Parameter adjustment device and method taking into account characteristic variations due to load device position

Abstract

To adjust a controller which can achieve stable high-speed and highly precise control without impairing control performance, without causing allophone or vibrations when a characteristic variation occurs due to the position of a load device.SOLUTION: A parameter adjustment system includes an actual machine frequency characteristic measurement unit that calculates a frequency characteristic of an actual machine on the basis of a motor input signal and an output of a detector, and a parameter adjustment unit that obtains a parameter of a controller, which is equal to or smaller than a specific reference, and applies a difference between an open loop characteristic, which is calculated from a result of the calculation by the actual machine frequency characteristic measurement unit and a frequency characteristic of the controller, and an ideal open loop characteristic to the controller.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a control device for a machine tool or the like, and in particular to an improvement in a parameter adjustment device and adjustment method. [Background technology]

[0002] In industrial servo systems, including machine tools, there is a growing demand for automating the adjustment process in order to achieve high-speed, high-precision positioning operations and reduce the labor required for adjusting control parameters.

[0003] In particular, the adjustment of a feedback controller, which is widely used in industry and combines a gain stabilization filter for broadband and a vibration suppression filter for canceling out resonance characteristics, affects not only the processing quality but also the stability of the control system, so it is desirable to perform the adjustment based on frequency characteristics, which can explicitly consider the resonance characteristics of the actual machine and the stability of the control system.

[0004] The frequency characteristics of the actual machine vary between machines due to machine variations such as assembly errors and variations in component dimensions.

[0005] FIG. 3 is a diagram showing an outline of a machine tool as an example equipped with a feed axis mechanism driven by a motor via a ball screw or the like.

[0006] A motor 33, a ball screw 34, a guide 35, and a spindle head 36 are arranged on the front of a structure 32 that is erected on a base 31 fixed to the floor. Torque generated by the motor 33 is converted into a linear force by the ball screw 34, which controls the movement of the spindle head 36, which is movably supported by the guide 35, in the left-right direction in the figure. A similar mechanism controls the movement of the spindle head 36 in the up-down direction in the figure, thereby controlling the movement of the tool attached to the tip. Furthermore, the table 37 arranged on the base 31 is controlled to move in a direction penetrating the page, thereby controlling the movement of a workpiece 38 attached to its upper surface. Furthermore, the workpiece 38 is machined by rotating the tool attached to the spindle head 36.

[0007] In addition, the torque generated by the motor 33 is controlled by a control device 310 incorporating a feedback controller so that the position detection value detected by a position detector (rotation angle detector) 39a attached to the motor 33, or a position detector (load position detector) 39b attached to the structure 32 so as to enable direct detection of the position of the spindle head 36, or both, matches the position command signal generated within the control device 310.

[0008] The actual frequency characteristics used to adjust the feedback controller can be calculated, for example, by performing frequency analysis on the torque command value generated to drive the motor 33 and the detected values ​​of the position detector (rotation angle detector) 39a and the position detector (load position detector) 39b, and then calculating the difference between the frequency spectra of both.

[0009] FIG. 4 is a diagram showing an example of how the actual frequency characteristics vary depending on the position of a load device driven by a motor, such as the spindle head 36 or table 37 in FIG.

[0010] As an example, the dashed line shows the actual frequency characteristics measured at a position (point a) when the spindle head 36 in Figure 3 is moved toward the motor 33, and the solid line shows the actual frequency characteristics measured at a position (point b) when the spindle head 36 is sufficiently far away from the motor 33. Depending on the position of the load device such as the table or spindle head, fluctuations in the motor shaft equivalent inertia of the ball screw 34 in particular will cause fluctuations in the actual frequency characteristics even within the same machine.

[0011] If the control parameters are adjusted based on the actual frequency characteristics measured with the load device at point a, when the load device is operated near point b, the gain stabilization filter and vibration suppression filter may not work effectively due to fluctuations in characteristics, which may result in the generation of abnormal noise and vibration and a decrease in control performance. Therefore, to enable stable, high-speed, and high-precision positioning operation, adjustments that take characteristic fluctuations into consideration are required.

