JPH04302309A - Servo controller - Google Patents

Abstract

PURPOSE:To always maintain the optimum control system by a moment of inertia identification function by a servo controller even when the fluctuation of inertia of a mechanical system occurs. CONSTITUTION:A feedback control part 8 which outputs a feedback signal UFB to a control system according to the target value omegar and the mounting value (omega) of a velocity signal, a feedforward control part 7 which outputs a feedforward signal taur to the control system according to the target value omegar' of an acceleration signal, and a disturbance compensation part 9 which outputs a disturbance compensation signal da to the control system according to the input value (u) of a mechanical load 15 and the mounting value (omega) of the velocity signal are provided. Furthermore, a moment of inertia identification part 3 which identifies the moment of inertia of the mechanical load 15 by the disturbance compensation signal da and approximate acceleration phi1' is provided, and the control parameters of the feedback control part 8, the feedforward control part 7, and the disturbance compensation part 9 can be adjusted at the optimum values by identification output.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a servo control system that can always maintain an optimal control system by using a moment of inertia identification function even if the moment of inertia of a mechanical system changes.

0002

2. Description of the Related Art Conventional servo control devices are tuned on the assumption that the moment of inertia of a mechanical system is constant.

0003

[Problem to be Solved by the Invention] When the moment of inertia of a mechanical system is assumed to be constant as in the above conventional method, when the inertial load is replaced, on-site adjustment must be made again.
There was a problem in that it required a lot of time and effort.

The present invention was made in view of the above-mentioned conventional drawbacks, and even if the inertial load of a mechanical system is replaced, the moment of inertia is identified, and feedforward control, feedback control, and disturbance compensation optimal for that value are performed. An object of the present invention is to provide a servo control device that can automatically perform the following steps.

[0005]

[Means for Solving the Problems] The present invention provides a feedback control unit that inputs a target value (ωr) and an actual value (ω) of a speed signal and outputs a feedback signal (UFB) to a servo device, and a target value of an acceleration signal. A feedforward control section inputs the value (ωr') and outputs the feedforward signal (τr) to the servo device, and inputs the measured value (ω) to the input value (u) and speed signal of the servo device to perform disturbance compensation. In a servo control device having a disturbance compensation section that outputs a signal (da) to a servo device, a disturbance compensation signal (da
) and approximate acceleration (φ1': differential value of ω) is provided with a moment of inertia identification section that identifies the moment of inertia (J) of the servo device, and the output of this moment of inertia identification section is sent to the feedback control section and the feedforward control section. The present invention is characterized in that it is configured to be output as a parameter to a control section and a disturbance compensation section.

[0006]

[Operation] With the above configuration, even if the inertia of the machine fluctuates, the feedforward control section, feedback control section, and disturbance compensation section New control parameters are set at , and the optimal control system for the load is always maintained.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be specifically described below based on embodiments shown in the drawings.

As shown in FIG. 1, the servo control device according to the present invention has a target value 1 (ωr) and an actual measured value 12 (
ω) and outputs the feedback signal 11 (UFB) to the mechanical load (servo device) 15; a feedback control unit 8 that inputs the target value 2 (ωr′) of the acceleration signal;
Feedforward signal (torque command) 10 (τr)
a feedforward control section 7 that outputs to the mechanical load 15;
, the input value 14 (u) of the mechanical load 15 and the actual measured value (angular velocity) 12 (ω) of the speed signal are input to generate the disturbance compensation signal 4 (d
a) is provided to the mechanical load 15, and a disturbance compensation section 9 is provided which outputs the disturbance compensation signal 4 (da) and the approximate acceleration 5.
(φ1': differential value of ω) is input, the moment of inertia (J) of the mechanical load 15 is identified, and the feedback control unit 8
, a feedforward control section 7, and a moment of inertia identification section 3 that outputs as a parameter to a disturbance compensation section 9. Note that the feedforward signal 10 (τr) output from the feedforward control unit 7 and the feedback signal 11 (U
FB) and the disturbance compensation signal 4 output from the disturbance compensation section 9
(da) is input to the adder/subtractor 13, and feed forward signal 10 (τr) and feedback signal 11 (UF
B) is added and the disturbance compensation signal 4 (da) is subtracted, and the calculation result is input to the mechanical load 15 and the disturbance compensation section 9 as the generated torque 14 (u). Next, the operation of the above embodiment will be explained. The disturbance compensation section 9, the moment of inertia identification section 3, the feedforward control section 7, and the feedback control section 8 will be explained in this order. <Disturbance compensation section 9>

The role of the disturbance compensator 9 is to estimate an unknown torque disturbance acting on the servo system and feed it back to cancel the torque disturbance. A disturbance observer can be used for this purpose. A disturbance observer is obtained using the following procedure. The equation of motion of the servo system is shown in equation (1) (for a rotating system). Jω′=u+d−Dω

