CN100524106C - Automatic regulating method and device for electromotor control device - Google Patents

Abstract

When adjusting the feedback control parameter, changing position instruction modes are continuously generated for adjusting running, the position instruction modes as the position instruction are provided to the position controller, the vibration amplitude of the position deviation is not exceeding the prescribed range, the response speeds of the position controller and velocity controller are improved, the maximum accelerating speed is automatically set. In another aspect, when adjusting the feedforward control parameter, multi-running action are executed for a plurality of position instruction modes, the feedforward control parameter is adjusted in order to make the overshooting not exceed the prescribed value. Thereby an automatic regulating method and device which can obtain feedback and/or feedforward control parameter in short time are provided.

Description

The automatic regulating method of control device of electric motor and device

Technical field

The present invention relates to the FEEDBACK CONTROL of the control device of electric motor that adopts FEEDBACK CONTROL and/or feedforward control and/or the automatic regulating method and the device of the control device of electric motor that the controlled variable in the feedforward control is automatically adjusted (tuning).

Background technology

Patent documentation 1 discloses a kind of the 1st prior art that feedback control parameters and feedforward control parameter are automatically adjusted.In the 1st prior art,,, then, determine best preceding feedforward parameter by fuzzy reasoning or neural network according to this response wave shape by Intelligent treatment supposition feedback control parameters according to the response wave shape of controlled quentity controlled variable with respect to control command value.

In addition, as the 2nd prior art, can exemplify patent documentation 2.The 2nd prior art, its principal character relate to the to feedover content of automatic regulating method is if controlled quentity controlled variable is with respect to control command value overshoot then reduce feedforward gain, if not overshoot then increase feedforward gain.In addition, introduce evaluation function, if its value is that setting is with next end adjustment.

Patent documentation 1: the spy opens flat 6-No. 102905 communiques (in full);

Patent documentation 2: the spy opens 2003-No. 61377 communiques (in full).

Summary of the invention

Usually, when feedback control parameters is adjusted, in order to improve control performance, be with the gain maximization of positioner and speed control.But in the mechanical system that is made of motor, driven object load and coupling shaft, there is intrinsic antiresonant frequency (antiresonance) in mechanical system, exists near the character that is very easy to resonance of this frequency.Therefore, in the adjustment of reality, almost adjust very lowly, so that near the gain of the described controller antiresonant frequency can be not too high.As the simple method that realizes trickle like this adjustment, generally acknowledging has the limit to make machinery practically with the antiresonant frequency vibration, and the method for the gain of positioner and speed control is improved on the limit in mechanical vibration are not excessive scope.In addition, the mechanical system of antiresonant frequency the unknown provides step (step) instruction as position command (control command value) with the most simple and reliable method of antiresonant frequency vibration.Yet, in the 1st prior art, because with the structural reason of control command value from the outside input, according to the control command value that provides, difficult situation about fully vibrating with antiresonant frequency may take place.Under such vibration condition, the limit positions action repeatedly, and aforesaid mechanical vibration are estimated on the limit, under the situation of the gain of raising positioner and speed control, even disturbance a little also might be set at the excessive gain that produces vibration.

In addition, when the feedforward control parameter adjustment,, overshoot is made as below the setting for a plurality of position command patterns, and then, about the ad-hoc location instruction mode of frequent use, be contemplated to be the minimized adjustment of setting time.But, in the 1st, 2 prior aries,, be difficult to set regularization condition aforesaid, that be associated with the position command pattern because do not have position command pattern generating unit and a plurality of position command pattern login feature in adjusting mechanism inside.

The objective of the invention is to, the automatic adjusting method and the device of the controlled variable that can obtain suitable feedback and/or feedforward in a kind of short time is provided.

The preferred embodiment of the present invention, connected positioner in cascade, in the control device of electric motor of speed control and current controller, generate the position command value of the position command pattern of continually varying adjustment adjustment running usefulness as described positioner, above-mentioned position command pattern is passed through displacement, decide with desirable maximal rate of the motor under the situation that allows the torque capacity acceleration and acceleration and deceleration time, when this position command pattern is provided as the position command of described positioner, vibration amplitude at the position deviation waveform is no more than in the scope of setting, increases the response frequency of described positioner and/or described speed control.

In addition, more specifically, the total moment of inertia value J based on described electric motor drive system determines described position command patterns of change degree.

Further, preferred, described position command patterns of change degree is no more than the mode that it allows maximal rate according to the speed of motor, and does not have the mode of carrying out acceleration and deceleration during the constant speed and determine.

In addition, in the preferred embodiment of the present invention, except described positioner, outside the cascade of speed control and current controller is connected, also according to the position feed-forward controller that comprises input position command value outgoing position feed-forward signal, in the control device of electric motor that the mode of the totalizer of the output of described positioner and described position feed-forward signal plus output speed command value is constituted, a plurality of position command patterns of login are in advance carried out a plurality of running actions as the position command of described positioner, in a plurality of described running actions, adjust the controlled variable of described position feed-forward controller, so that overshoot is no more than setting.

In the specific embodiment of the present invention, when feedback control parameters is adjusted, near available maximum output, provide the position command pattern of the short displacement that makes motor operation by its motor control system.Thus, mechanical system is fully vibrated near comprising the frequency of antiresonant frequency.And then, when described position command pattern generates, consider the total moment of inertia value J of drive system and the permission maximum torque value τ of control device of electric motor _Max, carry out automatic setting so that accelerated speed is being no more than described permission maximum torque value τ _MaxScope in be the maximum.And then, as the position command pattern of the maximal rate that is no more than motor.

Like this, generate automatically control system inside satisfy unsaturated condition near the position command pattern of step-like instruction, carry out the adjustment of feedback control parameters.

On the other hand, when the feedforward control parameter adjustment, generate a plurality of position command patterns, each position instruction mode is set regularization condition.

The invention effect

According to the preferred embodiment of the present invention, can adjust the feedback control parameters of the position command pattern of automatic generation at short notice.

In addition, according to the preferred embodiment of the present invention, a plurality of position command parameters for the user logins in advance can be controlled at overshoot below the setting.

Further, according to the preferred embodiment of the present invention,, can carry out the automatic adjustment of the minimized feedforward control parameter of setting time about the certain location instruction mode of frequent use.

Other purpose and feature of the present invention will be clear and definite in the explanation of following embodiment.

Description of drawings

Fig. 1 is the control block diagram of automatic adjusting method of feedback control parameters of the control device of electric motor of expression the 1st embodiment of the present invention.

Fig. 2 is the product process figure of the position command pattern among the embodiment of Fig. 1.

Fig. 3 is the oscillogram of the torque, speed and the position command pattern that are used for the genesis sequence of the self-adjusting position command pattern of feedback control parameters.

Fig. 4 is the automatic adjustment process flow diagram of the feedback control parameters of the 1st embodiment.

Fig. 5 is one of measurement processing flow chart of the vibration amplitude/setting time/overshoot among the 1st embodiment.

Fig. 6 be the vibration amplitude/setting time/overshoot among the 1st embodiment the measurement processing flow chart two.

Fig. 7 is the measuring method explanation oscillogram of the vibration amplitude/setting time/overshoot among the 1st embodiment.

Fig. 8 is the control block diagram of automatic regulating apparatus of control device of electric motor of automatic mechanism that has added the feedforward control parameter of the 2nd embodiment of the present invention.

Fig. 9 is the key diagram that concerns of feedforward gain among the embodiment of Fig. 8 and setting time/overshoot.

Figure 10 is the processing flow chart of the automatic adjusting method of the feedforward gain Kff among the 2nd embodiment.

Figure 11 is the graph of a relation of feedforward time constant and limit feedforward gain Kff_bd and setting time among the 2nd embodiment.

Figure 12 is one of limit feedforward gain Kff_db search routine.

Figure 13 is two of a limit feedforward gain Kff_db search routine.

Figure 14 is three of a limit feedforward gain Kff_db search routine.

Figure 15 is best feedforward time constant Tff_opt search data structure show figure.

Figure 16 is best feedforward time constant Tff_opt search procedure key diagram.

Figure 17 is that feedback control parameters is adjusted function setting picture example automatically.

Figure 18 is that the position feed-forward controlled variable is adjusted function setting picture example automatically.

Figure 19 is the total system one routine structural drawing that can adopt control device of electric motor of the present invention.

The explanation of figure Chinese words:

1-motor, the load of 2-driven object, 3-coupling shaft, 4-power converter, 5-position detector, 6-subtracter, 7-positioner, 8-speed arithmetical unit, 9-subtracter, 10-speed control, 11-current detector, 12-subtracter, 13-current controller, 14-feedback control parameters adjustment part, 15-position feed-forward controller, 16-totalizer, 17-feedback and position feed-forward controlled variable adjustment part, θ _M ^*-position command pattern, θ _M-position probing value, θ _e-position deviation, ω _M ^*-speed value, ω _M-speed detected value, ω _e-velocity deviation, I _q ^*-torque current command value, I _q-torque current detected value, I _e-current deviation, ω _FB ^*-positioner output signal, ω _FF ^*-position feed-forward controller output signal.

Embodiment

Following with reference to accompanying drawing, the preferred embodiments of the present invention are described.

