Workpiece mass estimation device

Abstract

The purpose of the present invention is to enable accurate estimation of the mass of a workpiece. This workpiece mass estimation device comprises a vibrating unit, a calculation unit, an identification unit, and an estimation unit. The vibration unit commands a motor control unit to perform vibration to agitate output of a motor and to perform a sweep vibration to vary the vibration frequency. The calculation unit, on the basis of motor information detected at a plurality of detection times within the period of execution of the sweep vibration, calculates the overall inertia as the inertia of a driven body for each detection time. The identification unit identifies the time, within the execution period, that has the vibration frequency for which the change in the overall inertia associated with the change to the adjacent vibration frequency is determined to be the least. The estimation unit, by calculating a value obtained by subtracting the inertia of a general driven body that includes a rotor, a transmission mechanism, and a workpiece loading unit from the overall inertia corresponding to the identified time, calculates the inertia of a workpiece to estimate the mass of the workpiece.

Description

[Technical Field]

[0001] This disclosure relates to a workpiece mass estimation device for estimating the mass of a workpiece loaded onto various machines such as machine tools. [Background technology]

[0002] Some machines, such as machine tools, are equipped with a workpiece loading section, a motor, a sensor, and a motor control unit. Workpieces are loaded onto the workpiece loading section. The motor drives the workpiece loading section. The sensor detects the motor's state. The motor's state includes, for example, the motor's current value and rotational speed. The motor control unit provides feedback control to the motor based on the detected motor state. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2015-55923 [Overview of the project] [Problems that the invention aims to solve]

[0004] The total mass of the driven body driven by the motor changes depending on the mass of the workpiece loaded on the workpiece loading section. Therefore, the inertia of the driven body also changes depending on the mass of the workpiece. This change in inertia changes the acceleration and deceleration of the driven body by the motor. Excessive acceleration and deceleration places an excessive load on the machine and can lead to machine failure, while insufficient acceleration and deceleration worsens the machine's work efficiency. Therefore, it is necessary to adjust the acceleration and deceleration of the driven body by the motor to the optimal level. For this reason, some machines adjust the acceleration and deceleration of the driven body by the motor to the optimal level based on the mass of the workpiece entered by the operator.

[0005] While such technology allows for adjustment of the acceleration and deceleration of the driven object, it requires the operator to manually input the workpiece mass each time. Therefore, a method for automatically estimating the workpiece mass has been proposed as follows: First, an excitation command is sent to the motor control unit of the machine to vibrate the motor output. The inertia of the driven object at this time is identified based on motor information such as current feedback values ​​and rotational speed feedback values ​​detected by sensors. The inertia of the workpiece is calculated by subtracting the inertia caused by factors other than the workpiece from the identified inertia of the driven object, and the workpiece mass is estimated.

[0006] However, the Disclosers have noted that such a configuration may lead to the following problems. Specifically, when the excitation frequency is a predetermined frequency such as the resonance frequency with the machine, a large error occurs in the motor information detected by the sensor, resulting in a large error in the calculated inertia. As a result, it becomes impossible to accurately estimate the mass of the workpiece.

[0007] This disclosure is made in view of the above circumstances and aims to enable accurate estimation of the mass of a workpiece. [Means for solving the problem]

[0008] The workpiece mass estimation apparatus disclosed herein is The machine comprises a workpiece loading section on which workpieces are loaded, a motor having a rotor and a stator that rotates the rotor, a transmission mechanism that transmits the motor output as a force to rotate the rotor to the workpiece loading section, a sensor that detects motor information indicating the state of the motor, and a motor control unit that controls the motor based on the detected motor information, and the motor output drives a driven body including the rotor, the transmission mechanism, the workpiece loading section, and the workpiece. A workpiece mass estimation device for estimating the mass of the aforementioned workpiece, An excitation unit commands the motor control unit to perform a sweep excitation that vibrates the motor output and varies the excitation frequency, A calculation unit calculates the overall inertia as the inertia of the driven body for each detection point based on the motor information detected at multiple detection points within the sweep excitation execution period, A calculation unit that determines the time point in time within the execution period at which the change in the overall inertia associated with the change to an adjacent excitation frequency is minimized, An estimation unit calculates the inertia of the workpiece by subtracting the inertia of the general driven body, including the rotor, the transmission mechanism, and the workpiece loading section, from the total inertia corresponding to the determined point in time, and estimates the mass of the workpiece. It is equipped with. [Brief explanation of the drawing]

