Frequency characteristic measurement method and frequency characteristic measurement device

Abstract

The frequency measurement method comprises: a first output step for outputting a first excitation signal in a first frequency range to a servo system (20); a first frequency characteristic calculation step for calculating a first frequency characteristic of the servo system (20) within a first frequency range on the basis of a first identification input signal and a first identification output signal acquired at a first sampling period from the servo system (20) to which the first excitation signal is input; a second output step for outputting a second excitation signal to the servo system (20), said second excitation signal having a second frequency range that is a partial range of the first frequency range and being generated on the basis of the first frequency characteristics; and a second frequency characteristic calculation step for calculating a second frequency characteristic of the servo system (20) within a second frequency range on the basis of a second identification input signal and a second identification output signal acquired at a second sampling period equal to or greater than the first sampling period from the servo system (20) to which the second excitation signal is input.

Description

Technical Field

[0001] This disclosure relates to a method for measuring the frequency characteristics of a servo system. Background Technology

[0002] Previously, techniques for measuring the frequency characteristics of a servo system were known (for example, see Patent Document 1).

[0003] Prior art literature

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2019-91141 Summary of the Invention

[0006] The problem that the invention aims to solve

[0007] The goal is to measure the frequency characteristics of the servo system with high precision.

[0008] Therefore, the purpose of this disclosure is to provide a method for measuring frequency characteristics that can obtain high-precision frequency characteristics of a servo system.

[0009] Methods for solving problems

[0010] One aspect of this disclosure relates to a frequency characteristic measurement method for measuring the frequency characteristics of a servo system. The frequency characteristic measurement method includes: a first output step of outputting a first excitation signal of a first frequency range to the servo system; and a first frequency characteristic calculation step of calculating a first frequency characteristic of the servo system within the first frequency range based on a first identification input signal input to the servo system for obtaining the frequency characteristics of the servo system, and a first identification output signal output from the servo system for obtaining the frequency characteristics of the servo system, wherein the first identification input signal and the first identification output signal are signals acquired from the servo system to which the first excitation signal is input during a first sampling period. The second output step involves outputting a second excitation signal, generated based on the first frequency characteristic, to the servo system within a second frequency range that is part of the first frequency range; and the second frequency characteristic calculation step involves calculating the second frequency characteristic of the servo system within the second frequency range based on a second identification input signal input to the servo system for obtaining the frequency characteristic of the servo system, and a second identification output signal output from the servo system for obtaining the frequency characteristic of the servo system, wherein the second identification input signal and the second identification output signal are signals acquired from the servo system to which the second excitation signal is input for a second sampling period greater than or equal to the first sampling period.

[0011] One aspect of this disclosure relates to a frequency characteristic measuring device for measuring the frequency characteristics of a servo system. The frequency characteristic measuring device includes: a first output unit that outputs a first excitation signal within a first frequency range to the servo system; and a frequency characteristic calculation unit that calculates a first frequency characteristic of the servo system within the first frequency range based on a first identification input signal input to the servo system for obtaining the frequency characteristics of the servo system, and a first identification output signal output from the servo system for obtaining the frequency characteristics of the servo system, wherein the first identification input signal and the first identification output signal are signals acquired from the servo system to which the first excitation signal is input during a first sampling period; and a first... The output unit outputs a second excitation signal, generated based on the first frequency characteristic, to the servo system, which is a second frequency range that is part of the first frequency range. The frequency characteristic calculation unit further calculates the second frequency characteristic of the servo system within the second frequency range based on a second identification input signal input to the servo system for obtaining the frequency characteristic of the servo system and a second identification output signal output from the servo system for obtaining the frequency characteristic of the servo system. The second identification input signal and the second identification output signal are signals acquired from the servo system to which the second excitation signal is input for a second sampling period of more than the first sampling period.

[0012] Invention Effects

[0013] According to the frequency characteristic measurement method and the like disclosed herein, high-precision frequency characteristics of a servo system can be obtained. Attached Figure Description

[0014] Figure 1 This is a block diagram showing the structure of the frequency characteristic measurement system according to Embodiment 1.

[0015] Figure 2A This is a waveform diagram illustrating an example of a white noise signal.

[0016] Figure 2B This is a waveform diagram illustrating an example of a multiple sinusoidal signal.

[0017] Figure 2C This is a waveform diagram illustrating an example of a sinusoidal sweep signal.

[0018] Figure 2D This is a waveform diagram showing an example of a step signal.

[0019] Figure 2E This is a waveform diagram illustrating an example of a ramp signal.

[0020] Figure 3This is a Bode plot showing an example of the first frequency characteristic calculated by the frequency characteristic calculation unit according to Embodiment 1.

[0021] Figure 4 This is a Bode plot showing an example of the frequency characteristics calculated by the frequency characteristic calculation unit according to Embodiment 1.

[0022] Figure 5 This is an example of a Bode plot that superimposes the first frequency characteristic calculated by the frequency characteristic calculation unit according to Embodiment 1 and the frequency characteristic of the servo system of an approximate rigid body.

[0023] Figure 6 This is a Bode plot showing an example of the second frequency characteristic calculated by the frequency characteristic calculation unit according to Embodiment 1.

[0024] Figure 7 This is a flowchart of the first frequency characteristic measurement process involved in Implementation Method 1.

[0025] Figure 8 This is a diagram used to explain why the accuracy of the second frequency characteristic is higher than that of the first frequency characteristic if the second frequency range is narrower than the first frequency range.

[0026] Figure 9 This is a block diagram showing the structure of the frequency characteristic measurement system according to Embodiment 2.

[0027] Figure 10 This is a Bode plot showing an example of the third frequency characteristic calculated by the frequency characteristic calculation unit according to Embodiment 2.

[0028] Figure 11 This is a Bode plot showing an example of the fourth frequency characteristic calculated by the frequency synthesis unit according to Embodiment 2.

[0029] Figure 12 This is a flowchart of the second frequency characteristic measurement process involved in Implementation Method 2.

[0030] Figure 13 This is an example of a Bode plot that superimposes the first frequency characteristic calculated by the frequency characteristic calculation unit according to Embodiment 2 and the frequency characteristic of the servo system of an approximate rigid body.

[0031] Figure 14 This is a block diagram showing the structure of the frequency characteristic measurement system involved in Modification Example 1.

[0032] Figure 15 This is a block diagram showing the structure of the frequency characteristic measurement system involved in Modification Example 2.

[0033] Figure 16 This is a block diagram showing the structure of the frequency response measurement system involved in Modification Example 3. Detailed Implementation

[0034] (The process by which this disclosure was obtained)

[0035] When measuring the frequency characteristics of a servo system by inputting an excitation signal containing frequency components of the frequency band of the object being measured, it is necessary to pre-set conditions such as the sampling period, frequency resolution, and frequency range.

[0036] If these conditions are not set properly, it may be impossible to accurately determine the resonant frequency and / or anti-resonant frequency, or to grasp the frequency characteristics in these frequency bands with fine resolution.

[0037] In addition, when measuring frequency characteristics, the amplitude of the excitation signal needs to be preset.

[0038] In servo systems with high inertia or high friction, if the amplitude of the excitation signal is small, the signal strength of the excitation signal will be insufficient, and sometimes it will be impossible to measure the frequency characteristics with high precision.

[0039] Furthermore, in servo systems containing anti-resonance, the excitation signal is attenuated in the anti-resonance frequency band with a relatively low gain, resulting in insufficient signal strength of the excitation signal, which sometimes makes it impossible to measure the frequency characteristics with high accuracy.

[0040] Conversely, in servo systems that incorporate resonance, when the amplitude of the excitation signal is set relatively large in order to measure frequency characteristics with high accuracy, the excitation signal is amplified in the resonant frequency band with relatively high gain, which can sometimes damage the components of the servo system.

[0041] In particular, compared to using an excitation signal with all frequency components, such as a pseudo-random signal, or an excitation signal with a wide frequency range set as the target frequency range, such as a sine wave sweep signal, to measure the frequency characteristics of a wide frequency range at once, when using an excitation signal with only frequency components of a specific frequency range to measure the frequency characteristics of a specific frequency range, the signal strength of the frequency components in the resonant frequency band is greater, thus increasing the likelihood of damage to the device constituting the servo system.

[0042] In order to obtain high-precision frequency characteristics, the amplitude of the excitation signal needs to be set to be relatively large to measure the frequency characteristics. However, on the other hand, if the amplitude of the excitation signal is set to be relatively large, the risk of damage to the devices constituting the servo system increases.

[0043] Therefore, in order to obtain high-precision frequency characteristics, it is necessary to maximize the amplitude of the excitation signal as much as possible without damaging the components of the servo system. However, setting such an amplitude is not easy.

[0044] Therefore, the inventors have conducted in-depth research and experimentation on a frequency characteristic measurement method that can obtain high-precision frequency characteristics of a servo system even when the servo system, which is the object of frequency characteristic measurement, has large inertia or friction, or when the servo system contains resonance or anti-resonance.

[0045] As a result, the inventors obtained the following knowledge: Initially, the amplitude of the excitation signal is automatically calculated based on the total inertia or friction characteristics of the servo system. An excitation signal containing the frequency components of the entire frequency band of the measurement object is input, and the frequency characteristics of the servo system are measured, thereby obtaining an overall understanding of the servo system's frequency characteristics. Next, based on the obtained overall understanding of the servo system's frequency characteristics, for a portion of the measured frequency band, a new excitation signal containing more frequency components of that band is input at a sampling period longer than the initial measurement, and the frequency characteristics of the servo system are measured. Thus, the inventors obtained the following knowledge: even when the inertia or friction of the servo system, the object of frequency characteristic measurement, is relatively large, or when resonance or anti-resonance is involved, a high-precision frequency characteristic of the servo system can be obtained based on the initially measured frequency characteristics and the subsequently measured frequency characteristics.

[0046] Moreover, based on this knowledge, the inventors conducted further research and experiments, and came up with the frequency characteristic measuring device disclosed herein.

[0047] One aspect of this disclosure relates to a frequency characteristic measurement method for measuring the frequency characteristics of a servo system. The frequency characteristic measurement method includes: a first output step of outputting a first excitation signal of a first frequency range to the servo system; and a first frequency characteristic calculation step of calculating a first frequency characteristic of the servo system within the first frequency range based on a first identification input signal input to the servo system for obtaining the frequency characteristics of the servo system, and a first identification output signal output from the servo system for obtaining the frequency characteristics of the servo system, wherein the first identification input signal and the first identification output signal are signals acquired from the servo system to which the first excitation signal is input during a first sampling period. The second output step involves outputting a second excitation signal, generated based on the first frequency characteristic, to the servo system within a second frequency range that is part of the first frequency range; and the second frequency characteristic calculation step involves calculating the second frequency characteristic of the servo system within the second frequency range based on a second identification input signal input to the servo system for obtaining the frequency characteristic of the servo system, and a second identification output signal output from the servo system for obtaining the frequency characteristic of the servo system, wherein the second identification input signal and the second identification output signal are signals acquired from the servo system to which the second excitation signal is input for a second sampling period greater than or equal to the first sampling period.

[0048] According to the frequency characteristic measurement method described above, the second frequency range of the second excitation signal is narrower than the first frequency range of the first excitation signal. Therefore, when the energy supplied to the servo system through the first and second excitation signals is the same, the signal strength of the second excitation signal is stronger than that of the first excitation signal within the second frequency range.

[0049] Therefore, based on the frequency characteristic measurement method described above, a second frequency characteristic with higher accuracy than the first frequency characteristic can be obtained in the second frequency range.

[0050] Furthermore, according to the frequency characteristic measurement method described above, the second sampling period is greater than or equal to the first sampling period.

[0051] Generally, the frequency resolution in a frequency response is the value obtained by dividing the sampling frequency (which is the reciprocal of the sampling period) by the number of measured data points. Therefore, when the number of measured data points is the same when calculating the first frequency response and the second frequency response, the fineness of the frequency resolution in the second frequency response is greater than that in the first frequency response.