Embodiment 1

Fig. 1 is the control block diagram of automatic regulating method of the control device of electric motor of expression the 1st embodiment of the present invention.The purpose of the 1st embodiment is, adjusts feedback control parameters automatically.In Fig. 1,1 is motor, and 2 is the driven object load that is driven by described motor 1, and 3 for connecting the coupling shaft of described motor 1 and described driven object load 2, and 4 for driving the power converter of described motor 1.5 for being installed in the turning axle of described motor 1, exports the position probing value θ of the turning axle of described motor 1 _MPosition detector, 6 is calculating location command value θ _M ^*Described position probing value θ with described motor 1 _MBetween position deviation θ _eSubtracter, 7 is according to described position deviation θ _eOutput speed command value ω _M ^*Positioner, 8 are the position probing value θ of input by 5 outputs of described position detector _M, export the speed detected value ω of the turning axle of described motor 1 _MThe speed arithmetical unit, 9 for calculating described speed value ω _M ^*Speed detected value ω with described motor 1 _MBetween velocity deviation ω _eSubtracter, 10 is according to described velocity deviation ω _e, output torque current command value I _q ^*Speed control, 11 for detecting the torque current detected value I supply to described motor 1 _qCurrent detector, 12 for calculating described torque current command value I _q ^*With the described torque current detected value I that supplies to described motor 1 _qBetween current deviation I _eSubtracter, 13 is according to described current deviation I _eAdjust the current controller of the output current of described electric power variation device 4,14 is the feedback control parameters adjustment part of adjusting the parameter of described positioner 7 and speed control 10 automatically.Described position deviation θ is imported in this adjustment part _e, outgoing position command value θ _M ^*With position response frequency that is set in described positioner 7 and the speed responsive frequency that is set in described speed control 10.

Inside in feedback control parameters adjustment part 14, the needed step-like position command of the automatic adjustment pattern with the automatic adjustment algorithm of feedback control parameters described later and feedback control parameters generates mechanism.Its elemental motion is the position deviation θ when the step-like position command applies _eVibration amplitude be no more than setting that the user sets for example in the scope of permissible value, with position response frequency and the maximization of speed responsive frequency.

Fig. 2 is the product process figure of position instruction mode among the embodiment of presentation graphs 1.

Fig. 3 is the oscillogram that expression is used for output torque, speed and the position command pattern in proper order of the self-adjusting position command pattern of feedback control parameters.

The oscillogram of using Fig. 3 generates mechanism automatically with the flow process of Fig. 2 and described position command pattern and specifically describes.Formation speed pattern in the flow process of Fig. 2, this velocity mode can not rotate amount of movement (the being the anglec of rotation) θ that is instructed to for torque and the speed value with maximum saturatedly concerning motor _Max, finally generate the position command pattern.At this moment, do not establish in the velocity mode during the constant speed, further, its policy is to set displacement short as far as possible.Thus, in the instruction mode of position, can comprise the higher frequency composition.

In Fig. 2,, carry out amount of movement θ in processing 21 from 20 beginning flow processs _MaxCalculating.Amount of movement θ has been narrated in the front _MaxShort more good more, still, terrifically with amount of movement θ _MaxSet in short-term, exist and position deviation θ _eWith respect to amount of movement θ _MaxThe ratio of vibration amplitude irrelevant, the situation that its absolute value is too small.At this moment, described vibration amplitude can not detect vibration less than the resolution of described position detector 5.Therefore, by formula (1) decision amount of movement θ _Max

(formula 1)

θ _max＝α·θ _vib………………(1)

θ in (formula 1) _VibBe the position deviation θ that the user sets _eThe allowable value of vibration amplitude.That is, if expression is used for position deviation θ _eVibration amplitude less than θ _Vib, then judge position deviation θ _eBe non-vibrating threshold value.In addition, α is the positive constant that the user sets, expression amount of movement θ _MaxDescribed relatively position deviation θ _eThe allowable value θ of vibration amplitude _VibRatio.For example, if set α=100, press θ _VibBe amount of movement θ _Max1/100 calculate amount of movement θ _Max, therefore can detect amount of movement θ reliably _MaxThe vibration of the amplitude more than 1/100.Usually, setting α on the experience is value more than 100.

Then, handling in 22, allowed maximum torque value τ by what formula (2) calculating motor control device and motor combination determined _MaxThe desirable maximal rate ω of motor when quickening _Peak

(formula 2)

${\mathrm{\ω}}_{\mathrm{peak}}=\sqrt{\frac{{\mathrm{\τ}}_{\max }\·{\mathrm{\θ}}_{\max }}{J}}............\left(2\right)$

But, in formula (2), τ _MaxBe the above-mentioned maximum torque value of allowing, θ _MaxBe above-mentioned amount of movement, J is the total moment of inertia value after moment of inertia (moment) value of above-mentioned motor, above-mentioned coupling shaft and above-mentioned driven object load adds up to.Then, in judgment processing 23, the desirable maximal rate ω of more above-mentioned motor _PeakWith motor allow maximal rate ω _Max, at desirable maximal rate ω _PeakAllow maximal rate ω less than motor _MaxThe time, enter and handle 24.Handle 24, even for to allow torque capacity τ _MaxQuicken, what also can not surpass motor allows maximal rate ω _MaxSituation.

Fig. 3 (a) is even the torque capacity τ of expression to allow _MaxQuicken, also be no more than the maximal rate ω that motor allows _MaxSituation.That is, as mentioned above, at first, according to the permission vibration amplitude θ that has set _Vibα doubly determine amount of movement θ _MaxIn this distance to allow torque capacity τ _MaxQuicken, be not provided with during the constant speed, immediately torque capacity τ to allow _MaxSlow down, then obtain velocity mode 40, thus, obtain amount of movement θ according to positive and negative torque 41 _MaxInterior position command pattern 42.The desirable maximal rate ω of the motor of this moment _PeakBe no more than the permission maximal rate ω of motor _MaxThe time, can use such velocity mode 40, therefore,, maximal rate is set at described desirable maximal rate ω handling in 24 _Peak, enter and handle 25.Handling 25, calculating to allow maximum torque value τ by formula (3) _MaxAccelerate to described desirable maximal rate ω _PeakAcceleration and deceleration time t _a, after setting, entering 26, processing finishes.

(formula 3)

$t_a=\sqrt{\frac{J\·{\mathrm{\θ}}_{\max }}{{\mathrm{\τ}}_{\max }}}............\left(3\right)$

Then,, suppose that the permission maximal rate of motor is extremely little situation, describe with reference to Fig. 3 (b) for the ease of judging.Now, by supposition, in described judgment processing 23, because desirable maximal rate ω _PeakGreater than the smaller permission maximal rate ω of motor _Max, therefore enter and handle 27.Handling 27 is to allow torque capacity τ _MaxDuring acceleration, surpass the permission maximal rate ω of motor _MaxSituation, can't shown in Fig. 3 (b), maximal rate be set at the permission maximal rate ω of described motor _Max, enter and handle 28.That is, velocity mode is shown in the symbol 43, and it is carried out displacement θ after the time integral _SLCan not reach described amount of movement θ _Max Handling 28, calculating to allow maximum torque value τ by formula (4) _MaxAccelerate to the permission maximal rate ω of motor _MaxAcceleration and deceleration time t _bAnd after setting, enter and handle 26, end process.

(formula 4)

t _b＝J·ω _max/τ _max………(4)

This moment motor amount of movement θ _SLCan use formula (5) expression.

(formula 5)

${\mathrm{\θ}}_{\mathrm{SL}}={\left(\frac{{\mathrm{\ω}}_{\max }}{{\mathrm{\ω}}_{\mathrm{peak}}}\right)}^2\·{\mathrm{\θ}}_{\max }\·\·\·\·\·\·\·\·\·\·\·\·\left(5\right)$

In addition, at this moment, the big or small τ of the torque that motor is exported _SLCan be with following formula (6) expression, as can be known than allowing torque capacity τ _MaxLittle.

(formula 6)

${\mathrm{\τ}}_{\mathrm{SL}}={\left(\frac{{\mathrm{\ω}}_{\max }}{{\mathrm{\ω}}_{\mathrm{peak}}}\right)}^2\·{\mathrm{\τ}}_{\max }\·\·\·\·\·\·\·\·\·\·\·\·\left(6\right)$

Fig. 4 is the automatic adjustment process flow diagram of the feedback control parameters of the 1st embodiment.Use this figure, the automatic adjustment algorithm of the feedback control parameters in the feedback control parameters adjustment part 14 is described.In Fig. 4, Fs_min represents the minimum speed response frequency that the user sets, and Fs_max represents the maximal rate response frequency that the user sets, and div_Fs is the speed responsive frequency increment step-length that the user sets.When adjustment described later, expression is increased to Fs_max with the div_Fs step-length from Fs_min with speed responsive frequency limit, and the limit is adjusted.Equally, Fp_min is the minimum position response frequency that the user sets, and Fp_max is the maximum position response frequency that the user sets, and div_Fp is the step-length of the position response frequency increment set of user.When adjustment described later, expression is increased to Fp_max with the step-length of div_Fp from Fp_min with response frequency limit, position, and the limit is adjusted.In addition, θ _VibBe that the user sets position deviation θ _eThe vibration amplitude permissible value.Fs is the current speed responsive frequency that sets in the described speed control 10, and Fp is the current position response frequency that sets in the described positioner 7.Fp_v0 is the maximum position response frequency of vibration mark, and the vibration amplitude of record position deviation waveform is permissible value θ _VibFollowing maximum position response frequency.Fs_v1 is the optimum velocity response frequency finally try to achieve by adjusting, and Fp_v1 is the optimum position response frequency of finally trying to achieve by adjusting, and it represents that the vibration amplitude of position deviation waveform is at permissible value θ during for " 1 " flag_fpmax _VibSign (flag) when following state is issued to the maximum position response.

According to the basic consideration method of the adjustment of this process flow diagram, be in the not vibrative scope of position deviation waveform, to get position response frequency and speed responsive frequency setting high as far as possible.In addition, rule of thumb there is following situation as can be known: in the process that the position response frequency is raise slowly,,, can reduce the vibration of described position deviation waveform by the speed responsive frequency is improved a little even in the position deviation waveform, produce vibration.Therefore, make full use of this experience, when there is vibration in the position deviation waveform, the speed responsive frequency is only improved with div_Fs, carry out revaluing of described vibration.

