JPS6120102A - Adjusting device of feedforward gain - Google Patents

Abstract

PURPOSE:To remove the influence of response delay and to obtain a feedforward gain adjusting device allowed to follow the status change of a system by automatically adjusting the transmitting function of a disturbance compensating element. CONSTITUTION:A stabilization detecting circuit 5 in the feedforward gain adjusting device discriminates whether a control system is stabilized or turned to transient status in accordance with the change of a set point SP and a control variable PV in the control system, the change of disturbance D, or the change of a manipulated variable C based upon feedback control. A transmitting function adjusting circuit 6 generates a feedback control initializing command 7 on the basis of a stabilization signal outputted from a stabilization detecting circuit 5 to initialize a transmitting function GB of a control device 1 and sends an adjusting command 8 for a transmitting function GF to a disturbance compensating element 4 to execute readjustment. When a deviation between the set point SP and the control variable PV is compensated so as to be ''0'' under feedback control, the transmitting function GF of the element 4 is adjusted by the compensated value.

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention relates to feedforward gain adjustment [
In particular, the present invention relates to a feedforward gain adjustment device suitable for automatically adjusting the gain of a system including a transfer function of a disturbance compensation element of a feedforward control system.

[Technical background of the invention]

In general, the feedforward control device is
It has a configuration as shown in the diagram. In the figure, there is jf2, and control device 1 is set! 11S P and the feedback control an m p v are matched and the difference between them is expressed as a transfer function G
The control file C is processed by B and sent to process 2.

The control is affected by disturbances, which are added to the output of the PROHIS 2 via the transfer function CD of the disturbance element 3, and this becomes the mi control fit PV.

In order to reduce the influence of this disturbance (MG), the disturbance is compensated for through the transfer function OF of the compensation prefecture 4.
This is added to the manipulated variable C, and this is added to PROHIS2.

As is clear from Fig. 1, this feed-for-two control device has a J effect of disturbance on the control amount Pv, Δ=D-G
In order to cancel D in advance and cancel J-, the disturbance is detected and the necessary compensation value B = D - G E
A control method is adopted in which the amount is added to the manipulated variable C and the result is given to PROHIS2. For this reason, the control fM given by the configuration shown in FIG.
PV is PV = (C+B)・GP10D-GO −(C+D-GF)・GP10D-GD −D・(Gr knee Gp+Go)+C-Gp...
... (1) Therefore, when the disturbance o h < a, the control 1d
In order to prevent P V from being affected, G F
−G p + G O= O・・・・・・・・・ (2
), so GF=-G, ]/GP ・・・・・・・・・ (
3) is derived. By setting J and the transfer function G += of the disturbance compensation element 4 that satisfy this equation (3), the influence on the control ia Pv caused by the disturbance can be preempted and applied H'5-= I can do things.

[Problems with back ode technology]

However, generally bu[1t? It is difficult to completely understand the transfer function Gp of the element 2 and the transfer function GD of the disturbance element 3, and their characteristics may change over time, so the optimum transfer function GE of the disturbance compensation element 4 is set. It is difficult to do so.

For this reason, feedforward control is generally rarely used alone, but is used in combination with feedback control as illustrated in FIG. In other words, the deviation between the setting +1i'l S P and the control amount PV that occurs because the gain setting of the transfer function GF of the disturbance compensation element 4 is not perfect [
Decrease J, urchin, and control device 1 corrected operation interval C.
A method is used in which τ is calculated and further correction is made using the manipulated variable [3] of the feedforward system.

However, in such a system, since the feedback control system is often unable to correct the effects of disturbances, it is difficult to overcome the inherent shortcomings of the feedback system. In other words, in the feedback system,
Since the existence of a deviation is the basis of the corrective action, the delay between R and the time when the corrective action is performed after a deviation occurs in the control amount Pv due to fluctuations in external disturbances such as load becomes a problem. In other words, the conventional control system illustrated in FIG. A control deviation occurs due to a change in the characteristics of the function GD,
Since a method is used to compensate for this using the feedback control system, even if the feedback control corrects the deviation that occurs during a certain disturbance so that the deviation becomes zero, the σ disturbance will occur again. When this occurs, the same deviation occurs again and must be corrected using feedback fi control. In order to deal with such problems, the transfer function 01 of the disturbance compensation tri element 4 should be optimally set so that the control deviation can be minimized even when the disturbance fluctuates. In reality, there is no established method for automatically and optimally adjusting this disturbance compensation function, and some countermeasures have been required.

[Purpose of the invention]

Therefore, the object of the present invention (1) is to eliminate the vacancies in the above-mentioned prior art, and to realize an optimal feedforward system by automatically adjusting the transfer function of the disturbance compensation element. We provide a forward gain adjustment device.

(Summary of the Invention) In order to achieve the above object, the present invention provides a feedback control means that generates a manipulated variable based on the deviation of the set planting control amount and applies it to the process, and a feedback control means that corrects the manipulated variable based on disturbance elements. a feedforward control means for adding the feedback control means; a detection means for detecting a small value when the feedback control means is in a stable state; and a detection means for adjusting the gain of the feedforward control means based on the output of the detection means and the operation amount of the feedback control means. The present invention also provides a feedforward gain adjustment device including adjustment means for initializing the feedback control means.

