Electricity: motive power systems – Positional servo systems – 'reset' systems
Patent
1992-02-14
1993-06-15
Ip, Paul
Electricity: motive power systems
Positional servo systems
'reset' systems
318609, 318561, 364148, 364165, G05B 2102
Patent
active
052202658
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to repetitive control for causing a controlled object to conduct a repetitive action in response to a cyclic target input, and more particularly, to a discrete-type repetitive control method and an apparatus therefor, having an excellent response to a cyclic target input and excellent flexibility to a variation in the cycle period of a target input.
2. Description of the Related Art
Conventionally, it has been known to control a controlled object in a manner causing the object to repeat a single patterned action, by the use of a discrete-type repetitive control system which is constituted in accordance with an internal model principle and adapted to receive a cyclic target input which changes in a single changing pattern periodically repeated. According to the internal model principle, it is possible to cause an output of the controlled object to follow a target input with no steady deviation, by incorporating a target input generation model in a closed loop of the control system. Conventionally, in view of the internal model principle, a discrete-type repetitive control system has been so designed as to stabilize an extended system which is configured by connecting a compensator, including the target input generation model, to the controlled object.
In this case, a cyclic target input generation model represented by 1/(Z.sup.L -1) is employed, in view of equation (3) obtained by substituting equation (2) into a Z transform Yr(Z) of a cyclic target input y(t) represented by equation (1). In equations (1)-(3), symbols Z.sup.-1 (i=1, 2, - - - , L, 2L, - - - ) each represent a time delay element, and symbol L represents the cycle of the cyclic target input. ##EQU1##
Namely, as shown in FIG. 1, the conventional disorete-type repetitive control system accommodates therein an internal model 1/(Z.sup.L -1) which is comprised of L' time delay elements Z.sup.-1. The number L' of the time delay elements Z.sup.-1 is equal to a value obtained by dividing the cycle period L of the cyclic target input by a sampling cycle period .tau. of the cyclic target input in the control system. In FIG. 1, symbol yr(i) represents the cyclic target input; u(i), an input to the controlled object; and y(i), an output of the controlled object. Symbol x(i) indicates a state variable of the controlled object; and e(i), a deviation between the output y(i) of the controlled object and the cyclic target input yr(i). Further, symbol h represents a feedback gain element having a properly selected gain, and K1-KL indicate gain compensators having properly selected gains.
According to the repetitive control system, a stable control system is provided. However, if the state variable x (i) cannot be observed directly, then the repetitive control system shown in FIG. 1 cannot be configured. Further, according to the control system, the number L' of the time delay elements Z.sup.-1 must be equal to the value obtained by dividing the cycle L of the target input by the sampling cycle .tau.. Therefore, the conventional repetitive control system lacks flexibility to a variation in the cycle period of a target input.
SUMMARY OF THE INVENTION
An object of the present invention is to provide a discrete-type repetitive control method and an apparatus therefor, which enable stable repetitive control even if a state variable of a controlled object is not observable, and which have an excellent response to a cyclic target input and excellent flexibility to a variation in the cycle period of a target input.
To achieve the aforementioned object, according to one aspect of the present invention, a discrete-type repetitive control method comprises the steps of: (a) periodically applying, to each of time delay elements which are equal to or greater in number than a value obtained by dividing a cycle period of a cyclic target input by a cycle period of sampling a deviation between an output of a controlled object and the cyclic target input, a sum of a product of the deviation pe
REFERENCES:
patent: 3988538 (1976-10-01), Patten
patent: 4904912 (1990-02-01), Yamamoto
patent: 4918584 (1990-04-01), Goff
patent: 5050119 (1991-09-01), Lish
patent: 5065263 (1991-11-01), Yoshida et al.
patent: 5119287 (1992-06-01), Nakamura et al.
Araki Mitsuhiko
Iwashita Yasusuke
Sakamoto Keiji
Fanuc Ltd.
Ip Paul
LandOfFree
Discrete-type repetitive control method and an apparatus therefo does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Discrete-type repetitive control method and an apparatus therefo, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Discrete-type repetitive control method and an apparatus therefo will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1045933