[0052] Therefore, based on the frequency characteristic measurement method described above, a second frequency characteristic with higher accuracy than the first frequency characteristic can be obtained in the second frequency range.

[0053] In this way, based on the frequency characteristic measurement method described above, high-precision frequency characteristics of the servo system can be obtained.

[0054] Alternatively, the second sampling period can be set to be greater than the first sampling period.

[0055] Alternatively, the first excitation signal can be any one of a pseudo-random signal, a multiple sine wave signal, a sine wave sweep signal, a step signal, and a ramp signal.

[0056] Alternatively, the second excitation signal can be set to be either a multiple sine wave signal or a sine wave sweep signal.

[0057] By setting the second excitation signal to a signal capable of specifying a frequency range, it is reliably possible to make the second frequency range narrower than the first frequency range. Therefore, when the energy supplied to the servo system via the first and second excitation signals is the same, it is reliably possible to make the signal strength of the second excitation signal stronger than the signal strength of the first excitation signal within the second frequency range.

[0058] Alternatively, it may also include: a first measurement condition calculation step, which calculates the amplitude of the first excitation signal based on at least one of the total inertia and friction characteristics of the servo system, and in the first output step, outputs the first excitation signal having the amplitude of the first excitation signal calculated by the first measurement condition calculation step to the servo system.

[0059] Therefore, a first excitation signal with a more appropriate amplitude corresponding to at least one of the total inertia and friction characteristics of the servo system can be output to the servo system.

[0060] Therefore, according to the frequency characteristic measurement method described above, it is possible to suppress the decrease in frequency characteristic accuracy caused by insufficient signal strength of the first excitation signal within the first frequency range, and / or suppress the risk of damage to the device constituting the servo system caused by excessive signal strength of the first excitation signal.

[0061] Alternatively, it may also include: a second measurement condition calculation step, which calculates the amplitude of the second excitation signal based on the first frequency characteristic, and in the second output step, outputs the second excitation signal having the amplitude of the second excitation signal calculated by the second measurement condition calculation step to the servo system.

[0062] Therefore, a second excitation signal with a more appropriate amplitude corresponding to the first frequency characteristic can be output to the servo system.

[0063] Therefore, based on the frequency characteristic measurement method described above, it is possible to suppress the risk of damage to the devices constituting the servo system within the second frequency range and obtain a second frequency characteristic with higher accuracy.

[0064] Furthermore, it may also include: a third output step, which outputs a third excitation signal generated based on the first frequency characteristic to the servo system in a third frequency range that is part of the first frequency range, the third frequency range being a range of the first frequency higher than the lower limit frequency of the second frequency range and the second frequency higher than the upper limit frequency of the second frequency range; and a third frequency characteristic calculation step, which calculates the third frequency characteristic of the servo system in the third frequency range based on a third identification input signal input to the servo system for obtaining the frequency characteristic of the servo system and a third identification output signal output from the servo system for obtaining the frequency characteristic of the servo system, the third identification input signal and the third identification output signal being signals obtained from the servo system to which the third excitation signal is input at a third sampling period greater than the first sampling period.

[0065] According to the frequency characteristic measurement method described above, the third frequency range of the third excitation signal is narrower than the first frequency range of the first excitation signal. Therefore, when the energy supplied to the servo system through the first and third excitation signals is the same, the signal strength of the third excitation signal is stronger than that of the first excitation signal within the third frequency range.

[0066] Therefore, based on the frequency characteristic measurement method described above, a third frequency characteristic with higher accuracy than the first frequency characteristic can be obtained in the third frequency range.

[0067] Furthermore, according to the frequency characteristic measurement method described above, the third sampling period is above the first sampling period.

[0068] Generally, the frequency resolution in a frequency response is the value obtained by dividing the sampling frequency (which is the reciprocal of the sampling period) by the number of measured data points. Therefore, when the number of measured data points is the same when calculating the first frequency response and the third frequency response, the fineness of the frequency resolution in the third frequency response is greater than that in the first frequency response.

[0069] Therefore, based on the frequency characteristic measurement method described above, a third frequency characteristic with higher accuracy than the first frequency characteristic can be obtained in the third frequency range.

[0070] In addition, it can also be configured to include: a frequency characteristic synthesis step, which synthesizes the second frequency characteristic and the third frequency characteristic to calculate the fourth frequency characteristic of the servo system.

[0071] Therefore, it is possible to obtain frequency characteristics with higher accuracy than the first frequency characteristics in a fourth frequency range, which is wider than the second and third frequency ranges.

[0072] Alternatively, the third sampling period can be set to be greater than the first sampling period.

[0073] Alternatively, the second frequency range and the third frequency range can be configured to be continuous.

[0074] Therefore, since there is no overlap between the second and third frequency ranges, a higher precision frequency characteristic can be obtained in a wider fourth frequency range compared to the case where the second and third frequency ranges overlap.

[0075] Alternatively, the lower limit frequency of the third frequency range can be set within the second frequency range.

[0076] Alternatively, it can be configured such that at least one of the second frequency range and the third frequency range includes at least one of the resonant frequency and the anti-resonant frequency of the servo system.

[0077] Therefore, it is possible to obtain frequency characteristics with higher accuracy than the first frequency characteristic within a frequency range that includes at least one of the resonant frequency and the anti-resonant frequency.

[0078] In addition, it can also be configured to include: a fourth output step, which outputs at least one of the second frequency characteristic, the third frequency characteristic, and the fourth frequency characteristic to a display device.

[0079] Therefore, users utilizing the above-described frequency characteristic measurement method can obtain information about at least one of the second, third, and fourth frequency characteristics through a visual confirmation display device.

[0080] Alternatively, it can also be configured to include: a first measurement condition calculation step, calculating the amplitude of the first excitation signal based on at least one of the total inertia and friction characteristics of the servo system and the control parameter settings of the servo system; and a second measurement condition calculation step, calculating the amplitude of the second excitation signal based on the first frequency characteristic and the control parameter settings; in the first output step, outputting the first excitation signal having the amplitude of the first excitation signal calculated by the first measurement condition calculation step to the servo system; and in the second output step, outputting the second excitation signal having the amplitude of the second excitation signal calculated by the second measurement condition calculation step to the servo system.

[0081] Therefore, a first excitation signal with a more appropriate amplitude corresponding to at least one of the total inertia and frictional characteristics of the servo system and the control parameters of the servo system can be output to the servo system.

[0082] Therefore, according to the frequency characteristic measurement method described above, it is possible to suppress the decrease in frequency characteristic accuracy caused by insufficient signal strength of the first excitation signal within the first frequency range, and / or suppress the risk of damage to the device constituting the servo system caused by excessive signal strength of the first excitation signal.

[0083] Furthermore, a second excitation signal with a more appropriate amplitude, corresponding to the first frequency characteristics and the control parameters of the servo system, can be output to the servo system.

[0084] Therefore, based on the frequency characteristic measurement method described above, it is possible to further suppress the risk of damage to the devices constituting the servo system in the second frequency range and obtain a second frequency characteristic with higher accuracy.

[0085] Alternatively, the first identification output signal and the second identification output signal may be configured to include sensor information representing the state of the driven body obtained by sensing the driven body in the servo system.

[0086] Alternatively, the output may be configured such that, in the first output step, the first excitation signal is output multiple times; in the second output step, the second excitation signal is output multiple times; in the first frequency characteristic calculation step, the first frequency characteristic is calculated based on the first identification input signal and the first identification output signal obtained multiple times from the servo system that has received the first excitation signal multiple times; and in the second frequency characteristic calculation step, the second frequency characteristic is calculated based on the second identification input signal and the second identification output signal obtained multiple times from the servo system that has received the second excitation signal multiple times.

[0087] This allows for the acquisition of higher precision frequency characteristics for the servo system.

[0088] Alternatively, the first identification input signal and the second identification input signal may be configured to have the same number of samples.

[0089] Therefore, the storage area for storing the first identification input signal and the storage area for storing the second identification signal can be used as a shared storage area, so there is no need to ensure an additional storage area for storing the identification input signal.

[0090] One aspect of this disclosure relates to a frequency characteristic measuring device for measuring the frequency characteristics of a servo system. The frequency characteristic measuring device includes: a first output unit that outputs a first excitation signal within a first frequency range to the servo system; and a frequency characteristic calculation unit that calculates a first frequency characteristic of the servo system within the first frequency range based on a first identification input signal input to the servo system for obtaining the frequency characteristics of the servo system, and a first identification output signal output from the servo system for obtaining the frequency characteristics of the servo system, wherein the first identification input signal and the first identification output signal are signals acquired from the servo system to which the first excitation signal is input during a first sampling period; and a first... The output unit outputs a second excitation signal, generated based on the first frequency characteristic, to the servo system, which is a second frequency range that is part of the first frequency range. The frequency characteristic calculation unit further calculates the second frequency characteristic of the servo system within the second frequency range based on a second identification input signal input to the servo system for obtaining the frequency characteristic of the servo system and a second identification output signal output from the servo system for obtaining the frequency characteristic of the servo system. The second identification input signal and the second identification output signal are signals acquired from the servo system to which the second excitation signal is input for a second sampling period of more than the first sampling period.

[0091] According to the frequency characteristic measuring device described above, the second frequency range of the second excitation signal is narrower than the first frequency range of the first excitation signal. Therefore, when the energy supplied to the servo system through the first and second excitation signals is the same, the signal strength of the second excitation signal is stronger than that of the first excitation signal within the second frequency range.

[0092] Therefore, according to the frequency characteristic measuring device described above, a second frequency characteristic with higher accuracy than the first frequency characteristic can be obtained in the second frequency range.

[0093] Furthermore, according to the frequency characteristic measuring device described above, the second sampling period is greater than or equal to the first sampling period.

[0094] Generally, the frequency resolution in a frequency response is the value obtained by dividing the sampling frequency (which is the reciprocal of the sampling period) by the number of measured data points. Therefore, when the number of measured data points is the same when calculating the first frequency response and the second frequency response, the fineness of the frequency resolution in the second frequency response is greater than that in the first frequency response.

[0095] Therefore, according to the frequency characteristic measuring device described above, a second frequency characteristic with higher accuracy than the first frequency characteristic can be obtained in the second frequency range.

[0096] In this way, the frequency characteristics of the servo system can be obtained with high precision using the frequency characteristic measuring device described above.

[0097] The following description, with reference to the accompanying drawings, details a specific example of a frequency response measurement system according to one aspect of this disclosure. The embodiments shown herein are all examples of this disclosure. Therefore, the numerical values, shapes, constituent elements, configurations and connection methods of the constituent elements, and steps (processes) and their order shown in the following embodiments are merely examples and are not intended to limit this disclosure. Furthermore, the figures are schematic diagrams and are not necessarily strictly illustrated. In each figure, structures substantially the same as those in other figures are labeled with the same reference numerals, and repeated descriptions are omitted or simplified.

[0098] (Implementation Method 1)

[0099] The frequency response measurement system according to Embodiment 1 will be described below. This frequency response measurement system is a system for measuring the frequency response of a servo system.

[0100] <Structure>

[0101] Figure 1 This is a block diagram showing the structure of the frequency characteristic measurement system 1 according to Embodiment 1.

[0102] like Figure 1 As shown, the frequency characteristic measurement system 1 includes a frequency characteristic measurement device 10 and a servo system 20.

[0103] The servo system 20 is the object whose frequency characteristics are measured by the frequency characteristic measuring device 10, and includes a motor control device 21, a motor 22, a load 23, and a first state quantity measuring device 24.

[0104] Motor 22 is controlled by motor control device 21 and drives load 23. Here, motor 22 is described as a rotary motor. However, motor 22 is not necessarily limited to a rotary motor; for example, it could also be a linear motor.