...(1) However, J: moment of inertia, ω: angular velocity, D:
Viscous resistance, u: generated torque, d: disturbance torque In a linear system, equation (2) is obtained. MV'=F+R-DV

...(2) However, M: mass, V: velocity, F: generated force, R:
Both the disturbance force rotation system and the linear system have equations in the same form, and the following explanation will be valid regardless of which one is adopted, so the explanation will be given using the rotation system as an example. For an observer that estimates only the disturbance d, if the time constant required for estimation is To, then
It is expressed by the following Laplace transform formula. da(s)=(Ja s+Da)ω(s)
/(Tos+1)
−u(s)/(To s+1)
...(3) However, da: estimated disturbance value, Ja: moment of inertia value of observer model Da:
Observer model viscous resistance value

[0010] Here, in order to accurately determine the disturbance, "J=Ja" must be satisfied. The disturbance compensator 9 is configured as shown in FIG. 2, and generates a disturbance compensation signal (estimated disturbance) 4 (da) and an approximate acceleration signal 5 (φ1') from the angular velocity 12 (ω) and the generated torque 14 (u). It has a function to output. <Moment of Inertia Identification Unit 3> Disturbances include those in which a constant load is applied and those in which the direction of force changes depending on the direction of velocity, such as Coulomb friction. The disturbance d is expressed as equation (4). d=-Dω-dc-Tc sign(ω)
…(
4) However, dc: steady disturbance, Tc: magnitude of Coulomb friction. A configuration diagram of the controlled object (mechanical load 15) at this time is shown in FIG. Also, let the error between the observer model and the actual moment of inertia be δJ, and δJ=J−Ja

...(5) is defined. At this time, if the influence of the error δJ is also regarded as a disturbance, the block diagram of the controlled object is represented as shown in FIG. 4, and equation (6) holds true as an equivalent disturbance. d=-dc-Tc sign(ω)-Dω-
δJω'...(6) Since the disturbance estimated value is output with a first-order delay with respect to equation (6), da = (-dc Tc sign(ω)-D
ω−δJω′)/(To s+1)


...(7). From equation (7), it can be seen that the estimated disturbance value includes the error δJ of the moment of inertia. If the error δJ can be obtained using this, it means that the moment of inertia of the observer model can be identified by setting Ja ←Ja + δJ. Below, we will show how to obtain the error δJ. φ1 (t) and φ2 (t) are defined as new signals.

[0011]

[Equation 1] From the above equations (8) and (9), equation (7) is da
=-Dφ1 (t)-δJφ1 ′(t)-φ2 (t
) +dao exp(-t/to
)
...(10) However, dao: Disturbance estimation initial value In a certain time interval [t1, t2], both sides of (10) have φ
Multiplying by 1' and integrating

0012

[Equation 2] Here, since the target value is a periodic function, φ1 becomes a periodic function as a result. Also, depending on whether it is forward rotation or reverse rotation, φ1
can be either positive or negative. In this case, φ1 >0 or φ1 <0

...(12) φ1 (t+nT)=φ1 (t
+(n+1)T) …(1
3) However, (n=0, 1, 2,...) holds true. At this time, t1 = t+nT, t2 = t+
(n+1)T

[0013]

[Formula 3] Also, according to equation (12), φ2 = -dc -Tc : φ1 > 0
φ2 = dc + Tc: φ1 <0
…(
16). Therefore,

[0014]

[Equation 4] In this way, when φ1 (t) satisfies equations (12) and (13), the error δJ is found using the interval orthogonality of equations (15) and (17). Also, equation (11), (15
), from equation (17)

[0015]

[Formula 5] From the above formula, at t→∞, from exp (-t/to)→0,
In practical terms t》To

[0016]

[Equation 6] Then, the identification is completed as "Ja ←Ja + δJ". Figure 5 shows the configuration of the moment of inertia identification section 3.
Shown below. <Feedforward control section 7>

The feedforward control section 7 is configured as shown in FIG. 6, and has a target value 2 (ωr
') and the moment of inertia identification value 6 (Ja) from the moment of inertia identification section 3, a feedforward signal (torque command) 10 (τr) shown in equation (20) is output. τr = Ja ωr ′

(20) As a result, the moment of inertia identification value is always set, and the feedforward amount is optimally adjusted. <Feedback Control Unit 8> The feedback gain in the feedback control unit 8 is calculated based on the moment of inertia identification value 6 (Ja). Considering the target characteristics of the control system in terms of speed control system, G(s)=ωc/(s+ωc)
…(
21). Here, ωc represents the control band. In order to keep the control band ωc constant regardless of inertia fluctuations, the speed control gain Kv is changed based on the moment of inertia identification value 6 (Ja). The characteristics of the control system are as follows: When the feedback control section 8 is configured as shown in FIG.
) …From (22), Kv = Ja ωc

...(23) is calculated. This ensures that the characteristics are always constant. In addition to this, PI control or PID control is also possible, and the above is an example.