Then, the process flow diagram of key diagram 4 in order.In Fig. 4, flow process is from handling 50 beginnings, in the initialization of handling 51 enforcement variablees.In initialization process 51, substitution Fs_min in Fs sets according to the mode of the response frequency that begins from described minimum speed response frequency Fs_min to regulate the speed.Equally, with substitution Fp_min among the Fp, set according to the mode that begins to adjust the position response frequency from described minimum position response frequency Fp_min.And then the maximum position response frequency Fp_v0 of vibration mark carries out initialization by minimum position response frequency Fp_min, and described sign flag_fpmax carries out initialization by " 0 ".Then, enter and handle 52, the position command of the step-like that mensuration will generate in described mechanism is as the position command value θ among Fig. 1 _M ^*Position deviation θ when applying _eVibration amplitude.About this position deviation θ _eThe mensuration mechanism of vibration amplitude describe in the back.Then, enter judgment processing 53, judge described position deviation θ _eVibration amplitude whether at described permissible value θ _VibUnder.If vibration amplitude is at permissible value θ in judgment processing 53 _VibBelow, then there is the leeway that the position response frequency is further improved, enter judgment processing 54.In judgment processing 54, whether judge current position response frequency Fp less than maximum position response frequency Fp_max, if less than would enter and handle 55.Handling in 55, current position response frequency Fp is increased with position response frequency increment step-length div_Fp, enter into described processing 52.On the other hand, in described judgment processing 54, be that maximum position response frequency Fp_max enters when above and handles 56 at current position response frequency Fp.Handling in 56, described sign flag_fpmax is set at " 1 ", the vibration amplitude of expression position deviation waveform is with permissible value θ _VibFollowing state reaches the maximum position response, enters into judgment processing 57.In judgment processing 57, whether judge current speed responsive frequency Fs less than maximal rate response frequency Fs_max, if less than would enter and handle 58.Handling in 58, present speed response frequency Fs is increased with speed responsive frequency increment step-length div_Fs, the reduction of the vibration amplitude of position deviation waveform is sought on the limit, and the limit enters described processing 52.In described judgment processing 57, at current speed responsive frequency Fs when maximal rate response frequency Fs_max is above, expression position response frequency/speed responsive frequency together, with the vibration amplitude of position deviation waveform at permissible value θ _VibFollowing state reaches maximal value, enters to handle 59.Handling in 59,, respectively after current speed responsive frequency Fs of substitution and the current position response frequency Fp, enter and handle 60 and finish for described optimum velocity response frequency Fs_v1 and optimum position response frequency Fp_v1.

In the described judgment processing 53 at vibration amplitude not at permissible value θ _VibWhen following, enter judgment processing 61.In judgment processing 61, check described sign flag_fpmax, whether the vibration amplitude of confirming the position deviation waveform is with permissible value θ _VibFollowing state reaches the maximum position response.Here, be not 0 o'clock at flag_fpmax, the vibration amplitude of expression position deviation waveform is at permissible value θ _VibFollowing state reaches the maximum position response, then, surpasses permissible value θ at the vibration amplitude of position deviation waveform _VibBefore, improve the speed responsive frequency, enter and handle 67.Handling in 67, for described optimum velocity response frequency Fs_v1, because the vibration amplitude of position deviation waveform is set at permissible value θ _VibFollowing condition, so the value of substitution deduct speed responsive frequency increment step-length div_Fs in described current speed responsive frequency Fs after enter and handle 60 and finish.In judgment processing 61, when flag_fpmax=0, enter judgment processing 62, relatively the maximum position response frequency Fp_v0 of current location response frequency Fp and described vibration mark.Here, " Fp〉Fp_v0 " when being false, represent that there is not the condition above the maximum position response frequency Fp_v0 of vibration mark in current position response frequency Fp, enters described processing 67.In described judgment processing 62, " Fp〉Fp_v0 " when setting up, enter judgment processing 63.In judgment processing 63, whether judge current speed responsive frequency Fs less than maximal rate response frequency Fs_max, if handle 64 less than entering.Handling in 64,, seeking the reduction of the vibration amplitude of position deviation waveform the present speed response frequency Fs response frequency increment step-length div_Fs that gathers way.Further, current position response frequency Fp is reduced position response frequency increment step-length div_Fp, so that the vibration amplitude of position deviation waveform must be permissible value θ _VibBelow, enter and handle 65.Handling in 65, write down the maximum position response frequency Fp_v0 of current position response frequency Fp as vibration mark, response frequency further raising the in position entered described processing 52.In described judgment processing 63, be maximal rate response frequency Fs_max when above at current speed responsive frequency Fs, mean that current speed responsive frequency Fs can not be increased to this more than value, enter and handle 66.Handling in 66, for described optimum velocity response frequency Fs_v1, the speed responsive frequency Fs that substitution is current.Further, for described optimum position response frequency Fp_v1, must be according to the vibration amplitude of position deviation waveform at permissible value θ _VibFollowing mode, substitution will be current position response frequency Fp reduce value behind the response frequency increment step-length div_Fp of position, then, enter and handle 60 end.

Fig. 5 and Fig. 6 are the process flow diagrams that the measurement of vibration amplitude in the 1st embodiment and setting time and overshoot is handled.Use this flow process, the method for the vibration amplitude of measuring necessary described position deviation waveform in feedback control parameters adjustment part 14 and position feedback control parameter adjustment described later portion 17, setting time (setting time), overshoot is described.In Fig. 5 and Fig. 6, timeout is that the user sets the supervision time-out time of adjusting, and is the position deviation waveform with zero cross point to start with starting point constantly, only measures the adjustment parameter of setting time in timeout.Posin_pls is that adjusting of setting of user judged and to be used deviation, stabilizes to posin_pls at the absolute value of position deviation and adjusts when following.In addition, poserr is described position deviation, poserr_work is that vibration is measured with overshoot and use position deviation, and the position deviation after the step-like instruction applies and the irrelevant to rotation direction of motor are the quantity of states behind the positive side begin symbol adjustment position deviation poserr all the time.Flag_poserr_plus is the sign that the position deviation after expression step-like position command applies begins from positive side, in position deviation when positive side begins, if flag_poserr_plus=1 when position deviation begins from minus side, establishes flag_poserr_plus=0.Thereby, if flag_poserr_plus=1 then poserr_work=poserr, if flag_poserr_plus=0 then poserr_work=-poserr.Poserr_min is the position deviation minimum value, is after the position command of step-like change to finish, till finishing to the measurement of setting time during, remain the variable of minimum value of the described poserr_work in this moment.Poserr_vib is the vibration amplitude maximal value in the position deviation, is to remain peaked variable from the vibration amplitude of the poserr_work that calculates according to " poserr_work-poserr_min ".Flag_plsin adjusts to judge with deviation arrival sign, if the absolute value of position deviation is used below the deviation posin_pls in the judgement of adjusting, then establishes flag_plsin=1, if not so, then establishes flag_plsin=0.Flag_plsin_bk adjusts to judge with deviation and arrive the last value of sign, is the flag_plsin that sets with the last time sign with value.Time_s measures the elapsed time, is that position command with step-like changes and finishes the elapsed time of back as starting point, and time_w adjusts to monitor the elapsed time, and expression position deviation waveform is the elapsed time that starting point begins with zero cross point constantly.St is a setting time, begins to become posin_pls when following from judging than adjusting with the big value of deviation posin_pls in that the absolute value of position deviation is each, will measure elapsed time time_s substitution St.Thus, if described adjusting monitors that time-out time timeout sets suitably longly, the needed time till the offset difference that begins to put in place after then the position command variation of record step-like finishes in St is finally adjusted.Over_shoot is the overshoot of overshoot with respect to the position command value of motor position, and the back of adjusting equates with the opposite in sign value of described poserr_min.

