Structural control design: a reliability-based approach

This paper develops probabilistically-based optimal control strategies which focus on increasing the safety and reliability of a seismically excited structure by minimizing the extreme responses of the structure. A performance function is chosen that is directly related to the failure probability for the structure. Optimization is performed over a parameterized class of controllers, subject to a probabilistic constraint on the available control effort. System uncertainty is modelled parametrically and explicitly incorporated into the optimization algorithm. First and second order reliability methods (FORM/SORM) are employed for calculation of the required probabilities. Numerical examples using a single story structure subject to seismic excitation are provided to illustrate the approach. Comparisons are made with extensive Monte Carlo simulation to demonstrate the efficiency and accuracy of the approach.