Optimal Design of Active Controlled System for Torsionally Coupled Framed Structure Using Metaheuristic Algorithms

This study presents the optimal design of an active controlled system for a torsionally coupled framed structure using metaheuristic optimization algorithms. The considered control system includes active control using a linear quadratic regulator (LQR) algorithm, a hydraulic actuator, and sensors. A two-stage optimization approach is introduced: the first stage determines the optimal actuator location using the controllability index method, while the second stage optimizes the LQR weighting matrices through six metaheuristic algorithms. The optimization process employs six metaheuristic techniques: particle swarm optimization, genetic algorithm, grey wolf optimizer, whale optimization algorithm, ant lion optimizer, and improved grey wolf optimizer. A key novelty of this work lies in the comprehensive evaluation of LQR controller tuning for torsionally irregular structures, using a range of objective functions that include nondimensionalized lateral, torsional, and combined displacements. Additionally, constraints based on response effectiveness factors for displacement, acceleration, and story drift are incorporated to enhance the control design’s robustness. To study the efficacy of an optimized active control system compared to the exhaustive search method, the various controlled responses of the system are compared with the corresponding uncontrolled system. The study demonstrates the capability of each algorithm to fine-tune the LQR weighting matrices to achieve superior dynamic response control under varying seismic excitations.