JP2007135344A - Method and apparatus for automatically adjusting motor controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and an apparatus for quickly obtaining a proper feedback and/or feedforward control parameters. <P>SOLUTION: When the feedback control parameter is adjusted, a position instruction pattern used for a continuously changed adjustment operation is generated and provided as a position instruction of a position controller, a response frequency of the position controller and a speed controller is increased within a range in which a vibrational amplitude of a position error waveform is prevented from exceeding a predetermined value, and the maximum acceleration rate is automatically set. When the feedforward control parameter is adjusted, a plurality of operations are performed in a plurality of the position instruction pattern, and a control parameter of a position feedforward controller is adjusted so as to prevent an overshoot from exceeding the predetermined value. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an automatic adjustment method and apparatus for an electric motor control apparatus that automatically tunes control parameters in the feedback control and / or feed forward control of an electric motor control apparatus that employs feedback control and / or feed forward control.

  Patent Document 1 is known as a first conventional technique for automatically tuning a feedback control parameter and a feedforward control parameter. In the first prior art, a feedback control parameter is estimated by knowledge processing from a response shape of a control amount with respect to a control command value, and then an optimum feedforward parameter is determined from the response shape by fuzzy inference or a neural network. .

  Patent Document 2 can be cited as a second conventional technique. The second conventional technique is particularly related to the feedforward automatic adjustment method, and is characterized in that if the control amount for the control command value overshoots, the feedforward gain is decreased, and if not, the feedforward gain is increased. It is said. In addition, when an evaluation function is introduced and the value is equal to or less than a predetermined value, the adjustment is terminated.

JP-A-6-102905 (Overall) JP 2003-61377 A (Overall)

  Normally, when adjusting the feedback control parameter, the gains of the position controller and the speed controller are maximized in order to improve the control performance. However, a mechanical system composed of an electric motor, a load to be driven, and a connecting shaft has an anti-resonance frequency unique to the mechanical system, and has a property that it is very likely to vibrate near that frequency. For this reason, in actual adjustment, the controller gain is adjusted slightly lower so that the gain of the controller in the vicinity of the anti-resonance frequency does not become too high. As a simple means for realizing such delicate adjustment, there is a method of increasing the gains of the position controller and speed controller within a range where mechanical vibration does not become excessive while actually vibrating the machine at an anti-resonance frequency. Conceivable. The simplest and most reliable method for exciting a mechanical system with an unknown anti-resonance frequency at the anti-resonance frequency is to give a step command as a position command (control command value). However, in the first prior art, since the control command value is input from the outside, the anti-resonance frequency may not be sufficiently vibrated depending on the given control command value. If the gains of the position controller and speed controller are increased while evaluating the mechanical vibration as described above while repeating the positioning operation under such an excitation condition, even a slight disturbance may oscillate. There was a possibility of an excessive gain setting.

  When adjusting the feedforward control parameter, the overshoot amount is set to a predetermined value or less for a plurality of position command patterns, and the settling time is minimized for a specific position command pattern that is frequently used. Adjustment was desired. However, since the first and second prior arts do not have a position command pattern generation unit and a plurality of position command pattern registration functions in the tuning means, the adjustment conditions associated with the position command pattern as described above are set. It was difficult to do.

  An object of the present invention is to provide an automatic adjustment method and apparatus capable of obtaining appropriate feedback and / or feedforward control parameters in a short time.

  In a preferred embodiment of the present invention, in a motor control device in which a position controller, a speed controller, and a current controller are connected in cascade, a position command for adjustment operation that continuously changes as a position command value of the position controller. When the position command pattern is generated and given as a position command of the position controller, the position controller and / or the speed controller is within a range in which the vibration amplitude of the position deviation waveform does not exceed a predetermined value. Increase the response frequency.

  More specifically, the degree of change in the position command pattern is determined based on the total moment of inertia value J of the motor drive system.

  More preferably, the degree of change in the position command pattern is determined so that the speed of the motor does not exceed the allowable maximum speed, and acceleration / deceleration is performed without a constant speed period.

  In a preferred embodiment of the present invention, in addition to the cascade connection of the position controller, the speed controller, and the current controller, a position feedforward controller that inputs a position command value and outputs a position feedforward signal; In the motor control device configured to output the speed command value by adding the output of the position controller and the position feedforward signal, a plurality of position command patterns registered in advance are used as the position commands of the position controller. A plurality of driving operations are performed, and the control parameter of the position feedforward controller is adjusted so that the overshoot amount does not exceed a predetermined value in the plurality of driving operations.

In a specific embodiment of the present invention, when the feedback control parameter is adjusted, a position command pattern of a short movement distance for operating the motor near the maximum output that can be output by the motor control system is provided. Thereby, the mechanical system is sufficiently vibrated including the vicinity of the anti-resonance frequency. Further, in generating the position command pattern, the total moment of inertia value J of the drive system, taking into account the allowable maximum torque value tau _max of a motor control device, not the acceleration rate exceeds the allowable maximum torque value tau _max range Automatically set to the maximum. Further, the position command pattern does not exceed the maximum speed of the electric motor.

  As described above, the position command pattern closest to the stepped command that satisfies the condition that the inside of the control system is not saturated is automatically generated, and the feedback control parameter is adjusted.

  On the other hand, when adjusting the feedforward control parameter, a plurality of position command patterns are generated, and adjustment conditions are set for each position command pattern.

  According to the preferred embodiment of the present invention, it is possible to adjust the feedback control parameter based on the automatically generated position command pattern in a short time.

  Further, according to a preferred embodiment of the present invention, the overshoot amount can be set to a predetermined value or less for a plurality of position command patterns registered in advance by the user.

  Furthermore, the preferred embodiment of the present invention allows automatic adjustment of feedforward control parameters that minimizes settling time for specific position command patterns that are frequently used.

  Other objects and features of the present invention will become apparent from the description of the embodiments described below.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a control block diagram showing an automatic adjustment method for an electric motor control device according to a first embodiment of the present invention. The first embodiment aims at automatic adjustment of feedback control parameters. In FIG. 1, 1 is an electric motor, 2 is a drive target load driven by the electric motor 1, 3 is a connecting shaft that connects the electric motor 1 and the drive target load 2, and 4 is a power converter that drives the electric motor 1. is there. Reference numeral 5 denotes a position detector that is attached to the rotating shaft of the electric motor 1 and outputs a detected position value θ _M of the rotating shaft of the electric motor 1, and 6 is a position command value θ _M ^* and the detected position value θ _{M of the} electric motor 1. it is a subtracter for calculating a positional deviation theta _e with. 7 is a position controller that outputs a speed command value ω _M ^* in accordance with the position deviation θ _e , and 8 is a position controller θ _M that outputs a position detection value θ _M that is output from the position detector 5. a velocity calculator for outputting a speed detection value omega _M. 9 is a subtractor that calculates a speed deviation ω _e between the speed command value ω _M ^* and the detected speed value ω _M of the motor 1, and 10 is a torque current command value I _q ^* that is output according to the speed deviation ω _e. Speed controller. 11 is a current detector for detecting a torque current detection value I _q to be supplied to the electric motor 1, 12 is a current and said torque current detection value I _q to be supplied to the electric motor 1 and the torque current command value I _q ^* It is a subtractor for calculating the deviation _Ie . Reference numeral 13 denotes a current controller that adjusts the output current of the power converter 4 according to the current deviation _Ie . A feedback control parameter tuning unit 14 automatically adjusts parameters of the position controller 7 and the speed controller 10. The tuning unit inputs the position deviation θ _e described above, and outputs the position command value θ _M ^* , the position response frequency set in the position controller 7 and the speed response frequency set in the speed controller 10.

The feedback control parameter tuning unit 14 includes a feedback control parameter automatic adjustment algorithm, which will be described later, and stepped position command pattern generation means necessary for automatic adjustment of the feedback control parameter. Its basic operation is intended that the vibration amplitude of the positional deviation theta _e during stepwise position command applies the position response frequency and velocity response frequency is a predetermined value set by the user, to maximize the extent that for example, does not exceed the allowable value is there.

  FIG. 2 is a flow chart for generating a position command pattern in the embodiment of FIG.

  FIG. 3 is a waveform diagram of torque, speed, and position command pattern in a procedure for generating a position command pattern for automatic feedback control parameter adjustment.

  The above-described position command pattern automatic generation means will be specifically described with reference to the flowchart of FIG. 2 and the waveform diagram of FIG. In the flowchart of FIG. 2, a speed pattern that rotates the instructed movement amount (rotation angle for an electric motor) θmax with the maximum torque and without saturating the speed command value is generated, and finally, the position command pattern Is generated. At this time, the policy is to set a movement distance as short as possible without providing a constant speed period in the speed pattern. Thereby, it becomes possible to include a higher frequency component in the position command pattern.

In FIG. 2, the flowchart starts at 20, and the movement amount θ _max is calculated at processing 21. The amount of movement theta _max above has dealt with better short as possible, setting extremely the amount of movement theta _max short, regardless of the ratio of the vibration amplitude of the positional deviation theta _e with respect to the moving amount theta _max, the absolute value It may become too small. In this case, the vibration amplitude falls below the resolution of the position detector 5, and vibration cannot be detected. Therefore, the movement amount θ _max is determined by the equation (1).

θ _max = α · θ _vib (1)
In equation (1), θ _vib is an allowable value of the vibration amplitude of the positional deviation θ _e set by the user. That is, the vibration amplitude of the positional deviation theta _e is less than theta _vib, positional deviation theta _e denotes the threshold value for determining that not vibrating. Further, α is a positive constant set by the user, and means the ratio of the movement amount θ _max to the allowable value θ _vib of the vibration amplitude of the position deviation θ _e . For example, if α = 100 is set, the movement amount θ _max is calculated so that θ _vib is 1/100 of the movement amount θ _max , and therefore vibration with an amplitude of 1/100 or more of the movement amount θ _max is certain. Can be detected. Usually, a value of 100 or more is set to α empirically.

Next, in the process 22, the ideal maximum speed ω _peak of the motor when accelerating at the allowable maximum torque value τ _max determined by the combination of the motor control device and the motor is calculated by the equation (2).