Further, the concrete operation of the present invention will be explained. Here, we will take a speed control system as an example. As for the position control system, it is sufficient to add one more position control system to the outside, and this speed control system is necessary in any case. Figure 1
When the target value 1 (ωr ) of the speed signal and the target value 2 (ωr ′) of the acceleration signal are given, the disturbance compensation unit 9 sends the disturbance compensation signal 4 (da ) and approximate acceleration 5 (φ1') are input.

In the moment of inertia identifying section 3, first, the disturbance compensation signal 4 (da) is multiplied by the approximate acceleration 5 (φ1'), and the resulting value is integrated to become Wa. Further, after the approximate acceleration 5 (φ1') is squared, the value is integrated to become Wb.

[0020] The error δJ is determined by Wa /Wb,
Ja is updated by "Ja ← Ja + δJ" and is output as the moment of inertia identification value 6 (Ja). This moment of inertia identification value 6 (Ja) is input to the feedforward control section 7, the feedback control section 8, and the disturbance compensation section 9.

The feedforward control section 7, whose internal calculations are shown in FIG. and outputs its output as a feedforward signal 10 (τr).

In the feedback control section 8, internal calculations as shown in FIG. 7 are performed. Target value of speed signal 1 (ω
r) and the actual measured value (angular velocity) 12(ω) are input, and the deviation is calculated by "ωr - ω". This deviation is input to the compensator Gc (s), and the control amount is output as a feedback signal 11 (UFB). Here Gc (s)
The parameters within are the moment of inertia identification value 6 (Ja)
determined according to the size of the These are arbitrarily determined by the designer based on machine characteristics and control specifications. As an example, when changing only the speed control gain Kv, the above (23
) is calculated using the formula. As shown in FIG. 2, the disturbance compensator 9 uses the moment of inertia identification value 6 (Ja).

[0023] Then, in the adder/subtractor 13, the feedforward signal 10 (τr) and the feedback signal 1
1 (UFB) is added, and the disturbance compensation signal 4 (da
) is calculated and output as generated torque 14(u). The mechanical load 15 is shown in Fig. 3, and the actual measurement value (angular velocity) 1
2(ω) is output. Even if the mechanical load changes in this way, the moment of inertia identification section 3 calculates the moment of inertia and outputs it as the moment of inertia identification value 6 (Ja).Based on this, the feedforward control section 7 performs feedback control. Since the section 8 and the disturbance compensator 9 are adjusted, an optimal control system can be maintained at all times.

[0024]

[Effects of the Invention] As detailed above, according to the present invention, even if the inertia of a machine fluctuates, the moment of inertia identification section calculates the fluctuation and identifies the value of the moment of inertia.
In addition, the control parameters of the feedforward control section, feedback control section, and disturbance compensation section are updated in accordance with this value to optimally adjust to the magnitude of the moment of inertia, so when fluctuations in machine inertia occur. However, it is possible to maintain an optimal control system at all times.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram of a servo control device according to an embodiment of the present invention.

FIG. 2 is a configuration diagram of a disturbance compensation section in FIG. 1.

FIG. 3 is a configuration diagram of the mechanical load in FIG. 1.

FIG. 4 is a configuration diagram of equivalent disturbance in FIG. 1.

FIG. 5 is a configuration diagram of a moment of inertia identification section in FIG. 1.

FIG. 6 is a configuration diagram of a feedforward control section in FIG. 1.

FIG. 7 is a configuration diagram of a feedback control section in FIG. 1.

FIG. 8 is a configuration diagram showing a conventional speed servo control device.

[Explanation of symbols]

1(ωr)...Target value of speed signal, 2(ωr')...Target value of acceleration signal, 3...Moment of inertia identification unit, 4(d
a)...Disturbance compensation signal, 5 (φ1')...Approximate acceleration, 6
(Ja)...Moment of inertia identification value, 7...Feedforward control unit, 8...Feedback control unit, 9...Disturbance compensation unit, 10(τr)...Feedforward signal, 11(
UFB)...Feedback signal, 12(ω)...Actual measurement value (
angular velocity), 13...adder/subtractor, 14(u)...generated torque,
15...Mechanical load.

Claims (1)

[Claims] 1. A feedback control section that inputs a target value (ωr) and an actual measured value (ω) of a speed signal and outputs a feedback signal (UFB) to a servo device, and a feedback control unit that inputs a target value (ωr') of an acceleration signal. , a feedforward control unit that outputs a feedforward signal (τr) to the servo device, and a disturbance compensation signal (da) that inputs the input value (u) of the servo device and the actual measured value (ω) of the speed signal to the servo device. In a servo control device having a disturbance compensation section that outputs a disturbance compensation signal (da) and an approximate acceleration (φ
1': differential value of ω), identifies the moment of inertia (J) of the servo device, and outputs it as a parameter to the feedback control section, feedforward control section, and disturbance compensation section. Characteristic servo control device.