JP5404507B2 - Correction parameter adjustment device - Google Patents

Description

  The present invention relates to a correction parameter adjusting device for adjusting a correction parameter for improving the movement accuracy of a machine in a numerically controlled machine tool or a robot.

  In numerically controlled machine tools and robots, a motor is driven to control the position of the machine as faithfully as possible to the command position. In that case, an error occurs between the command position and the actual machine position due to the influence of the vibration of the machine or the frictional force at the time of reversal of the movement direction, and for example, the processed surface is damaged. When such a problem occurs, the movement locus of the machine is measured to find the cause of the problem, and various correction parameters of the controller that controls the motor are adjusted. For this purpose, a plurality of methods are known for measuring and displaying the machine motion trajectory.

  In the method disclosed in Patent Document 1, two high-precision steel balls are connected via a displacement meter, and a motion (arc) is performed to keep the relative distance between the two balls constant. It reads the displacement when it is applied. This method is called a ball bar method and is widely used as a measurement method for adjusting correction parameters.

  When adjusting the correction parameters using the ballbar method, measure the motion trajectory of the machine, infer the error factor from the result, change the correction parameter, measure the motion trajectory of the machine again, and change the correction parameter. Check the effect. If satisfactory results cannot be obtained with a single change, the above procedure is generally repeated.

  Patent Document 2 describes a method and system for measuring a mechanical error using the ball bar method and proposing a change of a correction parameter of a control device in order to reduce a burden on an operator who adjusts a correction parameter. ing. The point of this technology is to automatically generate an NC program so that the test is properly performed based on the input test conditions, instruct the operator to set up the measuring instrument, and to execute the test. Analyze and suggest changes to calibration parameters and display expected effects.

JP-A 61-209857 JP-T-2004-525467

  However, in the method described in Patent Document 1, the adjustment operator estimates the error factor from the measurement result of the machine motion trajectory and adjusts the correction parameter. In order to be able to infer the error factor from the measurement result, it takes a lot of experience and expertise, so there are limited personnel who can adjust the correction parameter, and it takes a lot of time to adjust the correction parameter There are many.

  In the method and system described in Patent Document 2, the setup of the measuring instrument is instructed, so that even an operator who does not have expertise can adjust the correction parameter. However, since the ball bar method is used for measuring the mechanical motion trajectory, there is a problem that only an arc trajectory having a predetermined radius can be measured, and one or more workers are always required during the adjustment work.

  In common with the above method, in order to measure a movement locus by attaching a displacement meter to a machine, for example, when a jig or tool is attached to a machine tool, it is necessary to remove them for measurement. For this reason, for example, even if there is a defect in the machining result after trial cutting, it is necessary to remove the tool or workpiece in order to adjust the correction parameter. There was a problem that it was difficult to return an object to its original state.

  In addition, there is a problem that the measurement range is limited to a relatively narrow range in common with the above method. That is, if the machine moves accidentally beyond the measuring range, the measuring instrument will be damaged. In addition, when trying to measure the motion trajectory of a machine with a large stroke, it was necessary to remove the measuring device once, shift the place, and install it again. In addition, there are machines that cannot be installed due to structural and dimensional problems even if it is desired to install a measuring instrument. In such a case, there has been a problem that correction parameters cannot be adjusted.

  In addition, in common with the above method, when adjusting the correction parameter using only the measurement result of the machine motion trajectory, the measurement result includes, for example, a trajectory error due to a motor motion error and a trajectory error due to a machine error. Even if this is done, the two errors cannot be separated, making adjustment work more difficult.

  The present invention has been made in view of the above, and can measure the movement trajectory of a machine in actual use without removing a jig or a tool, and can appropriately set correction parameters by separating error factors. An object of the present invention is to obtain a corrective correction parameter adjusting device.

  In order to solve the above-described problems and achieve the object, the present invention feeds back the detected position detected from the motor driving the movable shaft, and drives the motor so that the detected position follows the command position. An accelerometer for measuring the acceleration of the machine in the device for controlling the position of the machine, a machine motion analysis unit for analyzing the machine motion from the acceleration and the detected position, and improving the motion accuracy of the machine from the analysis result of the machine motion And a correction parameter calculation unit for determining a correction parameter for this purpose.

  According to the present invention, it is possible to measure the movement trajectory of a machine in an actual use state without removing a jig or a tool, and it is possible to separate error factors and appropriately set correction parameters.