但し、（２）式において、τ_ｍａｘは前記許容最大トルク値、θ_ｍａｘは前記移動量、Ｊは前記電動機と前記連結軸と前記駆動対象負荷の慣性モーメント値を合計した合計慣性モーメント値である。次に、判定処理２３において、前記電動機の理想最大速度ω_ｐｅａｋと電動機の許容最大速度ω_ｍａｘを比較し、理想最大速度ω_ｐｅａｋが電動機の許容最大速度ω_ｍａｘを下回る場合には、処理２４に移行する。処理２４は、許容最大トルクτ_ｍａｘで加速しても、電動機の許容最大速度ω_ｍａｘを超えない場合である。

In the equation (2), τ _max is the allowable maximum torque value, θ _max is the movement amount, and J is a total inertia moment value obtained by summing the inertia moment values of the electric motor, the connecting shaft, and the drive target load. . Then, in the determination process 23, when comparing the permissible maximum speed omega _max ideal maximum speed omega _peak and the motor of the electric motor, the ideal maximum speed omega _peak is below the permissible maximum speed omega _max of the motor, the processing 24 Transition. Process 24 is a case where even if acceleration is performed with the allowable maximum torque τ _max , the allowable maximum speed ω _max of the electric motor is not exceeded.

FIG. 3A shows a case where the maximum allowable speed ω _max of the electric motor is not exceeded even if the maximum allowable torque τ _max is accelerated. That is, as described above, first, the movement amount θ _max is determined by α times the set allowable vibration amplitude θ _vib . The distance to accelerate the allowable maximum torque tau _max, without placing the constant speed period and is decelerated at the maximum permissible torque tau _max without a missing a beat, the positive and negative torque 41, the speed pattern 40 is obtained, whereby The position command pattern 42 within the movement amount θ _max is obtained. If the ideal maximum speed ω _peak of the motor at this time does not exceed the allowable maximum speed ω _max of the motor, the speed pattern 40 can be used as it is. Therefore, in process 24, the maximum speed is set to the ideal maximum speed ω _peak. Then, the process proceeds to process 25. In step 25, it calculates the deceleration time _{t a} to accelerate the allowable maximum torque value tau _max until the ideal maximum speed ω _peak (3) by the formula, after the setting, the process proceeds to 26, the process ends.

次に、判り易くするため、電動機の許容最大速度が極端に小さいケースを想定して、図３（ｂ）を参照して説明する。今、仮定により、前記判定処理２３において、理想最大速度ω_ｐｅａｋが、電動機の小さな許容最大速度ω_ｍａｘを上回ってしまうので、処理２７に移行する。処理２７では、許容最大トルクτ_ｍａｘで加速すると、電動機の許容最大速度ω_ｍａｘを超えてしまう場合であり、止む無く図３（ｂ）に示すように、最大速度として前記電動機の許容最大速度ω_ｍａｘに設定し、処理２８に移行する。すなわち、速度パターンは符号４３で示すようになり、これを時間積分した移動距離θ_ＳＬは、前記移動量θ_ｍａｘには達し得ない。処理２８では、許容最大トルク値τ_ｍａｘで電動機の許容最大速度ω_ｍａｘまで加速する加減速時間ｔ_ｂを（４）式で算出し設定後、２６に移行し、処理を終了する。

Next, in order to make it easy to understand, the case where the allowable maximum speed of the electric motor is extremely small is assumed and described with reference to FIG. Assuming that the ideal maximum speed ω _peak exceeds the small allowable maximum speed ω _max of the electric motor in the determination process 23, the process _proceeds to the process 27. In the process 27, when the maximum allowable torque τ _max is accelerated, the maximum allowable speed ω _max of the electric motor is exceeded, and as shown in FIG. 3B, the maximum allowable speed ω of the electric motor is used as the maximum speed. Set to _max and _proceed to processing 28. That is, the speed pattern is as indicated by reference numeral 43, and the movement distance θ _SL obtained by time integration of the speed pattern cannot reach the movement amount θ _max . In process 28, after setting calculates the deceleration time t _b to accelerate the allowable maximum torque value tau _max maximum allowed speed omega _max of the motor (4) below, the process proceeds to 26, the process ends.

t _b = J · ω _max / τ _max (4)
The amount of movement θ _SL of the electric motor at this time can be expressed by equation (5).

また、このときに、電動機が出力するトルクの大きさτ_ＳＬは、下記（６）式で表すことができ、許容最大トルクτ_ｍａｘよりも小さくなることが分かる。

Further, at this time, the magnitude τ _SL of the torque output from the electric motor can be expressed by the following equation (6), and it can be seen that it is smaller than the allowable maximum torque τ _max .

図４は、第１の実施例によるフィードバック制御パラメータの自動調整フローチャートである。この図を用いて、フィードバック制御パラメータチューニング部１４におけるフィードバック制御パラメータの自動調整アルゴリズムを説明する。図４において、Ｆｓ＿ｍｉｎは、ユーザが設定する最小速度応答周波数、Ｆｓ＿ｍａｘはユーザが設定する最大速度応答周波数、ｄｉｖ＿Ｆｓはユーザが設定する速度応答周波数増分ステップである。後述する調整時には、速度応答周波数をＦｓ＿ｍｉｎからＦｓ＿ｍａｘまでｄｉｖ＿Ｆｓのステップで増加しながら調整することを意味する。同様に、Ｆｐ＿ｍｉｎは、ユーザが設定する最小位置応答周波数、Ｆｐ＿ｍａｘはユーザが設定する最大位置応答周波数、ｄｉｖ＿Ｆｐはユーザが設定する位置応答周波数増分ステップである。後述する調整時には、位置応答周波数をＦｐ＿ｍｉｎからＦｐ＿ｍａｘまでｄｉｖ＿Ｆｐのステップで増加しながら調整することを意味する。また、θ_ｖｉｂは、ユーザが設定する位置偏差θ_ｅの振動振幅許容値である。Ｆｓは、前記速度制御器１０に設定されている現在の速度応答周波数、Ｆｐは前記位置制御器７に設定されている現在の位置応答周波数である。Ｆｐ＿ｖ０は、振動限界の最大位置応答周波数であり、位置偏差波形の振動振幅が許容値θ_ｖｉｂ以下となる最大の位置応答周波数を記録する。Ｆｓ＿ｖｌは、調整により最終的に求まる最適速度応答周波数、Ｆｐ＿ｖｌは調整により最終的に求まる最適位置応答周波数、ｆｌａｇ＿ｆｐｍａｘは‘１’の場合に位置偏差波形の振動振幅が許容値θ_ｖｉｂ以下のまま最大位置応答に達したことを示すフラグである。

FIG. 4 is a flowchart for automatically adjusting the feedback control parameter according to the first embodiment. The feedback control parameter automatic adjustment algorithm in the feedback control parameter tuning unit 14 will be described with reference to FIG. In FIG. 4, Fs_min is a minimum speed response frequency set by the user, Fs_max is a maximum speed response frequency set by the user, and div_Fs is a speed response frequency increment step set by the user. In the adjustment described later, this means that the speed response frequency is adjusted while increasing in steps of div_Fs from Fs_min to Fs_max. Similarly, Fp_min is a minimum position response frequency set by the user, Fp_max is a maximum position response frequency set by the user, and div_Fp is a position response frequency increment step set by the user. In the adjustment described later, this means that the position response frequency is adjusted while increasing in steps of div_Fp from Fp_min to Fp_max. Further, θ _vib is a vibration amplitude allowable value of the position deviation θ _e set by the user. Fs is a current speed response frequency set in the speed controller 10, and Fp is a current position response frequency set in the position controller 7. Fp_v0 is the maximum position response frequency at the vibration limit, and records the maximum position response frequency at which the vibration amplitude of the position deviation waveform is equal to or less than the allowable value θ _vib . Fs_vl is the optimum speed response frequency finally obtained by adjustment, Fp_vl is the optimum position response frequency finally obtained by adjustment, and flag_fpmax is maximum when the vibration amplitude of the position deviation waveform remains below the allowable value θ _vib when “1”. A flag indicating that the position response has been reached.

  The basic idea of the adjustment according to this flowchart is to set the position response frequency and the velocity response frequency as high as possible within a range that does not induce vibration in the position deviation waveform. In addition, in the process of gradually increasing the position response frequency, even if vibration occurs in the position deviation waveform, it may be possible to reduce the vibration of the position deviation waveform by slightly increasing the speed response frequency. I know it. Therefore, in order to make use of this empirical rule, when vibration occurs in the position deviation waveform, the speed response frequency is increased by div_Fs and the vibration is re-evaluated.

Next, the flowchart of FIG. 4 will be described in order. In FIG. 4, the flowchart starts at 50, and variable initialization is performed at process 51. In the initialization process 51, Fs_min is substituted for Fs, and the speed response frequency is set to start adjustment from the minimum speed response frequency Fs_min. Similarly, Fp_min is substituted for Fp, and the position response frequency is set to start adjustment from the minimum position response frequency Fp_min. Further, the maximum position response frequency Fp_v0 of the vibration limit is initialized with the minimum position response frequency Fp_min, and the flag flag_fpmax is initialized with “0”. Then, the process proceeds to the process 52, to measure the vibration amplitude of the positional deviation theta _e in applying as a position command value theta _M ^* in FIG. 1 a stepwise position command generated in the above-mentioned means. The measuring means of the vibration amplitude of the position deviation theta _e will be described later. Then, the process proceeds to the determination process 53, the vibration amplitude of the positional deviation theta _e determines whether the allowable value theta _vib below. If the vibration amplitude is equal to or _smaller than the allowable value θ _{vib in the} determination process 53, there is room for further increasing the position response frequency, and the process proceeds to the determination process 54. In the determination process 54, it is determined whether or not the current position response frequency Fp is less than the maximum position response frequency Fp_max. In process 55, the current position response frequency Fp is increased by the position response frequency increment step div_Fp, and the process proceeds to the process 52. On the other hand, when the current position response frequency Fp is greater than or equal to the maximum position response frequency Fp_max in the determination process 54, the process proceeds to process 56. In the process 56, the flag flag_fpmax is set to “1”, it is clearly shown that the maximum position response has been reached while the vibration amplitude of the position deviation waveform is not more than the allowable value θ _vib , and the process proceeds to the determination process 57. In the determination process 57, it is determined whether or not the current speed response frequency Fs is less than the maximum speed response frequency Fs_max. In process 58, the current speed response frequency Fs is increased by the speed response frequency increment step div_Fs, and the process shifts to the process 52 while reducing the vibration amplitude of the position deviation waveform. When the current speed response frequency Fs is greater than or equal to the maximum speed response frequency Fs_max in the determination process 57, the vibration amplitude of the position deviation waveform reaches the maximum value with the allowable value θ _vib or less for both the position response frequency and the speed response frequency. The process proceeds to processing 59. In the process 59, the current speed response frequency Fs and the current position response frequency Fp are assigned to the optimum speed response frequency Fs_vl and the optimum position response frequency Fp_vl, respectively.