[0012] Patent Document 1 discloses a technology in which the actual frequency characteristics of a machine are measured using a sine wave sweep method, and reference frequency characteristics are obtained by successively changing the parameter set of the physical model so that the frequency characteristics calculated using the physical model match the actual frequency characteristics. When changes occur to the machine's workpiece, jigs, or machine configuration, the actual machine's frequency characteristics are measured again to determine the difference from the previously determined reference frequency characteristics, and the gain of the control system is automatically adjusted to achieve the same control performance as before the change. This makes it possible to solve the problem of deterioration in control performance due to characteristic variations that occur between machines. [Prior art documents] [Patent documents]

[0013] [Patent Document 1] Japanese Patent Application Publication No. 2018-128734 Summary of the Invention [Problem to be solved by the invention]

[0014] The technology of Patent Document 1 requires a physical model to be designed in advance to obtain reference frequency characteristics. Therefore, if the mass or inertial system of each machine type differs, a physical model tailored to that type is required, which takes time to implement. Furthermore, if a significant amount of mechanical resonance characteristics that affect control performance are included, the inertia of the physical model must be increased, which may be affected by limitations in the computing power of the DSP (digital signal processor). Furthermore, while it is possible to address characteristic variations between machines, characteristic variations that occur within the same machine are difficult to address because only one physical model can be specified for each machine. [Means for solving the problem]

[0015] The parameter adjustment device of the present invention is a servo control device that drives a load device using a servo motor, and includes: a command generation unit that generates a command value for the servo motor; a detector that detects a state quantity of the servo motor or the load device being driven; a controller that controls a motor input signal so that the command value and the output of the detector coincide; an actual machine frequency characteristic measurement unit that calculates the frequency characteristics of the actual machine from the motor input signal and the output of the detector; and a difference between an open loop characteristic calculated from the calculation result of the actual machine frequency characteristic measurement unit and the frequency characteristic of the controller and an ideal open loop characteristic, By solving the optimization problem using sequential quadratic programming as the objective function, and a parameter adjustment unit that solves and applies the parameters of the controller. [Effects of the Invention]

[0016] According to the control device of the present invention, when characteristic fluctuations occur due to the position of the load device, it is possible to adjust the controller to enable stable, high-speed, and high-precision control without generating abnormal noise or vibration and without impairing control performance. [Brief explanation of the drawings]

[0017] [Figure 1] FIG. 2 is a block diagram showing the configuration of a servo control unit of the control device of the present invention. [Figure 2] 3 is a flowchart showing the operation of the control device according to the present invention. [Figure 3] FIG. 1 is a diagram showing an example of a general machine tool. [Figure 4] FIG. 10 is a diagram showing an example in which the actual frequency characteristics vary depending on the position of the load device. DETAILED DESCRIPTION OF THE INVENTION

[0018] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0019] FIG. 1 is a block diagram showing the configuration of a servo control unit of a control device according to the present invention.

[0020] A command generation unit 17 generates a movement command for the load device 16 according to a program provided by an operator or the like. The controller 13 controls the torque generated by the motor 15 so that the generated command matches the value detected by the detector 14. The actual machine frequency characteristic measurement unit 11 receives the outputs of the controller 13 and the detector 14 as inputs, calculates the frequency characteristics of the frequency characteristic measurement target unit 18, and outputs the calculated frequency characteristics to the parameter adjustment unit 12. The parameter adjustment unit 12 finds optimal parameters for the controller 13 from the frequency characteristics calculated by the actual machine frequency characteristic measurement unit 11, and updates the controller 13. The controller 13, the actual machine frequency characteristic measurement unit 11, and the parameter adjustment unit 12 are physically configured as a computer having a processor and memory. The detector 14 detects the state quantity of the motor 15 or the load device 16, and is, for example, a position sensor or a vibration sensor.

[0021] FIG. 2 is a flowchart showing the operation of the control device of the present invention, where S1 in FIG. 2 shows the operation of the actual frequency characteristic measurement unit 11 in FIG. 1, and S2 in FIG. 2 shows the operation of the parameter adjustment unit 12 in FIG. 1.

[0022] First, in steps S11 and S12, the actual frequency characteristics of the frequency characteristic measurement target 18 in FIG. 1 are measured while changing the position of the load device. Here, n represents the number of measurement points and is a value determined in advance so as to maximize the range of characteristic fluctuation. For example, if the load device is the table 37 shown in FIG. 3, the difference in motor-shaft-equivalent inertia of the ball screw 34 is greatest when the load device is near the plus end and the minus end of its movable range. Furthermore, because the elastic deformation of the ball screw 34 is greatest when the load device is near the center of the movable range, it is preferable to measure the frequency characteristics at three points: near the plus end, the minus end, and the center of the movable range. After processing in step S11, in step S12, the number of measurements m is incremented from 0. If the number of measurements m is equal to or greater than the number of measurement points n, the process proceeds to step S21.

[0023] In step S21, the structure of the speed controller C(s) is determined based on the actual frequency characteristics obtained in steps S11 and S12. In the present invention, compensator elements included in the structure of the speed controller C(s) include, for example, a PID compensator, a PI compensator, a PD compensator, a P compensator, a phase lead / lag compensator, a notch filter, an all-pass filter, a low-pass filter, and a high-pass filter. In this embodiment, a PI compensator and an X-stage notch filter C k The following formula (1) is used as an example to describe the configuration in which (s) are combined. k (s) can be expressed as the following equation (2).