FIG. 1 is an external perspective view showing a schematic configuration of a numerically controlled machine tool to which a correction parameter adjusting device according to Embodiment 1 of the present invention is applied. FIG. 2 is a diagram schematically showing a cross-sectional view of the Y-axis drive mechanism of the numerically controlled machine tool shown in FIG. 1 when viewed from the side. FIG. 3 is a block diagram illustrating a schematic configuration of a correction parameter adjusting device and the like included in the numerically controlled machine tool. FIG. 4 is a flowchart illustrating an outline of processing for adjusting the backlash correction parameter in the mechanical motion analysis unit and the correction parameter calculation unit. FIG. 5 is a diagram showing an outline of processing in the mechanical motion analysis unit in step S3. FIG. 6 is a diagram illustrating an example of calculation results of the detection position, the machine position, and the machine error when the reciprocating motion is performed. FIG. 7 is a diagram showing a mechanical error waveform of 0.5 mm before and after reversing the motion direction. FIG. 8A is a diagram illustrating an arc trajectory of an actual machine before automatically adjusting the backlash correction parameter. FIG. 8-2 is a diagram illustrating an arc trajectory of the actual machine after automatically adjusting the backlash correction parameter. FIG. 9 is a flowchart showing an outline of processing in the mechanical motion analysis unit and the correction parameter calculation unit in the first modification of the first embodiment, and shows an outline of processing for adjusting the pitch error correction parameter. is there. FIG. 10A is a diagram for explaining a mechanism of occurrence of a trajectory error due to elastic deformation. FIG. 10-2 is a diagram illustrating the amplitude of the driven body when a sine wave reciprocating motion is commanded. FIG. 11 is a flowchart for explaining the procedure for calculating the natural angular frequency in the second modification of the first embodiment. FIG. 12 is a flowchart for explaining the procedure for calculating the natural angular frequency of the third modification of the first embodiment using Equation (8). FIG. 13 is a diagram for explaining an overview of the processing in step S9. FIG. 14A is a diagram illustrating a detection trajectory and a mechanical trajectory before adjusting an elastic deformation amount correction parameter. FIG. 14B is a diagram of the detection locus and the mechanical locus after adjusting the elastic deformation amount correction parameter. FIG. 15 is a flowchart showing an outline of the adjustment process of the drive unit correction parameter in the fourth modification of the first embodiment. FIG. 16A is a diagram illustrating a detection locus and a machine locus before adjusting the drive unit correction parameter. FIG. 16B is a diagram illustrating a detection locus and a machine locus after adjusting the drive unit correction parameter.

  Embodiments of a correction parameter adjusting apparatus according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

Embodiment 1 FIG.
FIG. 1 is an external perspective view showing a schematic configuration of a numerically controlled machine tool to which a correction parameter adjusting device according to Embodiment 1 of the present invention is applied. The numerically controlled machine tool 50 has a plurality of movable shafts whose movement is guided along the X-axis, Y-axis, and Z-axis directions, and each movable shaft is driven by a drive mechanism having a motor 1 and a feed screw 2. The The rotation angle of the motor is detected by the rotation angle detector 3 and fed back to the control device (for example, the motor drive unit 12). As a method for driving each movable shaft, a linear motor may be used instead of the motor 1 and the feed screw 2, or a linear scale may be used instead of the rotation angle detector 3.

  In the numerically controlled machine tool 50, the work table 4 is driven by the Y-axis drive mechanism, and the column 5 is driven by the X-axis drive mechanism. The spindle head 7 is driven via the ram 6 by the Z-axis drive mechanism attached to the column 5. As a result of these drivings, a three-dimensional shape is created between the tool attached to the tip of the spindle head 7 and the workpiece placed on the work table 4.

  FIG. 2 is a diagram schematically showing a cross-sectional view of the Y-axis drive mechanism of the numerically controlled machine tool 50 shown in FIG. 1 when viewed from the side. Although only the Y-axis drive mechanism is shown here, the X-axis and Z-axis drive mechanisms have the same configuration. The rotational movement of the motor 1 is transmitted to the feed screw 2 through the coupling 8 and converted into straight movement through the nut 9. The rotation axis of the feed screw 2 is constrained by the support bearing 10 and enables the work table 4 to move straight. A command position is output from a controller (for example, the command generation unit 11), and the command position is transmitted to the motor drive unit 12. In the motor drive unit 12, an error between the detected position obtained by converting the motor rotation angle detected by the rotation angle detector 3 into a position and the transmitted command position is as small as possible, that is, the detected position is commanded. The motor 1 is driven so as to follow the position.

  In addition to the rotation angle of the motor 1, a linear scale or a laser displacement meter may be added to detect the position of the work table 4 (machine position) and feed back the detected position to the motor drive unit 12. A linear motor may be used instead of the motor 1 and the feed screw 2. As for the X-axis and Z-axis, similarly to the Y-axis, the motor 1 is driven so that the detected position follows the command position, and the position of the tool attached to the tip of the spindle head 7 (machine position) is Be controlled.

  In the numerically controlled machine tool 50, the relative displacement between the tool attached to the tip of the spindle head 7 and the work table 4 is important. The relative displacement is measured in advance, and the command generator 11 or the motor drive unit 12 In general, an error existing in the numerically controlled machine tool 50 is corrected. First of all, there are static errors such as perpendicularity between movable shafts and lead screw pitch errors, which are measured and adjusted at the stage of assembling the machine and rarely change during normal use. .

  On the other hand, it is also known that dynamic errors such as elastic deformation and backlash mainly occurring in the coupling 8, feed screw 2, and support bearing 10 portions, errors due to posture changes of the columns 5 and rams 6, and errors due to frictional forces occur. ing. The characteristics of these dynamic errors greatly change depending on the use state of the numerically controlled machine tool 50, the load mass on the work table 4, the secular change and wear of the machine, and the like. Therefore, it is desirable to be able to suppress the dynamic error by measuring the machine motion trajectory periodically or continuously while using the numerically controlled machine tool 50 and adjusting various correction parameters of the controller.