[0009] [Figure 1] This is a configuration diagram showing the workpiece mass estimation device and machine tool of this embodiment. [Figure 2] This graph shows the relationship between the excitation frequency and the estimated overall inertia. [Figure 3] This graph shows the waveforms of each detected value in frequency bands other than the first frequency band. [Figure 4] This graph shows the waveform of each detected value in the first frequency band. [Figure 5] This graph shows the trend in the detected current value. [Figure 6] This graph shows the change in waveform due to sweep excitation. [Figure 7] This graph shows the relationship between the excitation frequency and the estimated overall inertia. [Figure 8] This is an enlarged view of section VIII shown in Figure 7. [Figure 9] This is a flowchart showing the procedure for estimating the mass of a workpiece. [Modes for carrying out the invention]

[0010] [First Embodiment] As shown in FIG. 1, the workpiece mass estimation device 100 of the present embodiment is installed on the machine tool 200. The machine tool 200 includes a motor control unit 50, a sensor 60, a motor 70, a transmission mechanism 80, and a workpiece loading unit 90.

[0011] The workpiece loading unit 90 is provided so as to be movable in the rotational direction or the linear direction. A workpiece W is loaded on the workpiece loading unit 90. The motor 70 has a rotor 78 and a stator 76 that rotates the rotor 78. Hereinafter, the output from the stator 76 to the rotor 78 is referred to as "motor output Mo". The transmission mechanism 80 transmits the motor output Mo from the rotor 78 to the workpiece loading unit 90. The transmission mechanism 80 may include a speed reducer such as a gear.

[0012] Hereinafter, the component group driven by the motor output Mo is referred to as "driven body Db". The driven body Db includes the rotor 78, the transmission mechanism 80, the workpiece loading unit 90, and the workpiece W. Also hereinafter, the one excluding the workpiece W from the driven body Db is referred to as "general driven body". Also hereinafter, the inertia of the driven body Db is referred to as "total inertia J", and the inertia of the general driven body is referred to as "general inertia".

[0013] The sensor 60 detects motor information Mi indicating the state of the motor 70. The motor information Mi includes the current value of the motor 70 and the rotational speed of the rotor 78. The motor control unit 50 performs feedback control of the motor 70 based on the motor information Mi detected by the sensor 60.

[0014] Next, the workpiece mass estimation device 100 will be described. The workpiece mass estimation device 100 comprises an excitation unit 10, a calculation unit 20, an indexing unit 30, and an estimation unit 40. The excitation unit 10, the calculation unit 20, the indexing unit 30, and the estimation unit 40 are, for example, mainly composed of the same computer. The computer has, for example, a CPU, ROM, RAM, memory, etc. Although the workpiece mass estimation device 100 and the machine tool 200 are shown separately in Figure 1, the workpiece mass estimation device 100 may be incorporated into the machine tool 200.

[0015] The vibration unit 10 outputs a vibration command Vc to the motor control unit 50. The vibration command Vc is a command to vibrate the motor output Mo. More specifically, the vibration command Vc is a command that adds a sinusoidal command of a constant frequency to the position command, torque command, or rotational speed command issued by the motor control unit 50.

[0016] The calculation unit 20 calculates the total inertia J based on the current value Iq of the motor 70 and the rotational speed ω of the rotor 78 detected by the sensor 60 during the vibration execution period. Specifically, for example, the total inertia J is calculated based on the following equation 1.

[0017]

number

[0018] In Equation 1, "Kt" is the torque constant Kt of motor 70, "Iq" is the current value Iq of motor 70, "ω" is the rotational speed ω of rotor 78, and "t" is the elapsed time t. Also, in Equation 1, "￣" indicates the average value of values ​​sampled within a certain number of seconds.

