شماره ركورد كنفرانس :
3926
عنوان مقاله :
MISO GPC for Blood Glucose Control in T1DM
پديدآورندگان :
Mohammadi Zahrani Mina mina.mohammadi@ec.iut.ac.ir Electrical and Computer engineering Department Isfahan University of Technology Isfahan, Iran, 83111-84156 , Zekri Maryam mzekri@cc.iut.ac.ir Electrical and Computer engineering Department Isfahan University of Technology Isfahan, Iran, 83111-84156 , Taghian Dinani Soudabeh s.taghian@ec.iut.ac.ir Electrical and Computer engineering Department Isfahan University of Technology Isfahan, Iran, 83111-84156 , Kamali Marzieh m.kamali@cc.iut.ac.ir Electrical and Computer engineering Department Isfahan University of Technology Isfahan, Iran, 83111-84156
كليدواژه :
T1DM , GPC , adaptive control , system identification
عنوان كنفرانس :
بيست و چهارمين كنفرانس مهندسي برق ايران
چكيده فارسي :
In this paper, a new two-input one-output model, derived from Hovorka model -a famous SISO model of type-I diabetics- is proposed, with both the insulin and glucose as inputs. For the new model, a multiple GPC combined with a system identification system is also designed. This controller is used to regulate the B.G. level within the normal band in less than 2 hours after meal intake. It can handle the inherent time delays existed in the system dynamics. In comparison to the single-GPC used for Hovorka mode, the proposed procedure removes the need for the feed-forward controller. The new procedure can also regulate the blood glucose automatically and more accurately and prevent diabetics from hypoglycemia, a fatal phenomenon. The validation of the new procedure is shown via simulations and via the Clarke error grid analysis, an accepted clinical analysis.