[0105] The load 23 is the driven body in the servo system 20 and is driven by the motor 22.

[0106] The first state quantity measuring device 24 measures the state quantity of the servo system 20 from the servo system 20 and outputs a state signal representing the measured state quantity of the servo system 20. Here, it will be described as the first state quantity measuring device 24 outputting the state quantity related to the motor 22 as a state signal. More specifically, it will be described as an encoder that measures the rotational position of the motor 22 and outputs a state signal representing the rotational position of the motor 22. However, the first state quantity measuring device 24 is only necessary if it is a device capable of measuring the state quantity of the servo system 20 and outputting a state signal representing the measured state quantity of the servo system 20; it is not necessarily limited to the encoder described above. Alternatively, the first state quantity measuring device 24 may be, for example, a sensor that measures the rotational speed of the motor 22 and outputs a state signal representing the rotational speed of the motor 22.

[0107] The motor control device 21 performs control so that the input state quantity command and the motor state quantity are consistent. In this embodiment 1, the motor 22 is controlled based on the excitation signal (the first excitation signal and the second excitation signal described later) output from the frequency characteristic measuring device 10. Here, it is explained that the excitation signal output from the frequency characteristic measuring device 10 is a torque command (hereinafter also referred to as "input torque command") indicating the torque of the motor 22. Furthermore, it is explained that if an input torque command output from the frequency characteristic measuring device 10 is input, the motor control device 21 uses the first state signal output from the first state quantity measuring device 24 to output a feedback torque command that makes the input torque command consistent with the torque generated by the motor to the motor 22. However, the motor control device 21 does not necessarily need to be limited to the above structure as long as it can control the motor 22 based on the excitation signal output from the frequency characteristic measuring device 10. For example, it may be possible to output the excitation signal output from the frequency characteristic measuring device 10 directly to the motor 22 without feedback control.

[0108] Furthermore, this description assumes that the excitation signal output from the frequency characteristic measuring device 10 is an input torque command. However, the excitation signal output from the frequency characteristic measuring device 10 does not necessarily have to be limited to an input torque command. For example, it could also be an input speed command indicating the speed of the motor 22, or an input position command indicating the position of the motor 22.

[0109] The frequency response measuring device 10 includes a first output unit 11, a second output unit 12, a frequency response calculation unit 13, a frequency response output unit 15, and an excitation signal selection unit 16, and measures the frequency response of the servo system 20.

[0110] Alternatively, the frequency characteristic measuring device 10 can be implemented, for example, by the processor executing a program stored in the memory in a computer device having a processor, memory and input / output interface.

[0111] The first output unit 11 has a first measurement condition calculation unit 111 and a first excitation signal generation unit 112, and outputs a first excitation signal in a first frequency range.

[0112] The first measurement condition calculation unit 111 is input with total inertia information, which represents the sum of the inertia of the motor 22 and the load 23, i.e., the total inertia of the servo system 20. Based on the total inertia represented by the input total inertia information, the amplitude of the first excitation signal is calculated.

[0113] The input of total inertia information to the first measurement condition calculation unit 111 is achieved, for example, by the user of the frequency characteristic measuring device 10 inputting total inertia information, which represents the total inertia calculated based on the design information of the servo system 20, to the first measurement condition calculation unit 111 via the input / output interface provided by the frequency characteristic measuring device 10.

[0114] When outputting an excitation signal to the servo system 20 for excitation, the greater the total inertia of the servo system 20, the larger the amplitude of the excitation signal is required. Therefore, it can also be configured such that the first measurement condition calculation unit 111 calculates the amplitude of the first excitation signal, for example, by multiplying the total inertia represented by the total inertia information by a given coefficient.

[0115] In addition, here, it is explained that the first measurement condition calculation unit 111 is input with total inertia information and the amplitude of the first excitation signal is calculated based on the total inertia expressed by the total inertia information. However, for example, it may also be explained that the first measurement condition calculation unit 111 is input with friction characteristic information representing the friction characteristics of the servo system 20 and the amplitude of the first excitation signal is calculated based on the friction characteristics expressed by the friction characteristic information. For example, it may also be explained that the first measurement condition calculation unit 111 is input with total inertia information and friction characteristic information and the amplitude of the first excitation signal is calculated based on the total inertia expressed by the total inertia information and the friction characteristic information expressed by the friction characteristic information.

[0116] For example, if the amplitude of the first excitation signal is small compared to the frictional torque of the load 23, the load 23 will not operate sufficiently even if the first excitation signal is output to the servo system 20, thus making it impossible to accurately measure the frequency characteristics of the servo system 20. Therefore, by calculating the amplitude of the first excitation signal based on the frictional characteristics represented by frictional characteristic information, the frequency characteristics of the servo system 20 can be accurately measured.

[0117] Alternatively, when frictional characteristic information representing the frictional torque of the load 23 is input to the first measurement condition calculation unit 111, the first measurement condition calculation unit 111 calculates the amplitude of the first excitation signal, for example, by multiplying the frictional torque represented by the frictional characteristic information by a given coefficient.

[0118] Alternatively, when the first measurement condition calculation unit 111 is input with total inertia information representing the total inertia of the servo system 20 and friction characteristic information representing the friction torque of the load 23, the first measurement condition calculation unit 111 may calculate, for example, the sum of the value obtained by multiplying the total inertia represented by the total inertia information by a given coefficient and the value obtained by multiplying the friction torque represented by the friction characteristic information by a given coefficient, as the amplitude of the first excitation signal.

[0119] The first excitation signal generation unit 112 generates a first excitation signal having the amplitude of the first excitation signal calculated by the first measurement condition calculation unit 111, and outputs the generated first excitation signal to the excitation signal selection unit 16. Here, the first frequency range is described as the entire frequency range of the frequency characteristics measured by the frequency characteristic measuring device 10. Furthermore, as mentioned above, the first excitation signal is described as a torque command (first torque command) indicating the torque of the motor 22.

[0120] The first excitation signal output by the first excitation signal generation unit 112 is any one of white noise signal, multiple sine wave signal, sine wave sweep frequency signal, step signal and ramp signal.

[0121] Figure 2A This is a waveform diagram illustrating an example of a white noise signal. Figure 2B This is a waveform diagram illustrating an example of a multiple sinusoidal signal. Figure 2C This is a waveform diagram illustrating an example of a sinusoidal sweep frequency signal. Figure 2D This is a waveform diagram illustrating an example of a step signal. Figure 2E This is a waveform diagram illustrating an example of a ramp signal.

[0122] exist Figures 2A-2E In the diagram, the horizontal axis represents time, and the vertical axis represents torque.

[0123] Figure 2A The illustrated white noise signal is a type of pseudo-random signal that roughly uniformly contains all frequency components within a frequency band less than half the sampling period. White noise signals have the following characteristics: although noise exists in the high-frequency band, they contain frequency components with wide amplitude and are easy to generate.

[0124] exist Figure 2B The illustrated multiple sinusoidal signal has the following characteristics: although a frequency range needs to be set and the signal generation is relatively complex, a strong signal strength can be obtained within the set frequency range, and it has a very uniform spectrum throughout the set frequency range.

[0125] exist Figure 2CThe illustrated sinusoidal sweep signal has the following characteristics: although a frequency range needs to be set, a strong signal strength can be obtained within the set frequency range, and a uniform spectrum is present in most of the set frequency range.

[0126] exist Figure 2D The illustrated step signal has the following characteristics: although the signal strength weakens at high frequencies, it is easy to generate the signal.

[0127] exist Figure 2E The ramp signal illustrated has the following characteristics: although the signal strength weakens at high frequencies, it is easy to generate the signal.

[0128] Here, we will explain this by assuming that the first excitation signal output by the first output unit 11 is a white noise signal.

[0129] Back to Figure 1 Continuing with the explanation of frequency response measurement system 1.

[0130] The excitation signal selection unit 16 selects either the first excitation signal output from the first output unit 11 or the second excitation signal output from the second output unit 12 (described later), and outputs the selected first excitation signal or the second excitation signal as an excitation signal to the servo system 20.

[0131] The frequency response calculation unit 13 calculates the frequency response of the servo system 20 based on the identification input signal input to the servo system 20 for obtaining the frequency response of the servo system 20, and the identification output signal output from the servo system 20 for obtaining the frequency response of the servo system 20. At this time, the frequency response calculation unit 13 samples to acquire the identification input signal and the identification output signal.

[0132] Here, the identification input signal can be, for example, the excitation signal output from the frequency response measuring device 10 itself, or a control signal generated in the motor control device 21 based on the excitation signal output from the frequency response measuring device 10 for driving the motor 22. Here, the control signal generated in the motor control device 21 based on the excitation signal output from the frequency response measuring device 10 for driving the motor 22 can be, for example, a feedback torque command. Here, we will assume that the identification input signal is the excitation signal input to the servo system 20.

[0133] Furthermore, the identification output signal may, for example, be a state signal output from the first state quantity measuring device 24. Here, we will explain this as if the identification output signal were a state signal output from the first state quantity measuring device 24.

[0134] That is, (1) when a first excitation signal in a first frequency range output from the frequency characteristic measuring device 10 is input to the servo system 20, the frequency characteristic calculation unit 13 calculates the first frequency characteristic of the servo system 20 in the first frequency range based on the first excitation signal and the first status signal acquired at a first sampling period. Furthermore, (2) when a second excitation signal in a second frequency range output from the frequency characteristic measuring device 10 is input to the servo system 20, the frequency characteristic calculation unit 13 calculates the second frequency characteristic of the servo system 20 in the second frequency range based on the second excitation signal and the second status signal acquired at a second sampling period greater than or equal to the first sampling period.

[0135] More specifically, (1) when the first excitation signal is input to the servo system 20, the frequency characteristic calculation unit 13 calculates the first frequency characteristic of the servo system 20 within a first frequency range based on the first input torque command obtained from the motor control device 21 in the servo system 20 where the first excitation signal is input and the first state signal obtained from the first state quantity measuring device 24 in the servo system 20 where the first excitation signal is input and the first state signal obtained from the first state quantity measuring device 24 in the servo system 20 where the first excitation signal is input and the first state signal obtained from the first state quantity measuring device 24 in the servo system 20 where the first excitation signal is input and the first state signal obtained from the first state quantity measuring device 24 in the servo system 20 where the first excitation signal is input and the second excitation signal obtained from the motor control device 21 in the servo system 20 where the second excitation signal is input and the second state signal obtained from the first state quantity measuring device 24 ...

[0136] Here, the frequency characteristic calculated by the frequency characteristic calculation unit 13 is described as the frequency characteristic of the rotational speed [rpm] of the motor 22 relative to the input torque [Nm] input to the servo system 20.

[0137] Figure 3 This is a Bode plot showing an example of the first frequency characteristic calculated by the frequency characteristic calculation unit 13 when the frequency characteristic measuring device 10 outputs the first excitation signal.

[0138] exist Figure 3 In the figure, the horizontal axis represents the frequency, and the vertical axis represents the gain G shown in equation (1). Here, the responsiveness of the rotational speed to the input torque command at each frequency is shown as the gain.

[0139] [Mathematical Expression 1]

[0140]

[0141] exist Figure 3In the diagram, each black circle represents the gain at each frequency calculated based on equation (1), and lines connect these points.

[0142] exist Figure 3 Within the first frequency range, there exists a point where the gain variation with increasing frequency changes from increasing to decreasing; that is, a point where the gain variation with increasing frequency bulges upwards. The frequency at this point is the resonant frequency of the servo system 20. Furthermore, within the first frequency range, there exists a point where the gain variation with increasing frequency changes from decreasing to increasing; that is, a point where the gain variation with increasing frequency bulges downwards. The frequency at this point is the anti-resonant frequency of the servo system 20.

