Connecting two buildings has been proved as an effective method of structural control for alleviating seismic responses. Researchers have proposed that two adjacent buildings through supplemental energy dissipating devices to mitigate the buildings’ responses. Numerous researchers have proposed various methods: active, passive, and semi-active control strategies. In Japan, some applications of coupled buildings control have been successfully implemented by utilizing passive and active control technology. Magnetorheological (MR) dampers have been identified as semi-active devices that can be used to reduce the vibration of the seismic structures during various types of ground motions. They can offer the adaptability of active devices, stability, and reliability of passive devices. Nevertheless, one of the difficulties in application of the MR dampers is the development of the appropriate control algorithms. Accordingly, this study presents the implementation of the adaptive neuro-fuzzy inference system (ANFIS) controller for earthquake hazard mitigation under coupled buildings control system with base-isolated building connecting to the free wall by MR dampers. The ANFIS whose training data is based on the Linear Quadratic Regulator (LQR) method is conducted to modify the parameters of the fuzzy logic controller and optimize the fuzzy rules. The performance of MR dampers is evaluated under seismic response. It is compared under four methods, including passive-off, passive-on, and two semi-active control strategies: ANFIS and LQR. Besides, various types of feedback of the ANFIS operated as two-input single output feedback system are investigated to assess the performance of the developed control scheme for structural vibration control. The numerical simulation results show that the proposed semi-active control system consisting of coupled buildings system and MR dampers by utilizing ANFIS can be effective in mitigating seismic responses of structures.
Read full abstract