed θ ^· _m is differentiated by the calculation element e519, multiplied by the nominal value J _{mn of the} motor inertia, and output to the subtraction element e522. Further, the motor position θ _m is converted into an estimated value T ^ _reac of the shaft torsional reaction force by the transfer function shown in Expression (4) in the calculation element e521, and output to the subtraction element e522.

The subtracting element e522 subtracts the outputs of the computing elements e519 and e521 from the output of the computing element e518, and a low-frequency band component is extracted by the low-pass filter e523, and this is output as an estimated value T ^{^} _dism of the disturbance torque. . Estimate T ^{^} _dism of disturbance torque is divided by the nominal value K _tn torque constant by the computing element E524, it is converted to a compensation current value I _cmp for compensating for the disturbance. The compensation current value I _cmp is fed back to the addition element e506 and added to the command current value I _ref . Thus, in the disturbance observer 50 according to the present embodiment, the compensation current value I _cmp considering the shaft torsion reaction force T _reac is fed back to the command current value I _ref .

Therefore, even in the semi-closed control in which the load position θ _l is not directly detected by a linear encoder or the like, it is possible to compensate for the vibration caused by the shaft torsion reaction force T _reac by the disturbance observer 50 and reduce the overshoot at the time of positioning. ing. Further, even when the spring constant (rigidity) between the motor and the load is low, the observer gain G _dis can be increased. Furthermore, since the operation on the load side can be reflected in the control through the axial torsional reaction force _Treac , it is not affected by the fluctuation of the nominal value of the load inertia.

  In full-closed control, positioning was settled by applying a shaft torsion reaction force feedback disturbance observer, and the simulation results shown in FIGS. 14 to 19 were obtained. Here, as shown in FIG. 14A to FIG. 19A, the parameters of the motor control system are set, and the simulation is performed for each of the case where the normal disturbance observer is used and the case where the shaft torsion reaction force feedback type disturbance observer is used. 14 to 19, the simulation result on the right side of the drawing is a partially enlarged view of the simulation result on the left side of the drawing.

14 to 16 show the simulation results of positioning settling when the spring constant _Kf is adjusted. Figure 14 is a simulation result in the case of setting the spring constant K _f 1250 (corresponding to the resonance frequency 236Hz). Figure 15 is a simulation result in the case of setting the spring constant K _f to 120 (corresponding to the resonance frequency 73 Hz). Figure 16 is a simulation result in the case of setting the spring constant K _f to 10 (corresponding to the resonance frequency 21 Hz).

As shown in FIGS. 14B and FIG. 14C, the higher the rigidity by setting the spring constant K _f in 1250, overshoot when using the disturbance observer of the torsional reaction force feedback type even when an ordinary disturbance observer There is no big difference in size and settling time. As the spring constant _Kf is set to be smaller from this state, the difference in overshoot size and settling time increases when the normal disturbance observer is used and when the shaft torsion reaction force feedback type disturbance observer is used. .

That is, as shown in FIGS. 15B and 15C, by setting the spring constant K _f to 120, as compared with the case of using a conventional disturbance observer, in the case of using the disturbance observer of the torsional reaction force feedback type The convergence of the vibration of the output waveform becomes faster. As further shown in FIGS. 16B and 16C, by setting the spring constant K _f to 10, the vibration of the output waveform does not converge in the case of using the conventional disturbance observer, a disturbance observer of the torsional reaction force feedback type When used, the vibration of the output waveform converges.

FIG. 17 is a simulation result of positioning settling when the observer gain is adjusted. In the case of using the conventional disturbance observer as shown in FIG. 17B, when the spring constant K _f is set to 10, it can not converge the vibration of the output waveform to be lowered the observer gain G _dis about 20 rad / s. On the other hand, as shown in FIG. 17C, when a shaft torsion reaction force feedback type disturbance observer is used, the observer gain G _dis can be increased to 600 rad / s even if the spring constant K _f is set to 10. . For this reason, when the shaft torsional reaction force feedback type disturbance observer is used, the settling time can be shortened as compared with the case of using a normal disturbance observer. Further, by increasing the observer gain G _dis , it is possible to improve the suppression effect and robustness against disturbances (friction, torque ripple, parameter fluctuation) other than the axial torsional reaction force.

18 and 19 show the simulation results of positioning settling when the nominal value of the load inertia fluctuates. FIG. 18 shows a simulation result when the nominal value J _ln of the load inertia is set to 1.5 times the load inertia J _l . FIG. 19 shows a simulation result when the nominal value J _ln of the load inertia is set to 0.5 times the load inertia J _l .

Since the normal disturbance observer is a model of one inertia system, the nominal value of the motor inertia is set to J _mn = J _m + J _l . On the other hand, the shaft torsion reaction force feedback type disturbance observer can reflect the operation on the load side in the control through the shaft torsion reaction force, so that J _mn = J _m is set. Therefore, while a normal disturbance observer is affected by fluctuations in the nominal value J _ln of the load inertia, a shaft torsion reaction force feedback type disturbance observer is not affected by fluctuations in the nominal value J _ln of the load inertia.

As shown in FIGS. 18B and 19B, when a normal disturbance observer is used, if the nominal value J _ln of the load inertia is changed from 1.5 times to 0.5 times the load inertia J _l , the output waveform Changes significantly. On the other hand, as shown in FIGS. 18C and 19C, in the case of using a shaft torsion reaction force feedback type disturbance observer, the nominal value J _ln of the load inertia is 1.5 to 0.5 times the load inertia J _l. The output waveform does not change even if it is varied.