[Embodiments of the invention]

Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 2 is a block diagram of a feedforward gain adjustment device according to an embodiment of the present invention. In the same figure, the stability detection circuit 5 detects the set value SP and all controls in the control system.
It has the function of determining whether the control system is stable or in a transient state based on changes in mPV, changes in disturbance, changes in operation IC due to feedback control, etc. On the other hand, the transfer function adjustment circuit 6 generates a feedback control initialization command 7 based on the stability signal 9 from the stability detection circuit 5, initializes the transfer function GB of the control device 1 that performs feedback control, and also initializes the transfer function GB of the control device 1 that performs feedback control.
By sending the adjustment command 8 of the transfer function GF to the GF, the function of readjusting the transfer function 01 of the disturbance compensation element 4 is improved.

The operation of this configuration will be explained next.

When there is a disturbance D1, the stability detection circuit 5 detects that there is no deviation [1 (El 0) from the set value SP1 and the control amount PV1 at that time, and that the manipulated variable C1 due to feedback control is Based on a constant value, it is detected that the current control system is stable. That is, at this time, disturbance D
1, the optimum operation amount is the feedforward system operation amount, that is, the sum B1+C1 of the compensation value B1 and the feedforward operation amount C1. In this state, the optimum output of the transfer function G[- of the optimum disturbance compensation element 1a summary 4 is 81 +01, not the compensation value B1, and therefore the transfer function adjustment circuit 6 of the disturbance compensation element 4 receives the disturbance J. is D
l's Togi disturbance compensation 2! The transfer function G1 is adjusted so that the output of element 4 becomes Bl-toC1.

At the same time, initialize (CI = O) the output of the control variable [transfer function "[3] in the feedback control system, stabilize the feedback system 911 U, and automatically change the transfer function GE of the disturbance compensation element 4 in this state. The reason why it is necessary to detect the stability of the control system using the stability detection circuit 5 is that when the control system is in a transient state, for example, when disturbances are fluctuating or when the operation IC of the feedforward system During the change in , it is difficult to optimally adjust the transfer function GF of the disturbance compensation element 4 due to the dynamic characteristics of the process.

Next, a specific example in which the transfer function GF of the disturbance compensation element 4 is obtained using a 10-fold line and a specific example of its adjustment method will be described with reference to the characteristic diagram in FIG. Incidentally, in FIG. 3, the horizontal axis represents the disturbance D1, and the vertical axis represents the output B of the disturbance compensation element 4. Furthermore,
(1)0,130).

(DI, Bl)...(Dlo, BIO
) is the transfer function GF of the disturbance compensation element 4 set in advance.
This is the turning point where the decision is made.

Now, let us say b that an adjustment range of ±α is set for each point of the disturbance. For example, if the disturbance that keeps the control system in a stable state is D4' and the sum of the manipulated variables of the feedforward system and the feedback system at that time is I34', then 04' is the adjustment of D4. It is within the range, that is, D4-α≦D4' ≦D4+α... (4) is established, then 1) D4°B4), where the broken line of 4 is dotted, is newly converted to D4'. Broken line points (r)4', r34')
Then, the broken line connecting (D3, 83) and (05, 135> is determined as the transfer function GF of the new disturbance compensation element 4.

Therefore, in the control system of this embodiment, when a deviation occurs between the setting l1isp and the control ff1PV when a certain disturbance D1 occurs, and the feedback υ control is used to correct the Q difference to zero, this correction Since the transfer function G of the disturbance compensation element 4 is adjusted according to the value, the disturbance is again reduced to Dl.
When this happens, there is an advantage that no deviation occurs between the set value SP and the control amount PV, and the transfer function Gp of process 2 and the transfer function GD of disturbance element 3 will change in characteristics or change over time. 4Tbu[]t? At that time, the disturbance compensation is 1$
The adjustment of the transfer function GE in line 4 can be automatically performed online.

By the way, the adjustment range of the transfer function GF of the disturbance compensation 11 cable 4 may be limited to some extent. In this case, if the limit value of the adjustment range is α, then when C1>α, the transfer function G1 is adjusted to make the output of the disturbance compensation element 4 81+α, and the transfer function G13 of the control device 1 is adjusted to make the output Initialize 11 to 01-α.

〔Effect of the invention〕

As described above, according to the present invention, in a feedforward system that also has a feedback system, the transfer function of the disturbance compensation element can be automatically adjusted according to the system state, so that the response of the feedback system is It is possible to obtain a feedforward gain adjustment device that is not affected by delays and can appropriately follow changes in the state of the system.

[Brief explanation of drawings]

Fig. 1 is a block diagram of feedforward gain adjustment KFff according to an embodiment of the present invention, Fig. 2 is a block diagram of a conventional feedforward control ill device, and Fig. 3 explains the operation 1j of the configuration shown in Fig. 1. It is a characteristic diagram for J. DESCRIPTION OF SYMBOLS 1...Control device, 2...Process, 3...Disturbance element, 4...Disturbance compensation element, 5...Stability detection circuit,
6...Transfer function adjustment circuit. Applicant's representative Ino +1Q Kiyoshi U-→-

Claims (1)

[Scope of Claims] 1. Feedback control means that generates a manipulated variable based on the deviation between the set value and the controlled variable and applies it to the process, feedforward control means that corrects the manipulated variable based on disturbance elements, and feedback A detection means for detecting that the control means is in a stable state, and an adjustment means for adjusting the gain of the feedforward control means and initializing the feedback control means based on the output of the detection means and the operation amount of the feedback control means. A feedforward gain adjustment device featuring: 2. The feedforward gain according to claim 1, wherein the detection means determines the set value, the controlled amount, the manipulated amount, and the disturbance over time to determine whether the feedback control system is in a stable state. Adjustment device.