[0143] The resonance and anti-resonance of the servo system 20 are vibration characteristics that significantly affect the motion control of the motor 22 performed by the motor control device 21. Accurately determining the resonance and anti-resonance frequencies is extremely important in measuring the frequency characteristics of the servo system 20. Therefore, it is desirable to measure the frequency characteristics in the resonance and anti-resonance frequency bands with high precision.

[0144] Here, the frequency response calculation unit 13 is described as calculating the gain characteristic, which is one type of Bode plot, as the frequency response. However, the frequency response calculation unit 13 is not limited to a structure that uses the output gain characteristic as the frequency response. For example, it can be configured to calculate the phase characteristic, which is another type of Bode plot, as the frequency response, or it can be configured to calculate both the gain characteristic and the phase characteristic as the frequency response. Furthermore, the frequency response calculation unit 13 can also be configured to calculate the Nyquist plot instead of the Bode plot as the frequency response.

[0145] Back to Figure 1 Continuing with the explanation of frequency response measurement system 1.

[0146] like Figure 1 As shown, the frequency response calculation unit 13 can also be implemented, for example, by a structure having a Fourier transform unit 17 and a frequency response calculation unit 18. Here, the Fourier transform unit 17 performs a Fourier transform on the input torque command output from the motor control device 21 and the rotational speed of the motor 22 obtained by time differentiation of the state signal representing the rotational position of the motor 22 output from the first state quantity measuring device 24. Furthermore, the frequency response calculation unit 18 calculates the gain of the rotational speed relative to the input torque command for each frequency component by substituting the input torque command for each frequency component and the rotational speed for each frequency component obtained by the Fourier transform unit 17 into equation (1).

[0147] The second output unit 12 has a second measurement condition calculation unit 121 and a second excitation signal generation unit 122, and outputs a second excitation signal of a second frequency range that is part of the first frequency range.

[0148] The second excitation signal can, for example, be used as an excitation signal for more refined and accurate re-measurement of the frequency characteristics near the resonant frequency or anti-resonant frequency in the servo system 20.

[0149] Here, using Figure 4 To explain in Figure 3 The accuracy of the frequency characteristics in the data may become insufficient, as they are based on the frequency characteristics near the resonant frequency or the anti-resonant frequency.

[0150] Figure 4 This is a Bode plot showing an example of the frequency characteristics calculated by the frequency characteristic calculation unit 13 when the frequency characteristic measuring device 10 outputs an excitation signal.

[0151] exist Figure 4 In the diagram, the spacing between the circles with diagonal lines and the spacing between the black circles represent the frequency resolution in the frequency response calculation.

[0152] Here, each circle with a diagonal line and Figure 3 The same applies to each black circle. On the other hand, each black circle is an example of the frequency response calculated by the frequency response calculation unit 13 when the frequency resolution is finer compared to each circle with diagonal lines.

[0153] It can be seen that, in Figure 4 In each of the black circles, the point where the gain changes and bulges downwards represents the change in anti-resonant frequency from the anti-resonant frequency in the circles with diagonal lines. This is because the frequency resolution in the frequency response shown by the diagonal-lined circles differs from that shown by the black circles, and the frequency resolution shown by the diagonal-lined circles is coarser than that shown by the black circles. Therefore, the anti-resonant frequency in the frequency response shown by the diagonal-lined circles is less precise compared to that shown by the black circles. In other words, in this case, to calculate the resonant and anti-resonant frequencies with high accuracy, the frequency resolution needs to be refined.

[0154] The second excitation signal is used to obtain frequency characteristics with a finer frequency resolution than the first excitation signal. Frequency resolution is generally defined as the value obtained by dividing the sampling frequency (which is the reciprocal of the sampling period) by the number of measurement data points. The number of data points is limited by the capacity of the memory storing the measured data; therefore, to obtain frequency characteristics with finer frequency resolution, the second excitation signal needs to be generated with a sampling period greater than that of the first excitation signal. Furthermore, here, the measured frequency band is narrowed to further improve signal strength; that is, the second frequency range is set as a portion of the first frequency range. Moreover, setting it to the second frequency range is predetermined.

[0155] Back to Figure 1 Continuing with the explanation of frequency response measurement system 1.

[0156] The second measurement condition calculation unit 121 receives the first frequency characteristic calculated by the frequency characteristic calculation unit 13, and calculates the amplitude of the second excitation signal based on the input first frequency characteristic. Alternatively, it can also be configured to input total inertia information to the second measurement condition calculation unit 121.

[0157] An example of the method for calculating the amplitude of the second excitation signal using the second measurement condition calculation unit 121 is as follows: Figure 5 To illustrate.

[0158] Figure 5 It is Figure 3 The Bode plot shown is an overlap of the first frequency characteristic and the frequency characteristic of the approximately rigid servo system 20 when the servo system 20 is approximately a rigid body.

[0159] Here, the so-called frequency characteristics of the near-rigid body servo system 20 refer to the frequency characteristics calculated solely based on the inertia of the motor 22 and the load 23, assuming that the servo system 20 is an ideal servo system without resonance or anti-resonance.

[0160] The frequency characteristics of the approximate rigid body servo system 20 can be calculated based on the total inertia represented by the total inertia information.

[0161] The second measurement condition calculation unit 121 compares the first frequency characteristic within the second frequency range with the frequency characteristic of the near-rigid servo system 20. For the frequency band where the gain of the first frequency characteristic is less than the gain of the near-rigid servo system 20, it calculates the amplitude of the second excitation signal, making the amplitude of the second excitation signal greater than the amplitude of the first excitation signal. For the frequency band where the gain of the second frequency characteristic is greater than the gain of the near-rigid servo system 20, it calculates the amplitude of the second excitation signal, making the amplitude of the second excitation signal less than the amplitude of the first excitation signal.

[0162] Alternatively, the second measurement condition calculation unit 121 may calculate the amplitude of the second excitation signal, for example, by multiplying the amplitude of the first excitation signal by the ratio of the gain of the first frequency characteristic to the gain of the servo system 20, which is approximately a rigid body.

[0163] The second excitation signal generation unit 122 generates a second excitation signal having the amplitude of the second excitation signal calculated by the second measurement condition calculation unit 121, and outputs the generated second excitation signal to the excitation signal selection unit 16. Here, it will be described as the second excitation signal being a torque command (second torque command) indicating the torque of the motor 22, just like the first excitation signal.

[0164] The second excitation signal output by the second excitation signal generation unit 122 is either a multiple sine wave signal or a sine wave sweep frequency signal.

[0165] As mentioned earlier, these two signals are signals whose frequency ranges can be set. Therefore, by narrowing the frequency range relative to the white noise signal used for the first excitation signal, even if the amplitudes of the first and second excitation signals are the same, the signal strength of the second excitation signal is higher than that of the first excitation signal within the second frequency range. Therefore, a second frequency characteristic with higher accuracy than the first frequency characteristic can be obtained within the second frequency range.

[0166] Here, we will explain the second excitation signal output by the second output unit 12 as a multiple sine wave signal. The reason for using a multiple sine wave signal as the second excitation signal is that, since a multiple sine wave signal has a very uniform spectrum across the entire specified frequency range compared to a sine wave sweep signal, it meets the objective of measuring frequency characteristics with higher accuracy within the specified frequency range.

[0167] Back to Figure 1 Continuing with the explanation of frequency response measurement system 1.

[0168] If a second excitation signal is input from the second output unit 12, the excitation signal selection unit 16 outputs the second excitation signal to the servo system 20.

[0169] As mentioned above, when the second excitation signal is input to the servo system 20, the frequency characteristic calculation unit 13 calculates the second frequency characteristic.

[0170] Figure 6 This is a Bode plot showing an example of the second frequency characteristic calculated by the frequency characteristic calculation unit 13 when the frequency characteristic measuring device 10 outputs the second excitation signal.

[0171] exist Figure 6In the diagram, the horizontal axis represents frequency, and the vertical axis represents the gain G shown in equation (1). Here, in... Figure 6 In the diagram, the dashed line represents an example of the first frequency characteristic calculated by the frequency characteristic calculation unit 13 when the frequency characteristic measuring device 10 outputs the first excitation signal. Figure 3 The same applies to the first frequency characteristic shown.

[0172] like Figure 6 As shown, the frequency response calculation unit 13 calculates the second frequency response with a frequency resolution that is finer than the frequency resolution used when calculating the first frequency response.

[0173] Back to Figure 1 Continuing with the explanation of frequency response measurement system 1.

[0174] The frequency response output unit 15 outputs the first frequency response, the second frequency response, or both the first frequency response and the second frequency response calculated by the frequency response calculation unit 13 to the outside.

[0175] Alternatively, the frequency characteristic output unit 15 may output these frequency characteristics to the outside based on the user input operation via the frequency characteristic measuring device 10 through the input / output interface provided by the frequency characteristic measuring device 10.

[0176] Alternatively, the frequency characteristic output unit 15 may output these frequency characteristics to an external display device, for example, whereby the external display device generates and displays an image based on these frequency characteristics output from the frequency characteristic output unit 15. In this case, the user of the frequency characteristic measuring device 10 can visually confirm the image displayed on the external display device to understand the frequency characteristics of the servo system 20.

[0177] <Action>

[0178] The operation of the frequency characteristic measuring device 10 with the above structure will be described below.

[0179] The frequency response measuring device 10 performs a first frequency response measurement process to measure the frequency response of the servo system 20. The first frequency response measurement process is started, for example, by a user of the frequency response measuring system 1 instructing the frequency response measuring device 10 to begin the first frequency response measurement process.

[0180] Figure 7 This is a flowchart of the first frequency characteristic measurement process performed by the frequency characteristic measurement device 10.

[0181] like Figure 7As shown, if the first frequency characteristic measurement process starts, the first output unit 11 generates a first excitation signal in the first frequency range, and outputs the generated first excitation signal to the servo system 20 via the excitation signal selection unit 16 (step S10).

[0182] Next, the frequency characteristic calculation unit 13 calculates the first frequency characteristic based on the first identification input signal and the first identification output signal acquired from the servo system 20 which is input with the first excitation signal during the first sampling period (step S20).

[0183] If the first frequency characteristic is calculated, the second output unit 12 outputs the second excitation signal of the second frequency range, which is part of the first frequency range, to the servo system 20 via the excitation signal selection unit 16 based on the first frequency characteristic (step S30).

[0184] Next, the frequency characteristic calculation unit 13 calculates the second frequency characteristic based on the second identification input signal and the second identification output signal acquired from the servo system 20, which is input with the second excitation signal, for a second sampling period of more than the first sampling period (step S40).

[0185] If the second frequency characteristic is calculated, the frequency characteristic output unit 15 outputs the first frequency characteristic, the second frequency characteristic, or both the first frequency characteristic and the second frequency characteristic to the outside (step S50).

[0186] If the processing in step S50 is completed, the frequency characteristic measuring device 10 will end the first frequency characteristic measuring process.

[0187] <Inspection>

[0188] The frequency response measuring device 10 described above initially uses a first excitation signal in a first frequency range, which is the object of frequency response measurement, to calculate the first frequency response of the servo system 20, thereby obtaining an overall understanding of the frequency response of the servo system 20. Next, using a second excitation signal in a second frequency range, which is a portion of the first frequency range, the second frequency response of the servo system 20 is calculated more precisely and with higher accuracy within this second frequency range at a second sampling period greater than or equal to the first sampling period.

[0189] Therefore, based on the frequency characteristic measuring device 10 with the above structure, even when the inertia or friction of the servo system 20 is relatively large, or when resonance or anti-resonance is involved, the frequency characteristic of the servo system 20 can be obtained with high precision.