[0024] The number of stages X of the notch filter may be determined in advance taking into consideration the calculation capacity of the DSP and the number of resonance characteristics at which the notch filter effectively functions for widening the bandwidth, and may also be appropriately changed in step S21 to match the resonance characteristics of the actual frequency characteristics.

[0025]

number

[0026] At this time, K p , K. i are the parameters of the PI compensator, and α k , ζ k , ω k is a parameter of the notch filter. Also, s means the Laplace operator, which is equivalent to the product of the imaginary unit j and the angular frequency ω. Therefore, speed The controller C(s) can be expressed by a complex number as in the following equation (3).

[0027]

number

[0028] In step S22, the actual frequency characteristics measured in steps S11 and S12 are compared with the frequency characteristics designed in step S21. speed From the controller C(s), open-loop characteristics L1(jω),L2(jω),…,L nDesign (jω).

[0029] The actual frequency characteristics measured in steps S11 and S12 are expressed as P1(jω), P2(jω), ..., P n (jω), the open loop characteristics for each actual frequency characteristic are L1(jω), L2(jω), ..., L n (jω) can be expressed as in the following equation (4).

[0030]

number

[0031] As in the above equation (4), by using a complex number expression using jω instead of a transfer function expression using the Laplace operator s, it is possible to obtain the open-loop characteristics without modeling the actual frequency characteristics.

[0032] Next, in step S23, an optimization problem is calculated to find the parameters of the speed controller C(s) based on the open-loop characteristics of the above equation (4). The optimization algorithm used here is preferably one that can impose constraints, since constraints using gain margins and phase margins are required to operate the controller without generating abnormal noise or vibration. Furthermore, if the speed controller C(s) is expressed in a linear form with respect to the parameters to be solved, it can be formulated using linear matrix inequalities (LMIs). In this embodiment, since the open-loop characteristics of the above equation (4) are expressed in a nonlinear form, the optimization problem is formulated using sequential quadratic programming (SQP). When the optimization problem is formulated using SQP, the following equation (5) is obtained.

[0033]

number

[0034] where L d(s) is the ideal open loop characteristic, and can be arbitrarily determined depending on the performance that the designer desires for the control system. If the objective is simply to widen the bandwidth, it is sufficient to use the following formula (6) as an example. In this case, α and β are arbitrary coefficients. Also, while there are as many open loop characteristics using the actual frequency characteristics as there are measurement points n, there are only n open loop characteristics using the ideal open loop characteristic L d (s) can be one type.

[0035]

number

[0036] In the above formula (5), ω p is the angular frequency at which the phase characteristic is -180°, and the gain margin GM is the angular frequency at which ω p This shows the gain characteristics of the open loop when ω g is the angular frequency at which the gain characteristic becomes 0 dB, and the phase margin PM is the angular frequency ω g Furthermore, T is the number of data points of the frequency characteristics used to formulate the optimization problem, and can be set to any value within the number of data points measured in step S11.

[0037] The gain margin GM, phase margin PM, and ideal open-loop characteristics L d (s) must be determined in advance when formulating the above equation (5).

[0038] In the above equation (5), the ideal open-loop characteristic L_d(s) and the open-loop characteristic L1(jω), L2(jω), ..., L n The decision variable for the optimization problem that makes the difference between (jω) less than the specified standard is K p , K. i , α k , ζ k , ω k (k=1, 2, ..., X), and if the optimal solution for these 2 + 3X decision variables is obtained, the optimal open-loop characteristics and controller parameters can be obtained.

[0039] In step S24, the speed controller parameters are updated to the parameters solved in step S23, so that the load device 16 can be controlled with an optimal controller.

[0040] As a result, a controller capable of wideband operation is designed while satisfying the gain and phase margins for all frequency characteristics when the load device is located at any point, enabling stable, high-speed, and high-precision control that does not generate abnormal noise or vibration, even when the actual frequency characteristics fluctuate depending on the position of the load device.At the same time, the data of the actual frequency characteristics can be directly formulated into the optimization problem, eliminating the effort required for modeling and enabling the design of a controller that takes all resonance characteristics into account.