Then, in order the flow process of Fig. 5 and Fig. 6 is described.In Fig. 5 and Fig. 6, flow process is confirmed position command value θ from handling 80 beginnings in judgment processing 81 _M ^*Variation.Here, because before the step-like position command is transfused to, the position command no change, thus repeat judgment processing 81, but during input step shape position command, judge " position command value θ _M ^*Exist and change ", enter judgment processing 82.In judgment processing 82, judge whether position command is constant.Here, in the moment after the position command of step-like rises fully, position command is constant, enters judgment processing 83.On the other hand, in the step-like instruction is risen, repeat described judgment processing 82.In judgment processing 83, confirm the positive and negative of position deviation poserr, handle 84 if canonical enters.Handling in 84, described position deviation poserr is changed to " 1 " from the sign flag_poserr_plus that is just beginning with expression, enters to handle 85.Handling in 85, poserr_min enters into initialization process 86 by the described position deviation minimum value of position deviation poserr initialization.In described judgment processing 83, if being negative, position deviation poserr enters processing 87, the described position deviation poserr of expression is changed to " 0 " from the sign flag_poserr_plus that is just beginning, enter processing 88.Handling in 88, described position deviation minimum value poserr_min is being carried out initialization with opposite in sign value-poserr of position deviation poserr, entering into initialization process 86.In the initialization process 86, vibration amplitude maximal value poserr_vib in the position deviation, measurement elapsed time time_s, the supervision elapsed time time_w that adjusts, the judgement of adjusting are arrived sign last time value flag_plsin_bk with deviation and carry out initialization with zero, enter into processing 90.Handling in 90, measuring elapsed time time_s and add 1, entering judgment processing 91.In judgment processing 91, if the absolute value of position deviation poserr is used below the deviation posin_pls in the judgement of adjusting, then enter into and handle 92, the judgement of adjusting arrives with deviation and indicates that flag_plsin is changed to 1, enters into judgment processing 93.In judgment processing 93, judging adjusts judges whether be zero, if flag_plsin_bk=0 enters into and handles 95, setting time St record is measured elapsed time time_s and entered into processing 95 if arriving sign last time value flag_plsin_bk with deviation.Handling in 95, to possess in the position deviation judgment processing 91 next time, adjust and judge with deviation and arrive this adjust and judge with deviation and arrive the value that indicates flag_plsin of the last value flag_plsin_bk of sign substitution, enter into judgment processing 97.In described judgment processing 91, if the absolute value of position deviation poserr is not to adjust to judge when using deviation posin_pls following, enters and handle 96, will adjust and judge that arriving sign flag_plsin with deviation is changed to " 0 ", enters into described processing 95.In addition, in described judgment processing 93,, enter described processing 95 if flag_plsin_bk=1 then because adjust, will not upgrade by setting time St.In addition, in judgment processing 97, the described position deviation of check expression is from the sign flag_poserr_plus on the occasion of beginning.If flag_poserr_plus=1, position deviation from the occasion of, the described vibration of the direct substitution of position deviation poserr/overshoot measured uses position deviation poserr_work, enter into judgment processing 99.In judgment processing 97, if flag_poserr_plus=0, then position deviation is used the described vibration of opposite in sign value-poserr substitution/overshoot measurement of position deviation poserr position deviation poserr_work, is entered into judgment processing 99 from negative value.In judgment processing 99, judge whether vibration/overshoot is measured with position deviation poserr_work is below the position deviation minimum value poserr_min.If below it, then enter and handle 101, measure with position deviation poserr_work with current vibration/overshoot and upgrade position deviation minimum value poserr_min, enter into judgment processing 102.In described judgment processing 99, if vibration/overshoot is measured with position deviation poserr_work not when position deviation minimum value poserr_min is following, needn't upgrade position deviation minimum value poserr_min, enter into judgment processing 102.In judgment processing 102, vibration amplitude maximal value poserr_vib and " poserr_work-poserr_min " in position deviation of measuring before this relatively.And, under the little situation of poserrr_vib one side, upgrade poserr_vib with " poserr_work-poserr_min " and enter into judgment processing 104.In described judgment processing 102, under the big situation of poserrr_vib one side, do not upgrade poserr_vib, enter into judgment processing 104.In judgment processing 104, carried out zero crossing in order to detect position deviation, whether estimate position deviation minimum value poserr_min below zero.Here, if poserr_min≤0 judges that then position deviation has been carried out zero crossing in the past.Enter into and handle 105.Handling in 105, moment of the initial zero crossing of expression position deviation waveform is being monitored that as the adjusting of elapsed time of starting point elapsed time time_w adds 1, entering judgment processing 106.In judgment processing 104, when be false in poserr_min≤0, judge that the position deviation waveform did not carry out zero crossing as yet one time, enter into to handle 90.In judgment processing 106, relatively the position command variation with step-like finishes to monitor elapsed time time_w as adjusting of starting point afterwards and adjust supervision time-out time timeout, if time_w 〉=timeout then enters into and monitor the processing 107 that finishes with adjusting.On the other hand, in described judgment processing 106, when time_w 〉=timeout is false, enter into the processing 90 that continue to adjust and monitor.In described processing 107, opposite in sign value-poserr_min of desired location deviation minimum value poserr_min is overshoot over_shoot, enters to handle 108 the processing of the vibration amplitude of the described position deviation waveform of end measurement, setting time, overshoot.

Fig. 7 is the explanation oscillogram of the measuring method of the vibration amplitude/setting time/overshoot among explanation first embodiment, uses the waveform of each quantity of state, with the result of flow of concrete example presentation graphs 5 and Fig. 6.In Fig. 7, waveform 120 is position command value θ _M ^*, waveform 121 is position deviation poserr and vibration/overshoot measurement position deviation poserr_work.Waveform 122 is to adjust to judge that with deviation arrival sign flag_plsin, waveform 123 are position deviation minimum value poserr_min, and waveform 124 is poserr_work-poserr_min.In addition, the transverse axis of each waveform is to measure elapsed time time_s, and the moment that the waveform 120 of representing position command is risen fully is made as time_s=0.Because position command waveform 120 is transformed to positive dirction at this moment, so time_s 〉=0, poserr_work equates with position deviation poserr.Here, if position command waveform 120 is changed to negative direction, the waveform 121 of expression poserr_work is constant, just with the waveform 121 relative time axis of the position deviation poserr waveform after the conversion symmetrically.In addition, if be present in the curve at the waveform 121 of expression poserr_work-posin_pls is between the posin_pls, then expression is adjusted and is judged that the waveform 122 that arrives sign flag_plsin with position deviation gets " 1 ", in addition, gets " 0 ".Shown in this example, under the situation of position deviation waveform, adjust and judge that arriving sign flag_plsin with deviation gets " 0 " and " 1 " for many times afterwards repeatedly, especially for vibration, monitor if adjust time-out time timeout set long enough, then fall within the most at last " 1 ".In addition, arrive the rising edge (arrow) of sign flag_plsin with deviation with measuring elapsed time time_s renewal setting time St in each judgement of adjusting.Adjusting monitors that elapsed time time_w is used to measure above-mentioned adjusting to monitor the elapsed time of time-out time timeout, shown in time shaft among the figure 125, with the moment after the position deviation waveform 121 first zero crossings as starting point.Like this, elapsed time time_s is different with measuring, and resets with the adjust supervision elapsed time time_w of the moment after the position deviation waveform 121 first zero crossings as starting point.Its reason is the influence that not too is subjected to the such controlled variable of position response frequency or speed responsive frequency till the adjust needed time of beginning in the moment after self-alignment offset difference waveform zero is intersected.Thus, when adjusting of user monitors the setting of time-out time timeout, needn't consider position response frequency or the such controlled variable of speed responsive frequency, only be set at certain value, just can correctly measure setting time St.In addition, waveform 123 is position deviation minimum value poserr_min, if compare with position deviation, then understand easily with vibration/overshoot measurement, but each minimum value constantly among its expression poserr_work.In addition, waveform 124 is poserr_work-poserr_min, is only to extract vibration/overshoot to measure with the waveform behind the vibration component included among the position deviation poserr_work.If poserr_work is a waveform of not following vibration to decay gradually, it is understandable like this that poserr_work-poserr_min is always zero.Vibration maximal value poserr_vib is the peak value retention value of waveform 124 in the position deviation of finally trying to achieve in addition.

As previously discussed, by using the flow process of Fig. 5 and Fig. 6, can measure vibration amplitude, setting time, the overshoot of needed described position deviation waveform in the feedback control parameters adjustment part 14.In addition, even in the position feed-forward controlled variable adjustment part 17 of the 2nd embodiment described later, also can utilize vibration amplitude, setting time, the overshoot of above-mentioned position deviation waveform.

(embodiment 2)

Fig. 8 is the controlling party block diagram of automatic mechanism of feedforward control parameter that comprises the control device of electric motor of the present invention the 2nd embodiment.In the 2nd embodiment, except the automatic adjustment of position feed-forward controlled variable is also carried out in the automatic adjustment of the feedback control parameters of Fig. 1.In Fig. 8, portion gives identical symbol with Fig. 1 identical functions, thereby avoids repeat specification.The 7th, correspondence position deviation θ _eOutgoing position controller output signal ω _FB ^*Positioner, the 15th, input position command value θ _M ^*Outgoing position feedforward controller output signal ω _FF ^*The position feed-forward controller.Its input/output relation during as s, uses feedforward gain Kff and feedforward time constant Tff as controller parameter in Laplace operator, can use formula (7) to represent.

(formula 7)

${\mathrm{\ω}}_{\mathrm{FF}}^{*}=\frac{\mathrm{Kff}\·{\mathrm{\θ}}_M^{*}}{1+s\·\mathrm{Tff}}............\left(7\right)$

17 14 identical with Fig. 1 except the parameter of automatic adjustment positioner 7 and speed control 10, are still adjusted feedforward gain Kff in the position feed-forward controller 15 and the controlled variable adjustment part of time constant Tff automatically.Therefore, as function additional on 14 basis, input position deviation θ _e, outgoing position command value θ _M ^*With feedforward gain Kff that sets in the position feed-forward controller 15 and feedforward time constant Tff.In addition, the inside in controlled variable adjustment part 17 also has a plurality of position command patterns that automatic adjustment algorithm and user with position feed-forward controlled variable described later login in advance successively as position command value θ _M ^*The function of output.Its elemental motion is moved for the actual running of all carrying out of a plurality of position command patterns, satisfies the condition that overshoot is no more than permissible value about all position command patterns, and, the setting time in the certain location instruction mode is minimized.At this moment, carry out the automatic search of the feedforward gain Kff and the feedforward time constant Tff of position feed-forward controller.

Then, specify the automatic adjusting method of feedforward gain Kff, at first this method is described with reference to Fig. 9.

Fig. 9 is the key diagram that concerns of feedforward gain among the embodiment of Fig. 8 and setting time and overshoot, and this figure (a) is the relation of feedforward gain and setting time, and this figure (b) is the relation of expression feedforward gain and overshoot.Among Fig. 9 (a), transverse axis is feedforward gain Kff, and the longitudinal axis is a setting time, has represented the more little trend of the big more setting time of feedforward gain Kff basically.As shown in the figure, be worth when above the setting time sharp increase at certain if represented feedforward gain Kff.On the other hand, in Fig. 9 (b), transverse axis is feedforward gain Kff, and the longitudinal axis is an overshoot, and feedforward gain Kff is little, can not produce overshoot fully.But feedforward gain Kff reaches certain and is worth when above, begins to produce overshoot, after, its generating capacity increases about Kff is dull.