[0019] Furthermore, in Equation 1, "F(ω)" is the frictional force F(ω). The frictional force F(ω) is the sum of viscous friction and Coulomb friction, and is expressed as a function of the rotational speed ω of the rotor 78. In Equation 1, the latter term, the frictional force F(ω), can be ignored if it is small compared to the first term, the product of the torque constant Kt and the average current value. If it is not small, for example, a known frictional force F(ω) can be substituted into F(ω) in Equation 1.

[0020] The estimation unit 40 calculates the inertia of the workpiece W by subtracting the pre-stored general inertia from the total inertia J calculated by the calculation unit 20, and then estimates the mass of the workpiece W.

[0021] Next, with reference to Figures 2 to 5, the problems to be solved in this embodiment will be explained. Hereinafter, the frequency at which the motor output Mo is excited will be referred to as the "excitation frequency fv". Depending on the excitation frequency fv, a large error may occur in the motor information Mi detected during the execution of the excitation, and a large error may also occur in the calculated overall inertia J. As a result, it becomes impossible to accurately estimate the mass of the workpiece W.

[0022] Specifically, the frequency bands that are likely to produce large errors are the following first to third frequency bands B1 to B3.

[0023] First, let's explain the first frequency band B1 shown on the left side of Figure 2. The first frequency band B1 is the frequency band in which the movement speed of the workpiece loading section 90 becomes so low that Coulomb friction becomes dominant. At frequencies higher than this first frequency band B1, the Coulomb friction F(ω) becomes sufficiently small. Therefore, as shown in Figure 3, there is no dead zone between the current value Iq and the rotational speed ω. Note that the dead zone here refers to the interval in which the values ​​hardly change.

[0024] On the other hand, in the first frequency band B1, stick-slip-like behavior occurs in the movement of the workpiece loading section 90 due to Coulomb friction. As a result, as shown in Figure 4, a dead zone Dz occurs in one or both of the current value Iq and the rotational speed ω. Due to this dead zone Dz, as shown in Figure 2, the error in the calculated overall inertia J becomes large in the first frequency band B1, making it impossible to accurately estimate the mass of the workpiece W.

[0025] Next, we will explain the second frequency band B2 shown in the left-right center of Figure 2. The second frequency band B2 is the frequency band in which the commanded excitation frequency fv overlaps with the natural frequency of the entire machine tool 200. In this second frequency band B2, as shown in Figure 5, the current value Iq is amplified. As a result, in this second frequency band B2, as shown in the left-right center of Figure 2, the error in the calculated total inertia J becomes large, making it impossible to accurately estimate the mass of the workpiece W.

[0026] Next, we will explain the third frequency band B3 shown on the right side of Figure 2. The third frequency band B3 is the frequency at which the movement speed of the workpiece loading section 90 becomes so fast that it falls outside the control response band in the control of the motor 70. In this third frequency band B3, the current value becomes unstable as shown in Figure 5. As a result, in this third frequency band B3, the error in the overall inertia J becomes large, as shown on the right side of Figure 2, and it becomes impossible to accurately estimate the mass of the workpiece W.

[0027] Therefore, if we were to randomly select an excitation frequency fv as shown in Figure 2, there is a risk that we would select an excitation frequency fv belonging to one of the first to third frequency bands B1, B2, or B3. In that case, it would be impossible to accurately estimate the mass of the workpiece W.

[0028] To solve the above problems, in this embodiment, the vibration unit 10 and calculation unit 20 shown in Figure 1 are configured as follows, and the workpiece mass estimation device 100 further includes an indexing unit 30.

[0029] As shown in Figure 6, the excitation unit 10 commands the motor control unit 50 to perform sweep excitation, which varies the excitation frequency fv within a predetermined range vR. Specifically, the excitation unit 10 commands the motor control unit 50 to perform sweep excitation by inputting a torque command value and varying the excitation frequency fv of the torque command value.

[0030] In this case, the excitation unit 10 varies the excitation frequency fv from the low-frequency end of the variation range vR toward the high-frequency end. However, instead, the excitation frequency fv may be varied from the high-frequency end of the variation range vR toward the low-frequency end. The excitation unit 10 is configured so that the variation range vR over which the excitation frequency fv is varied in sweep excitation can be changed by an operator or the like.