[0041] In this embodiment, a machine tool having an axis that drives a table from a motor via a ball screw has been described as an example, but the present invention can also be applied to an axis in which the load device is a spindle head, or to machine tools driven by linear motors, and can also be applied to industrial machines other than machine tools, etc. Also, although the above embodiment has been described using a speed control system as an example, the configuration of the speed control system is not important, and the present invention can also be applied to a position control system or a cascade control system that combines position and speed control systems. [Explanation of symbols]

[0042] 11 Actual machine frequency characteristic measurement unit, 12 Parameter adjustment unit, 13 Controller, 14 Detector, 15, 33 Motor, 16 Load device, 17 Command generation unit, 18 Frequency characteristic measurement target unit, 31 Base, 32 Structure, 34 Ball screw, 35 Guide, 36 Spindle head, 37 Table, 38 Workpiece, 39a Position detector (rotation angle detector), 39b Position detector (load position detector), 310 Control device.

Claims

1. In a servo control device that drives a load device using a servo motor, a command generating unit that generates a command value for the servo motor; a detector for detecting a state quantity of the servo motor or the load device to be driven; a controller that controls a motor input signal so that the command value and the output of the detector coincide with each other; an actual machine frequency characteristic measuring unit that calculates the frequency characteristics of the actual machine from the motor input signal and the output of the detector; a parameter adjustment unit that solves an optimization problem by sequential quadratic programming using a difference between an open-loop characteristic calculated from the calculation result of the actual device frequency characteristic measurement unit and the frequency characteristic of the controller and an ideal open-loop characteristic as an objective function, and applies the solution to parameters of the controller; Equipped with the actual device frequency characteristic measurement unit, when calculating the frequency characteristics of the actual device, arbitrarily changes the position of the load device and calculates the frequency characteristics of the actual device for each position; the parameter adjustment unit determines a plurality of open-loop characteristics corresponding to the frequency characteristics of a plurality of actual devices, and sets a total value of differences between each of the plurality of open-loop characteristics and the ideal open-loop characteristic as the objective function. A parameter adjustment device comprising:

2. 2. The parameter adjustment device according to claim 1, the parameter adjustment unit uses the frequency characteristics of the plurality of actual devices calculated by the actual device frequency characteristic measurement unit as measured values ​​without modeling, formulates open-loop characteristics combining the controllers as an optimization problem, and solves the optimization problem to obtain parameters to be applied to the controllers.

3. A parameter adjustment method for a servo control device that drives a load device using a servo motor, the method including: a command generation unit that generates a command value for the servo motor; a detector that detects a state quantity of the movable servo motor or the load device; and a controller that controls a motor input signal so that the command value and an output of the detector coincide with each other, an actual machine frequency characteristic measuring step of calculating frequency characteristics of the actual machine from the motor input signal and the output of the detector; a parameter adjustment step of solving an optimization problem by sequential quadratic programming using a difference between an open-loop characteristic calculated from the calculation result of the actual device frequency characteristic measurement step and the frequency characteristic of the controller and an ideal open-loop characteristic as an objective function, thereby obtaining and applying parameters of the controller; Equipped with When calculating the frequency characteristics of the actual device, the position of the load device is arbitrarily changed and the frequency characteristics of the actual device for each position are calculated; In the parameter adjustment step, a plurality of open-loop characteristics corresponding to the frequency characteristics of a plurality of actual devices are obtained, and the total value of the differences between each of the plurality of open-loop characteristics and the ideal open-loop characteristic is set as the objective function. A parameter adjustment method comprising:

4. A servo control device that drives a load device using a servo motor, a command generating unit that generates a command value for the servo motor; a detector for detecting a state quantity of the servo motor or the load device to be driven; a controller that controls a motor input signal so that the command value and the output of the detector coincide with each other; an actual machine frequency characteristic measuring unit that calculates the frequency characteristics of the actual machine from the motor input signal and the output of the detector; a parameter adjustment unit that solves an optimization problem by sequential quadratic programming using a difference between an open-loop characteristic calculated from the calculation result of the actual device frequency characteristic measurement unit and the frequency characteristic of the controller and an ideal open-loop characteristic as an objective function, and applies the solution to parameters of the controller; Equipped with the actual device frequency characteristic measurement unit, when calculating the frequency characteristics of the actual device, arbitrarily changes the position of the load device and calculates the frequency characteristics of the actual device for each position; the parameters of the controller are a plurality of variables included in an equation representing the controller; the ideal open-loop characteristics are expressed by one type of equation that has predetermined coefficients and a Laplace operator as elements and that is determined based on the performance of the controller that is determined at the time of design; the parameter adjustment unit applies the controller to a plurality of actual machine frequencies corresponding to a plurality of positions, respectively, to calculate a plurality of open-loop characteristics; A combination of the plurality of variables is obtained such that the sum of the differences between each of the plurality of open-loop characteristics and the ideal open-loop characteristic is equal to or less than a predetermined reference value; applying the determined variables to the controller as the controller parameters; A parameter adjustment device comprising:

Images