Neural Network Based Active Control of Base Isolated Structure Considering Isolator Nonlinearity

Hybrid control systems that combine passive and active strategies have emerged as effective solutions to enhance structural resilience during earthquakes. In parallel, recent advancements in smart structures have integrated active tuned mass dampers (ATMDs) to precisely control dynamic responses and mitigate seismic hazards. Simultaneously, the rise of Artificial Intelligence (AI), particularly machine learning algorithms, has opened new frontiers in structural control. This study proposes a novel AI-based approach for controlling structures equipped with nonlinear base isolation and an ATMD. Specifically, an artificial neural network (ANN) is employed, trained through supervised learning using the Levenberg-Marquardt backpropagation algorithm, to minimize displacement demands during strong earthquakes. The ANN-driven controller effectively reduces seismic responses, accounting for the nonlinear hysteretic behavior of the isolation system. The main objectives of the study are to achieve significant response reduction with fewer sensors compared to traditional control algorithms, increase system robustness against signal time delay and white noise contamination. To validate the proposed methodology, an ATMD is installed at the base isolation layer of an 8-story benchmark building. The ANN-based controller's performance is evaluated under both near-field and far-field seismic excitations and is compared with a conventional linear quadratic regulator (LQR) controlled ATMD. Additional robustness tests consider time delays and white noise in the input signals. Results demonstrate that the ANN-driven ATMD notably reduces key dynamic response parameters, including peak base acceleration, displacement, velocity, inter-story drift, maximum drift and base shear. The proposed ANN controller achieves performance comparable to the full-state LQR controller but requires significantly fewer sensors, enhancing both practicality and cost-effectiveness for real-world applications.