If the vibration amplitude is not equal to or _smaller than the allowable value θ _{vib in the} determination process 53, the process _proceeds to the determination process 61. In the determination process 61, the flag flag_fpmax is checked to check whether or not the vibration amplitude of the position deviation waveform has reached the maximum position response with the allowable value θ _vib or less. Here, when flag_fpmax = 0, the vibration amplitude of the position deviation waveform remains below the allowable value θ _vib and reaches the maximum position response, and then the speed response until the vibration amplitude of the position deviation waveform exceeds the allowable value θ _vib. This means that the frequency has been increased, and the processing shifts to processing 67. In the process 67, in order to set a condition that the vibration amplitude of the position deviation waveform is equal to or less than the allowable value θ _vib with respect to the optimum speed response frequency Fs_vl, only the speed response frequency increment step div_Fs from the current speed response frequency Fs. After subtracting the subtracted value, the process proceeds to 60 and ends. If flag_fpmax = 0 in the determination process 61, the process proceeds to the determination process 62, and the current position response frequency Fp is compared with the maximum position response frequency Fp_v0 of the vibration limit. Here, when “Fp> Fp_v0” is not satisfied, it means that there is no condition that the current position response frequency Fp exceeds the maximum position response frequency Fp_v0 of the vibration limit, and the process proceeds to the process 67. If “Fp> Fp_v0” is established in the determination process 62, the process proceeds to the determination process 63. In the determination process 63, it is determined whether or not the current speed response frequency Fs is less than the maximum speed response frequency Fs_max. In the process 64, the current speed response frequency Fs is increased by the speed response frequency increment step div_Fs to reduce the vibration amplitude of the position deviation waveform. Further, the current position response frequency Fp is decreased by the position response frequency increment step div_Fp so that the vibration amplitude of the position deviation waveform is always equal to or less than the allowable value θ _vib , and the process proceeds to processing 65. In process 65, the current position response frequency Fp is recorded as the maximum position response frequency Fp_v0 of the vibration limit, and the process proceeds to process 52 in order to further increase the position response frequency. If the current speed response frequency Fs is greater than or equal to the maximum speed response frequency Fs_max in the determination process 63, it means that the current speed response frequency Fs cannot be increased any further, and the process proceeds to process 66. In process 66, the current speed response frequency Fs is substituted for the optimum speed response frequency Fs_vl. Furthermore, a value obtained by reducing the current position response frequency Fp by the position response frequency increment step div_Fp is substituted for the optimum position response frequency Fp_vl so that the vibration amplitude of the position deviation waveform is always equal to or less than the allowable value θ _vib. After that, the process proceeds to 60 and ends.

  FIG. 5 and FIG. 6 are measurement processing flowcharts of vibration amplitude, settling time, and overshoot amount in the first embodiment. A method of measuring the vibration amplitude, settling time, and overshoot amount of the position deviation waveform required in the feedback control parameter tuning unit 14 and the position feedforward control parameter tuning unit 17 described later will be described using this flow. 5 and 6, timeout is a settling monitoring timeout time set by the user, and is an adjustment parameter for measuring the settling time by timeout from the time point when the position deviation waveform first crosses zero. Posin_pls is a settling determination deviation set by the user, and is set when the absolute value of the position deviation is stably equal to or less than posin_pls. Further, the position error is the position deviation, and the position error poser_work is a position deviation for vibration / overshoot measurement. The position deviation immediately after the stepwise position command is applied always starts from the positive side regardless of the rotation direction of the motor. Is a state quantity obtained by adjusting the sign. flag_poserr_plus is a flag indicating that the positional deviation immediately after application of the step-like position command starts from positive. When the positional deviation starts from positive, flag_poserr_plus = 1, and when the positional deviation starts from negative, flag_poserr_plus = 0. And Therefore, when flag_poser_plus = 1, poser_work = poserr, and when flag_poser_plus = 0, poser_work = −poserr. Poserr_min is a minimum position deviation value, and is a variable that always holds the minimum value of the poser_work at that point in time from immediately after the change of the step-like position command to the end of the settling time measurement. “poserr_vib” is a vibration amplitude maximum value in the positional deviation, and is a variable that always holds the maximum value from the vibration amplitude of the poser_work calculated by “poserr_work-poserr_min”. flag_plsin is a settling judgment deviation arrival flag. If the absolute value of the position deviation is equal to or less than the settling judgment deviation posin_pls, flag_plsin = 1 is set. Otherwise, flag_plsin = 0 is set. flag_plsin_bk is the previous value of the settling determination deviation arrival flag, and is a flag equivalent to the flag_plsin set previously. time_s is the measurement elapsed time, the elapsed time starting immediately after the stepped position command has been changed, and time_w is the settling monitoring elapsed time, and the elapsed time starting from the time when the position deviation waveform first crosses zero. Show. St is the settling time, and the measured elapsed time time_s is substituted for St each time the absolute value of the position deviation becomes less than posin_pls from a value larger than the settling determination deviation posin_pls. Thereby, if the settling monitoring timeout time timeout is set appropriately long, the time required until the position deviation is finally settled from immediately after the stepped position command is changed is recorded in St. It will be. “over_shot” is an overshoot amount with respect to the position command value of the motor position, and is equal to the sign inversion value of the above-mentioned porr_min after settling.

Next, the flowcharts of FIGS. 5 and 6 will be described in order. 5 and 6, the flowchart is started at 80, and a change in the position command θ _M ^* is confirmed in the determination process 81. Here, since the position command does not change until the step-shaped position command is input, the determination processing 81 is repeatedly executed. However, when the step-shaped position command is input, the “position command θ _M ^* is set. It is determined that there is a change ”, and the process proceeds to determination process 82. In the determination process 82, it is determined whether or not the position command is constant. Here, the position command becomes constant when the step-shaped position command completely rises, and the process proceeds to the determination process 83. On the other hand, while the step-like position command is rising, the determination process 82 is repeatedly executed. In the determination process 83, the sign of the positional deviation poser is confirmed. If it is positive, the process proceeds to process 84. In process 84, “1” is set to a flag flag_poserr_plus indicating that the position deviation starts from positive, and the process proceeds to process 85. In the process 85, the position deviation minimum value poerr_min is initialized with the position deviation poerr, and the process proceeds to an initialization process 86. In the determination process 83, if the position deviation “poserr” is negative, the process proceeds to process 87, “0” is set to a flag flag_poserr_plus indicating that the position deviation starts from positive, and the process proceeds to process 88. In the process 88, the position deviation minimum value poerr_min is initialized with the sign inversion value -poserr of the position deviation poser, and the process proceeds to an initialization process 86. In the initialization process 86, the vibration amplitude maximum value during the position deviation poser_vib, the measurement elapsed time time_s, the settling monitoring elapsed time time_w, and the settling judgment deviation arrival flag previous value flag_plsin_bk are initialized to zero, and the process proceeds to the process 90. In the process 90, 1 is added to the measurement elapsed time time_s, and the process proceeds to the determination process 91. In the determination process 91, if the absolute value of the position deviation poser is equal to or less than the settling determination deviation posin_pls, the process proceeds to process 92, the settling determination deviation arrival flag flag_plsin is set to “1”, and the process proceeds to the determination process 93. In the determination process 93, it is determined whether or not the settling determination deviation arrival flag previous value flag_plsin_bk is zero. If flag_plsin_bk = 0, the process proceeds to process 95, the measured elapsed time time_s is recorded in the settling time St, and the process proceeds to process 95. To do. In the process 95, in preparation for the next position deviation determination process 91, the value of the current settling determination deviation arrival flag flag_plsin is substituted into the previous set flag determination deviation flag flag_plsin_bk, and the process proceeds to the determination process 97. In the determination process 91, when the absolute value of the position deviation poser is not less than or equal to the settling determination deviation posin_pls, the process proceeds to process 96, the settling determination deviation arrival flag flag_plsin is set to “0”, and the process 95 is set. Transition. In the determination process 93, if flag_plsin_bk = 1, the settling time St does not need to be updated because the settling has already been completed, and the process proceeds to the process 95. In the determination process 97, a flag flag_poserr_plus indicating that the position deviation starts from positive is checked. If flag_poserr_plus = 1, it is determined that the positional deviation starts from positive, and the positional deviation poerr is directly substituted into the vibration / overshoot measurement positional deviation poerr_work. In the determination process 97, if flag_poserr_plus = 0, it is determined that the position deviation starts from a negative value, and the sign-inverted value -poserr of the position deviation poser is substituted into the vibration / overshoot measurement position deviation poerr_work. . In the determination process 99, it is determined whether or not the vibration / overshoot measurement position deviation poerr_work is equal to or smaller than the position deviation minimum value poerr_min. If it is below, the process proceeds to process 101, the position deviation minimum value poerr_min is updated with the current position deviation poerr_work for vibration / overshoot measurement, and the process proceeds to the determination process 102. In the determination process 99, when the vibration / overshoot measurement position deviation poerr_work is not less than or equal to the minimum position deviation value poerr_min, the process proceeds to the determination process 102 without updating the minimum position deviation value poerr_min. In the determination process 102, the vibration amplitude maximum value poerr_vib in the positional deviations measured so far is compared with “poserr_work-poserr_min”. If the position of the poser_vib is smaller, the position of the porrer_vib is updated with “poser_work-poserr_min” and the process proceeds to the determination process 104. In the determination process 102, if the position of the poerr_vib is larger, the process proceeds to the determination process 104 without updating the position of the poser_vib. In the determination process 104, in order to detect that the position deviation has crossed zero, it is evaluated whether or not the position deviation minimum value poerr_min is equal to or less than zero. Here, if poser_min ≦ 0, it is determined that the position deviation has crossed zero in the past, and the process proceeds to processing 105. In the process 105, 1 is added to the settling monitoring elapsed time time_w representing the elapsed time starting from the time when the position deviation waveform first crosses zero, and the process proceeds to the determination process 106. If the determination process 104 does not satisfy poser_min ≦ 0, it is determined that the position deviation waveform has never zero-crossed, and the process proceeds to process 90. In the determination process 106, the settling monitoring elapsed time time_w starting from immediately after the change of the stepped position command is compared with the settling monitoring timeout time timeout, and if time_w ≧ timeout, the process proceeds to process 107 to end the settling monitoring. To do. On the other hand, in the determination process 106, when time_w ≧ timeout is not established, the process proceeds to process 90 in order to continue the settling monitoring. In the process 107, the sign inversion value -poserr_min of the position deviation minimum value poerr_min is set as the overshoot amount over_shot, and the process proceeds to 108, where the process of measuring the vibration amplitude, settling time, and overshoot amount of the position deviation waveform is completed. To do.

