TY - GEN
T1 - A distributed and cooperative supervisory estimation of multi-agent systems - Part II
T2 - 2009 Canadian Conference on Electrical and Computer Engineering, CCECE '09
AU - Tousi, M. M.
AU - Azizi, S. M.
AU - Khorasani, K.
PY - 2009
Y1 - 2009
N2 - A framework for supervisory cooperative estimation in multi-agent linear time-invariant (LTI) systems is presented in the companion work (Part I). We introduced a set of sub-observers such that each estimates some states with a given set of input, output, and state information. A discerete-event system (DES) supervisory control framework is used for cooperation among the sub-observers. The supervisor selects a set of sub-observers to successfully estimate all states of the multi-agent system. In addition, in presence of a fault in the system, the supervisor reconfigures the set of selected sub-observers to minimize the fault impact on the estimation performance. Our general framework can be applied to any multi-agent system including industrial processes. In the companion paper (Part I), our proposed framework for the supervisory estimation is developed based on the notion of sub-observers and DES supervisory control. In this paper (Part II), a DES-based combinatorial optimization method for selection of an optimal set of sub-observers is presented, the feasibility of the overall integrated sub-observers is validated, and the application of our proposed method in a practical industrial process is demonstrated through numerical simulations.
AB - A framework for supervisory cooperative estimation in multi-agent linear time-invariant (LTI) systems is presented in the companion work (Part I). We introduced a set of sub-observers such that each estimates some states with a given set of input, output, and state information. A discerete-event system (DES) supervisory control framework is used for cooperation among the sub-observers. The supervisor selects a set of sub-observers to successfully estimate all states of the multi-agent system. In addition, in presence of a fault in the system, the supervisor reconfigures the set of selected sub-observers to minimize the fault impact on the estimation performance. Our general framework can be applied to any multi-agent system including industrial processes. In the companion paper (Part I), our proposed framework for the supervisory estimation is developed based on the notion of sub-observers and DES supervisory control. In this paper (Part II), a DES-based combinatorial optimization method for selection of an optimal set of sub-observers is presented, the feasibility of the overall integrated sub-observers is validated, and the application of our proposed method in a practical industrial process is demonstrated through numerical simulations.
UR - http://www.scopus.com/inward/record.url?scp=70350220722&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=70350220722&partnerID=8YFLogxK
U2 - 10.1109/CCECE.2009.5090285
DO - 10.1109/CCECE.2009.5090285
M3 - Conference contribution
AN - SCOPUS:70350220722
SN - 9781424435081
T3 - Canadian Conference on Electrical and Computer Engineering
SP - 1022
EP - 1027
BT - 2009 Canadian Conference on Electrical and Computer Engineering, CCECE '09
Y2 - 3 May 2009 through 6 May 2009
ER -