[0190] Furthermore, in this embodiment 1, it is assumed that the first identification input signal and the second identification input signal are used respectively as the first identification input signal and the second identification input signal for the description of frequency characteristic calculation. However, the first identification input signal and the second identification input signal may not necessarily be the first identification input signal and the second identification input signal respectively. For example, when the first identification input signal and the second identification input signal are input speed commands, it is also possible to use a feedback torque command for the first identification input signal and the second identification input signal. With such a structure, the frequency characteristic measuring device 10 can control the speed state of the motor 22 when the first identification input signal and the second identification input signal are output, and obtain the frequency characteristic of the servo system 20 relative to the motor torque, without the influence of the speed control of the motor 22.

[0191] Furthermore, in Embodiment 1, it is explained that by setting the second sampling period to be greater than or equal to the first sampling period, the frequency resolution in the second frequency characteristic is finer than or equal to the frequency resolution in the first frequency characteristic.

[0192] In contrast, this section explains why, even when the second sampling frequency is equal to the first sampling frequency (i.e., the frequency resolution in the second frequency characteristic is equal to the frequency resolution in the first frequency characteristic), the accuracy of the second frequency characteristic is higher than that of the first frequency characteristic, as long as the second frequency range is narrower than the first frequency range.

[0193] Figure 8 This diagram illustrates why, even when the frequency resolution in the second frequency characteristic is equal to that in the first frequency characteristic, the accuracy of the second frequency characteristic is higher than that of the first frequency characteristic, as long as the second frequency range is narrower than the first frequency range.

[0194] Figure 8 The section (A1) is a graph showing the time series data of the white noise signal used as the first excitation signal. In this graph, the horizontal axis represents time, and the vertical axis represents state variables (torque).

[0195] Figure 8 Section (B1) is a graph showing the time series data of the multiple sinusoidal wave signals that serve as the second excitation signal. In this graph, the horizontal axis represents time, and the vertical axis represents state variables (torque).

[0196] Here, it is assumed that the amplitude in the time series data of the first excitation signal is equal to the amplitude in the time series data of the second excitation signal.

[0197] Figure 8Section (A2) is a graph showing the frequency components of the first excitation signal obtained by performing a Fourier transform on the first excitation signal. In this graph, the horizontal axis represents frequency, and the vertical axis represents gain. The gain of the excitation signal represents the signal strength of the excitation signal; the greater the gain, the greater the signal strength.

[0198] Figure 8 Section (B2) is a graph showing the frequency components of the second excitation signal obtained by performing a Fourier transform on the second excitation signal. In this graph, the horizontal axis represents frequency, and the vertical axis represents gain.

[0199] Comparing the diagrams of part (B1) and part (B2), it can be seen that the range of frequency components contained in the second excitation signal is narrower than that contained in the first excitation signal. On the other hand, the signal strength of the second excitation signal is greater than that of the first excitation signal.

[0200] In this way, even when the frequency resolution in the second frequency characteristic is equal to that in the first frequency characteristic, the signal strength of the second excitation signal can be made greater than that of the first excitation signal within the second frequency range by making the second frequency range narrower than the first frequency range. Therefore, within the second frequency range, the second frequency characteristic has higher accuracy compared to the first frequency characteristic.

[0201] In Embodiment 1, the first measurement condition calculation unit 111 is described as calculating the amplitude of the first excitation signal based on at least one of total inertia and friction characteristics. Alternatively, the first measurement condition calculation unit 111 may calculate the amplitude of the first excitation signal based not only on total inertia and friction characteristics, but also on control parameters of the servo system 20. This would allow the first measurement condition calculation unit 111 to calculate a more appropriate amplitude of the first excitation signal.

[0202] Furthermore, in Embodiment 1, the second measurement condition calculation unit 121 is described as calculating the amplitude of the second excitation signal based on the second frequency characteristic. Alternatively, the second measurement condition calculation unit 121 may calculate the amplitude of the first excitation signal based not only on the first frequency characteristic but also on the control parameters of the servo system 20. This would allow the second measurement condition calculation unit 121 to calculate a more appropriate amplitude of the second excitation signal.

[0203] Alternatively, in Embodiment 1, the number of samples for the first identification input signal and the second identification input signal can be set to be equal. Therefore, the storage area for storing the first identification input signal and the storage area for storing the second identification signal can be used as a shared storage area, thus eliminating the need to ensure an additional storage area for storing the identification input signal.

[0204] (Implementation Method 2)

[0205] Hereinafter, we will describe the frequency characteristic measurement system according to Embodiment 2, which is constructed by changing a part of the structure of the frequency characteristic measurement system 1 according to Embodiment 1.

[0206] Regarding the frequency response measurement system according to Embodiment 2, the same components as those in Frequency Response Measurement System 1 have been described, and therefore the same reference numerals are used and detailed descriptions are omitted. The description will focus on the differences from Frequency Response Measurement System 1.

[0207] Figure 9 This is a block diagram showing the structure of the frequency characteristic measurement system 1A according to Embodiment 2.

[0208] like Figure 9 As shown, the frequency response measurement system 1A is configured by changing the frequency response measurement device 10 to the frequency response measurement device 10A according to Embodiment 1. Furthermore, the frequency response measurement device 10A is configured by changing the excitation signal selection unit 16 to the excitation signal selection unit 16A, changing the frequency response output unit 15 to the frequency response synthesis unit 19, and adding a third output unit 14, as described in Embodiment 1.

[0209] The third output unit 14 includes a third measurement condition calculation unit 141 and a third excitation signal generation unit 142, and outputs a third excitation signal within a third frequency range, which is a part of a first frequency range. The third frequency range is the range of the first frequency (higher than the lower limit frequency of the second frequency range) and the second frequency (higher than the upper limit frequency of the second frequency range). Alternatively, the second and third frequency ranges can be assumed to be continuous, with the lower limit frequency of the third frequency range falling within the second frequency range; that is, a part of the second frequency range overlaps with a part of the third frequency range. Here, the explanation assumes a part of the second frequency range overlaps with a part of the third frequency range.

[0210] The third excitation signal, like the second excitation signal, is used to obtain frequency characteristics with a finer frequency resolution than the first excitation signal.

[0211] The third measurement condition calculation unit 141 is input with the first frequency characteristic calculated by the frequency characteristic calculation unit 13, and the amplitude of the third excitation signal is calculated based on the input first frequency characteristic. Alternatively, total inertia information can also be input to the third measurement condition calculation unit 141.

[0212] The method for calculating the amplitude of the third excitation signal by the third measurement condition calculation unit 141 is the same as the method for calculating the amplitude of the second excitation signal by the second measurement condition calculation unit 121.

[0213] The third excitation signal generation unit 142 generates a third excitation signal having the amplitude of the third excitation signal calculated by the third measurement condition calculation unit 141, and outputs the generated third excitation signal to the excitation signal selection unit 16A. Here, similar to the cases of the first and second excitation signals, the third excitation signal is assumed to be a torque command (third torque command) indicating the torque of the motor 22 for explanation.

[0214] The third excitation signal output by the third excitation signal generation unit 142 is either a multiple sine wave signal or a sine wave sweep frequency signal.

[0215] As mentioned earlier, these two signals are signals whose frequency ranges can be set. Therefore, by narrowing the frequency range relative to the white noise signal used for the first excitation signal, even if the amplitudes of the first and third excitation signals are the same, the signal strength of the third excitation signal is higher than that of the first excitation signal within the third frequency range. Therefore, a third frequency characteristic that represents a frequency characteristic with higher precision than the first frequency characteristic can be obtained within the third frequency range.

[0216] Here, we will explain this by assuming that the third excitation signal output by the third output unit 14 is a multi-sine wave signal.

[0217] The excitation signal selection unit 16A selects any one of the first excitation signal output from the first output unit 11, the second excitation signal output from the second output unit 12, and the third excitation signal output from the third output unit 14, and outputs the selected first excitation signal, second excitation signal, or third excitation signal as an excitation signal to the servo system 20.

[0218] When the third excitation signal of the third frequency range output from the frequency characteristic measuring device 10A is input to the servo system 20, the frequency characteristic calculation unit 13 calculates the third frequency characteristic of the servo system 20 in the third frequency range based on the third excitation signal and the third state signal acquired at the third sampling period.

[0219] More specifically, when the third excitation signal is input to the servo system 20, the frequency characteristic calculation unit 13 calculates the third frequency characteristic of the servo system 20 in the third frequency range based on the third input torque command output from the motor control device 21 in the servo system 20 where the third excitation signal is input and the third state signal output from the first state quantity measuring device 24 in the servo system 20 where the third excitation signal is input and the third state signal is input and the third state signal is obtained from the first state quantity measuring device 24 in the servo system 20 where the third excitation signal is input and the third frequency characteristic calculation unit 13 is input to the servo system 20 where the third excitation signal is input and the third frequency characteristic calculation unit 13 is input to the servo system 20 where the third excitation signal is input and the third frequency characteristic calculation unit 13 is input to the servo system 20 where the third frequency characteristic calculation unit 13 is input to the servo system 20 where the third excitation signal is input and the third frequency characteristic calculation unit 13 is input to the servo system 20 where the third frequency characteristic calculation unit 13 is input to the servo system 20 where the third excitation signal is input and the third state signal is obtained from the first state quantity measuring device 24 in the servo system 20 where the third excitation signal is input and the third state signal is obtained to ...

[0220] Figure 10 This is a Bode plot showing an example of the third frequency characteristic calculated by the frequency characteristic calculation unit 13 when the frequency characteristic measuring device 10A outputs the third excitation signal.

[0221] exist Figure 10 In the diagram, the horizontal axis represents frequency, and the vertical axis represents the gain G shown in equation (1). Here, in... Figure 10 In the diagram, the thick dashed line represents an example of the first frequency characteristic calculated by the frequency characteristic calculation unit 13 when the frequency characteristic measuring device 10A outputs the first excitation signal. Figure 3 The first frequency characteristic illustrated is the same, and the thin dashed line is an example of the second frequency characteristic calculated by the frequency characteristic calculation unit 13 when the frequency characteristic measuring device 10A outputs the second excitation signal. Figure 6 The second frequency characteristic illustrated is the same.

[0222] like Figure 10 As shown, the frequency response calculation unit 13 calculates the third frequency response with a finer frequency resolution than when calculating the first frequency response.

[0223] Back to Figure 9 Continue with the explanation of the frequency response measurement system 1A.

[0224] The frequency response synthesis unit 19 synthesizes the second and third frequency responses to calculate the fourth frequency response of the servo system 20. Then, the frequency response synthesis unit 19 outputs at least one of the second, third, and fourth frequency responses to the outside.

[0225] Figure 11 This is a Bode plot showing an example of the fourth frequency characteristic synthesized by the frequency characteristic synthesis unit 19.

[0226] like Figure 10 as well as Figure 11 As shown, for the portion of the frequency ranges that do not overlap in the second and third frequency ranges, the frequency characteristic synthesis unit 19, for example, directly calculates the second and third frequency characteristics as the fourth frequency characteristic.

[0227] Furthermore, for the portion of the frequency ranges that overlap in the second and third frequency ranges, the frequency characteristic synthesis unit 19 calculates, for example, the average of the second and third frequency characteristics as the fourth frequency characteristic.

[0228] Alternatively, the frequency response synthesis unit 19 may output these frequency characteristics to the outside based on user input via the frequency response measurement device 10A through an input / output interface provided by the frequency response measurement device 10A.

[0229] Alternatively, the frequency response synthesis unit 19 may output these frequency characteristics to an external display device, for example, whereby the external display device generates and displays an image based on these frequency characteristics output from the frequency response synthesis unit 19. In this case, the user of the frequency response measurement device 10A can visually confirm the image displayed on the external display device to understand the frequency characteristics of the servo system 20.

[0230] <Action>

[0231] The operation of the frequency characteristic measuring device 10A with the above structure will be described below.

[0232] Instead of the first frequency characteristic measurement process performed by the frequency characteristic measurement device 10 according to Embodiment 1, the frequency characteristic measurement device 10A performs a second frequency characteristic measurement process obtained by modifying part of the first frequency characteristic measurement process.