More than, according to Fig. 9 (a) and (b), be zero or very hour in overshoot, we can say that side's setting time that feedforward gain Kff sets greatly is short.In addition, according to feedforward gain Kff being set to such an extent that big more overshoot is as can be known big more: minimize in order to make setting time, feedforward gain Kff is no more than permissible value in overshoot gets in the scope maximization and get final product.

Thereby, in the following description treatment scheme,, satisfy the search of maximum feedforward gain Kff that overshoot is no more than the condition of permissible value about all position command patterns of login in advance.

Figure 10 is the process flow diagram of processing of the automatic regulating method of the feedforward gain Kff among the 2nd embodiment.In addition, about the automatic regulating method of feedforward time constant Tff, in its adjustment process, because need the automatic adjustment of feedforward gain kff, so describe in the back.

In Figure 10, Kff_ini is the feedforward gain initial value that the user sets, and dmax_Kff is that the feedforward gain that the user sets is searched for maximum step-length (step), and dmin_Kff is the feedforward gain search minimum step that the user sets.Kff is the current feedforward gain that is set in the position feed-forward controller 15, and div_Kff is current feedforward gain step-size in search, and Kff_bd is the limit feedforward gain that finally will try to achieve.When adjustment described later, begin search from Kff=Kff_ini, if the Kff=Kff_ini overshoot less than permissible value, increases Kff with the step-length of dmax_Kff.In addition, Kff=Kff_ini is if overshoot more than permissible value, then reduces Kff with the step-length of dmax_Kff.And then in the renewal of the Kff of above-mentioned dmax_Kff step-length, overshoot perhaps, when overshoot is changed to less than permissible value, is implemented following the processing from changing to more than the permissible value less than permissible value more than permissible value.That is, the limit reduces by half above-mentioned current feedforward gain step-size in search div_Kff, and the increase and decrease search processing of Kff is carried out on the limit, till div_Kff is less than feedforward gain search minimum step dmin_Kff.

Then, the flow sequence according to Figure 10 describes.In Figure 10,, in initialization process 151, current feedforward gain Kff is set at feedforward gain initial value Kff_ini from 150 beginning flow processs.And current feedforward gain step-size in search div_kff is set at feedforward gain and searches for maximum step-length dmax_Kff, enters into to handle 152.Handling in 152, the running of all position command patterns that the execution user logins is in advance measured setting time, overshoot by said mechanism in each running, enter into judgment processing 153.In described judgment processing 153, the overshoot of each operation mode is estimated, under the overshoot of any operation mode is situation more than the permissible value, enters into and handle 154.Handling in 154, current feedforward gain Kff reduces after the current feedforward gain step-size in search div_Kff, enters into to handle 155.Handling in 155, carrying out the running of all position command patterns that the user logins in advance equally, in each running, measuring setting time, overshoot, entering into judgment processing 156 with above-mentioned processing 152.In described judgment processing 156, carry out the evaluation of the overshoot of each operation mode,, enter into and handle 157 during in the overshoot of all operation modes less than permissible value.On the other hand, in described judgment processing 156, the overshoot of arbitrary operation mode is a purpose to reduce current feedforward gain Kff further when permissible value is above, enters into described processing 154.In described processing 157, less than one of candidate of the maximum Kff of permissible value, upgrade limit feedforward gain Kff_bd as the overshoot of all operation modes with current feedforward gain Kff, enter into and handle 158.Handling in 158, consider the flow process of back, current feedforward gain Kff is added current gain step-size in search div_Kff, the overshoot that resets in arbitrary operation mode reaches the above condition of permissible value, enters into and handles 159.In described judgment processing 153, the overshoot in all operation modes enters into and handles 160 during less than permissible value.In described processing 160, less than one of candidate of the maximum Kff of permissible value, upgrade limit feedforward gain Kff_bd as the overshoot of all operation modes with current feedforward gain Kff, enter into and handle 161.In described processing 161, current feedforward gain Kff is added after the current feedforward gain step-size in search div_Kff, enter into and handle 162.Handling in 162, carrying out the running of all position command patterns of logining in advance with the same user of described processing 155, in each running, measuring setting time, overshoot, entering into judgment processing 163.In described judgment processing 163, carry out the evaluation of each overshoot of each operation mode, reach permissible value when above in any overshoot of operation mode, carry out described processing 159.On the other hand, in described judgment processing 163, the overshoot in all operation modes is a purpose to increase current feedforward gain kff further during all less than permissible value, enters described processing 160.Follow the flow process of described processing 159, the described current feedforward gain step-size in search div_Kff search that the limit increases and decreases Kff that reduces by half is handled on the limit, handling in 159, described current feedforward gain step-size in search div_Kff is reduced by half, and enters into judgment processing 165.In described judgment processing 165, judge whether the size of current feedforward gain step-size in search div_Kff is more than the feedforward gain search minimum step dmin_Kff.And, if more than it, then enter into and handle 166, if less than it then enter into 176, finish the search of feedforward gain Kff and handle.Handling in 166, behind the feedforward gain step-size in search div_Kff that current feedforward gain Kff minimizing is current, entering into and handle 167.Handling in 167, carrying out the running of all position command patterns that the user logins in advance equally, measuring setting time, overshoot in each running, entering into judgment processing 168 with described processing 162.In described judgment processing 168, carry out the overshoot evaluation of each operation mode,, enter into and handle 169 during in the overshoot of all operation modes less than permissible value.On the other hand, in described judgment processing 168, the overshoot of arbitrary operation mode reaches permissible value when above, is purpose to reduce current feedforward gain Kff further, enters into described processing 159.In described processing 169, less than one of candidate of the maximum Kff of permissible value, upgrade limit feedforward gain Kff_bd as the overshoot of all operation modes with current feedforward gain Kff, enter into and handle 170.In described processing 170, described current feedforward gain step-size in search div_Kff is reduced by half, enter into judgment processing 171.In described judgment processing 171, judge whether the size of current feedforward gain step-size in search div_Kff is more than the feedforward gain search minimum step dmin_Kff.And, handle 172 if enter into more than it, if less than would enter into 176, finish the search of feedforward gain Kff and handle.Handling in 172, current feedforward gain Kff is being added enter into and handle 173 after the current feedforward gain step-size in search div_Kff.Handling in 173, same with described processing 167, carry out the running of all position command patterns that the user logins in advance, in each running, measure setting time, overshoot, enter into judgment processing 174.In described judgment processing 174, carry out the evaluation of the overshoot of each operation mode, reach permissible value when above in the overshoot of any operation mode, be purpose further to reduce current feedforward gain Kff, enter into and handle 159.On the other hand, in described judgment processing 174, under the situation of overshoot as all operation modes, enter and handle 175 less than permissible value.In described processing 175, upgrade limit feedforward gain Kff_bd with current feedforward gain, less than one of candidate of the maximum Kff of runtime value, is purpose to increase current feedforward gain Kff as the overshoot of all operation modes further, enters into described processing 170.

More than, if use the flow process of Figure 10, then all the position command patterns about logining are in advance found out and are satisfied the not maximum feedforward gain Kff of the condition of super permissible value of overshoot, as limit feedforward gain Kff_db.In addition, be minimum if the permissible value of the setting time overshoot of this moment is enough little.

In addition, the concrete expression of described Fig. 9 (a) and (b) the adjustment process of carrying out with the flow process of Figure 10.Each of this moment adjusted parameter K ff_ini=0.10, dmin_Kff=0.005, dmax_Kff=0.10, overshoot permissible value=2[pulse].Search for to the order of (6) by (1) among Fig. 9, finally handling in 170 set div_Kff=0.003125.Therefore, in ensuing judgment processing 171, be "No", enter into end.

Figure 11 is the key diagram of the relation of feedforward time constant among the 2nd embodiment and limit feedforward gain Kff_db and setting time.Illustrated the setting time in feedforward time constant shown in this figure (a) and the limit feedforward gain (maximal value) relation and, the feedforward time constant shown in this figure (b) and the relation of limit feedforward gain.

Figure 11 (a) transverse axis is feedforward time constant Tff, and the longitudinal axis is the setting time of limit feedforward gain, and Figure 11 (b) transverse axis is feedforward time constant Tff, and the longitudinal axis is a feedforward gain.In addition, as the parameter of each curve, select three kinds of different position command patterns.Here, so-called limit feedforward gain Kff_db is about each position instruction mode, satisfies the maximum feedforward gain Kff that overshoot is no more than the condition of permissible value.Specifically, the numerical value on each waveform of Figure 11 (b) under the fixing condition of the time constant Tff that will feedover, is carried out the flow process of described Figure 10 for each independent position command pattern, and the numerical value of trying to achieve.Numerical value on each waveform of Figure 11 (a) is the numerical value that writes down the setting time St_bd among the limit feedforward gain Kff_bd that tries to achieve like this.Here, according to Figure 11 (a),

(b) as can be known, the best feedforward time constant of giving minimum setting time is according to the position command pattern and different.For example, according to Figure 11 (a), the best of position command pattern 1 feedforward time constant is about 5ms, and the best feedforward time constant of position command pattern 2 and position command mode 3 is respectively 9ms, 10ms.In addition, Fig. 1 (b) expression limit feedforward gain Kff_db is also according to the position command pattern and difference.

Shown in above concrete example, best feedforward time constant and limit feedforward gain Kff_bd change according to the position command pattern.Therefore, when the feedforward control parameter adjustment of position, use the position command pattern of the actual use of user and implement.In addition, for a plurality of position command patterns, when determining suitable position feed-forward controlled variable, particularly the user can specify the minimized position command pattern of setting time.And, find the minimized feedforward time constant of the setting time that is implemented in this position command pattern.

