GA-optimized FLC-driven semi-active control for phase-II smart nonlinear base-isolated benchmark building

Abstract

An optimized fuzzy logic control (FLC) algorithm is developed for the phase-II smart base-isolated benchmark building with nonlinear isolation system. A restart genetic algorithm-based optimization strategy has been used to change the fuzzy system properties like the fuzzy rule base, pre-scale gains, membership function type and parameters at every simulation step. Acceleration and relative velocity responses at the damper location have been taken as inputs to the FLC system. Voltage required by the magneto-rheological (MR) damper is obtained as an output from the FLC. The use of MR dampers in the benchmark study as a control device along with isolation bearings in the building renders the overall system nonlinear. The advantage of using a fuzzy rule base is its inherent ability to handle nonlinearities and uncertainties in structural behavior, input excitation, sensor, and actuator dynamics. As a consequence, FLC provides robustness to the control mechanism. Moreover, FLC-driven MR damper voltage monitoring provides a gradual and smooth change of voltage. In the present study, the number of sensors and actuators and their locations have been kept unchanged as in the sample controller provided in the benchmark study. Simulation results for FLC, the adaptive rule and fixed rule base type, and results from the sample controller provided have been tabulated and compared. Results obtained indicate improvement using the proposed control approach without considering a multi-objective nondominated solution, where the weight of objective functions can be varied. A stability test for the proposed adaptive rule base FLC has been shown.