On the Use of Complex Gains in Virtual Feedback for Model Updating

It has recently been shown that a signal-processing implementation can replace real-time operation using actuators in the design of closed-loop eigenstructures. This virtual implementation eliminates the practical issues associated with closed-loop testing, removes the constraint of closed-loop stability, and allows computation of several closed-loop eigenstructures based on a single open-loop realization. Additionally, it is noted that virtual implementation opens the possibility of using complex-valued gains. No examination of the merit of complex gains has, however, been carried out, thus the focus in the present paper is to explore this and to establish recommendations on how to select the complex gains for use in closed-loop model updating. A numerical examination with a structural system is presented that demonstrates the basic points outlined in the paper.