[0233] Figure 12 This is a flowchart of the second frequency characteristic measurement process performed by the frequency characteristic measurement device 10A.

[0234] The processing of steps S110 to S140 in the second frequency characteristic measurement process is the same as the processing of replacing the excitation signal selection unit 16 with the excitation signal selection unit 16A in the processing of steps S10 to S40 in the first frequency measurement process.

[0235] Therefore, this explanation will focus on the processes in steps S150 to S170.

[0236] If the processing of step S140 is completed, the third output unit 14 outputs the third excitation signal of the third frequency range, which is part of the first frequency range, to the servo system 20 via the excitation signal selection unit 16 based on the first frequency characteristics. The third frequency range is the range of the first frequency which is higher than the lower limit frequency of the second frequency range and the second frequency which is higher than the upper limit frequency of the second frequency range (step S150).

[0237] Next, the frequency characteristic calculation unit 13 calculates the third frequency characteristic based on the third identification input signal and the third identification output signal acquired from the servo system 20, which is input with the third excitation signal, for a third sampling period of more than the first sampling period (step S160).

[0238] If the third frequency characteristic is calculated, the frequency characteristic synthesis unit 19 combines the second and third frequency characteristics to calculate the fourth frequency characteristic of the servo system 20, and outputs the calculated fourth frequency characteristic to the outside (step S170).

[0239] If the processing in step S170 is completed, the frequency characteristic measuring device 10A will end the second frequency characteristic measuring process.

[0240] <Inspection>

[0241] The frequency characteristic measuring device 10A according to the above structure, similar to the frequency characteristic measuring device 10 according to Embodiment 1, can obtain high-precision frequency characteristics of the servo system 20 even when the inertia or friction of the servo system 20 is relatively large, or when resonance or anti-resonance is involved.

[0242] Furthermore, in Embodiment 1 and Embodiment 2, the methods for calculating the amplitude of the second excitation signal performed by the second measurement condition calculation unit 121 and the methods for calculating the amplitude of the third excitation signal performed by the third measurement condition calculation unit 141 were described using the case where the second frequency range or the third frequency range includes a resonant frequency and / or an anti-resonant frequency as examples. However, even if the second frequency range or the third frequency range does not include a resonant frequency or an anti-resonant frequency, the amplitude of the second excitation signal and the amplitude of the third excitation signal can be calculated using the same calculation method.

[0243] Figure 13 It is Figure 3 The Bode plot showing the first frequency characteristic overlapping with the frequency characteristic of the approximately rigid servo system 20 when the servo system 20 is approximately a rigid body is the Bode plot when the second frequency range does not include the resonant frequency and the anti-resonant frequency.

[0244] like Figure 13 As shown, even when the second frequency range does not include the resonant frequency and the anti-resonant frequency, a gain difference will still occur between the first frequency characteristic and the frequency characteristic of the approximately rigid servo system 20. Therefore, even when the second frequency range does not include the resonant frequency and the anti-resonant frequency, the second measurement condition calculation unit 121 can calculate the amplitude of the second excitation signal using the same calculation method, and even when the third frequency range does not include the resonant frequency and the anti-resonant frequency, the third measurement condition calculation unit 141 can calculate the amplitude of the third excitation signal using the same calculation method.

[0245] (Variation Example 1)

[0246] Hereinafter, we will describe the frequency characteristic measurement system according to Modified Example 1, which is a frequency characteristic measurement system modified from the frequency characteristic measurement system 1A according to Embodiment 2 by changing a part of its structure.

[0247] Frequency response measurement system 1A is an example of a structure that measures the frequency response of the motor 22 of the servo system 20 in relation to its state variables. In contrast, the frequency response measurement system of Modified Example 1 is an example of a structure that measures the frequency response of the load 23, which is the driven body, of the servo system 20B (described later) of Modified Example 1 in relation to its state variables.

[0248] Hereinafter, the frequency response measurement system involved in Modification Example 1 has been described with respect to the same components as the frequency response measurement system 1A, and the same reference numerals are used to mark them and detailed descriptions are omitted. The description will focus on the differences from the frequency response measurement system 1A.

[0249] Figure 14 This is a block diagram showing the structure of the frequency characteristic measurement system 1B involved in Modification Example 1.

[0250] like Figure 14 As shown, the frequency response measurement system 1B is configured differently from the frequency response measurement system 1A according to Embodiment 2 by changing the frequency response measurement device 10A to the frequency response measurement device 10B and the servo system 20 to the servo system 20B. Furthermore, the frequency response measurement device 10B is configured differently from the frequency response measurement device 10A by changing the frequency response calculation unit 13 to the frequency response calculation unit 13B and the Fourier transform unit 17 to the Fourier transform unit 17B. Additionally, the servo system 20B is configured by adding a second state quantity measurement device 25 to the servo system 20.

[0251] The second state quantity measuring device 25 measures the state quantity related to the load 23, which is the driven body in the servo system 20B, and outputs the measured state quantity related to the load 23 as a state signal representing the state quantity of the servo system 20B. Here, the second state quantity measuring device 25 is described as a sensor that senses the load 23 and outputs the state quantity related to the load 23. That is, the state signal output by the second state quantity measuring device 25 is described as including sensor information representing the state quantity of the load 23 obtained by sensing the load 23, which is the driven body in the servo system 20B. More specifically, the second state quantity measuring device 25 is described as a sensor that measures the rotational position of the load 23 and outputs a state signal representing the rotational position of the load 23. However, the second state quantity measuring device 25 does not necessarily need to be limited to the above-described sensor, as long as it is a device capable of measuring the state quantity related to the load 23 and outputting the measured state quantity related to the load 23.

[0252] The frequency response calculation unit 13B calculates the frequency response of the servo system 20B based on the identification input signal input to the servo system 20B for obtaining the frequency response of the servo system 20B, and the identification output signal output from the servo system 20B for obtaining the frequency response of the servo system 20B. At this time, the frequency response calculation unit 13B samples to acquire the identification input signal and the identification output signal.

[0253] Here, the identification input signal can be, for example, the excitation signal output from the frequency response measuring device 10B itself, or a control signal generated in the motor control device 21 based on the excitation signal output from the frequency response measuring device 10B in order to drive the motor 22. Here, in the motor control device 21, the control signal generated based on the excitation signal output from the frequency response measuring device 10B in order to drive the motor 22 can be, for example, a feedback torque command. Here, we will explain this as the identification input signal being the excitation signal input to the servo system 20B.

[0254] Furthermore, the identification output signal may, for example, be a state signal output from the second state quantity measuring device 25. Here, we will explain this as if the identification output signal were a state signal output from the second state quantity measuring device 25.

[0255] That is, (1) when a first excitation signal in a first frequency range output from the frequency characteristic measuring device 10B is input to the servo system 20B, the frequency characteristic calculation unit 13B calculates the first frequency characteristic of the servo system 20B in the first frequency range based on the first excitation signal and the first status signal acquired at a first sampling period. In addition, (2) when a second excitation signal in a second frequency range output from the frequency characteristic measuring device 10B is input to the servo system 20B, the frequency characteristic calculation unit 13B calculates the second frequency characteristic of the servo system 20B in the second frequency range based on the second excitation signal and the second status signal acquired at a second sampling period of more than the first sampling period. (3) when a third excitation signal in a third frequency range output from the frequency characteristic measuring device 10B is input to the servo system 20B, the frequency characteristic calculation unit 13B calculates the third frequency characteristic of the servo system 20B in the third frequency range based on the third excitation signal and the third status signal acquired at a third sampling period of more than the first sampling period.

[0256] More specifically, (1) when the first excitation signal is input to the servo system 20B, the frequency characteristic calculation unit 13B calculates the first frequency characteristic of the servo system 20B within the first frequency range based on the first input torque command obtained from the motor control device 21 in the servo system 20B with the first sampling period, and the first state signal obtained from the second state quantity measuring device 25 in the servo system 20B with the first excitation signal, obtained within the first sampling period. Furthermore, (2) when the second excitation signal is input to the servo system 20B, the frequency characteristic calculation unit 13B calculates the second frequency characteristic of the servo system 20B within the second frequency range based on the second input torque command obtained from the motor control device 21 in the servo system 20B with the second sampling period greater than or equal to the first sampling period, and the second state signal obtained from the second state quantity measuring device 25 in the servo system 20B with the second excitation signal, obtained within the second sampling period greater than or equal to the first sampling period. (3) When the third excitation signal is input to the servo system 20B, the frequency characteristic calculation unit 13B calculates the third frequency characteristic of the servo system 20B in the third frequency range based on the third input torque command obtained from the motor control device 21 in the servo system 20B with a third sampling period of more than the first sampling period and the third state signal obtained from the second state quantity measuring device 25 in the servo system 20B with a third sampling period of more than the first sampling period.

[0257] Here, the frequency characteristic calculated by the frequency characteristic calculation unit 13B is described as the frequency characteristic of the load 23's rotational speed [rpm] relative to the input torque [Nm] input to the servo system 20B.

[0258] like Figure 14 As shown, the frequency response calculation unit 13B can also be implemented, for example, by having a structure that includes a Fourier transform unit 17B and a frequency response calculation unit 18. Here, the Fourier transform unit 17B performs a Fourier transform on the input torque command output from the motor control device 21 and the rotational speed of the load 23 obtained by time differentiation of the state signal representing the rotational position of the load 23 output from the second state quantity measuring device 25.

[0259] <Inspection>

[0260] Generally, understanding the frequency characteristics of the load 23's state relative to the state command (e.g., torque command) given to the motor 22 is extremely useful for the control and adjustment of the servo system 20B. The frequency characteristic measuring device 10B with the above-described structure can measure these frequency characteristics with high accuracy. Therefore, the frequency characteristic measuring device 10B with the above-described structure can be used extremely usefully in the control and adjustment of the servo system 20B.

[0261] (Variation Example 2)

[0262] Hereinafter, we will describe the frequency characteristic measurement system according to Modified Example 2, which is a frequency characteristic measurement system modified from the frequency characteristic measurement system 1A according to Embodiment 2 by changing a part of its structure.

[0263] Frequency response measurement system 1A is an example of a structure that measures the frequency response of the state quantity of the motor 22 of the servo system 20. In contrast, the frequency response measurement system according to Modified Example 2 is an example of a structure that measures the frequency response of the state quantity of the motor 22 and the frequency response of the state quantity of the load 23 as a driven body in the servo system 20C (described later) according to Modified Example 2.

[0264] Hereinafter, the frequency response measurement system involved in Modification Example 2 has been described with respect to the same components as the frequency response measurement system 1A, and the same reference numerals are used to mark them and detailed descriptions are omitted. The description will focus on the differences from the frequency response measurement system 1A.

[0265] Figure 15 This is a block diagram showing the structure of the frequency characteristic measurement system 1C involved in Modified Example 2.

[0266] like Figure 15 As shown, the frequency response measurement system 1C is configured by changing the frequency response measurement device 10A to the frequency response measurement device 10C and the servo system 20 to the servo system 20C, which is different from the frequency response measurement system 1A according to Embodiment 2. Furthermore, the frequency response measurement device 10C is configured by changing the frequency response calculation unit 13 to the frequency response calculation unit 13C and the Fourier transform unit 17 to the Fourier transform unit 17C, which is different from the frequency response measurement device 10A. Additionally, the servo system 20C is configured by adding a second state quantity measurement device 25 to the servo system 20.

[0267] The frequency characteristic calculation unit 13C calculates the frequency characteristics of the servo system 20C related to the motor 22 (hereinafter also referred to as "motor frequency characteristics") and the frequency characteristics related to the load 23 (hereinafter also referred to as "load frequency characteristics") based on the identification input signal input to the servo system 20C for obtaining the frequency characteristics of the servo system 20C, and the identification output signal output from the servo system 20C for obtaining the frequency characteristics of the servo system 20C. At this time, the frequency characteristic calculation unit 13C samples to acquire the identification input signal and the identification output signal.