  The correction parameters for improving the motion accuracy of the numerically controlled machine tool 50 correspond to a parameter group constituting a static or dynamic error model of the numerically controlled machine tool 50. If the error of the numerically controlled machine tool 50 including the control device can be completely modeled, the motion error will approach zero as much as possible by the correction. In other words, the correction function for improving the motion accuracy of the machine and its correction parameter are in a relationship with the mathematical model of the machine, and the correction parameter also changes with the progress of the correction function.

  In order to improve the motion accuracy of the numerically controlled machine tool 50, the correction parameters of the current general controller are roughly classified into a command generation unit correction parameter and a drive unit correction parameter. The command generation unit correction parameter has an effect of correcting the command position in the command generation unit 11, and the drive unit correction parameter has an effect of correcting the command speed or the command torque in the motor drive unit 12.

  Examples of the command generation unit correction parameter include a backlash correction parameter, a pitch error correction parameter, an elastic deformation correction parameter, a prefilter parameter, and the like. Examples of the drive unit correction parameter include a friction compensation parameter, a backlash acceleration parameter, and a vibration suppression control parameter. In order to improve the mechanical motion accuracy, it is necessary to appropriately set the command generation unit correction parameter and the drive unit correction parameter after separating the error factors.

  FIG. 3 is a block diagram showing a schematic configuration of the correction parameter adjusting device 40 and the like included in the numerically controlled machine tool 50. As shown in FIG. 3, the correction parameter adjusting device 40 includes a tool attached to the tip of the spindle head 7, an accelerometer 13 for measuring the acceleration of the work table 4, and the acceleration and detection position measured in synchronization. And a correction parameter calculation unit 18 for determining a correction parameter for improving the movement accuracy of the machine from the analysis result of the machine movement. The numerically controlled machine tool 50 is also provided with a motion trajectory display unit 15 that displays a mechanical motion trajectory and the like.

  First, as an example of command generation unit correction parameter adjustment, adjustment of backlash correction parameters will be described as an example. FIG. 4 is a flowchart showing an outline of processing for adjusting the backlash correction parameter in the mechanical motion analysis unit 14 and the correction parameter calculation unit 18. First, a test motion is executed on the numerically controlled machine tool 50 (step S1). Next, the detected position and acceleration during the test motion are acquired by the mechanical motion analysis unit 14 (step S2). Here, as the test motion for adjusting the backlash correction parameter, a circular motion by simultaneous two-axis control is suitable, and the detected position and acceleration at that time are simultaneously measured as time-synchronized data. In addition, execution of the test exercise | movement in step S1 may be performed by an operator operating a control apparatus, and may be performed by the instruction | indication from a higher-order apparatus.

  Next, the machine position is calculated by the machine motion analysis unit 14 from the detected position and acceleration acquired in step S2 (step S3). Here, the process in the mechanical motion analysis part 14 in step S3 is demonstrated using drawing. FIG. 5 is a diagram showing an outline of processing in the mechanical motion analysis unit 14 in step S3. In step S3, an error waveform between the position obtained by integrating the acceleration twice and the detected position is approximated with a predetermined approximate expression. Then, the machine position is calculated by subtracting the solution based on the approximate expression from the result of integrating the acceleration twice. As the predetermined approximate expression, for example, an 8th order polynomial can be used, and each coefficient of the 8th order polynomial can be determined by a known method such as a least square method or a downhill simplex method.

  Returning to FIG. 4, the mechanical error is calculated by the correction parameter calculation unit 18 by subtracting the detection position acquired in step S <b> 2 from the machine position calculated in step S <b> 3. FIG. 6 is a diagram illustrating an example of calculation results of the detection position, the machine position, and the machine error when the reciprocating motion is performed. According to FIG. 6, a mechanical error of about ± 2 μm is generated during a reciprocating motion with an amplitude of 10 mm, and the mechanical error changes in a step shape at a position where the motion direction is reversed. This step-like change is mainly caused by backlash existing in the drive mechanism.

  When the machine position and the detection position are acquired using different measurement systems, it is difficult to calculate an accurate machine error because the sampling interval and acquisition timing of the two data are shifted. However, in the present embodiment, since the detected position and acceleration are acquired at the same time in step S2, accurate machine error calculation can be performed relatively easily in step S4 without considering the sampling interval and acquisition timing.

  Returning to FIG. 4, the movement parameter reversal point is extracted by the correction parameter calculation unit 18 from the detection position acquired in step S2 (step S5). Further, based on the relationship between the machine error calculated in step S4 and the detected position, a backlash amount and a rising coefficient are calculated as a backlash feature amount from a machine error waveform within a predetermined displacement before and after reversing the motion direction. 18 (step S6).

  FIG. 7 is a diagram showing a mechanical error waveform of 0.5 mm before and after reversing the motion direction. According to FIG. 7, the mechanical error after reversing the motion direction increases with a certain inclination, and converges to a constant value when the displacement increases. In the present embodiment, such characteristics are extracted as a backlash magnitude A and a backlash rising coefficient B.