FIG. 7 is a waveform diagram for explaining the measurement method of the vibration amplitude, settling time, and overshoot amount in the first embodiment, and the results of the flowcharts of FIGS. Show. In FIG. 7, a waveform 120 is a position command value θ _M ^* , and a waveform 121 is a position deviation poser and a position deviation poerr_work for vibration / overshoot measurement. A waveform 122 is a settling judgment deviation arrival flag flag_plsin, a waveform 123 is a position deviation minimum value porrer_min, and a waveform 124 is poser_work-poserr_min. The horizontal axis of each waveform is the measurement elapsed time time_s, and the time point when the waveform 120 indicating the position command completely rises is time_s = 0. In this case, since the position command waveform 120 changes in the positive direction, the position error posr_time when time_s ≧ 0 is equal to the position deviation positionr. Here, if the position command waveform 120 has changed in the negative direction, the waveform 121 indicating the position error_work is not changed, and only the position deviation position is a waveform obtained by converting the waveform 121 into line symmetry with respect to the time axis. In addition, the waveform 122 indicating the settling judgment deviation arrival flag flag_plsin takes “1” if the waveform 121 showing the porrer_work exists between −posin_pls and posin_pls in the graph, and takes “0” otherwise. When the position deviation waveform is oscillating as in this example, the settling judgment deviation arrival flag flag_plsin is set to a sufficiently long settling monitoring timeout time timeout after the round trip between “0” and “1” several times. If set, it will eventually settle to “1”. Further, the settling time St is updated with the measured elapsed time time_s for each rising edge (arrow) of the settling determination deviation arrival flag flag_plsin. The settling monitoring elapsed time time_w is an elapsed time for measuring the settling monitoring timeout time timeout, and starts from the time when the position deviation waveform 121 first crosses zero as shown by the time axis 125 in the figure. In this way, apart from the measurement elapsed time time_s, a settling monitoring elapsed time time_w starting from the time point when the position deviation waveform 121 first zero-crosses is newly provided. The reason is that the time required from the time when the position deviation waveform zero-crosses to the settling time is not significantly affected by the control parameters such as the position response frequency and the speed response frequency. Thereby, when setting the settling monitoring timeout time timeout by the user, it is possible to accurately measure the settling time St only by setting a constant value without considering control parameters such as position response frequency and speed response frequency. . The waveform 123 is a position deviation minimum value poerr_min, which is easy to understand when compared with the vibration / overshoot measurement position deviation poerr_work, but indicates the minimum value of each position of the poser_work. The waveform 124 is poser_work-poserr_min, and is a waveform obtained by extracting only the vibration component included in the vibration / overshoot measurement position deviation poerr_work. This can be understood from the fact that poser_work-poserr_min is always zero if poser_work is a gradually decreasing waveform without vibration. In addition, the vibration amplitude maximum value poerr_vib in the position deviation to be finally obtained is a peak hold value of the waveform 124.

  As described above, by using the flowcharts of FIGS. 5 and 6, it is possible to measure the vibration amplitude, settling time, and overshoot amount of the position deviation waveform necessary for the feedback control parameter tuning unit 14. In the position feedforward control parameter tuning unit 17 of the second embodiment described later, the vibration amplitude, settling time, and overshoot amount of the position deviation waveform can be used.

FIG. 8 is a control block diagram including an automatic adjustment unit for the feedforward control parameter of the motor control apparatus according to the second embodiment of the present invention. In the second embodiment, automatic adjustment of the position feedforward control parameter is also performed in addition to the automatic adjustment of the feedback control parameter of FIG. In FIG. 8, the same function parts as those in FIG. 7 is a position controller that outputs a position controller output signal ω _FB ^* in accordance with the position deviation θ _e , and 15 is a position that receives a position command value θ _M ^* and outputs a position feedforward controller output signal ω _FF ^*. This is a feedforward controller. The input / output relationship can be expressed by equation (7) using a feedforward gain Kff and a feedforward time constant Tff, which are parameters of the controller, where Laplace operator is s.

１７は、図１の１４と同様に、位置制御器７および速度制御器１０のパラメータを自動調整するほかに、位置フィードフォワード制御器１５におけるフィードフォワードゲインＫｆｆおよび時定数Ｔｆｆを自動調整する制御パラメータチューニング部である。したがって、１４に加わる機能として、位置偏差θ_ｅを入力し、位置指令値θ_Ｍ ^＊と位置フィードフォワード制御器１５に設定するフィードフォワードゲインＫｆｆおよびフィードフォワード時定数Ｔｆｆをも出力することとなる。また、制御パラメータチューニング部１７の内部には、後述する位置フィードフォワード制御パラメータの自動調整アルゴリズムと、予めユーザが登録した複数の位置指令パターンを順次、位置指令値θ_Ｍ ^＊として出力する機能をも有している。その基本動作は、複数の位置指令パターンの全てに対して、実際に運転動作を行い、全ての位置指令パターンに関してオーバーシュート量が許容値を超えない条件を満たし、且つ、特定の位置指令パターンにおける整定時間を最小化する。このとき、位置フィードフォワード制御器のフィードフォワードゲインＫｆｆおよびフィードフォワード時定数Ｔｆｆの自動探索を行うものである。

17 is a control parameter for automatically adjusting the feedforward gain Kff and the time constant Tff in the position feedforward controller 15 in addition to automatically adjusting the parameters of the position controller 7 and the speed controller 10 in the same manner as 14 in FIG. It is a tuning part. Therefore, as a function added to 14, the position deviation θ _e is input, and the position command value θ _M ^* and the feed forward gain Kff and the feed forward time constant Tff set in the position feed forward controller 15 are also output. Further, the control parameter tuning unit 17 has an automatic adjustment algorithm for a position feedforward control parameter, which will be described later, and a function for sequentially outputting a plurality of position command patterns registered in advance as a position command value θ _M ^*. Have. The basic operation is to actually perform the driving operation for all of the plurality of position command patterns, satisfy the condition that the overshoot amount does not exceed the allowable value for all the position command patterns, and in a specific position command pattern. Minimize settling time. At this time, an automatic search for the feedforward gain Kff and the feedforward time constant Tff of the position feedforward controller is performed.

  Next, an automatic adjustment method of the feedforward gain Kff will be specifically described. First, the concept will be described with reference to FIG.

  FIG. 9 is an explanatory diagram of the relationship between the feedforward gain and the settling time / overshoot in the embodiment of FIG. 8, where FIG. 9A shows the relationship between the feedforward gain and the settling time, and FIG. The relationship between the forward gain and the overshoot amount is shown. In FIG. 9A, the horizontal axis is the feedforward gain Kff, and the vertical axis is the settling time. Basically, the larger the feedforward gain Kff, the shorter the settling time. As shown in the figure, when the feedforward gain Kff exceeds a certain value, the settling time is abruptly increased. On the other hand, in FIG. 9B, the horizontal axis is the feedforward gain Kff, the vertical axis is the overshoot amount, and no overshoot occurs while the feedforward gain Kff is small. However, when the feedforward gain Kff becomes a certain value or more, overshoot starts to occur, and thereafter, the generation amount indicates a monotonous increase with respect to Kff.

  9A and 9B, when the overshoot amount is zero or sufficiently small, it can be said that the settling time is shorter when the feed forward gain Kff is set larger. Further, since the overshoot amount increases as the feedforward gain Kff is set larger, in order to minimize the settling time, the feedforward gain Kff must be maximized within a range where the overshoot amount does not exceed the allowable value. I understand that

  Therefore, in the processing flowchart described below, the maximum feedforward gain Kff that satisfies the condition that the overshoot amount does not exceed the allowable value is searched for all the position command patterns registered in advance.

  FIG. 10 is a process flow diagram of the automatic adjustment method of the feedforward gain Kff in the second embodiment. Note that the automatic adjustment method of the feedforward time constant Tff will be described later because automatic adjustment of the feedforward gain Kff is required in the adjustment process.

  In FIG. 10, Kff_ini is a feedforward gain initial value set by the user, dmax_Kff is a feedforward gain search maximum step set by the user, and dmin_Kff is a feedforward gain search minimum step set by the user. Kff is a current feedforward gain set in the position feedforward controller 15, div_Kff is a current feedforward gain search step, and Kff_bd is a limit feedforward gain to be finally obtained. At the time of adjustment to be described later, the search is started from Kff = Kff_ini, and if Kff = Kff_ini and the overshoot amount is less than the allowable value, Kff is increased by a step dmax_Kff. If Kff = Kff_ini and the overshoot amount is greater than or equal to the allowable value, Kff is decreased in a step dmax_Kff. Further, when the overshoot amount changes from less than the allowable value to the allowable value or more during the Kff update in the dmax_Kff step, or the overshoot amount changes from the allowable value to the allowable value, the following processing is performed. To do. That is, it means that the process of increasing / decreasing Kff while halving the current feedforward gain search step div_Kff is performed until div_Kff falls below the feedforward gain search minimum step dmin_Kff.

