14689

13223

كارشناسي ارشد

كنترل

اصفهان : دانشگاه صنعتي اصفهان

1398

يازده، 105 ص. : مصور، جدول، نمودار

جعفر قيصري، فريد شيخ الاسلام

محسن مجيري

1398/03/29

كتابنامه

مهندسي برق

مهندسي برق و كامپيوتر

1398/03/29

2542703

Fuzzy Event Triggered Control In Multi Agent Systems Faezeh Lal Kheftan f lal@ec iut ac ir May 21 2019 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language Farsi Supervisors Dr Jafar Ghaisari Dr Farid Sheikholeslam ghaisari@cc iut ac ir sheikh@cc iut ac ir Abstract Consensus control of multi agent systems has been intensively studied over the past two decades Distributed consensuscontrol of multi agent systems are employed in various industrial applications such as sensors data fusions cooperativerobots and vehicles Event triggered approaches usually reduce the frequency updates of input control signals as well assystem resource consumption In these methods controller s input signals are only calculated once and update momentsis reduced in compare with time triggered approaches However The bigger amplitude of applied demands may result inmore depreciation damage and maintenance costs to the actuators Therefore the design of event detection mechanism andcontrol strategy are very challenging In this thesis we proposed an event triggered consensus control protocol and an eventdetection mechanism to decrease control signal variations The distributed detection mechanism is defined based on controlvariations in each agent measurement errors and the state errors in respect to neighbors In the proposed event triggeredconsensus based protocol the difference between control signal of each agent and its latest update is used to reduce theamplitude of the steps that are generated in the input control The controllers gain is computed by a Riccati equation foreach agent separately In order to improve the efficiency of the proposed method several coefficients of the proposed eventdetection functions have been fuzzified using the Takagi Sugeno Fuzzy System These coefficients have been shown to agreater impact on the event detection function compared to the non fuzzified case To prove consensus for the proposeddetection function linear algebraic relations methods and graph theory have been used To evaluate performance of theproposed methods several simulations have been carried out and the results have been compared The proposed methodsdecrease control signal variations as well as number of events resulting in reduction of system wear The factor is greatimportance in systems comprising numerous agents Key Words Multi Agent System Event Triggered Control Consensus Problem Takagi SugenoFuzzy System

جعفر قيصري، فريد شيخ الاسلام

محسن مجيري