  Note that the form in which the feature amount is extracted depends on what correction function the controller to be adjusted has. For example, the relationship between the detection speed obtained by differentiating the detection position and the machine error, the command position and the machine The feature amount may be extracted from the relationship between the error, the relationship between the command speed obtained by differentiating the command position and the mechanical error, the relationship between the acceleration of the motion and the mechanical error, or the relationship between the time and the mechanical error. That is, even if the feature amount is extracted from the relationship between at least one of time, detection position, command position, detection speed or detection acceleration obtained by differentiating the detection position, and command speed or command acceleration obtained by differentiating the command position, and the machine error. Good. Further, the feature amount may be expressed by two or three or more parameters, may be expressed by a mathematical expression, or a mechanical error may be expressed in a lookup table format.

  Returning to FIG. 4, the backlash correction parameter as the command generation unit correction parameter is calculated by the correction parameter calculation unit 18 (step S7). Specifically, the backlash feature value extracted in step S6 is converted into a value to be set in the controller by the correction parameter calculation unit 18. In the controller, for example, the backlash correction parameter is set as the resolution of the rotation angle detector 3 and the unit of 0.01 μm.

  In the present embodiment, the effect when the backlash correction parameter is automatically adjusted will be described with reference to the drawings. FIG. 8A is a diagram illustrating an arc trajectory of an actual machine before automatically adjusting the backlash correction parameter. FIG. 8-2 is a diagram illustrating an arc trajectory of the actual machine after automatically adjusting the backlash correction parameter. In FIGS. 8A and 8B, the machine performs a circular motion with a radius of 10 mm and a feed speed of 1500 mm / min. In FIGS. 8A and 8B, the detection trajectory indicated by the broken line is obtained by synthesizing the detection positions for the two axes, and the mechanical trajectory indicated by the solid line is similarly obtained by synthesizing the machine positions for the two axes. And ask. Further, in FIGS. 8A and 8B, the trajectory error in the radial direction is enlarged and displayed.

  As shown in FIG. 8A, before adjusting the backlash correction parameter, a step-like trajectory error is observed on the mechanical trajectory of each quadrant switching unit (a portion surrounded by a circle). This is a trajectory error mainly due to the influence of backlash. As shown in FIG. 8B, by adjusting the backlash correction parameter by the method of the present invention, the step-like trajectory error appearing on the machine trajectory can be suppressed.

  When the same adjustment work was adjusted by a conventional manual method, it took about 2 hours for the adjustment to obtain a result equivalent to the adjustment result according to the present invention. On the other hand, when the backlash correction parameter was adjusted by the method according to the present invention, the time required for the adjustment was about 5 minutes, confirming the superiority of the correction parameter automatic adjustment method according to the present invention.

  In a similar manner, parameters other than the backlash correction parameter, for example, a pitch error correction parameter can be adjusted. Therefore, a first modification of the first embodiment will be described by taking adjustment of the pitch error correction parameter as an example. FIG. 9 is a flowchart showing an outline of processing in the mechanical motion analysis unit 14 and the correction parameter calculation unit 18 as Modification 1 of the first embodiment, and shows an outline of processing for adjusting the pitch error correction parameter. It is a flowchart.

  The difference from the case of adjusting the backlash correction parameter is that a pitch error correction amount is calculated by the correction parameter calculation unit 18 instead of steps S5 and S6 in the flowchart shown in FIG. 4 (step S8).

  As described above, only the mechanical error can be extracted by using the correction parameter adjusting device 40. Accordingly, there is an effect that correction parameters necessary for improving the mechanical motion accuracy can be calculated in accordance with the correction function of the controller.

  Next, a second modification of the first embodiment will be described by taking adjustment of an elastic deformation correction parameter as an example.

FIG. 10A is a diagram for explaining a mechanism of occurrence of a trajectory error due to elastic deformation. And detecting the position x _m as determined from signals detected by the rotation angle detector 3, between the machine position x _t at a point accelerometer 13 is installed, the feed screw 2 and the support bearing 10, column 5 and ram There is an elastic element such as 6, and its rigidity is K [N / m]. When the mass of the driven body (work table 4, column 5 or ram 6) driven by the motor 1 is M [kg], the driven body is moved during the exercise with a certain acceleration a [m / s ² ]. The elastic deformation amount Δx [m] generated by the acting inertia force F [N] is expressed by the following mathematical formula (1).

In Formula (1), ω _n is a natural angular frequency. That is, if the natural angular frequency ω _{n of} the mechanical element existing between the motor 1 and the place where the accelerometer 13 is installed is known, the elastic deformation amount is calculated by the equation (1) and corrected by the command generation unit 11. Can do. FIG. 10-2 is a diagram illustrating the amplitude of the driven body when a sine wave reciprocating motion is commanded. As shown in FIG. 10-2, when a sinusoidal reciprocating motion is commanded, the maximum acceleration is generated when the motion direction is reversed, so that the amplitude increases by the amount of elastic deformation.

Various correction methods using the elastic deformation amount Δx are known, and the command generation unit 11 may have a correction function, and the motor drive unit 12 may have a correction function. As a form of correction parameter, it may be set as a displacement depending on acceleration, or when calculating and setting a coefficient representing the amount of deformation of the machine with respect to the acceleration, such as the natural frequency of the machine element and the rigidity of the machine element There is also. In the second modification, the correction parameter calculation unit 18 converts the natural angular frequency ω _n of the machine element analyzed by the machine motion analysis unit 14 into a correction parameter having a form suitable for the correction function of the controller to be adjusted. The setting is described below. This setting is described below.

