Depot Institutionnel de l'UMBB >
Publications Scientifiques >
Publications Internationales >
Veuillez utiliser cette adresse pour citer ce document :
http://dlibrary.univ-boumerdes.dz:8080/handle/123456789/4786
|
Titre: | Advanced control algorithm : applications to industrial processes |
Auteur(s): | Ramdani, Ammar Grouni, Said Traïche, M. |
Mots-clés: | Dynamic matrix control predictive control Water heater |
Date de publication: | 2015 |
Collection/Numéro: | International Journal of Information and Electronics Engineering, Vol. 5, N°6 (2015);pp. 398-405 |
Résumé: | As in the most industrial systems, a control of the
input of the systems including a classic regulator is a key point.
The Proportional-Integral-Derivative controllers are commonly
used in many industrial control systems and appeared suitable
to stable the control of the majority of real processes. But in
some cases like a non-minimum-phase plant or a plant with a
dead-time proceed to a thin regulating of coefficients until to get
a system respecting the conditions specified. It is possible also to
present a problem of overtaking with the increase of the gain or
seems impotent for systems having a big delay and the use of
sophisticated process controllers is required. Model predictive
control is an important branch of automatic control theory, it
refers to a class of control algorithms in which a process model is
used to predict and optimize the process performance. MPC has
been widely applied in industry. Dynamic Matrix Control
Algorithm belongs to the family of Model predictive control
Algorithms where these algorithms only differ between
themselves in the model that represents the process, disruptions
and the function of cost. In this paper the study of the Dynamic
Matrix Control Algorithm are interested while applying him on
processes of water heating and mechanical rotations of steering
mirrors in a Light Detection and Ranging system as a second
application. The objective of this work consists of solving the
problem of prediction of the output and input of the process by
fixing a horizon finished N, and while considering the present
state like initial state, to optimize a cost function on this interval,
while respecting constraints. Therefore, the future reference is
known and the system behavior must be predictable by an
appropriate model. It results an optimal sequence of N control of
it among which alone the first value will be applied effectively.
As the time advances, the horizon of prediction slips and a new
problem of optimization is to solve while considering the state of
the system updating. In summary, every moment, it is necessary
to elaborate an optimal control sequence in open loop, refined
systematically by the present measure arrival |
URI/URL: | http://dlibrary.univ-boumerdes.dz:8080/handle/123456789/4786 |
Collection(s) : | Publications Internationales
|
Fichier(s) constituant ce document :
|
Tous les documents dans DSpace sont protégés par copyright, avec tous droits réservés.
|