[0031] When sweep excitation is performed, the sensor 60 shown in Figure 1 detects motor information Mi at multiple detection points P that fall within the sweep excitation execution period. Based on the motor information Mi at these multiple detection points P, the calculation unit 20 calculates the total inertia J for each detection point P, as shown in Figure 7.

[0032] The following describes the adjacent detection time point P. n-1 ,P n The change in the overall inertia J between two points is called the "rate of change ΔJ". As shown in Figure 8, the calculation unit 20 calculates the rate of change at each detection time point P. This calculation is performed, for example, based on the following equation 2.

[0033]

number

[0034] Here, "J n " is the nth detection time point P n The overall inertia J is "J n-1 " is the n-1 detection time point P n-1 This is the overall inertia J. "ΔJ n " is the n-1 detection time point Pn-1 and the n-th detection time point P n is the change rate ΔJ therebetween.

[0035] The determination unit 30 determines the detection time point P at which the change rate ΔJ is the minimum from among the plurality of detection time points P for which the change rate ΔJ has been calculated by the calculation unit 20. Thereby, from among the excitation frequencies fv within the execution period of the sweep excitation, that is, from within the predetermined fluctuation range vR shown in FIG. 7, the excitation frequency fv at which it is determined that the change in the total inertia J accompanying the change to the adjacent excitation frequency fv is minimized is determined. Therefore, the determination unit 30 can easily determine the excitation frequency fv that does not overlap any of the first to third frequency bands B1, B2, B3 shown in FIG. 7.

[0036] This is because, first, in the first frequency band B1, since the estimated total inertia J always decreases with a right shoulder, the change rate ΔJn tends to be large. Also, in the second and third frequency bands B3, since the estimated total inertia J turns to decrease with a right shoulder immediately after increasing with a right shoulder, the change rate ΔJn tends to be large. Therefore, at the detection time point P of the excitation frequency fv that does not belong to any of the first to third frequency bands, the change rate ΔJ is likely to be minimized.

[0037] However, in the second and third frequency bands B2, by chance, on both sides sandwiching the peak in the curve showing the transition of the estimated total inertia J, the total inertia J at the (n - 1)-th detection time point Pn-1 n-1 and the total inertia J at the n-th detection time point Pn n may become substantially equal. In this case, there is a possibility that the change rate ΔJ at the detection time point P near the peak in the second or third frequency band B2 may be minimized.

[0038] To completely avoid such drawbacks, the indexing unit 30 may be configured to determine the detection time point at which the rate of change ΔJ is minimized, based on additional conditions. Specifically, for example, in the above equation 2, after the detection time point P at which the rate of change ΔJ is minimized is determined, further, in the same equation 2, "n-1" is replaced with "n-2", and the rate of change ΔJ n Recalculate the rate of change ΔJ. n If the value exceeds a predetermined value, the system may be configured to determine the detection time P that minimizes the rate of change ΔJ from among the remaining multiple detection time Ps, excluding the detection time P in question.

[0039] The estimation unit 40 shown in Figure 1 calculates the inertia of the workpiece W by subtracting the general inertia from the total inertia J at detection time P, which is determined by the calculation unit 30, and then estimates the mass of the workpiece W.

[0040] Subsequently, the estimation unit 40 transmits, for example, the estimated mass of the workpiece W to the motor control unit 50. In this case, the motor control unit 50 changes the control mode of the motor 70 and the acceleration / deceleration conditions based on the received mass of the workpiece W.

[0041] Next, referring to Figure 9, we will explain the flow of mass estimation for workpiece W as described above. In the following, "S" is an abbreviation for "step".

[0042] First, in S1, the workpiece mass estimation device 100 determines whether or not sweep excitation is in progress. If a negative determination N (No) is made, the determination in S1 is repeated, for example, at predetermined intervals. On the other hand, if an affirmative determination Y (Yes) is made, the process proceeds to the next S2.

[0043] In S2, the calculation unit 20 acquires motor information Mi from the sensor 60. In the following S3, the calculation unit 20 calculates the overall inertia J based on the motor information Mi. In the following S4, the calculation unit 20 calculates the detection time P n In the overall Inertia J n And, at the previous detection point P n-1In the overall Inertia J n-1 Based on this, the rate of change ΔJ is calculated. In the following S5, the calculation unit 20 determines whether the current excitation frequency fv is within a predetermined fluctuation range vR. If a positive determination Y is made, the process returns to S2. On the other hand, if a negative determination N is made in S5, the process proceeds to S6.