[0268] Here, the identification input signal can be, for example, the excitation signal output from the frequency response measuring device 10C itself, or a control signal generated in the motor control device 21 based on the excitation signal output from the frequency response measuring device 10C for driving the motor 22. Here, in the motor control device 21, the control signal generated based on the excitation signal output from the frequency response measuring device 10C for driving the motor 22 can be, for example, a feedback torque command. Here, we will explain this as the identification input signal being the excitation signal input to the servo system 20C.

[0269] Furthermore, the identification output signal may be, for example, a status signal (hereinafter also referred to as "motor status signal") indicating the status quantity of the motor 22 and a status signal (hereinafter also referred to as "load status signal") indicating the status quantity of the load 23, output from the first status quantity measuring device 24. Here, we will describe the identification output signal as the motor status signal output from the first status quantity measuring device 24 and the load status signal output from the second status quantity measuring device 25.

[0270] That is, (1) when the first excitation signal of the first frequency range output from the frequency characteristic measuring device 10C is input to the servo system 20C, the frequency characteristic calculation unit 13C calculates the first motor frequency characteristic of the servo system 20C in the first frequency range based on the first excitation signal and the first motor status signal acquired in the first sampling period, and calculates the first load frequency characteristic of the servo system 20C in the first frequency range based on the first excitation signal and the first load status signal acquired in the first sampling period. In addition, (2) when the second excitation signal of the second frequency range output from the frequency characteristic measuring device 10C is input to the servo system 20C, the frequency characteristic calculation unit 13C calculates the second motor frequency characteristic of the servo system 20C in the second frequency range based on the second excitation signal and the second motor status signal acquired in the second sampling period or longer than the first sampling period, and calculates the second load frequency characteristic of the servo system 20C in the second frequency range based on the second excitation signal and the second load status signal acquired in the second sampling period. Furthermore, (3) when the third excitation signal of the third frequency range output from the frequency characteristic measuring device 10C is input to the servo system 20C, the frequency characteristic calculation unit 13C calculates the third motor frequency characteristic of the servo system 20C in the third frequency range based on the third excitation signal and the third motor status signal acquired in the third sampling period of more than the first sampling period, and calculates the third load frequency characteristic of the servo system 20C in the third frequency range based on the third excitation signal and the third load status signal acquired in the third sampling period.

[0271] More specifically, (1) when the first excitation signal is input to the servo system 20C, the frequency characteristic calculation unit 13C calculates the first motor frequency characteristic of the servo system 20C in the first frequency range based on the first input torque command output from the motor control device 21 in the servo system 20C with the first sampling period and the first motor state signal output from the first state quantity measuring device 24 in the servo system 20C with the first excitation signal and the first motor state signal output from the second state quantity measuring device 25 in the servo system 20C with the first excitation signal and the first load frequency characteristic of the servo system 20C in the first frequency range based on the first input torque command output from the motor control device 21 in the servo system 20C with the first excitation signal and the first load state signal output from the second state quantity measuring device 25 in the servo system 20C with the first excitation signal and the first load frequency characteristic of the servo system 20C in the first frequency range. Furthermore, (2) when the second excitation signal is input to the servo system 20C, the frequency characteristic calculation unit 13C calculates the second motor frequency characteristic of the servo system 20C in the second frequency range based on the second input torque command output from the motor control device 21 in the servo system 20C with a second sampling period of more than the first sampling period, and the second motor state signal output from the first state quantity measuring device 24 in the servo system 20C with the second excitation signal, with a second sampling period. It also calculates the second load frequency characteristic of the servo system 20C in the second frequency range based on the second input torque command output from the motor control device 21 in the servo system 20C with the second excitation signal, with a second sampling period, and the second load state signal output from the second state quantity measuring device 25 in the servo system 20C with the second excitation signal, with a second sampling period. Furthermore, (3) when the third excitation signal is input to the servo system 20C, the frequency characteristic calculation unit 13C calculates the third motor frequency characteristic of the servo system 20C in the third frequency range based on the third input torque command obtained from the motor control device 21 in the servo system 20C with a third sampling period of more than the first sampling period, and the third motor state signal obtained from the first state quantity measuring device 24 in the servo system 20C with the third excitation signal, which is obtained in the third sampling period. Based on the third input torque command obtained from the motor control device 21 in the servo system 20C with the third excitation signal, which is obtained in the third sampling period, and the third load state signal obtained from the second state quantity measuring device 25 in the servo system 20C with the third excitation signal, which is obtained in the third sampling period, the third load frequency characteristic of the servo system 20C in the third frequency range is calculated.

[0272] Here, the motor frequency characteristic calculated by the frequency characteristic calculation unit 13C is described as the frequency characteristic of the motor 22's rotational speed [rpm] relative to the input torque [Nm] input to the servo system 20C, and the load frequency characteristic calculated by the frequency characteristic calculation unit 13C is described as the frequency characteristic of the load 23's rotational speed [rpm] relative to the input torque [Nm] input to the servo system 20C.

[0273] like Figure 15 As shown, the frequency response calculation unit 13C can also be implemented, for example, by having a structure that includes a Fourier transform unit 17C and a frequency response calculation unit 18. Here, the Fourier transform unit 17C performs Fourier transforms on the input torque command output from the motor control device 21, the rotational speed of the motor 22 obtained by time differentiation of the state signal indicating the rotational position of the motor 22 output from the first state quantity measuring device 24, and the rotational speed of the load 23 obtained by time differentiation of the state signal indicating the rotational position of the load 23 output from the second state quantity measuring device 25.

[0274] <Inspection>

[0275] The frequency characteristic measuring device 10C with the above structure can measure the frequency characteristics of the servo system 20C, the state quantity of the motor 22, and the state quantity of the load 23 as the driven body with high precision.

[0276] (Variation Example 3)

[0277] The following describes a frequency response measurement system based on a modified version 3, which is a frequency response measurement system constructed by changing a portion of the structure of the frequency response measurement system 1A according to Embodiment 2.

[0278] The frequency response measurement system 1A is an example of the following structure: the frequency response measurement device 10 outputs the first excitation signal, the second excitation signal, and the third excitation signal once each, and measures the first frequency response, the second frequency response, and the third frequency response based on the signals of the first identification input signal, the second identification input signal, and the third identification signal, and the first identification output signal, the second identification output signal, and the third identification signal, which are input from the servo system 20 that has been output once. In contrast, the frequency response measurement system involved in Modification 3 is an example of the following structure: the frequency response measurement device involved in Modification 3 outputs the first excitation signal, the second excitation signal, and the third excitation signal multiple times, and measures the first frequency response, the second frequency response, and the third frequency response based on the signals of the first identification input signal, the second identification input signal, and the third identification signal, and the first identification output signal, the second identification output signal, and the third identification signal, which are repeatedly acquired from the servo system involved in Modification 3, which is input to and output multiple times.

[0279] Hereinafter, the frequency response measurement system involved in Modification Example 3 has been described for the same constituent elements as the frequency response measurement system 1A, and the same reference numerals are used to mark them and detailed descriptions are omitted. The description will focus on the differences from the frequency response measurement system 1A.

[0280] Figure 16 This is a block diagram showing the structure of the frequency characteristic measurement system 1D involved in Modification Example 3.

[0281] like Figure 16 As shown, the frequency response measurement system 1D is configured by changing the frequency response measurement device 10A to the frequency response measurement device 10D, which is different from the frequency response measurement system 1A according to Embodiment 2. Furthermore, the frequency response measurement device 10D is configured by changing the first output unit 11 to the first output unit 11D, the first excitation signal generation unit 112 to the first excitation signal generation unit 112D, the second output unit 12 to the second output unit 12D, the second excitation signal generation unit 122 to the second excitation signal generation unit 122D, the third output unit 14 to the third output unit 14D, and the frequency response calculation unit 13 to the frequency response calculation unit 13D.

[0282] The first output unit 11D has a first measurement condition calculation unit 111 and a first excitation signal generation unit 112D, and outputs the first excitation signal in the first frequency range multiple times.

[0283] The first excitation signal generation unit 112D generates a first excitation signal having the amplitude of the first excitation signal calculated by the first measurement condition calculation unit 111, and outputs the generated first excitation signal to the excitation signal selection unit 16 multiple times. Here, the first frequency range is described as the entire frequency range of the frequency characteristic measured by the frequency characteristic measuring device 10D. Furthermore, the first excitation signal is described as a torque command (first torque command) indicating the torque of the motor 22.

[0284] The first excitation signal output by the first excitation signal generation unit 112D is any one of white noise signal, multiple sine wave signal, sine wave sweep signal, step signal and ramp signal.

[0285] The second output unit 12D has a second measurement condition calculation unit 121 and a second excitation signal generation unit 122D, and outputs a second excitation signal of a second frequency range, which is part of the first frequency range, multiple times.

[0286] The second excitation signal generation unit 122D generates a second excitation signal having the amplitude calculated by the second measurement condition calculation unit 121 based on the amplitude of the second excitation signal, and outputs the generated second excitation signal to the excitation signal selection unit 16 multiple times. Here, it is assumed that the second excitation signal is, in the same case as the first excitation signal, a torque command (second torque command) indicating the torque of the motor 22.

[0287] The second excitation signal output by the second excitation signal generation unit 122D is either a multiple sine wave signal or a sine wave sweep frequency signal.

[0288] The third output unit 14D has a third measurement condition calculation unit 141 and a third excitation signal generation unit 142D, and outputs the third excitation signal in the third frequency range multiple times.

[0289] The third excitation signal generation unit 142D generates a third excitation signal with the amplitude calculated by the third measurement condition calculation unit 141 based on the amplitude of the third excitation signal, and outputs the generated third excitation signal to the excitation signal selection unit 16 multiple times. Here, it is assumed that the third excitation signal, like the first and second excitation signals, is a torque command (third torque command) indicating the torque of the motor 22.

[0290] The third excitation signal output by the third excitation signal generation unit 142D is either a multiple sine wave signal or a sine wave sweep frequency signal.

[0291] The frequency response calculation unit 13D calculates the frequency response of the servo system 20 based on the identification input signal input to the servo system 20 for obtaining the frequency response of the servo system 20, and the identification output signal output from the servo system 20 for obtaining the frequency response of the servo system 20. At this time, the frequency response calculation unit 13D samples to acquire the identification input signal and the identification output signal.

[0292] Here, the identification input signal can be, for example, the excitation signal output from the frequency response measuring device 10D itself, or a control signal generated in the motor control device 21 based on the excitation signal output from the frequency response measuring device 10D in order to drive the motor 22. Here, in the motor control device 21, the control signal generated based on the excitation signal output from the frequency response measuring device 10D in order to drive the motor 22 can be, for example, a feedback torque command. Here, we will explain this as the identification input signal being the excitation signal input to the servo system 20.

[0293] Furthermore, the identification output signal may, for example, be a state signal output from the first state quantity measuring device 24. Here, we will explain this as if the identification output signal were a state signal output from the first state quantity measuring device 24.

[0294] That is, (1) when the first excitation signal of the first frequency range output from the frequency characteristic measuring device 10D is input to the servo system 20 multiple times, the frequency characteristic calculation unit 13D calculates the first frequency characteristic of the servo system 20 in the first frequency range based on the first excitation signal and the first status signal acquired multiple times over the first sampling period. In addition, (2) when the second excitation signal of the second frequency range output from the frequency characteristic measuring device 10D is input to the servo system 20 multiple times, the frequency characteristic calculation unit 13D calculates the second frequency characteristic of the servo system 20 in the second frequency range based on the second excitation signal and the second status signal acquired multiple times over the second sampling period of more than the first sampling period. (3) when the third excitation signal of the third frequency range output from the frequency characteristic measuring device 10D is input to the servo system 20 multiple times, the frequency characteristic calculation unit 13D calculates the third frequency characteristic of the servo system 20 in the third frequency range based on the third excitation signal and the third status signal acquired multiple times over the third sampling period of more than the first sampling period.