The machine position x _t [m] and the machine speed V when the displacement amplitude of the machine is R [m], the speed amplitude is V [m / s], and an error (elastic deformation amount) due to elastic deformation is Δx [m]. _t [m / s] is expressed by Equation (2). Here, t is time [s].

Therefore, the velocity amplitude when there is no elastic deformation, that is, the velocity amplitude obtained by differentiating the detection position is A _v [m / s], and the velocity amplitude considering the increase in amplitude due to elastic deformation, ie, the accelerometer 13. _Assuming that the velocity amplitude obtained by integrating the acceleration measured by A _vt [m / s], the relationship between the two velocity amplitudes is shown in Equation (3).

Here, R _A in Equation (3) is defined as the velocity amplitude ratio. By substituting the result of the formula (1) into the formula (3) and arranging it, the formula (4) is obtained. Here, ω is an angular frequency [rad / s] of the sinusoidal reciprocating motion, and can be calculated as a ratio of the velocity amplitude V and the displacement amplitude R.

That is, the rate of increase in velocity amplitude due to elastic deformation is determined by the relationship between the natural angular frequency ω _n of the mechanical element and the angular frequency ω of the reciprocating motion. Therefore, the natural angular frequency ω _n of the machine element can be calculated by measuring the velocity amplitude ratio _RA and the angular frequency ω of the reciprocating motion. The following mathematical formula (5) is obtained by modifying the mathematical formula (4) to derive the natural angular frequency ω _n .

  In the configuration shown in FIG. 3, the procedure for calculating the natural angular frequency using Equation (5) will be described with reference to the drawings. FIG. 11 is a flowchart for explaining the procedure for calculating the natural angular frequency in the second modification of the first embodiment. First, a test exercise is executed in step S1. In step S <b> 2, the detected position and acceleration during the test exercise are acquired by the mechanical motion analysis unit 14. A sinusoidal reciprocating motion is suitable as the test motion for calculating the natural angular frequency. For example, the velocity amplitude may be set to 6000 mm / min and the displacement amplitude may be set to about 10 mm. The detected position and acceleration at that time are simultaneously measured as time-synchronized data. In addition, execution of the test exercise | movement in step S1 may be performed by an operator operating a control apparatus, and may be performed by the instruction | indication from a higher-order apparatus.

Next, a velocity amplitude ratio _RA is calculated by the mechanical motion analysis unit 14 from the detected position and acceleration acquired in step S2 (step S9). FIG. 13 is a diagram for explaining an overview of the processing in step S9. As shown in FIG. 13, in the process in step S9, the detection speed amplitude A _vfb and the machine speed amplitude A _vt are calculated from the detection speed obtained by differentiating the detection position and the machine speed obtained by integrating the acceleration. And calculate. When calculating the velocity amplitude, data immediately after the start of exercise and immediately before the end of exercise are removed.

  Returning to FIG. 11, the angular frequency ω of the reciprocating motion is calculated from the displacement amplitude R calculated from the detected position acquired in step S2 and the velocity amplitude V calculated from the detected speed obtained by differentiating the detected position. It is calculated by the mechanical motion analysis unit 14 (step S10). Also in this case, data immediately after the start of exercise and immediately before the end of exercise are removed.

The natural angular frequency ω _n is calculated by the mechanical motion analysis unit 14 by applying the velocity amplitude ratio _RA and the angular frequency ω calculated in steps S9 and S10 to the mathematical formula (5) (step S11). In step S7, the natural angular frequency ω _n calculated in step S11 is converted into a parameter corresponding to the correction function of the controller to be adjusted.

  When the motor driving unit 12 has a function of correcting the elastic deformation amount, the driving is performed as in step S14 (see FIG. 15), which will be described later, instead of calculating the command generation unit correction parameter in step S7. It may be configured to calculate partial correction parameters. Further, when the controller does not have a corresponding correction function, the target position may be corrected separately.

Further, if the natural angular frequency ω _n of the machine element is used, it is applicable not only to correction of the elastic deformation amount but also to parameter adjustment such as a pre-filter and a vibration suppression control function. As described above, since the mechanical motion analysis unit 14 according to the present invention can calculate the natural angular frequency ω _n of the machine element, it is possible to adjust correction parameters corresponding to various functions of the controller to be adjusted. Play.

  Furthermore, there is an effect that the natural angular frequency of the mechanical element can be calculated without performing the excitation test using the sweep sine wave or the M-sequence signal, which has been conventionally performed for identifying the natural angular frequency.

  Next, a third modification of the first embodiment will be described by taking adjustment of an elastic deformation correction parameter as an example.

When the accelerometer 13 is installed in a machine, it is difficult to completely match the moving direction of the movable axis to be measured with the sensitivity direction of the accelerometer. As a result, the true acceleration and the measured acceleration There is a subtle error between the two. Until now, the measurement results obtained when moving at a known acceleration were used to calibrate the accelerometer with high accuracy in advance, thereby avoiding problems due to measurement errors in acceleration. In the third modification, the natural angular frequency ω _n is calculated in the mechanical motion analysis unit 14 by the following method so that the problem due to the measurement error of the acceleration can be avoided without performing the calibration work of the accelerometer.