The nominal value J _mn of the motor inertia can be obtained from a motor specification or the like. However, the nominal value J _ln of the load inertia is often obtained from calculation and is often different from the actual value. Therefore, highly accurate positioning control can be performed without being affected by the nominal value J _ln of the load inertia. In the present embodiment, a simulation for positioning and setting in full-closed control is performed, but it is assumed that the same tendency can be obtained in semi-closed control.

As described above, according to the control device of the positioning device according to the present embodiment, since the disturbance due to the shaft torsional reaction force T _reac is estimated and fed back to the command value for the motor, the rigidity between the motor and the load Even if is small, the vibration can be converged early. In addition, since overshoot during positioning is suppressed, the settling time can be shortened and positioning can be performed at high speed and with high accuracy. In addition, since the operation on the load side can be reflected in the control through the axial torsional reaction force, it is not affected by fluctuations in the nominal value of the load inertia. Further, compared with state feedback control, H∞ control, and resonance ratio control, the calculation amount is small and the amount of calculation is small, and an expensive CPU or the like is not necessary. Furthermore, since overshoot is suppressed without increasing the rigidity of the apparatus, it is possible to reduce the weight of the entire apparatus and realize energy saving and cost reduction.

  The present invention is not limited to the second embodiment, and can be implemented with various modifications. In the above-described embodiment, the size, shape, and the like illustrated in the accompanying drawings are not limited to this, and can be appropriately changed within a range in which the effects of the present invention are exhibited. In addition, various modifications can be made without departing from the scope of the object of the present invention.

  For example, in the block diagram of the control device according to the second embodiment described above, P control is used for speed calculation, but PI control, PD control, and PID control are used according to speed deviation and speed response. May be. Further, the control device for the positioning device according to the present embodiment can be applied not only to a rotary motor but also to a linear motor.

  In the second embodiment, the angular position detector is a rotary encoder. However, any configuration may be used as long as the angular position of the output shaft can be detected. The angular position detector need not be fixed to the output shaft, and may be configured to detect the angular position via a belt or a gear. Further, although the moving position detector is a linear encoder, any structure may be used as long as the moving position (angular position) of the load can be detected. The movement position detector may be composed of, for example, a rotary encoder or an acceleration sensor.

  In the second embodiment described above, the continuous transfer function as shown in Expression (4) has been described as an example. However, the present invention is not limited to this configuration. In the case of digital control, a discrete transfer function may be used.

In the second embodiment described above, the rigidity between the motor and the load changes according to the moving position of the load. Therefore, the disturbance observer 40 and 50 of the torsional feedback type may be configured to correcting the spring constant K _f of the movement positions of the load. With this configuration, it is possible to perform higher vibration suppression control while maintaining higher robustness. Further, it may be configured that the disturbance observer 40, 50 of the shaft torsional Feedback is correcting the spring constant K _f over time.

  The positioning device may be controlled by combining the first and second embodiments described above. That is, the axial torsional reaction force feedback type disturbance observer may be designed based on a disturbance model including a friction parameter as a part of the disturbance.

  As described above, the present invention can suppress overshoot and can perform high-speed and high-accuracy positioning, and is used particularly in an electronic component mounting apparatus such as a chip mounter. It is useful for a control device of a positioning device.

1 Mounting device 2 Base 4 Mounting head (load)
5 Mounting head moving mechanism (positioning device)
7 Substrate transport section (positioning device)
11 X-axis table (load)
20, 40, 50 Disturbance observer

Claims (9)

A control device for a positioning device that positions a load at a predetermined position by a motor,
A control system for controlling the motor based on a current command value for the motor;
A disturbance observer that estimates disturbance from the current command value and motor speed for the motor and feeds back to the current command value for the motor,
The disturbance observer is a control device for a positioning apparatus, wherein a friction parameter included in a part of the disturbance is a function in which friction decreases with an increase in motor speed, and the disturbance observer is based on a disturbance model including the function. 2. The positioning according to claim 1, wherein the disturbance observer is based on a disturbance model including a function of Formula (1) as the friction parameter, where B is a friction coefficient, u is a motor speed, and a is a constant. Control device for the device.

2. The controller for a positioning apparatus according to claim 1, wherein the disturbance observer is based on a disturbance model including a function of Expression (2) as the friction parameter, where B is a friction coefficient and u is a motor speed. .

2. The positioning according to claim 1, wherein the disturbance observer is based on a disturbance model including a function of Equation (3) as the friction parameter, where B is a friction coefficient, u is a motor speed, and a is a constant. Control device for the device.

  5. The control device for a positioning device according to claim 2, wherein the disturbance observer minimizes the friction coefficient when the motor is started. 6.   6. The control device for a positioning apparatus according to claim 2, wherein the disturbance observer minimizes the friction coefficient when driven at a motor speed equal to or lower than a predetermined speed.   5. The disturbance observer according to claim 1, wherein the disturbance observer has a plurality of friction coefficients with respect to the function as the friction parameter, and dynamically switches the friction coefficient according to a positioning position. A control device for the positioning device according to claim 1. 5. The control device for a positioning apparatus according to claim 2 , wherein the disturbance observer dynamically adjusts a friction coefficient so that a position deviation amount is a predetermined value or less.   An electronic component mounting apparatus comprising the positioning device control device according to claim 1.

Images