Figure 12～Figure 14 is based on the limit feedforward gain Kff_bd search routine figure of the idea shown in above.In addition, in this flow process, pre-set parameter and setting time measurement result in order to preserve with evaluation limitation feedforward gain Kff_bd are efficiently used by data[0 shown in Figure 15], data[1], data[3] data structure of 3 structure arrange types that structure build key element constitutes.

Figure 15 is best feedforward time constant Tff_opt search data structure show figure, uses by data[0], data[1], data[3] data structure of 3 structure arrange types that structure build key element constitutes.In addition, the key element of structure is following four.At first being feedforward time constant Tff, then is the limit feedforward gain Kff_bd among this Tff.And then be setting time St_bd among this Tff, the Tff_bd with respect to the preferential position instruction mode, be the index nest that the next data of number are arranged in indication at last, the arrangement number of this arrangements number data that to be preservation relevant with inferior little Tff among this Tff.Sm_index is to be illustrated in described structure to arrange data[0], data[1], data[2] in, preserve the minimum Tff data of the data relevant in the arrangement of which number and indicate index with the Tff of current minimum.In addition, following, to each structure key element with reference to mechanism, describe equally with the description of C language.For example, the reference structure body is arranged data[1] key element St_bd the time, be described as data[1] .St_bd.In addition, as shown in figure 15, at data[0] during .next=1, even be data[data[0] .next] .St_bd also can be with reference to data[1] key element St_bd.

In the flow process of Figure 12～14, Tff_ini is the feedforward time constant initial value that the user sets, and dmin_Tff is the feedforward time constant search minimum step that the user sets.In addition, dmax_Tff is that the feedforward time constant that the user sets is searched for maximum step-length, and Tff is the current feedforward time constant of setting in the position feed-forward controller 15, and div_Tff is current feedforward time constant step-size in search.Further, Tff_opt is the best feedforward time constant that finally will try to achieve, and Kff_opt is best feedforward gain, equates with limit feedforward gain Kff_bd among the best feedforward time constant Tff_opt.In addition, dat_count initially obtains the data bulk counting, is used for determining the counting of obtaining data bulk existence 3 relevant with Tff, data[] and sm_index is as mentioned above.

Then, the limit illustrates the search summary of the best feedforward time constant Tff_opt of this flow process with reference to Figure 16 limit.

Figure 16 is best feedforward time constant Tff_pot search procedure key diagram.In the drawings, transverse axis is feedforward time constant Tff, and the longitudinal axis is the setting time St_bd for the preferential position instruction mode among the limit feedforward gain Kff_bd among this feedforward time constant Tff.In addition, in example shown in Figure 16, establish Tff_ini=Tff_1.In this flow process, 3 point data relevant with Tff are often estimated on the limit, the Tff that the limit decision should be searched for.Therefore, at first with the data (1) of Tff=Tff_1 as starting point, obtain data (2), data (3) at interval with dmax_Tff.Then, evaluating data (1), (2), (3), these St_bd reduce about Tff is dull, and therefore, expectation St_db further reduces, and obtains data (4) in the prolongation of data (3), abandons overflow data (1).Then, evaluating data (2), (3), (4) because these St_bd about Tff to lower convexity, therefore best feedforward time constant Tff_opt is positioned among " between data (2) and the data (3) " or " between data (3) and the data (4) " arbitrary.This can not conclude which side Tff_opt is present in constantly, but compares the data (2) at two ends and the St_bd of data (4), searches for a little side.Thereby, obtain the middle data (5) of data (3) and data (4), abandon data (2).Then, evaluating data (3), (5), (4) because these St_bd about Tff to lower convexity, with the St_bd of above-mentioned same comparing data (3) and data (4), obtain a little side data (6), abandon data (4).Then, data (7) because these St_bd reduce about Tff is dull, are therefore obtained in evaluating data (3), (6), (5) in the prolongation of data (5), abandon overflow data (3).Here, evaluating data (6), (5), (7), these St_bd about Tff to lower convexity.But, further search for the feedforward time constant search minimum step dmin_Tff that current feedforward time constant step-size in search div_Tff is set less than the user.Thereby, finishing search, the Tff of the data (5) of this smallest point constantly is as the best feedforward time constant Tff_pot.More than, the limit is with reference to the search summary of the flow process shown in explanation Figure 12～14, Figure 16 limit.

The flow process of Figure 12～14 then, is described in order.At first in 200 beginning flow processs.In initialization process 201, current feedforward time constant Tff is set at feedforward time constant initial value Tff_ini, current feedforward time constant step-size in search div_Tff is set at the feedforward time constant searches for maximum step-length dmax_Tff.In addition, described structure is arranged data[] formation be initialized as data[0 according to the little order of Tff of storage], data[1], data[2].Therefore, following data indication index next is made as data[0] .next=1, data[1] .next=2, data[2] .next=0, further, minimum Tff data indication index sm_index is made as sm_index=0.In addition, initially obtain data bulk counting dat_count=0, enter into and handle 202 for the current data bulk of obtaining being made as zero, establishing.Handling in 202, implementing the limit feedforward gain Kff_bd search routine of above-mentioned Figure 10, entering into and handle 203.Handling in 203, at data[dat_count] preserve current feedforward time constant Tff among the .Tff, at data[dat_count] preserve the limit feedforward gain Kff_db among the current Tff among the .kff_bd.In addition, at data[dat_count] preserve setting time St_bd among the .St_bd for current Tff and the preferential position instruction mode among the Kff_bd, enter into judgment processing 204.In judgment processing 204, estimate the value initially obtain data bulk counting dat_count, if be 3 about the quantity of the data of Tff then enter into judgment processing 205,, then enter into and handle 206 if data bulk does not have 3 points.Handling in 206, current feedforward time constant Tff adds current feedforward time constant step-size in search div_Tff, and the value that initially obtains data bulk counting dat_count is added 1, enters into judgment processing 202.In described judgment processing 205, judge setting time St_bd for the preferential position instruction mode about feedforward time constant Tff whether to lower convexity.The processing here specifically, if then judge to lower convexity like that according to following, enters into and handles 207.Be 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.Handling 207, current feedforward time constant step-size in search div__Tff is reduced by half, enter into judgment processing 208.In described judgment processing 208, whether the size of judging current feedforward time constant step-size in search div_Tff is more than the feedforward time constant search minimum step dmin_Tff, if more than it, enter into judgment processing 209, if littler then enter into and handle 210 than it.In described judgment processing 209, within 3 point data, compare magnitude relationship for the setting time St_bd and the St_bd among the maximum Tff of the preferential position instruction mode among the minimum Tff.