[0295] More specifically, (1) when the first excitation signal is input to the servo system 20 multiple times, the frequency characteristic calculation unit 13D calculates the first frequency characteristic of the servo system 20 in the first frequency range based on the multiple first input torque commands that are output multiple times from the motor control device 21 in the servo system 20 where the first excitation signal is input multiple times and the multiple first state signals that are output multiple times from the first state quantity measuring device 24 in the servo system 20 where the first excitation signal is input multiple times and are obtained in the first sampling period. Furthermore, (2) when the second excitation signal is input to the servo system 20 multiple times, the frequency characteristic calculation unit 13D calculates the second frequency characteristic of the servo system 20 within the second frequency range based on the multiple second input torque commands repeatedly output from the motor control device 21 in the servo system 20 where the second excitation signal is input multiple times, acquired at a second sampling period of more than the first sampling period, and the multiple second state signals repeatedly output from the first state quantity measuring device 24 in the servo system 20 where the second excitation signal is input multiple times, acquired at a second sampling period of more than the first sampling period. (3) When the third excitation signal is input to the servo system 20 multiple times, the frequency characteristic calculation unit 13D calculates the third frequency characteristic of the servo system 20 in the third frequency range based on the multiple third input torque commands that are output multiple times from the motor control device 21 in the servo system 20 where the third excitation signal is input multiple times and the multiple third state signals that are output multiple times from the first state quantity measuring device 24 in the servo system 20 where the third excitation signal is input multiple times and the multiple third state signals that are output multiple times from the first state quantity measuring device 24 in the servo system 20 where the third excitation signal is input multiple times.

[0296] Here, the frequency characteristic calculated by the frequency characteristic calculation unit 13D is described as the frequency characteristic of the rotational speed [rpm] of the motor 22 relative to the input torque [Nm] input to the servo system 20.

[0297] <Inspection>

[0298] Generally, when the first state quantity measuring device 24 measures the state quantity of the servo system 20, it also includes the influence of noise superimposed on the state quantity of the servo system 20. Therefore, the state quantity of the servo system 20 measured by the first state quantity measuring device 24 may deviate from the amount of superimposed noise influence in each measurement.

[0299] In contrast, the frequency characteristic measuring device 10D according to the above structure calculates the second frequency characteristic based on the first excitation signal and the first state signal acquired multiple times, calculates the second frequency characteristic based on the second excitation signal and the second state signal acquired multiple times, and calculates the second frequency characteristic based on the second excitation signal and the second state signal acquired multiple times.

[0300] At this time, in the frequency characteristic calculation unit 18, the measurement accuracy of the state signals can be determined by evaluating the average, variance, and / or coherence of multiple state signals. Furthermore, for example, the first, second, and third frequency characteristics can be calculated solely based on the average of the state signals determined to have high measurement accuracy. Therefore, according to the frequency characteristic measurement device 10D with the above structure, a higher accuracy frequency characteristic of the servo system 20 can be obtained, reducing the influence of noise in the measurement of state quantities of the servo system 20.

[0301] (Replenish)

[0302] As described above, embodiments 1 and 2, and variations 1 to 3, have been presented as examples of the technology disclosed in this application. However, this disclosure is not limited to these embodiments and variations. It is also possible that, without departing from the spirit of this disclosure, various modifications conceived by those skilled in the art to this embodiment or variation, and methods of constructing by combining constituent elements from different embodiments or variations, are also included within the scope of one or more of the methods disclosed.

[0303] This disclosure includes not only frequency response measurement system 1, but also frequency response measurement method in which the characteristic components of frequency response measurement system 1 are set as steps. Furthermore, this disclosure also includes a computer program that causes a computer to execute the characteristic steps of the frequency response measurement method. Additionally, this disclosure also includes a computer-readable non-transient recording medium on which such a computer program is recorded.

[0304] Industrial availability

[0305] This disclosure can be widely used in systems for measuring frequency characteristics, etc.

[0306] Explanation of reference numerals in the attached figures

[0307] 1, 1A, 1B, 1C, 1D: Frequency response measurement system

[0308] 10, 10A, 10B, 10C, 10D: Frequency response measuring device

[0309] 11, 11D: First Output Section

[0310] 12, 12D: Second Output Unit

[0311] 13, 13B, 13C, 13D: Frequency Response Calculation Department

[0312] 14, 14D: Third Output Unit

[0313] 15: Frequency response output section

[0314] 16, 16A: Excitation signal selection unit

[0315] 17, 17B, 17C: Fourier Transform Part

[0316] 18: Frequency Response Calculation Unit

[0317] 19: Frequency response synthesis section

[0318] 20, 20B, 20C: Servo System

[0319] 21: Motor control device

[0320] 22: Motor

[0321] 23: Load

[0322] 24: First state quantity measuring device

[0323] 25: Second state quantity measuring device

[0324] 111: First Measurement Condition Calculation Department

[0325] 112, 112D: First excitation signal generation unit

[0326] 121: Calculation Section for Second Measurement Conditions

[0327] 122, 122D: Second excitation signal generation unit

[0328] 141: Calculation Section for the Third Measurement Condition

[0329] 142, 142D: Third excitation signal generation unit.

Claims

1. A method for measuring frequency characteristics of a servo system. The frequency characteristic measurement method has the following features: The first output step involves outputting the first excitation signal of the first frequency range to the servo system. The first frequency characteristic calculation step calculates the first frequency characteristic of the servo system within the first frequency range based on the first identification input signal input to the servo system for obtaining the frequency characteristic of the servo system and the first identification output signal output from the servo system for obtaining the frequency characteristic of the servo system. The first identification input signal and the first identification output signal are signals acquired from the servo system to which the first excitation signal is input for a first sampling period. The second output step involves outputting a second excitation signal, which is a second frequency range that is part of the first frequency range and is generated based on the characteristics of the first frequency, to the servo system. and The second frequency characteristic calculation step calculates the second frequency characteristic of the servo system within the second frequency range based on the second identification input signal input to the servo system for obtaining the frequency characteristic of the servo system and the second identification output signal output from the servo system for obtaining the frequency characteristic of the servo system. The second identification input signal and the second identification output signal are signals obtained from the servo system to which the second excitation signal is input for a second sampling period of more than the first sampling period.

2. The frequency characteristic measurement method according to claim 1, wherein, The second sampling period is greater than the first sampling period.

3. The frequency characteristic measurement method according to claim 1, wherein, The first excitation signal is any one of the following: a pseudo-random signal, a multiple sine wave signal, a sine wave sweep signal, a step signal, and a ramp signal.

4. The frequency characteristic measurement method according to claim 1, wherein, The second excitation signal is either a multiple sine wave signal or a sine wave sweep signal.

5. The frequency characteristic measurement method according to claim 1, wherein, The frequency characteristic measurement method further includes: a first measurement condition calculation step, which calculates the amplitude of the first excitation signal based on at least one of the total inertia and friction characteristics of the servo system. In the first output step, the first excitation signal having the amplitude of the first excitation signal calculated by the first measurement condition calculation step is output to the servo system.

6. The frequency characteristic measurement method according to claim 1, wherein, The frequency characteristic measurement method further includes: a second measurement condition calculation step, which calculates the amplitude of the second excitation signal based on the first frequency characteristic. In the second output step, the second excitation signal, having the amplitude of the second excitation signal calculated by the second measurement condition calculation step, is output to the servo system.

7. The frequency characteristic measurement method according to claim 1, wherein, The frequency response measurement method also has the following characteristics: The third output step involves outputting a third excitation signal, generated based on the characteristics of the first frequency, within a third frequency range that is part of the first frequency range, to the servo system. The third frequency range is a range of first frequencies higher than the lower limit frequency of the second frequency range and second frequencies higher than the upper limit frequency of the second frequency range. The third frequency characteristic calculation step calculates the third frequency characteristic of the servo system within the third frequency range based on the third identification input signal input to the servo system for obtaining the frequency characteristic of the servo system and the third identification output signal output from the servo system for obtaining the frequency characteristic of the servo system. The third identification input signal and the third identification output signal are signals obtained from the servo system to which the third excitation signal is input for a third sampling period of more than the first sampling period.

8. The frequency characteristic measurement method according to claim 7, wherein, The frequency characteristic measurement method further includes a frequency characteristic synthesis step, which combines the second frequency characteristic and the third frequency characteristic to calculate the fourth frequency characteristic of the servo system.

9. The frequency characteristic measurement method according to claim 7, wherein, The third sampling period is greater than the first sampling period.

10. The frequency characteristic measurement method according to claim 7, wherein, The second frequency range and the third frequency range are continuous.

11. The frequency characteristic measurement method according to claim 7, wherein, The lower limit frequency of the third frequency range is within the second frequency range.

12. The frequency characteristic measurement method according to claim 7, wherein, At least one of the second frequency range and the third frequency range includes at least one of the resonant frequency and anti-resonant frequency of the servo system.

13. The frequency characteristic measurement method according to claim 8, wherein, The frequency characteristic measurement method further includes a fourth output step, in which at least one of the second frequency characteristic, the third frequency characteristic, and the fourth frequency characteristic is output to a display device.

14. The frequency characteristic measurement method according to claim 1, wherein, The frequency response measurement method also has the following characteristics: The first measurement condition calculation step involves calculating the amplitude of the first excitation signal based on at least one of the total inertia and friction characteristics of the servo system and the control parameter settings of the servo system. and The second measurement condition calculation step involves calculating the amplitude of the second excitation signal based on the first frequency characteristic and the control parameter settings. In the first output step, the first excitation signal, having the amplitude of the first excitation signal calculated through the first measurement condition calculation step, is output to the servo system. In the second output step, the second excitation signal, having the amplitude of the second excitation signal calculated by the second measurement condition calculation step, is output to the servo system.

15. The frequency characteristic measurement method according to claim 1, wherein, The first identification output signal and the second identification output signal contain sensor information representing the state of the driven body obtained by sensing the driven body in the servo system.

16. The frequency characteristic measurement method according to claim 1, wherein, In the first output step, the first excitation signal is output multiple times. In the second output step, the second excitation signal is output multiple times. In the first frequency characteristic calculation step, the first frequency characteristic is calculated based on the first identification input signal and the first identification output signal, which are obtained multiple times from the servo system that has been repeatedly input with the first excitation signal. In the second frequency characteristic calculation step, the second frequency characteristic is calculated based on the second identification input signal and the second identification output signal obtained multiple times from the servo system that has been repeatedly input with the second excitation signal.

17. The frequency characteristic measurement method according to claim 1, wherein, The number of samples of the first identification input signal and the second identification input signal are equal.

18. A frequency response measuring device for measuring the frequency response of a servo system. The frequency response measuring device includes: The first output unit outputs a first excitation signal in the first frequency range to the servo system; The frequency characteristic calculation unit calculates the first frequency characteristic of the servo system within the first frequency range based on a first identification input signal input to the servo system for obtaining the frequency characteristic of the servo system, and a first identification output signal output from the servo system for obtaining the frequency characteristic of the servo system. The first identification input signal and the first identification output signal are signals acquired from the servo system to which the first excitation signal is input, with a first sampling period. The second output unit outputs a second excitation signal, generated based on the characteristics of the first frequency, which is a second frequency range that is part of the first frequency range, to the servo system. The frequency characteristic calculation unit further calculates the second frequency characteristic of the servo system within the second frequency range based on the second identification input signal input to the servo system for obtaining the frequency characteristic of the servo system and the second identification output signal output from the servo system for obtaining the frequency characteristic of the servo system. The second identification input signal and the second identification output signal are signals obtained from the servo system to which the second excitation signal is input for a second sampling period of more than the first sampling period.

Images