  Next, the flowchart of FIG. 10 will be described in order. In FIG. 10, the flowchart is started at 150, and an initialization process 151 sets the feedforward gain initial value Kff_ini to the current feedforward gain Kff. Then, the feedforward gain search maximum step dmax_Kff is set in the current feedforward gain search step div_Kff, and the process proceeds to processing 152. In the process 152, the operation is performed with all the position command patterns registered in advance by the user, the settling time and the overshoot amount are measured by the above-mentioned means for each operation, and the process proceeds to the determination process 153. In the determination process 153, the overshoot amount for each operation pattern is evaluated. If the overshoot amount in any operation pattern is equal to or greater than the allowable value, the process proceeds to process 154. In process 154, the current feedforward gain Kff is reduced by the current feedforward gain search step div_Kff, and then the process proceeds to process 155. In the process 155, similarly to the process 152, the operation is performed with all the position command patterns registered in advance by the user, the settling time and the overshoot amount are measured for each operation, and the process proceeds to the determination process 156. In the determination process 156, the overshoot amount for each operation pattern is evaluated. If the overshoot amount in all operation patterns is less than the allowable value, the process proceeds to process 157. On the other hand, in the determination process 156, when the overshoot amount in any of the operation patterns is equal to or larger than the allowable value, the process proceeds to the process 154 for the purpose of further reducing the current feedforward gain Kff. In the process 157, the limit feedforward gain Kff_bd is updated with the current feedforward gain Kff as one of the maximum Kff candidates for which the overshoot amount in all operation patterns is less than the allowable value, and the process proceeds to the process 158. . In the process 158, considering the subsequent flow, the current gain search step div_Kff is added to the current feedforward gain Kff, and the condition that the overshoot amount in any one of the operation patterns is equal to or larger than the allowable value is reset. The process proceeds to process 159. In the determination process 153, when the overshoot amount in all operation patterns is less than the allowable value, the process proceeds to process 160. In the process 160, the limit feedforward gain Kff_bd is updated with the current feedforward gain Kff as one of the maximum Kff candidates for which the overshoot amount in all operation patterns is less than the allowable value, and the process proceeds to the process 161. . In the process 161, the current feedforward gain search step div_Kff is added to the current feedforward gain Kff, and then the process 162 is performed. In the process 162, similarly to the process 155, the operation is performed with all the position command patterns registered in advance by the user, the settling time and the overshoot amount are measured for each operation, and the process proceeds to the determination process 163. In the determination process 163, the overshoot amount for each operation pattern is evaluated. If the overshoot amount in any of the operation patterns is equal to or greater than the allowable value, the process proceeds to the process 159. On the other hand, in the determination process 163, when the overshoot amount in all operation patterns is less than the allowable value, the process proceeds to the process 160 for the purpose of further increasing the current feedforward gain Kff. In the flow following the process 159, the current feedforward gain search step div_Kff is halved while the Kff is increased or decreased. In the process 159, the current feedforward gain search step div_Kff is halved, and the determination process 165 is performed. Migrate to In the determination process 165, it is determined whether or not the current feedforward gain search step div_Kff is greater than or equal to the feedforward gain search minimum step dmin_Kff. If so, the process proceeds to process 166, and if less, the process proceeds to 176, and the search process for the feedforward gain Kff is terminated. In process 166, the current feedforward gain Kff is reduced by the current feedforward gain search step div_Kff, and then the process proceeds to process 167. In the process 167, similarly to the process 162, the operation is performed with all the position command patterns registered in advance by the user, the settling time and the overshoot amount are measured for each operation, and the process proceeds to the determination process 168. In the determination process 168, the overshoot amount for each operation pattern is evaluated, and if the overshoot amount in all operation patterns is less than the allowable value, the process proceeds to process 169. On the other hand, in the determination process 168, when the overshoot amount in any of the operation patterns is equal to or larger than the allowable value, the process proceeds to the process 159 for the purpose of further reducing the current feedforward gain Kff. In the process 169, the limit feedforward gain Kff_bd is updated with the current feedforward gain Kff as one of the candidates for the maximum Kff that makes the overshoot amount in all operation patterns less than the allowable value, and the process proceeds to the process 170. . In the process 170, the current feedforward gain search step div_Kff is halved, and the process proceeds to the determination process 171. In the determination process 171, it is determined whether or not the current feedforward gain search step div_Kff is greater than or equal to the feedforward gain search minimum step dmin_Kff. If so, the process proceeds to process 172. If less, the process proceeds to 176, and the feed forward gain Kff search process is terminated. In process 172, the current feedforward gain search step div_Kff is added to the current feedforward gain Kff, and then the process proceeds to process 173. In the process 173, similarly to the process 167, the operation is performed with all the position command patterns registered in advance by the user, the settling time and the overshoot amount are measured for each operation, and the process proceeds to the determination process 174. In the determination process 174, the amount of overshoot for each operation pattern is evaluated. If the amount of overshoot in any operation pattern exceeds the allowable value, the current feedforward gain Kff is further reduced. The process proceeds to process 159. On the other hand, in the determination process 174, when the overshoot amount in all operation patterns is less than the allowable value, the process proceeds to process 175. In the process 175, the limit feedforward gain Kff_bd is updated with the current feedforward gain Kff as one of the candidates for the maximum Kff that makes the overshoot amount in all operation patterns less than the allowable value, and the current feedforward is further updated. For the purpose of increasing the gain Kff, the processing shifts to the processing 170.

  As described above, using the flowchart of FIG. 10, the maximum feedforward gain Kff that satisfies the condition that the overshoot amount does not exceed the allowable value is found as the limit feedforward gain Kff_bd for all position command patterns registered in advance. Further, the settling time at that time is minimized if the allowable overshoot value is sufficiently small.

  9A and 9B specifically show the adjustment process according to the flowchart of FIG. The adjustment parameters at this time are Kff_ini = 0.10, dmin_Kff = 0.005, dmax_Kff = 0.10, and overshoot allowable value = 2 [pulse]. Searches are performed in the order of (1) to (6) in FIG. 9, and finally div_Kff = 0.003125 is set in processing 170. For this reason, it becomes No in the next determination processing 171 and shifts to END.

  FIG. 11 is an explanatory diagram of the relationship between the feedforward time constant, the limit feedforward gain Kff_bd, and the settling time in the second embodiment. The relationship between the feedforward time constant and the settling time in the limit feedforward gain (maximum value) shown in FIG. 11A and the relationship between the feedforward time constant and the limit feedforward gain shown in FIG.

  11A, the horizontal axis is the feedforward time constant Tff, the vertical axis is the settling time in the limit feedforward gain, and FIG. 11B is the horizontal axis, the feedforward time constant Tff, and the vertical axis is the feedforward gain. is there. Also, three different position command patterns are selected as parameters for each graph. Here, the limit feedforward gain Kff_bd is the maximum feedforward gain Kff that satisfies the condition that the overshoot amount does not exceed the allowable value for each position command pattern. Specifically, the numerical values on each waveform in FIG. 11B are obtained by executing the flowchart of FIG. 10 with respect to each individual position command pattern under the condition that the feedforward time constant Tff is fixed. . The numerical value on each waveform in FIG. 11A is a record of the settling time St_bd at the limit feedforward gain Kff_bd thus obtained. Here, it is clear from FIGS. 11A and 11B that the optimum feedforward time constant that gives the minimum settling time differs depending on the position command pattern. For example, from FIG. 11A, the optimum feedforward time constant of the position command pattern 1 is about 5 ms, but the optimum feedforward time constants of the position command pattern 2 and the position command pattern 3 are about 9 ms and 10 ms, respectively. Further, FIG. 11B shows that the limit feedforward gain Kff_bd also varies depending on the position command pattern.

  As shown in the above specific examples, the optimum feedforward time constant and the limit feedforward gain Kff_bd vary depending on the position command pattern. For this reason, when adjusting the position feedforward control parameter, the position command pattern actually used by the user is used. In addition, when an appropriate position feedforward control parameter is determined for a plurality of position command patterns, the user can designate a position command pattern for which the settling time is to be minimized. Then, consider finding a feed-forward time constant that minimizes the settling time in the position command pattern.

  12 to 14 are limit feedforward gain Kff_bd search flowcharts based on the above-described concept. Further, in this flowchart, in order to efficiently save and evaluate the parameter setting value and the settling time measurement result at the limit feedforward gain Kff_bd, the data [0], data [1], and data [2] shown in FIG. A data structure of a structure array type composed of three structure type elements is used.

  FIG. 15 is an explanatory diagram of the optimum feedforward time constant Tff_opt search data structure, which is a structure array type data structure composed of three structure type elements of data [0], data [1], and data [2]. Is used. The structure has the following four members. First, the feedforward time constant Tff, and then the limit feedforward gain Kff_bd at the Tff. Further, it is a settling time St_bd for the priority position command pattern at Tff and Kff_bd, and finally a next data instruction index next that indicates an array number for storing data related to Tff that is the next smaller Tff. The sm_index is a minimum Tff data indication index indicating which number of the array of the structure arrays data [0], data [1], and data [2] stores data related to the current minimum Tff. In the following description, the means for referring to each structure member will be described as being similar to the description in C language. For example, when referring to the member St_bd of the structure array data [1], data [1]. It is described as St_bd. As shown in FIG. 15, data [0]. When next = 1, data [data [0]. next]. The member St_bd of data [1] can also be referred to as St_bd.