Specifically, judge data[sm__index] .St__bd〉data[data[data[sm__index] .next] .next] whether .St__bd set up, and when setting up, enter into and handle 212.Handling in 212, in order to obtain new data, Tff is being made as value after second largest Tff and the div_Tff addition at the center of second largest Tff and maximum Tff.Specifically, use data[data[sm__index] .next] .Tff+div__Tff upgrades current feedforward time constant Tff, enters into to handle 213.Handling 213, implementing the limit feedforward gain Kff_bd search routine of described Figure 10, entering into and handle 214.Handling in 214, the new data of being obtained by described processing 213 is rewritten and is upgraded the arrangement key element of having preserved after the data relevant with the Tff of current minimum.Therefore, current feedforward time constant Tff is kept at data[sm__index] among the .Tff, limit feedforward gain Kff_bd among the current Tff is kept at data[sm__index] among the .Kff_bd.And, will be kept at data[sm__index for the setting time St_bd of current Tff and the preferential position instruction mode among the Kff_bd] .St__bd.And then, revise the confusion of next data indication index next, this next one data indication index next data that the new data of second largest Tff related data from now on upgrades the arrangement key element of preserving the data relevant with minimum so far Tff of serving as reasons are indicated index.Specifically, with data[sm__index] value of .next temporarily exists after the variable tmp, uses data[data[sm__index] .next] .next upgrades data[sm__index] .next.In addition, upgrade data[data[sm__index with sm_index] .next] .next, upgrade data[data[data[sm__index with the value of described temporary transient variable tmp] .next] .next] .next.At last, upgrade sm_index, enter described judgment processing 205 with the value of described temporary transient variable tmp.In described judgment processing 209, at data[sm__index] .St__bd〉data[data[data[sm__index] .next] .next] under the invalid situation of .St__bd, enter into and handle 215.Handling in 215, in order to obtain new data, Tff is being made as from second largest Tff deducts value behind the div_Tff at the center of the Tff of minimum and second largest Tff.Specifically, use data[data[sm__index] .next] .Tff-div__Tff upgrades current feedforward time constant Tff, enters into to handle 216.Handling 216, implementing the limit feedforward gain Kff_bd search routine of described Figure 10, entering into and handle 217.Handling in 217, for by the arrangement key element of handling after the data relevant with the Tff of current maximum have been preserved in the 216 new datas rewriting renewals that obtain, current feedforward time constant Tff is kept at data[data[data[sm__index] .next] .next] .Tff.And, limit feedforward gain Kff_bd among the current Tff is kept at data[data[data[sm__index] .next] .next] .Kff__bd.And, will be kept at data[data[data[sm__index for the setting time St_bd of current Tff and the preferential position instruction mode among the Kff_bd] .next] .next] .Kff__bd.And then, revise the confusion of next data indication index next, this next one data indication index next data that the new data of second largest Tff related data from now on upgrades the arrangement key element of preserving the data relevant with maximum so far Tff of serving as reasons are indicated index.Specifically, with data[sm__index] value of .next temporarily exists after the variable tmp, uses data[data[sm__index] .next] .next upgrades data[sm__index] .next.In addition, use data[data[data[sm__index] .next] .next] .next upgrades data[data[sm__index] .next] .next.And the value with described temporary transient variable tmp is upgraded data[data[data[sm__index] .next] .next] .next, enter described judgment processing 205.In described judgment processing 208, when div__Tff ≧ dmin__Tff is false, enters and handle 210.Handling in 210, in the best feedforward time constant Tff_opt, be provided as the data[data[sm__index of second largest Tff] .next] .Tff.And, be provided as the data[data[sm__index of the limit feedforward gain Kff_bd among second largest Tff at best feedforward gain Kff_opt] .next] .Kff__bd, enter into done state 211.On the other hand, in described judgment processing 205, judge for the setting time St_bd of preferential position instruction mode whether protruding downwards about feedforward time constant Tff.Here, if data[sm__index] .St__bd＜data[data[sm__index] .next] .St__bd＜data[data[data[sm__index] .next] .next] .St__bd then be judged as dull increasing, enter into and handle 219.Handling in 219, because dull increasing, so the minimizing of expectation St_bd obtains new data in the prolongation of minimum Tff data.Specifically, use data[sm__index] .Tff-current feedforward time constant Tff of div__Tff renewal, enter into and handle 220.Handling in 220, implementing the limit feedforward gain Kff_bd search routine of described Figure 10, entering into and handle 221.Handling in 221, be used in described processing 220 obtained new datas and rewrite the arrangement key element that the data relevant with the Tff of current maximum have been preserved in renewal.Therefore, current feedforward time constant Tff is kept at data[data[data[sm__index] .next] .next] .Tff.And limit feedforward gain Kff_bd among the current Tff is kept at data[data[data[sm__index] .next] .next] .Kff__bd.In addition, will be kept at data[data[data[sm__index for the setting time St_bd of current Tff and the preferential position instruction mode among the Kff_bd] .next] .next] .St__bd.Further, carry out the renewal of minimum Tff data indication index sm_index, this minimum Tff data indication index is to rewrite the minimum Tff data indication index that upgrades the data relevant with minimum Tff in the arrangement key element of the data that in store maximum so far Tff is correlated with.That is, pass through data[data[sm__index] .next] substitution of .next carries out, and enters into described judgment processing 205.In described judgment processing 218, be not judged as the monotone increasing added-time, think dull and reduce, enter into and handle 222.Handling in 222, because be dull the minimizing, the further minimizing of expectation St_bd obtains new data in the prolongation of maximum Tff data.Specifically, use data[sm__index] .Tff+3 * current feedforward time constant Tff of div__Tff renewal, enter into and handle 223.Handling in 223, implementing the limit feedforward gain Kff_bd search routine of described Figure 10, entering into and handle 224.Handling in 224, be used in described processing 223 obtained new datas and rewrite the arrangement key element that the data relevant with the Tff of current minimum have been preserved in renewal.Therefore, current feedforward time constant Tff is kept at data[sm__index] .Tff, limit feedforward gain Kff_bd among the current Tff is kept at data[sm__index] .Kff__bd.In addition, will be kept at data[sm__index for the setting time St_bd of current Tff and the preferential position instruction mode among the Kff_bd] among the .St__bd.Further, carry out the renewal of minimum Tff data indication index sm_index, this minimum Tff data indication index is to rewrite the minimum Tff data indication index that upgrades the data relevant with maximum Tff in the arrangement key element of the data that in store minimum so far Tff is correlated with.Promptly pass through data[sm__index] substitution of .next carries out, and enters into described judgment processing 205.

Like this,, overshoot is suppressed at regulation (permission) value with interior the time, can makes the automatic search of the minimized position feed-forward controlled variable of setting time of the certain location instruction mode of appointment for all position command patterns of login in advance.

In addition, handling 203, handling 214, handling 217, handling 221, handling in 224, at data[] the setting time St_bd for the preferential position instruction mode is set among the St_bd, still, also the longest setting time can be set within all position command patterns of estimating.

Then, the embodiment of expression adjustment parameter setting picture in Figure 17,18.

Figure 17 is that the user imports the picture configuration example of the personal computer of regularization condition when the feedback control parameters of the 1st embodiment of the present invention is adjusted automatically.Each cuit in Figure 17 is corresponding with described each adjustment parameter, and for example, 290 is corresponding with minimum position response frequency Fp_min, and 291 is corresponding with maximum position response frequency Fp_max, and 292 is corresponding with position response frequency increment step-length div_Fp.In addition, 293 is corresponding with minimum speed response frequency Fs_min, and 294 is corresponding with maximal rate response frequency Fs_max, and 295 is corresponding with speed responsive frequency increment step-length div_Fs, 296 with position deviation θ _eVibration amplitude permissible value θ _VibCorrespondence, 297 monitor that time-out time timeout is corresponding with adjusting.In addition, still manually set, exclusive selection check frame (box) 298,299 is set for the position command that can select at user side to generate step-like automatically.Here, selecting at 298 o'clock, the processing shown in the flow process of execution graph 2, automatic setting position command pattern.On the other hand, selecting at 299 o'clock, each project by the user sets acceleration and deceleration time 300, displacement 301, maximal rate 302 needs decision position instruction mode.

Figure 18 is the position feed-forward controlled variable in the 2nd embodiment of the present invention when adjusting automatically, and the user imports the personal computer picture configuration example of regularization condition.Each cuit in Figure 18, to adjust parameter corresponding with described each, for example, 311 with feedforward gain to search for minimum step dmin_Kff corresponding, 312 with feedforward gain to search for maximum step-length dmax_Kff corresponding.In addition, 314 is corresponding with feedforward time constant initial value Tff_ini, 316 is corresponding with identical time constant search minimum step dmin_Tff, 317 with same time constant to search for maximum step-length dmax_Tff corresponding, 318 with the supervision time-out time of adjusting timeout is corresponding.

In addition, 320 to 324 are respectively position command pattern 1 sets picture to the regularization condition of position instruction mode 5, the setting picture that can launch this pattern with 320 to 324 mark parts of numbering the MODE name of representing by click.In addition, among Figure 18, be the state behind the setting image expansion of position command pattern 1, but the setting picture of other position command pattern also is same formation.Thereby, in 300, set the acceleration and deceleration time in this position command pattern, in 301, set the displacement of this position command pattern, in 302, set the maximal rate in this position command pattern, determine this position command pattern.In addition, 325 is corresponding with the permissible value of overshoot in this position command pattern, 326 with this position command pattern in adjust judge with deviation posin_pls corresponding.Verification frame 327 is used to set the adjustment of whether using this position command pattern.In addition, for the enforcement that has or not the position feed-forward time constant to adjust is set, option 313 is set.In 313 the option, there be " adjustment " and " not adjusting ", when selecting " not adjusting ", carry out the flow process of Figure 10,, only adjust the position feed-forward gain for current position feed-forward time constant.On the other hand, when selecting " adjustment ", carry out the flow process of Figure 12～14, position feedforward gain/time constant both sides are adjusted.In addition, in order to specify, set option 319 with the minimized override of setting time position instruction mode.In 313 option, can select one with " setting time max model " is central from position command pattern with described check box 327 additional verifications.

Figure 19 is the total system pie graph that can adopt control device of electric motor of the present invention general among the 1st embodiment, the 2nd embodiment.In Figure 19, the 341st, ball-screw (ball screw) unit, the 342nd, motor, the 343rd, the position detector of motor 343, the 345th, carry the slider (silder) of load 344, the 339th, servoamplifier, the 346th, the position detection signal of motor 342 is sent to the cable of servoamplifier 339.In addition, the 347th, from the cable of servoamplifier 339 supply motor 342 driving electric, the 349th, to the cable of servoamplifier supply power.The 352nd, when the automatic adjustment of feedback control parameters and position feed-forward controlled variable, the user imports the personal computer of regularization condition, and the 353rd, be used for the regularization condition from personal computer 352 is sent to the telecommunication cable of servoamplifier 339.

Then, the corresponding relation between the symbol among the symbol of employed Fig. 1, Fig. 8 in the explanation of described the 1st embodiment, the 2nd embodiment and Figure 19 is described.Motor 1 is corresponding with 342, and position detector 5 is corresponding with 343, and load 2 is corresponding with 344 and 345, and driving shaft 3 is corresponding with the ball-screw of ball screw unit 341 inside.In addition, ensuing inscape is included in the servoamplifier 339.Be common control device such as power converter 4, subtracter 6, positioner 7, speed arithmetical unit 8, subtracter 9, speed control 10, current detector 11, subtracter 12 and current controller 13.In addition, except that the feedback control parameters adjustment part 14 or 17 that adds according to the present invention, position feed-forward controller 5, totalizer 16 etc. also is included in the servoamplifier 339.

Claims (23)

1, a kind of automatic adjusting method of control device of electric motor, this control device of electric motor possesses: the motor that combines with the driven object load by coupling shaft, drive the power converter of described motor, positioner according to the deviation output speed command value between the position probing value of position command value and described motor, speed control according to the deviation output torque current command value of the speed detected value of described speed value and motor, according to described torque current command value with supply to the current controller that deviation between the torque current detected value of described motor is adjusted the output current of described power converter, the automatic adjusting method of this control device of electric motor is characterised in that to have:

Generate the step of the position command pattern of continually varying adjustment running usefulness as the position command value of described positioner, above-mentioned position command pattern is by displacement, decide to allow desirable maximal rate of motor and acceleration and deceleration time under the situation that torque capacity quickens; And

When this position command pattern is provided as the position command value of described positioner, be no more than in the scope of permissible value, increase the step of the response frequency of described positioner and/or described speed control at the vibration amplitude of position deviation waveform.

2, the automatic adjusting method of control device of electric motor according to claim 1 is characterized in that,

Also have:, determine the step of the intensity of variation that described position command mode continuous changes based on the total moment of inertia value of the drive system that comprises described motor and described coupling shaft and described driven object load.

