부산대학교 도서관

Conventional methods of seismic rehabilitation with steel bracing could not be a proper method for upgrading of buildings with these methods due to the cost and time expenses. Therefore, an innovative passive control system called braced damper for steel frames has been introduced in this study. Thus, for investigating the behavior of this structural damper, a single-bay steel frame is formed without rotary brace damper device. Later, the frame is equipped with rotary brace damper and subjected to monotonic and cyclic loading. Few response parameters are regarded, and the performance of the rotary brace damper in reducing each response has been measured. In terms of methodology, ANFIS (Adaptive neuro-fuzzy inference system) is used in data analysis in a variable selectionprocessing to distinguish the predominant values that influence the strength of this damper. The process is on the basis of the variable search method that has been later applied to measure how the four parameters of (1) displacement, (2) diameter (3) thickness, and (4) plate thickness influence the strength of damper. As a result, displacement is the most effectual parameter in forcing prediction and best predictor of accuracy.