That is, a sinusoidal reciprocating motion is performed by changing at least one of the amplitude and the velocity in two ways, and the velocity amplitude ratio is calculated from each detected position and acceleration. The velocity amplitude ratio and angular frequency in the first measurement result are R _A (1) and ω (1), respectively, and the velocity amplitude ratio and angular frequency in the second measurement result are respectively R _A (2) and ω (2). In other words, the relationship shown in Equation (6) is established between each velocity amplitude ratio and each angular frequency, as in Equation (4).

Here, if there is a measurement error in acceleration, the error affects the velocity / amplitude ratio _RA , but since the rate of the effect is constant regardless of the motion conditions such as amplitude and velocity, the accelerometer remains attached. If the measurement is performed under two types of motion conditions, the influence of the measurement error of the accelerometer can be offset by calculating the ratio of the two velocity amplitude ratios as shown in Equation (7). Further, the mathematical formula (7) is solved for the natural angular frequency ω _n to obtain the mathematical formula (8).

As described above, according to the mechanical motion analysis unit 14 of the present invention, even if there is an acceleration measurement error, the measurement error is canceled by continuously measuring under two motion conditions, and the natural angular vibration of the mechanical element is obtained. The effect is that the number ω _n can be calculated.

  In the configuration shown in FIG. 3, a method for calculating the natural angular frequency using Equation (8) will be described with reference to the drawings. FIG. 12 is a flowchart for explaining the procedure for calculating the natural angular frequency of the third modification of the first embodiment using Equation (8). First, when the test motion is executed in step S1, the detected position and acceleration during the test motion are acquired by the mechanical motion analysis unit 14 in step S2. A sinusoidal reciprocating motion is suitable as a test motion for calculating the natural angular frequency according to the present embodiment. For example, the velocity amplitude under the first motion condition is set to 3000 mm / min, and the displacement amplitude is set to about 10 mm. Good. The detected position and acceleration at that time are simultaneously measured as time-synchronized data. In addition, execution of the test exercise | movement in step S1 may be performed by an operator operating a control apparatus, and may be performed by the instruction | indication from a higher-order apparatus.

The first velocity amplitude ratio R _A (1) is calculated by the mechanical motion analysis unit 14 from the detected position and acceleration acquired in step S2 (step S9). As described above, the outline of the processing in step S9 is as follows. From the detection speed obtained by differentiating the detection position and the machine speed obtained by integrating the acceleration, the detection speed amplitude A _vfb and the machine speed amplitude A _{vt are obtained.} Are calculated (see also FIG. 13). When calculating the velocity amplitude, data immediately after the start of exercise and immediately before the end of exercise are removed. In step S10, the angular frequency ω (1) of the first reciprocating motion is calculated. Also in this case, data immediately after the start of exercise and immediately before the end of exercise are removed.

  Since the two sets of data are necessary for the calculation of the natural angular frequency according to the equation (8), when the two sets of data have not been acquired (step S15, No), the amplitude and speed of the reciprocating motion Is changed (step S12), and the process returns to step S1. It is preferable to set the velocity amplitude under the second motion condition to 6000 mm / min and the displacement amplitude to about 10 mm. Steps S12 and S1 may be performed by an operator operating the control device, or may be performed by an instruction from a higher-level device.

The detected position and acceleration during the test motion are acquired synchronously in step S2, and the second velocity amplitude ratio R _A (2) is calculated in step S9 using the result. In step S10, the angular frequency ω of the second reciprocating motion is calculated from the displacement amplitude R calculated from the detection position acquired in step S2 and the speed amplitude V calculated from the detection speed obtained by differentiating the detection position. (2) is calculated.

When acquisition of two sets of data is completed (step S15, Yes), the velocity amplitude ratios R _A (1) and R _A (2) calculated in step S9 and the angular frequency ω (1) calculated in step S10 And ω (2), the natural angular frequency ω _n is calculated according to Equation (8) (step S11). Further, in step S7, the correction parameter calculation unit 18 converts the natural angular frequency calculated in step S11 into a parameter corresponding to the correction function of the controller to be adjusted.

  When the motor driving unit 12 has a function of correcting the elastic deformation amount, the driving is performed as in step S14 (see FIG. 15), which will be described later, instead of calculating the command generation unit correction parameter in step S7. It may be configured to calculate partial correction parameters. Further, when the controller does not have a corresponding correction function, the target position may be corrected separately.

  Further, if the natural angular frequency of the machine element is used, the present invention is not limited to correction of the elastic deformation amount but can be applied to parameter adjustment such as a pre-filter and a vibration suppression control function. As described above, since the mechanical motion analysis unit 14 according to the present invention can calculate the natural angular frequency of the machine element, it is possible to adjust correction parameters corresponding to various functions of the controller to be adjusted. .

  Furthermore, there is an effect that the natural angular frequency of the mechanical element can be calculated without performing the excitation test using the sweep sine wave or the M-sequence signal, which has been conventionally performed for identifying the natural angular frequency.