3, the automatic adjusting method of control device of electric motor according to claim 1 is characterized in that,

The intensity of variation of described position command pattern is no more than it according to the speed of motor and allows the mode of maximal rate to determine.

4, the automatic adjusting method of control device of electric motor according to claim 1 is characterized in that,

Setting is as the caused amount of movement of the position command of the end value of described position command pattern, and as the function of the permissible value of the vibration amplitude of described position deviation waveform, this function satisfies following formula (1),

θ _max＝α·θ _vib··················(1)。

5, the automatic adjusting method of control device of electric motor according to claim 1 is characterized in that,

The mode that is essentially the acceleration and deceleration time sum of motor according to whole period of described position command pattern generates described position command pattern.

6, the automatic adjusting method of control device of electric motor according to claim 1 is characterized in that,

Switch function that generates described position command pattern automatically and the function of manually setting according to user's selection input.

7, the automatic adjusting method of control device of electric motor according to claim 1 is characterized in that,

When the vibration amplitude of described position deviation waveform is in minimizing trend, keep the low level of this waveform, when the vibration amplitude of described position deviation waveform is in increase trend, keep the high position of this waveform, calculate the vibration amplitude of described position deviation waveform.

8, the automatic adjusting method of control device of electric motor according to claim 1 is characterized in that,

Have:

The response frequency of described positioner is increased to the vibration amplitude of described position deviation waveform above the 1st step till the permissible value with the increment of stipulating;

When the vibration amplitude of described position deviation waveform surpasses permissible value, the vibration amplitude that the response frequency of described positioner is turned back to described position deviation waveform is no more than after the scope of described permissible value, increases the 2nd step of the response frequency of described speed control with the increment of regulation; And

Repeat the described the 1st and the 2nd step, by this repetition, when the condition below the permissible value of can not reaching with the vibration amplitude of described position deviation waveform increased the response frequency of described positioner, the vibration amplitude that the response frequency of the response frequency of described speed control and described positioner is turned back to described position deviation waveform was no more than the 3rd step in the scope of permissible value.

9, the automatic adjusting method of control device of electric motor according to claim 1, this control device of electric motor possesses: according to the positioner of the output of the deviation outgoing position controller between the position probing value of position command value and described motor, with described position command value is input, the position feed-forward controller of outgoing position feed-forward signal, with described positioner output and described position feed-forward signal plus, the totalizer of output speed command value, speed control according to the deviation output torque current command value of the speed detected value of described speed value and motor, according to described torque current command value with supply to the current controller that deviation between the torque current detected value of described motor is adjusted the output current of described power converter, the automatic adjusting method of this control device of electric motor is characterised in that to have:

A plurality of position command patterns of login are in advance carried out the step of a plurality of running actions as the position command of described positioner; And

In a plurality of described running actions, adjust the controlled variable of described position feed-forward controller so that overshoot is no more than the step of permissible value.

10, the automatic adjusting method of control device of electric motor according to claim 9 is characterized in that,

Adjust the controlled variable of described position feed-forward controller, so that the setting time of the ad-hoc location instruction mode in a plurality of described position command patterns minimizes.

11, the automatic adjusting method of control device of electric motor according to claim 10 is characterized in that,

The minimized described certain location instruction mode of setting time is switched in appointment input according to the user.

12, the automatic adjusting method of control device of electric motor according to claim 10 is characterized in that,

In the middle of the position command pattern of login in advance, the position command pattern that the automatic setting setting time is the longest is as the minimized described certain location instruction mode of setting time.

13, the automatic adjusting method of control device of electric motor according to claim 9 is characterized in that,

Described adjustment object is θ establishing described position command pattern _M ^*, described position feed-forward signal is ω _FF ^*, when Laplace operator is s, be feedforward gain Kff and the feedforward time constant Tff represent with the transport function of formula (7).

${\mathrm{\ω}}_{\mathrm{FF}}^{*}=\frac{\mathrm{Kff}\·{\mathrm{\θ}}_M^{*}}{1+s\·\mathrm{Tff}}\·\·\·\·\·\·\·\·\·\·\·\·\left(7\right)$

14, the automatic adjusting method of control device of electric motor according to claim 9 is characterized in that,

Under the condition of the time constant of fixing described feedforward controller, upgrade the gain of described feedforward controller, for each feedforward gain, the feedforward gain of described feedforward controller is adjusted in running under a plurality of position command patterns of logining in advance action in overshoot is no more than the limit feedforward gain of permissible value about a plurality of described position command patterns.

15, the automatic adjusting method of control device of electric motor according to claim 14 is characterized in that,

Have:

Described feedforward time constant is upgraded on the limit, and the limit is adjusted into described maximal value corresponding to each time constant that feedovers with described feedforward gain, measures the step of the setting time in the limit feedforward gain corresponding with each feedforward time constant; With

Try to achieve the step of the described feedforward time constant of the setting time minimum in the limit feedforward gain corresponding with described feedforward time constant by extremum search.

16, the automatic adjusting method of control device of electric motor according to claim 15 is characterized in that,

Described extremum search more equally spaced provides the deviation of described feedforward time constant and setting time in the relevant described limit feedforward gain of 3 described feedforward time constants obtaining.

17, the automatic adjusting method of control device of electric motor according to claim 16 is characterized in that,

Have:

During 3 point data of the setting time in the relevant described limit feedforward gain of more described feedforward time constant, suppose that transverse axis is that described feedforward time constant, the longitudinal axis are the curve of the setting time of described limit feedforward gain, judgement be curve to lower convexity, or dullly increase, or the dull step that reduces;

If this judged result is to lower convexity, then in 3 point data, the mid point of the either party in intermediate data and remaining two side datas is obtained the step of 1 data again;

If described judged result is dull increasing, then do not change data break, in the middle of 3 point data, stress newly to obtain the step of 1 data at the minimum time constant; And

If judged result is dull the minimizing, then do not change data break, in 3 point data maximum time constant stress newly to obtain the step of 1 data.

18, the automatic adjusting method of control device of electric motor according to claim 15 is characterized in that,

Set the Looking Out Time that is used to measure setting time according to user's input, the starting point of described Looking Out Time is made as position deviation carries out near moment after the zero crossing or its.

19, a kind of automatic adjusting method of control device of electric motor, this control device of electric motor possesses: the motor that combines with the driven object load by coupling shaft, drive the power converter of described motor, positioner according to the output of the deviation outgoing position controller between the position probing value of position command value and described motor, with described position command value is input, the position feed-forward controller of outgoing position feed-forward signal, with described positioner output and described position feed-forward signal plus, the totalizer of output speed command value, speed control according to the deviation output torque current command value of the speed detected value of described speed value and motor, according to described torque current command value with supply to the current controller that deviation between the torque current detected value of described motor is adjusted the output current of described power converter, the automatic adjusting method of this control device of electric motor is characterised in that to have:

The vibration amplitude of the position deviation waveform when the position command pattern is provided is no more than in the scope of permissible value, increases the step of the response frequency of described positioner and/or described speed control;

The a plurality of described position command pattern of login in advance as the position command for described positioner, is carried out the step that a plurality of runnings are moved; And

In a plurality of described running actions, adjust the controlled variable of described position feed-forward controller so that overshoot is no more than the step of permissible value.

20, the automatic adjusting method of control device of electric motor according to claim 19 is characterized in that,

Have: the step that described overshoot is no more than the controlled variable of the little described position feed-forward controller of the condition of permissible value and the setting time in each position command pattern is satisfied in search.

21, the automatic adjusting method of control device of electric motor according to claim 19 is characterized in that,

Have:

Accept the step of the setting input of regularization condition from the user interface; With

Accept the step of the selection input of one of a plurality of described position command patterns.

22, a kind of automatic regulating apparatus of control device of electric motor, this control device of electric motor possesses: the motor that combines with the driven object load by coupling shaft, drive the power converter of described motor, positioner according to the deviation output speed command value between the position probing value of position command value and described motor, speed control according to the deviation output torque current command value of the speed detected value of described speed value and motor, according to described torque current command value with supply to the current controller that deviation between the torque current detected value of described motor is adjusted the output current of described power converter, the automatic regulating apparatus of this control device of electric motor is characterised in that to have:

The position command pattern generates mechanism, and it generates the position command value of the position command pattern of continually varying adjustment running usefulness as described positioner; And

Response frequency maximization mechanism, it is providing the position command value of this position command pattern as described positioner, when making the control device running of described motor, vibration amplitude at the position deviation waveform is no more than in the scope of permissible value, increases the response frequency of described positioner and/or described speed control.

23, a kind of automatic regulating apparatus of control device of electric motor, this control device of electric motor possesses: the motor that combines with the driven object load by coupling shaft, drive the power converter of described motor, positioner according to the output of the deviation outgoing position controller between the position probing value of position command value and described motor, with described position command value is input, the position feed-forward controller of outgoing position feed-forward signal, with described positioner output and described position feed-forward signal plus, the totalizer of output speed command value, speed control according to the deviation output torque current command value of the speed detected value of described speed value and motor, according to described torque current command value with supply to the current controller that deviation between the torque current detected value of described motor is adjusted the output current of described power converter, the automatic regulating apparatus of this control device of electric motor is characterised in that to have:

Response frequency maximization mechanism, the vibration amplitude of its position deviation waveform when the position command pattern is provided is no more than in the scope of permissible value, increases the response frequency of described positioner and/or described speed control;

Adjust works, a plurality of described position command pattern that it will be logined in advance provides as the position command for described positioner, carries out a plurality of running actions; And

Parameter adjustment mechanism, it adjusts the controlled variable of described position feed-forward controller so that overshoot is no more than permissible value in a plurality of described running actions.

Images