  12 to 14, Tff_ini is a feedforward time constant initial value set by the user, and dmin_Tff is a feedforward time constant search minimum step set by the user. Dmax_Tff is a feedforward time constant search maximum step set by the user, Tff is a current feedforward time constant set in the position feedforward controller 15, and div_Tff is a current feedforward time constant search step. Furthermore, Tff_opt is an optimum feedforward time constant to be finally obtained, Kff_opt is an optimum feedforward gain, and is equal to the limit feedforward gain Kff_bd at the optimum feedforward time constant Tff_opt. Further, data_count is an initial acquired data number counter, and counters for confirming that there are three acquired data numbers related to Tff, data [] and sm_index are as described above.

  Next, a search outline of the optimum feedforward time constant Tff_opt according to this flowchart will be described with reference to FIG.

  FIG. 16 is an explanatory diagram of the optimum feedforward time constant Tff_opt search process. In the figure, the horizontal axis represents the feedforward time constant Tff, and the vertical axis represents the settling time St_bd for the priority position command pattern at the limit feedforward gain Kff_bd at the feedforward time constant Tff. In the example shown in FIG. 16, Tff_ini = Tff_1. In this flowchart, Tff to be searched next is determined while always evaluating the three-point data regarding Tff. For this reason, first, data (2) and data (3) are acquired at dmax_Tff intervals starting from data (1) of Tff = Tff_1. Next, data (1), (2), and (3) are evaluated, and since these St_bd are monotonically decreasing with respect to Tff, expecting a further decrease in St_bd, and extending the data (3) Data (4) is acquired, and the overflowing data (1) is discarded. Next, the data (2), (3), and (4) are evaluated, and since these St_bd is convex downward with respect to Tff, the optimum feedforward time constant Tff_opt is “data (2) and data (3)”. "Between" and "between data (3) and data (4)". At this time, it cannot be determined on which side Tff_opt exists, but St_bd of data (2) and data (4) at both ends is compared to search for a smaller side. Therefore, data (5) at the midpoint between data (3) and data (4) is acquired, and data (2) is discarded. Next, the data (3), (5), and (4) are evaluated, and since these St_bd is downwardly convex with respect to Tff, the St_bd of the data (3) and the data (4) is compared as described above. The data (6) on the smaller side is acquired, and the data (4) is discarded. Next, the data (3), (6), (5) were evaluated, and since these St_bd were monotonically decreasing with respect to Tff, the data (7) was acquired on the extension of the data (5) and overflowed Discard data (3). Here, when the data (6), (5), and (7) are evaluated, these St_bd are convex downward with respect to Tff. However, if a further search is performed, the current feedforward time constant search step div_Tff will be lower than the feedforward time constant search minimum step dmin_Tff set by the user. Therefore, the search is terminated, and Tff of data (5), which is the local minimum point at this time, is set as the optimum feedforward time constant Tff_opt. The search outline according to the flowcharts shown in FIGS. 12 to 14 has been described above with reference to FIG.

  Next, the flowcharts of FIGS. 12 to 14 will be described in order. First, the flowchart starts at 200. In the initialization process 201, the feedforward time constant initial value Tff_ini is set to the current feedforward time constant Tff, and the feedforward time constant search maximum step dmax_Tff is set to the current feedforward time constant search step div_Tff. Further, the arrangement of the structure array data [] is initialized as data [0], data [1], data [2] in ascending order of Tff to be stored. For this purpose, the next data instruction index next is set to data [0]. next = 1, data [1]. next = 2, data [2]. Next = 0, and the minimum Tff data instruction index sm_index is set to sm_index = 0. Further, in order to set the current number of acquired data to zero, the initial acquired data number counter dat_count = 0 is set, and the process proceeds to processing 202. In the process 202, the limit feedforward gain Kff_bd search flowchart of FIG. In the process 203, the current feedforward time constant Tff is set to data [data_count]. And save the limit feedforward gain Kff_bd at the current Tff with data [dat_count]. Save to Kff_bd. Further, the settling time St_bd for the priority position command pattern at the current Tff and Kff_bd is set to data [dat_count]. The data is stored in St_bd, and the process proceeds to the determination process 204. In the determination process 204, the value of the initial acquisition data number counter dat_count is evaluated. If the number of data regarding Tff is 3, the process proceeds to the determination process 205. If there is no data, the process proceeds to process 206. In process 206, the current feedforward time constant search step div_Tff is added to the current feedforward time constant Tff, 1 is added to the value of the initial acquisition data number counter dat_count, and the process proceeds to the process 202. In the determination processing 205, it is determined whether or not the settling time St_bd for the priority position command pattern is convex downward with respect to the feedforward time constant Tff. Specifically, if the processing here is as follows, it is determined to be convex downward, and the processing proceeds to processing 207. That is, data [sm_index]. St_bd> data [data [sm_index]. next]. St_bd, and data [data [data [sm_index]. next]. next]. St_bd> data [data [sm_index]. next]. St_bd. In the process 207, the current feedforward time constant search step div_Tff is halved, and the process proceeds to the determination process 208. In the determination process 208, it is determined whether or not the current feedforward time constant search step div_Tff is greater than or equal to the feedforward time constant search minimum step dmin_Tff. The process proceeds to process 210. In the determination process 209, the magnitude relation between the settling time St_bd for the priority position command pattern at the minimum Tff and the St_bd at the maximum Tff is compared among the three-point data.

  Specifically, data [sm_index]. St_bd> data [data [data [sm_index]. next]. next]. It is determined whether or not St_bd is established. If it is established, the process proceeds to processing 212. In process 212, in order to acquire new data at the midpoint between the second largest Tff and the maximum Tff, Tff is set to a value obtained by adding the second largest Tff and div_Tff. Specifically, the current feedforward time constant Tff is set to data [data [sm_index]. next]. Update with Tff + div_Tff, and proceed to processing 213. In the process 213, the limit feedforward gain Kff_bd search flowchart of FIG. In process 214, the array element storing the data relating to the current minimum Tff is overwritten and updated with the new data acquired in process 213. For this purpose, the current feedforward time constant Tff is set to data [sm_index]. And save the limit feedforward gain Kff_bd at the current Tff data [sm_index]. Save to Kff_bd. Then, the settling time St_bd for the priority position command pattern at the current Tff and Kff_bd is set to data [sm_index]. Save to St_bd. Further, the disturbance of the next data instruction index next caused by overwriting and updating the array element that has stored the data relating to the minimum Tff so far with new data that will be the data relating to the second largest Tff will be corrected. Specifically, data [sm_index]. After saving the value of next in a temporary variable tmp, data [sm_index]. Next is data [data [sm_index]. next]. Update with next. In addition, data [data [sm_index]. next]. Next is updated with sm_index, and data [data [data [sm_index]. next]. next]. Next is updated with the value of the temporary variable tmp. Finally, sm_index is updated with the value of the temporary variable tmp, and the process proceeds to the determination process 205. In the determination process 209, data [sm_index]. St_bd> data [data [data [sm_index]. next]. next]. When St_bd is not established, the process proceeds to process 215. In process 215, in order to obtain new data at the midpoint between the minimum Tff and the second largest Tff, Tff is set to a value obtained by subtracting div_Tff from the second largest Tff. Specifically, the current feedforward time constant Tff is set to data [data [sm_index]. next]. Update with Tff-div_Tff, and proceed to processing 216. In the process 216, the above-described limit feedforward gain Kff_bd search flowchart of FIG. 10 is executed, and the process proceeds to the process 217. In process 217, the current feedforward time constant Tff is changed to data [data [data [sm_index] .data in order to overwrite and update the array element storing the data related to the current maximum Tff with the new data acquired in process 216. next]. next]. Save to Tff. Then, the limit feedforward gain Kff_bd at the current Tff is set to data [data [data [sm_index]. next]. next]. Save to Kff_bd. In addition, the settling time St_bd for the priority position command pattern at the current Tff and Kff_bd is changed to data [data [data [sm_index]. next]. next]. Save to St_bd. Furthermore, the disturbance of the next data instruction index next caused by overwriting and updating the array element that has stored the data related to the maximum Tff so far with new data that will be the data related to the second largest Tff will be corrected. Specifically, data [sm_index]. After saving the value of next in a temporary variable tmp, data [sm_index]. Next is data [data [sm_index]. next]. Update with next. In addition, data [data [sm_index]. next]. Next, data [data [data [sm_index]. next]. next]. Update with next. Furthermore, data [data [data [sm_index]. next]. next]. Next is updated with the value of the temporary variable tmp, and the process proceeds to the determination process 205. If it is determined in the determination process 208 that div_Tff ≧ dmin_Tff does not hold, the process proceeds to process 210. In the process 210, data [data [sm_index]... Is the second largest Tff in the optimum feedforward time constant Tff_opt. next]. Set Tff. Then, data [data [sm_index]... Is the limit feedforward gain Kff_bd at the Tff that is the second largest to the optimum feedforward gain Kff_opt. next]. Kff_bd is set, and the flow proceeds to the end state 211. On the other hand, in the determination process 205, when the settling time St_bd for the priority position command pattern is not convex downward with respect to the feedforward time constant Tff, the process proceeds to the determination process 218. In the determination process 218, it is determined whether or not the settling time St_bd for the priority position command pattern is monotonously increasing with respect to the feedforward time constant Tff. Here, data [sm_index]. St_bd <data [data [sm_index]. next]. St_bd <data [data [data [sm_index]. next]. next]. If it is St_bd, it is determined that the increase is monotonous, and the process proceeds to processing 219. In the process 219, since it is monotonically increasing, new data is acquired on the extension of the minimum Tff data in anticipation of a decrease in St_bd. Specifically, the current feedforward time constant Tff is set to data [sm_index]. Update with Tff-div_Tff, and proceed to processing 220. In the process 220, the limit feedforward gain Kff_bd search flowchart of FIG. 10 described above is executed, and the process proceeds to the process 221. In the process 221, the array element storing the data related to the current maximum Tff is overwritten and updated with the new data acquired in the process 220. For this purpose, the current feed-forward time constant Tff is set to data [data [data [sm_index]. next]. next]. Save to Tff. Then, the limit feedforward gain Kff_bd at the current Tff is set to data [data [data [sm_index]. next]. next]. Save to Kff_bd. In addition, the settling time St_bd for the priority position command pattern at the current Tff and Kff_bd is changed to data [data [data [sm_index]. next]. next]. Save to St_bd. Further, the minimum Tff data instruction index sm_index is updated by overwriting and updating the data related to the minimum Tff to the array element that has stored the data related to the maximum Tff so far. That is, data [data [sm_index]. next]. This is performed by assigning “next”, and the process proceeds to the determination process 205. If the determination process 218 does not determine that the increase is monotonous, the process proceeds to process 222 assuming that the decrease is monotonous. In the process 222, since it is a monotonic decrease, new data is acquired on the extension of the maximum Tff data in anticipation of a further decrease in St_bd. Specifically, the current feedforward time constant Tff is set to data [sm_index]. Update with Tff + 3 × div_Tff, and proceed to processing 223. In process 223, the limit feedforward gain Kff_bd search flowchart of FIG. 10 described above is implemented, and the process proceeds to process 224. In the process 224, the array element storing the data related to the current minimum Tff is overwritten and updated with the new data acquired in the process 223. For this purpose, the current feedforward time constant Tff is set to data [sm_index]. And save the limit feedforward gain Kff_bd at the current Tff data [sm_index]. Save to Kff_bd. Further, the settling time St_bd for the priority position command pattern at the current Tff and Kff_bd is set to data [sm_index]. Save to St_bd. Further, the minimum Tff data instruction index sm_index is updated by overwriting and updating the data related to the maximum Tff on the array element that has been storing data related to the minimum Tff so far. That is, data [sm_index]. This is performed by assigning “next”, and the process proceeds to the determination process 205.

