13 th World Conference on Earthquake Engineering Vancouver, B.C., Canada August 1-6, 2004 Paper No. 1562 SENSORS/ACTUATORS PLACEMENT IN CONTROLLING A MDOF FRAME Alessandro BARATTA 1 , Claudia CENNAMO 2 and Ottavia CORBI 3 SUMMARY In the paper one focuses on the possibility of individuating a criterion for optimally positioning at suitable locations of a structure a prefixed number of sensors and actuators of a control system of active type. This problem is of central importance when dealing with structural control systems with reduced monitoring and actuation equipment, which is the usual case when addressing some economical objectives. To this purpose a theoretical approach is proposed, whose effectiveness is proved by numerical investigation. INTRODUCTION The possibility of controlling seismic vibrations in civil structures has attracted in the last decades a wide attention from the world-wide scientists [1-2]. Actually mitigation of the structural response, under seismic action, can be achieved bringing suitable changes to the mechanical parameters of the system, in particular acting on the terms of the stiffness and damping matrices. To obtain the optimal solution for this problem, it is often necessary to calibrate the coefficients of the motion equation in measure incompatible with the properties of the common structural materials. This can be carried out by active control techniques. Actually a number of papers can be mentioned to this purpose, reporting original contributions to control strategies, emphasizing basic control peculiarities or focusing on specific control features [3-13]. Since most of the control approaches are based on a feedback of some structural response variables monitored during the dynamic motion, the performance of the preferred control strategy (besides the limits intrinsic to the procedure itself) mainly relies upon two features: i) the availability of an adequate set of data, i.e. the presence of a suitable number of sensors able to give a complete representation of the structure behaviour; ii) the possibility of properly generating and exerting the desired (theoretically defined) control action, which, in turns, mainly depends on the availability of a suitable number of actuators. On the other hand, the higher the number of the control system components is the more expensive the system results, both in terms of cost of the components and of required actuation energy. 1 Full Professor, Università di Napoli Federico II, Napoli, Italia. Email: alessandro.baratta@unina.it 2 Researcher, Seconda Università di Napoli, Aversa, Italia. Email: ccennamo@unina.it 3 Researcher, Università di Napoli Federico II, Napoli, Italia. Email: ottavia.corbi@unina.it