  An example of the effect obtained by adjusting the elastic deformation amount correction parameter by the correction parameter adjustment device 40 according to the present invention will be described with reference to the drawings. FIG. 14A is a diagram illustrating a detection trajectory and a mechanical trajectory before adjusting an elastic deformation amount correction parameter. FIG. 14B is a diagram of the detection locus and the mechanical locus after adjusting the elastic deformation amount correction parameter. 14A and 14B are measurement results of an arc locus in a mechanical device having characteristics in which the natural frequency in the X-axis direction is significantly smaller than the natural frequency in the Y-axis direction, and are synthesized from the detection position. The detected trajectory is indicated by a broken line, and the mechanical trajectory synthesized from the machine position calculated by the mechanical motion analysis unit 14 is indicated by a solid line. The results are displayed with the radial error enlarged.

  Looking at the results when correction is not performed (FIG. 14-1), the detection trajectory is almost a perfect circle, except for the projection-like trajectory error (quadrant projection) generated in the quadrant switching unit. The machine trajectory has an elliptical shape having a long axis in the X-axis direction. This is a trajectory error due to elastic deformation due to the small rigidity in the X-axis direction. On the other hand, when the correction parameter is adjusted by the method of the present invention (FIG. 14-2), the machine locus is close to a perfect circle except for the quadrant projection.

  When the same adjustment is performed by the conventional method, the ball bar method cannot measure at a high acceleration that causes elastic deformation, so it was necessary to use a measuring instrument called a grid encoder (cross grid method). Machines that can measure with an encoder are limited, and adjustment work is often impossible. Moreover, since adjustment was made by trial and error while observing the measurement results, it took a lot of time.

  On the other hand, the correction parameter adjustment apparatus according to the present invention measures the machine position using an accelerometer whose installation target is not limited, and thus can be applied to most machine apparatuses, and reduces the natural angular frequency of machine elements. Since the time can be calculated, it has an excellent effect that the time required for adjusting the correction parameter can be greatly shortened.

  Next, a fourth modification of the first embodiment will be described using adjustment of the drive unit correction parameter as an example. FIG. 15 is a flowchart showing an outline of the adjustment process of the drive unit correction parameter in the fourth modification of the first embodiment.

  First, when the test motion is executed in step S1, the detected position and acceleration during the test motion are acquired by the mechanical motion analysis unit 14 in step S2. In addition, execution of the test exercise | movement in step S1 may be performed by an operator operating a control apparatus, and may be performed by the instruction | indication from a higher-order apparatus.

  The machine position is calculated from the detected position and acceleration acquired in step S2 (step S3). Since the process in step S3 is the same as that described above (see also FIG. 5), detailed description thereof is omitted. From the machine position calculated in step S3, a trajectory error within a predetermined range in which improvement is expected by parameter adjustment is calculated by the machine motion analysis unit 14 (step S13). The trajectory error is calculated as, for example, the difference between the command position and the machine position, the amplitude of the vibration that occurs during corner motion, or the difference between the average radius and the radius of the machine trajectory if it is a circular motion with simultaneous biaxial control. it can.

  The correction parameter calculation unit 18 determines whether the trajectory error amount calculated in step S13 is equal to or smaller than a preset setting value. If the trajectory error amount is larger than the set value (No in step S16), The drive parameter correction parameter is adjusted by the correction parameter calculator 18 (step S14), and the process returns to step S1. The above steps are repeated until the trajectory error amount is equal to or less than the set value (step S16, Yes).

  As for the adjustment method of the drive unit correction parameter in step S14, various methods such as the downhill simplex method are known, and the locus error amount is calculated using the detection locus obtained from the detection position, and the drive unit correction is performed. Methods for adjusting parameters are also known. However, in the above known technology, since adjustment based on the machine locus obtained from the machine position is impossible, the accuracy of the machine locus that is truly required to be improved can be improved by automatic adjustment of correction parameters. There wasn't.

  On the other hand, the correction parameter adjustment device 40 according to the present invention measures the machine position using an accelerometer, calculates the trajectory error amount from the machine trajectory synthesized from the machine position, and adjusts the correction parameter, which is a real improvement. It is possible to improve the accuracy of the desired machine trajectory by automatic adjustment. In addition, since the machine position is measured using an accelerometer whose installation target is not limited, there is an effect that it can be applied to almost all mechanical devices.

  In this modification, the adjustment of the drive unit correction parameter is described as an example. However, the command generation unit correction parameter may be adjusted by a similar method, or may be applied to the correction of the target position. Is possible.

  An example of the effect of the automatic adjustment of the drive unit correction parameter by the correction parameter automatic adjustment device according to the present invention will be described with reference to the drawings. FIG. 16A is a diagram illustrating a detection locus and a machine locus before adjusting the drive unit correction parameter. FIG. 16B is a diagram illustrating a detection locus and a machine locus after adjusting the drive unit correction parameter. FIGS. 16A and 16B show arc trajectories when the feed rate is 3000 mm / min and the radius is 25 mm. The detected trajectory synthesized from the detection position is calculated by a broken line, a mechanical motion analysis unit. The machine trajectories synthesized from the machine positions are indicated by solid lines. The results are displayed with the radial error enlarged.