[0044] In S6, the indexing unit 30 identifies the detection time P from among multiple detection time points P for which the rate of change ΔJ has been calculated, which is the detection time point P that minimizes the rate of change ΔJ.

[0045] In the following step S7, the estimation unit 40 calculates the total inertia J at the determined detection time P. In the following step S8, the estimation unit 40 calculates the inertia of the workpiece W by subtracting the general inertia from the calculated total inertia J, and estimates the mass of the workpiece W.

[0046] The configuration and effects of this embodiment are summarized below.

[0047] The excitation unit 10 inputs a torque command value to the motor control unit 50 and, by varying the excitation frequency fv of the torque command value, can cause the motor control unit 50 to perform sweep excitation.

[0048] The calculation unit 20 calculates the total inertia for each detection point P based on the motor information Mi detected at multiple detection points P during the sweep excitation execution period. The indexing unit 30 identifies the time point at which the change in total inertia associated with the change to an adjacent excitation frequency fv is minimized, within the sweep excitation execution period. This makes it possible to efficiently identify the time point at which the excitation frequency fv does not overlap with any of the first to third frequency bands B1, B2, and B3, as described above.

[0049] The estimation unit 40 calculates the inertia of the workpiece W by subtracting the general inertia from the total inertia J corresponding to the time point determined in this way, and estimates the mass of the workpiece W. This allows for an accurate estimation of the mass of the workpiece W. Based on the accurately estimated mass of the workpiece W, the machine tool 200 can, for example, adjust the motor output Mo to an appropriate value without excess or deficiency, thereby adjusting the acceleration and deceleration of the driven body Db to an appropriate value without excess or deficiency.

[0050] Furthermore, the excitation unit 10 is configured to allow changing the variation range vR for varying the excitation frequency fv in sweep excitation. Therefore, the operator can set the variation range vR as appropriate.

[0051] [Other embodiments] The embodiments described above can be modified as follows, for example: The workpiece mass estimation device 100 may be installed on a machine other than the machine tool 200. The vibration unit 10 may receive position commands or rotational speed commands as vibration input instead of torque command values.

[0052] According to the above embodiments, the workpiece mass estimation devices described in Appendices 1 to 4 below can be realized.

[0053] [Note 1] A machine (200) that drives a driven body (Db) including the rotor (78), the transmission mechanism (80), the workpiece (W), and the workpiece (W) using the motor output (Mo), a motor output (Mo) as the force that rotates the rotor (78) to the workpiece (90), a sensor (60) that detects motor information (Mi) indicating the state of the motor (70), and a motor control unit (50) that controls the motor (70) based on the detected motor information (Mi), wherein the motor output (Mo) drives a driven body (Db) including the rotor (78), the transmission mechanism (80), the workpiece (90), and the workpiece (W), A workpiece mass estimation device (100) for estimating the mass of the workpiece (W), An excitation unit (10) commands the motor control unit (50) to perform a sweep excitation that vibrates the motor output (Mo) and varies the excitation frequency (fv), A calculation unit (20) calculates the total inertia (J) as the inertia of the driven body (Db) for each detection time (P) based on the motor information (Mi) detected at multiple detection time points (P) within the sweep excitation execution period, A calculation unit (30) determines the time point within the execution period at which the change in the overall inertia (J) associated with the change to an adjacent excitation frequency (fv) is minimized, An estimation unit (40) calculates the inertia of the workpiece (W) and estimates the mass of the workpiece (W) by subtracting the inertia of the general driven body (Db), including the rotor (78), the transmission mechanism (80), and the workpiece loading section (90), from the total inertia (J) corresponding to the determined point in time, A workpiece mass estimation device (100) equipped with the following.

[0054] [Note 2] The vibration unit (10) is a workpiece mass estimation device (100) as described in Appendix 1, which inputs a torque command value to the motor control unit (50) and commands the sweep vibration by varying the vibration frequency (fv) of the torque command value.

