DocumentCode
2856016
Title
LQG based robust tracking control of blood gases during extracorporeal membrane oxygenation
Author
Smith, D.J. ; Porumamilla, Hemanth V.
Author_Institution
Mech. Eng., Cal Poly State Univ., San Luis Obispo, CA, USA
fYear
2011
fDate
June 29 2011-July 1 2011
Firstpage
324
Lastpage
329
Abstract
The paper presents three different types of LQG based controllers designed for tracking control of the arterial partial pressures of the blood gases; oxygen and carbon dioxide (O2 and CO2) during extracorporeal membrane oxygenation (ECMO). ECMO is a method of support for the heart and or lungs in severely ill patients. In this procedure, the blood is circulated out of the body and into an ECMO machine where the O2 and CO2 gas levels are restored before being pumped back into the body. The performance of each of the controllers was ascertained both from a tracking as well as disturbance rejection standpoint to step commands in both these inputs. Robustness analysis was also performed on all the closed loop configurations using the structured singular value analysis. The uncertainty was considered to be of a structured parametric type and was captured by the perturbation of the system "A" matrix. Performance of all the controllers was tuned so as to ensure robust stability to these parametric uncertainties. The LQG and LQG/LTR with feed forward control were able to achieve good tracking performance. But, only the LQG augmented with integral control was able to achieve accurate tracking of the arterial partial pressures of the blood gases in the presence of a step input disturbance in the blood gases flow rates.
Keywords
carbon compounds; closed loop systems; control system synthesis; feedforward; linear quadratic Gaussian control; medical control systems; oxygen; position control; robust control; singular value decomposition; LQG based robust tracking control; arterial partial pressures; blood gases flow rates; closed loop configurations; disturbance rejection standpoint; extracorporeal membrane oxygenation machine; feedforward control; integral control; parametric uncertainties; robust stability; step input disturbance; structured parametric type; structured singular value analysis; system A matrix; Blood; Closed loop systems; Equations; Feedforward neural networks; Gases; Mathematical model; Uncertainty;
fLanguage
English
Publisher
ieee
Conference_Titel
American Control Conference (ACC), 2011
Conference_Location
San Francisco, CA
ISSN
0743-1619
Print_ISBN
978-1-4577-0080-4
Type
conf
DOI
10.1109/ACC.2011.5991330
Filename
5991330
Link To Document