Burkert Type 1110 - Handbuch (Seite 74)

72 - 1110
DIGITAL INDUSTRIAL CONTROLLER
english
To achieve a stable response of the control loop, use must be made of the controller
that matches the given controlled system. If this is not done, stable operation of the
control loop will not be possible (e.g. it will oscillate) and control will also not be
possible. Therefor, the structure of the controller must be adapted to the
characteristics of the controlles system and its parameters must be chosen so as to
ensure that a control progression will be achieved for the controlled variable that has
a short stabilisation time, little overshoot and good attenuation.
The controller parameters can be set on the basis of setting rules (see Annex).
The Digital Controller has a self-optimisation function that assumes the frequently
time-consuming task of adapting the controller’s parameters to the process. Two
self-optimisation algorithms have been implemented, an adaption algorithm and a
tuning algorithm.
7 SELF-OPTIMISATION
7.1 Stability and control quality
The core of self-optimisation by adaption consists of a fuzzy logic module. In
analogy to the procedure followed by an experienced control technician, conclusions
as to the quality of the set controller parameters are drawn on the basis of the
characteristic attributes of the transition response in the closed control loop. The
expert knowledge required to do this is stored in the form of linguistic rules (rule
base) in the controller’s EPROM and is used by the fuzzy logic algorithm during the
course of adaption (Figure 38).
A „tune“ module is provided in addition to adaption for non-recurring and direct
determination of the controller parameters. The controller parameters are calculated
on the basis of a modified Ziegler-Nichols process (Figure 39 and Annex).
7.2 Principle of self-optimisation by adaption
7.3 Principle of self-optimisation by tuning