[0055] [Note 3] The workpiece mass estimation device (100) according to Appendix 1 or 2, wherein the excitation unit (10) is configured to change the variation range (vR) for varying the excitation frequency (fv) in the sweep excitation.

[0056] [Note 4] The indexing unit (30) determines the time point at which the excitation frequency (fv) is determined to be the minimum, As Coulomb friction becomes dominant, the movement speed of the workpiece loading section (90) decreases, resulting in a first frequency band (B1) where a dead zone occurs between the current value of the motor (70) and the rotational speed of the rotor (78). A second frequency band (B2) in which resonance with the aforementioned machine occurs, A third frequency band (B3) in which the movement speed of the workpiece loading section (90) increases as it goes outside the control response band of the motor (70), Determine the time point at which the excitation frequency (fv) does not overlap with any of the others. A workpiece mass estimation device (100) as described in any one of the appendices 1 to 3.

[0057] According to the workpiece mass estimation device (100) described in appendices 1 to 4 above, the mass of the workpiece (W) can be estimated with high accuracy.

[0058] Although the present disclosure has been described in detail above, it is not limited to the individual embodiments described above. These embodiments can be added, replaced, modified, partially deleted, etc., in any way that does not depart from the gist of the present disclosure or from the spirit of the present disclosure derived from the claims and their equivalents. Furthermore, these embodiments can be implemented in combination. For example, the order of operations and processes in the embodiments described above are shown as examples only and are not limited thereto. The same applies when numerical values ​​or mathematical formulas are used in the description of the embodiments described above. [Explanation of Symbols]

[0059] 10 Vibration section 20 Calculation Section 30 Indexing part 40 Estimation part 50 Motor control unit 60 sensors 70 Motor 76 stata 78 Rotor 80 Transmission mechanism 90 Work Loading Section 100 Workpiece Mass Estimation Device 200 Machine tools B1 First frequency band B2 Second frequency band B3 Third frequency band Db Driven Unit fv excitation frequency J General Inertia Mi Motor Information Mo motor output vR variation range Double job

Claims

1. The machine comprises a workpiece loading section on which workpieces are loaded, a motor having a rotor and a stator that rotates the rotor, a transmission mechanism that transmits the motor output as a force to rotate the rotor to the workpiece loading section, a sensor that detects motor information indicating the state of the motor, and a motor control unit that controls the motor based on the detected motor information, and the motor output drives a driven body including the rotor, the transmission mechanism, the workpiece loading section, and the workpiece. A workpiece mass estimation device for estimating the mass of the aforementioned workpiece, An excitation unit commands the motor control unit to perform a sweep excitation that vibrates the motor output and varies the excitation frequency, A calculation unit calculates the overall inertia as the inertia of the driven body for each detection point based on the motor information detected at multiple detection points within the sweep excitation execution period, A calculation unit that determines the time point in time within the execution period at which the change in the overall inertia associated with the change to an adjacent excitation frequency is minimized, An estimation unit calculates the inertia of the workpiece by subtracting the inertia of the general driven body, including the rotor, the transmission mechanism, and the workpiece loading section, from the total inertia corresponding to the determined point in time, and estimates the mass of the workpiece. A workpiece mass estimation device equipped with the following features.

2. The workpiece mass estimation device according to claim 1, wherein the vibration unit inputs a torque command value to the motor control unit and commands sweep vibration by varying the vibration frequency of the torque command value.

3. The workpiece mass estimation device according to claim 1 or 2, wherein the excitation unit is configured to change the range of variation in which the excitation frequency is varied in the sweep excitation.

4. The indexing unit determines the time point at which the excitation frequency is determined to be the minimum, The first frequency band is such that the movement speed of the workpiece loading section becomes so slow as Coulomb friction becomes dominant that a dead zone occurs between the motor current value and the rotor rotation speed, A second frequency band in which resonance with the aforementioned machine occurs, A third frequency band in which the movement speed of the workpiece loading section increases as it moves outside the control response band of the motor control, Determine the time point at which the excitation frequency does not overlap with any of the above. The workpiece mass estimation apparatus according to claim 1 or 2.

Images