  From the result before adjusting the correction parameter (FIG. 16-1), it can be seen that a protrusion-like locus error (quadrant protrusion) having a height of about 5 μm is generated on the machine locus at each quadrant switching portion. This trajectory error is mainly due to the frictional force generated on the guide surface for guiding the work table movement, and adjustment of the friction compensation parameter, the backlash acceleration parameter, etc. is effective for reducing the error amount. However, in the conventional automatic correction parameter adjusting device, since the drive unit correction parameter is adjusted based on the quadrant protrusion generated on the detection locus, it is difficult to reduce the quadrant protrusion generated on the mechanical locus.

  On the other hand, when the result (FIG. 16-2) when the drive unit correction parameter is adjusted by the correction parameter adjusting device according to the present invention is seen, the quadrant protrusion generated on the machine locus is significantly reduced compared to the result before the adjustment. I understand that. That is, in the fourth modification, the region where the quadrant protrusions appear before the adjustment is set within a predetermined range where improvement is expected by adjusting the parameters. The adjustment of the drive unit correction parameter is performed only for the quadrant protrusions near 90 degrees and 270 degrees, and there is no difference between the quadrant protrusions occurring near 0 degrees and 180 degrees before and after the adjustment. . Of course, quadrant protrusions occurring near 0 degrees and 180 degrees can be reduced by the correction parameter adjusting apparatus of the present invention.

  As described above, the correction parameter adjusting apparatus according to the present invention adjusts the correction parameter based on the error amount generated on the machine locus that really needs improvement, and thus has an excellent effect of improving the accuracy of the machine locus.

  As described above, the correction parameter adjustment device according to the present invention is suitable for a correction parameter adjustment device that adjusts a correction parameter that improves the motion accuracy of a machine in a numerically controlled machine tool or a robot.

DESCRIPTION OF SYMBOLS 1 Motor 2 Feed screw 3 Rotation angle detector 4 Worktable 5 Column 6 Ram 7 Spindle head 8 Coupling 9 Nut 10 Support bearing 11 Command generation part 12 Motor drive part 13 Accelerometer 14 Machine motion analysis part 15 Motion track display part 18 Correction parameter calculation unit 40 Correction parameter adjustment device 50 Numerical control machine tool

Claims (7)

In order to measure the acceleration of the machine in a device that controls the position of the machine by feeding back the detected position detected from the motor that drives the movable shaft and driving the motor so that the detected position follows the command position. Accelerometer
A mechanical motion analysis unit for analyzing mechanical motion from the acceleration and the detection position;
And a correction parameter calculation unit that determines a correction parameter for improving the movement accuracy of the machine from the analysis result of the machine movement. The acceleration and the detection position are measured in synchronization,
The mechanical motion analysis unit obtains a mechanical position from the acceleration and the detection position, and obtains a mechanical error that is a difference between the mechanical position and the detection position,
The correction parameter adjustment apparatus according to claim 1, wherein the correction parameter calculation unit determines the correction parameter using the mechanical error as the analysis result.   The correction parameter calculation unit includes at least one of time, the detection position, the command position, a detection speed or detection acceleration obtained by differentiating the detection position, and a command speed or command acceleration obtained by differentiating the command position, and the mechanical error. The correction parameter adjusting apparatus according to claim 2, wherein a feature amount representing the relationship is calculated and converted into the correction parameter.   The mechanical motion analysis unit is configured to detect an amplitude of a detected speed obtained by differentiating the detected position from the acceleration and the detected position when a reciprocating motion is commanded to the machine, and an amplitude of a mechanical speed obtained by integrating the acceleration. 2. The correction parameter according to claim 1, wherein a ratio is calculated, and a natural frequency of a mechanical structure is calculated from the amplitude ratio, or a coefficient representing a deformation amount of the machine with respect to the acceleration is calculated. Adjustment device.   The mechanical motion analysis unit detects the detected position obtained by differentiating the detected position in each reciprocating motion from the acceleration and the detected position when commanding two reciprocating motions with different amplitude and speed of the machine. The amplitude ratio between the amplitude of the speed and the amplitude of the machine speed obtained by integrating the acceleration is calculated, and the natural frequency of the mechanical structure is calculated from the two amplitude ratios, or the deformation of the machine with respect to the acceleration The correction parameter adjusting apparatus according to claim 1, wherein a coefficient representing a quantity is calculated.   The correction parameter calculation unit determines a correction parameter for improving the motion accuracy of the machine, using a natural frequency calculated by the machine motion analysis unit or a coefficient representing a deformation amount of the machine with respect to the acceleration. 6. The correction parameter adjusting device according to claim 4 or 5, characterized in that: The mechanical motion analysis unit obtains a machine position from the acceleration and the detected position, obtains a command locus synthesized from the command position and a machine locus synthesized from the machine position, and is within a preset range. Calculate the amount of trajectory error between the command trajectory and machine trajectory at
The correction parameter calculation unit changes the correction parameter so as to reduce the error,
2. The correction parameter adjusting apparatus according to claim 1, wherein the calculation of the trajectory error and the change of the correction parameter are repeated until the trajectory error is equal to or less than a preset set value.

Images