  As described above, the automatic position feedforward control parameter that minimizes the settling time for the specified specific position command pattern while suppressing the overshoot amount within a predetermined (allowable) value for all the previously registered position command patterns. Search can be performed.

  In the processing 203, processing 214, processing 217, processing 221, and processing 224, data []. In St_bd, the settling time St_bd for the priority position command pattern is set. However, the longest settling time among all the evaluated position command patterns may be set.

  Next, examples of adjustment parameter setting screens are shown in FIGS.

FIG. 17 is a screen configuration example of a personal computer in which a user inputs adjustment conditions when the feedback control parameter is automatically adjusted according to the first embodiment of the present invention. Each input item in FIG. 17 corresponds to each adjustment parameter described above. For example, 290 corresponds to the minimum position response frequency Fp_min, 291 corresponds to the maximum position response frequency Fp_max, and 292 corresponds to the position response frequency increment step div_Fp. Also, 293 to a minimum speed response frequency Fs_min, 294 is the maximum speed response frequency Fs_max, 295 the speed response frequency increment Div_Fs, 296 the oscillation amplitude tolerance theta _vib of the positional deviation theta _e, 297 is settling monitor timeouts Corresponds to time timeout. In addition, check boxes 298 and 299 that can be exclusively selected are provided so that the user can select whether to automatically generate a step-like position command or manually set the position command. Here, when 298 is selected, the processing shown in the flowchart of FIG. 2 is executed, and the position command pattern is automatically set. On the other hand, when 299 is selected, it is necessary for the user himself / herself to set the acceleration / deceleration time 300, the moving distance 301, and the maximum speed 302 and determine the position command pattern.

  FIG. 18 is a screen configuration example of a personal computer in which the user inputs adjustment conditions when the position feedforward control parameter is automatically adjusted according to the second embodiment of the present invention. Each input item in FIG. 18 corresponds to each adjustment parameter described above. For example, 311 corresponds to the minimum feedforward gain search step dmin_Kff, and 312 corresponds to the maximum feedforward gain search step dmax_Kff. Further, 314 corresponds to the feedforward time constant initial value Tff_ini, 316 also corresponds to the time constant search minimum step dmin_Tff, 317 similarly corresponds to the time constant search maximum step dmax_Tff, and 318 corresponds to the settling monitoring timeout time timeout.

  Reference numerals 320 to 324 are adjustment condition setting screens for the position command pattern 1 to the position command pattern 5, respectively. By clicking a pattern name notation indicated by the numbers 320 to 324 with the mouse, the setting of the pattern is performed. You can open the screen. In FIG. 18, the position command pattern 1 setting screen is open, but the other position command pattern setting screens have the same configuration. Therefore, the acceleration / deceleration time in the position command pattern is set in 300, the moving distance in the position command pattern is set in 301, and the maximum speed in the position command pattern is set in 302, and the position command pattern is determined. Further, 325 corresponds to an allowable value of the overshoot amount in the position command pattern, and 326 corresponds to a settling determination deviation posin_pls in the position command pattern. A check box 327 is used to set whether or not to perform adjustment with the position command pattern. In addition, a selection item 313 is provided to set whether or not position feedforward time constant adjustment is performed. The selection item 313 includes “adjustment” and “no adjustment”. When “no adjustment” is selected, the flowchart of FIG. 10 is executed, and only the position feedforward gain is compared with the current position feedforward time constant. Is adjusted. On the other hand, when “with adjustment” is selected, the flowcharts of FIGS. 12 to 14 are executed to adjust both the position feedforward gain and the time constant. In addition, a selection item 319 is provided for designating a top priority position command pattern that minimizes the settling time. In the selection item 313, one of the position command pattern checked with the check box 327 and the “maximum settling time pattern” can be selected.

  FIG. 19 is an overall system configuration diagram of an electric motor control apparatus that can employ the present invention common to the first and second embodiments. In FIG. 19, 341 is a ball screw unit, 342 is an electric motor, 343 is a position detector of the electric motor 343, 345 is a slider on which a load 344 is mounted, 339 is a servo amplifier, 346 transmits a position detection signal of the electric motor 342 to the servo amplifier 339. Cable. Reference numeral 347 denotes a cable for supplying driving power from the servo amplifier 339 to the electric motor 342. Reference numeral 349 denotes a cable for supplying power to the servo amplifier. A personal computer 352 is used to input adjustment conditions when the feedback control parameter and the position feedforward control parameter are automatically adjusted. A communication cable 353 is used to transmit the adjustment conditions from the personal computer 352 to the servo amplifier 339.

  Next, the correspondence relationship between the reference numerals in FIGS. 1 and 8 and the reference numerals in FIG. 19 used in the description of the first and second embodiments will be described. The electric motor 1 corresponds to 342, the position detector 5 corresponds to 343, the load 2 corresponds to 344 and 345, and the drive shaft 3 corresponds to a ball screw inside the ball screw unit 341. Further, the following components are included in the servo amplifier 339. That is, with the power converter 4 at the top, the subtractor 6, the position controller 7, the speed calculator 8, the subtractor 9, the speed controller 10, the current detector 11, the subtractor 12, and the current controller 13 are usually used. It is a control device. In addition to the feedback control parameter tuning unit 14 or 17 added by the present invention, the position feedforward controller 15 and the adder 16 are also included in the servo amplifier 339.

The control block diagram which shows the automatic adjustment method of the feedback control parameter of the motor control apparatus by 1st Example of this invention. FIG. 2 is a flow chart for generating a position command pattern in the embodiment of FIG. 1. FIG. 6 is a waveform diagram of torque, speed, and position command pattern in a procedure for generating a position command pattern for automatic adjustment of feedback control parameters. The automatic adjustment flowchart of the feedback control parameter by a 1st Example. FIG. 1 is a measurement processing flow diagram 1 of vibration amplitude, settling time, and overshoot amount in the first embodiment. FIG. 2 is a measurement process flow diagram 2 for vibration amplitude, settling time, and overshoot amount in the first embodiment. FIG. 6 is a waveform diagram for explaining a measurement method of vibration amplitude, settling time, and overshoot amount in the first embodiment. The control block diagram of the automatic adjustment apparatus of the motor control apparatus which added the automatic adjustment means of the feedforward control parameter by 2nd Example of this invention. FIG. 9 is an explanatory diagram of the relationship between the feedforward gain and settling time / overshoot in the embodiment of FIG. 8. The processing flow figure of the automatic adjustment method of the feedforward gain Kff in a 2nd Example. The relationship figure of the feedforward time constant in 2nd Example, limit feedforward gain Kff_bd, and settling time. Limit feed forward gain Kff_bd search flowchart part 1. Limit feedforward gain Kff_bd search flowchart part 2. Limit feedforward gain Kff_bd search flowchart part 3. The data structure explanatory drawing for optimal feedforward time constant Tff_opt search. Explanatory drawing of the optimal feedforward time constant Tff_opt search process. Example of feedback control parameter automatic adjustment function setting screen. Position feed forward control parameter automatic adjustment function setting screen example. BRIEF DESCRIPTION OF THE DRAWINGS FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Electric motor, 2 ... Drive target load, 3 ... Connection shaft, 4 ... Power converter, 5 ... Position detector, 6 ... Subtractor, 7 ... Position controller, 8 ... Speed calculator, 9 ... Subtractor, 10 ... speed controller, 11 ... current detector, 12 ... subtractor, 13 ... current controller, 14 ... feedback control parameter tuning unit, 15 ... position feedforward controller, 16 ... adder, 17 ... feedback and position feedforward Control parameter tuning unit, θ _M ^* ... position command pattern, θ _M ... position detection value, θ _e ... position deviation, ω _M ^* ... speed command value, ω _M ... speed detection value, ω _e ... speed deviation, I _q ^* ... Torque current command value, I _q ... Torque current detection value, I _e ... Current deviation, ω _FB ^* ... Position controller output signal, ω _FF ^* ... Position feedforward controller output signal.

Claims (24)

  An electric motor coupled with a load to be driven via a connecting shaft; a power converter that drives the electric motor; and a position controller that outputs a speed command value according to a deviation between a position command value and a position detection value of the electric motor; A speed controller that outputs a torque current command value according to a deviation between the speed command value and a detected speed value of the motor, and a deviation between the torque current command value and a detected torque current value supplied to the motor. In a motor control device including a current controller that adjusts an output current of the power converter, a step of generating a position command pattern for adjustment operation that continuously changes as a position command value of the position controller; and When the position command pattern is given as the position command value of the position controller, the response of the position controller and / or the speed controller is within a range where the vibration amplitude of the position deviation waveform does not exceed a predetermined value. Automatic adjustment method of a motor control device characterized by comprising the step of increasing the frequency.   2. The method according to claim 1, further comprising: determining a degree of change in which the position command pattern continuously changes based on a total moment of inertia value of the drive system including the electric motor, the connecting shaft, and the load to be driven. A method for automatically adjusting an electric motor control device.   2. The motor control device automatic adjustment method according to claim 1, wherein the degree of change of the position command pattern is determined so that the speed of the motor does not exceed an allowable maximum speed.   2. The automatic adjustment method for an electric motor control device according to claim 1, wherein a movement amount by a position command which is a final value of the position command pattern is set as a function of an allowable value of a vibration amplitude of the position deviation waveform.   2. The automatic adjustment of the motor control device according to claim 1, wherein the position command pattern is generated so that the entire period of the position command pattern is substantially the sum of an acceleration period and a deceleration period of the motor. Law.   The automatic adjustment method for an electric motor control device according to claim 1, wherein a function for automatically generating the position command pattern and a function for manual setting are switched in accordance with a user's selection input.   2. The position deviation waveform according to claim 1, wherein when the vibration amplitude of the position deviation waveform tends to decrease, a low level of the waveform is held, and when the vibration amplitude of the position deviation waveform tends to increase, the high level of the waveform is held, An automatic adjustment method for an electric motor control device, characterized by calculating a vibration amplitude of a waveform.   2. The first step of increasing the response frequency of the position controller in a predetermined increment until the vibration amplitude of the position deviation waveform exceeds a predetermined value, and the vibration amplitude of the position deviation waveform exceeds an allowable value. When returning the response frequency of the position controller to a range where the vibration amplitude of the position deviation waveform does not exceed the predetermined value, the second step of increasing the response frequency of the speed controller by a predetermined increment, When the first and second steps are repeated, and the repetition frequency of the position controller cannot be increased under the condition that the vibration amplitude of the position deviation waveform is less than or equal to a predetermined value, the response frequency of the speed controller and the position An automatic adjustment method for an electric motor control device comprising a third step of returning a response frequency of a controller to a range in which a vibration amplitude of the position deviation waveform does not exceed a predetermined value.   An electric motor coupled to a drive target load via a connecting shaft, a power converter that drives the electric motor, and a position controller that outputs a position controller output in accordance with a deviation between a position command value and a position detection value of the electric motor A position feedforward controller that receives the position command value as an input and outputs a position feedforward signal; an adder that adds the position controller output and the position feedforward signal and outputs a speed command value; and the speed A speed controller that outputs a torque current command value according to a deviation between the command value and a detected speed value of the motor; and the power conversion according to a deviation between the torque current command value and a detected torque current value supplied to the motor. In a motor control device including a current controller that adjusts an output current of the controller, a plurality of position command patterns registered in advance are used as a position command of the position controller. A step of performing a rolling operation, and a step of adjusting a control parameter of the position feedforward controller so that an overshoot amount does not exceed a predetermined value in the plurality of driving operations. Automatic adjustment method.   The motor control device according to claim 9, wherein the control parameter of the position feedforward controller is adjusted so as to minimize a settling time in a specific position command pattern among the plurality of position command patterns. Automatic adjustment method.   The automatic adjustment method for an electric motor control device according to claim 10, wherein the specific position command pattern that minimizes the settling time is switched according to a user's designated input.   11. The electric motor control device according to claim 10, wherein the specific position command pattern for minimizing settling time is automatically set to a position command pattern having the longest settling time from among previously registered position command patterns. Automatic adjustment method. 10. The adjustment target according to claim 9, wherein the position command pattern is θ _M ^* , the position feed-forward signal is ω _FF ^* , and the Laplace operator is s, and is expressed by a transfer function of equation (7). An automatic adjustment method for an electric motor control device, characterized by a feedforward gain Kff and a feedforward time constant Tff.

  10. The operation according to claim 9, wherein the gain of the feedforward controller is updated under a condition in which the time constant of the feedforward controller is fixed, and a plurality of position command patterns registered in advance for each feedforward gain. And adjusting the feedforward gain of the feedforward controller to a limit feedforward gain in which the overshoot amount does not exceed a predetermined value with respect to the plurality of position command patterns.   15. The settling at a limit feedforward gain corresponding to each feedforward time constant according to claim 14, wherein the feedforward time constant is updated, the feedforward gain is adjusted to the maximum value for each feedforward time constant. An electric motor control device comprising: a step of measuring time; and a step of obtaining the feedforward time constant that minimizes a settling time in a limit feedforward gain corresponding to the feedforward time constant by an extreme value search. Automatic adjustment method.   16. The extreme value search according to claim 15, wherein settling times in the limit feedforward gain are compared with respect to three feedforward time constants obtained by giving deviations of the feedforward time constant at equal intervals. Automatic adjustment method for motor controller.   17. In the comparison of three points of settling time in the limit feedforward gain with respect to the feedforward time constant, the horizontal axis is the feedforward time constant and the vertical axis is the settling time in the limit feedforward gain. Assuming a graph, the step of determining whether the graph is convex downward, monotonically increasing or monotonic decreasing, and if this determination result is convex downward, either of the intermediate data and the remaining two-sided data of the three-point data A step of newly acquiring one data at one midpoint, and a step of newly acquiring one data on the smallest time constant side among the three points of data without changing the data interval if the determination result is monotonously increased. If the determination result is monotonically decreasing, one new data is acquired on the maximum time constant side among the three points without changing the data interval. Automatic adjustment method of a motor control device characterized by comprising the steps.   16. The motor control device according to claim 15, wherein a monitoring time for measuring the settling time is set according to a user input, and a starting point of the monitoring time is set to a time point when a position deviation is zero-crossed or in the vicinity thereof. Automatic adjustment method.   An electric motor coupled to a drive target load via a connecting shaft, a power converter that drives the electric motor, and a position controller that outputs a position controller output in accordance with a deviation between a position command value and a position detection value of the electric motor A position feedforward controller that receives the position command value as an input and outputs a position feedforward signal; an adder that adds the position controller output and the position feedforward signal and outputs a speed command value; and the speed A speed controller that outputs a torque current command value according to a deviation between the command value and a detected speed value of the motor; and the power conversion according to a deviation between the torque current command value and a detected torque current value supplied to the motor. In a motor control device having a current controller for adjusting the output current of the generator, the vibration amplitude of the position deviation waveform when the position command pattern is given does not exceed a predetermined value. Increasing the response frequency of the position controller and / or the speed controller in a plurality of steps, performing a plurality of driving operations using a plurality of pre-registered position command patterns as position commands for the position controller, and a plurality of steps Adjusting the control parameter of the position feedforward controller so that the amount of overshoot does not exceed a predetermined value in the driving operation.   20. The method of claim 19, further comprising a step of searching for a control parameter of the position feedforward controller that satisfies a condition that the amount of overshoot does not exceed a predetermined value and that reduces a settling time in each position command pattern. An automatic adjustment method for a motor control device.   20. The automatic adjustment method for an electric motor control device according to claim 19, further comprising a step of receiving an adjustment condition setting input from a user interface and a step of receiving one selection input of the plurality of position command patterns.   An electric motor coupled with a load to be driven via a connecting shaft; a power converter that drives the electric motor; and a position controller that outputs a speed command value according to a deviation between a position command value and a position detection value of the electric motor; A speed controller that outputs a torque current command value according to a deviation between the speed command value and a detected speed value of the motor, and a deviation between the torque current command value and a detected torque current value supplied to the motor. In an automatic adjustment device for a motor control device including a current controller for adjusting an output current of the power converter, a position command pattern for adjustment operation that continuously changes as a position command value of the position controller is generated. When a position command pattern generating means and this position command pattern are given as a position command value of the position controller and the motor controller is operated, the vibration amplitude of the position deviation waveform exceeds a predetermined value. In the stomach range, automatic adjustment device of a motor control apparatus characterized by comprising a response frequency up means for increasing the response frequency of the position controller and / or the speed controller.   An electric motor coupled to a drive target load via a connecting shaft, a power converter that drives the electric motor, and a position controller that outputs a position controller output in accordance with a deviation between a position command value and a position detection value of the electric motor A position feedforward controller that receives the position command value as an input and outputs a position feedforward signal; an adder that adds the position controller output and the position feedforward signal and outputs a speed command value; and the speed A speed controller that outputs a torque current command value according to a deviation between the command value and a detected speed value of the motor; and the power conversion according to a deviation between the torque current command value and a detected torque current value supplied to the motor. In an automatic adjustment device for an electric motor control device provided with a current controller for adjusting the output current of the device, a plurality of position command patterns registered in advance are assigned to the position finger of the position controller. And adjusting operation means for performing a plurality of driving operations, and parameter adjusting means for adjusting the control parameter of the position feedforward controller so that the overshoot amount does not exceed a predetermined value in the plurality of driving operations. An automatic adjustment device for an electric motor control device.   An electric motor coupled to a drive target load via a connecting shaft, a power converter that drives the electric motor, and a position controller that outputs a position controller output in accordance with a deviation between a position command value and a position detection value of the electric motor A position feedforward controller that receives the position command value as an input and outputs a position feedforward signal; an adder that adds the position controller output and the position feedforward signal and outputs a speed command value; and the speed A speed controller that outputs a torque current command value according to a deviation between the command value and a detected speed value of the motor; and the power conversion according to a deviation between the torque current command value and a detected torque current value supplied to the motor. In an automatic adjustment device for a motor control device having a current controller for adjusting the output current of the generator, the vibration amplitude of the position deviation waveform when a position command pattern is given is predetermined. A response frequency maximizing means for increasing the response frequency of the position controller and / or the speed controller within a range not exceeding, and giving a plurality of pre-registered position command patterns as position commands to the position controller, Adjusting operation means for performing a plurality of driving operations; and parameter adjusting means for adjusting a control parameter of the position feedforward controller so that an overshoot amount does not exceed a predetermined value in the plurality of driving operations. An automatic adjustment device for a motor control device.

Images