A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Structural Engineering, School of Engineering and Technology

Idealizing hysteretic behaviour of Modal Cyclic pushover curves from High-rise RC shear wall buildings is one of the important factor that determines the accuracy of predicted inelastic demands using Uncoupled Modal response history Analysis (UMRHA). Recent studies adopt Self Centering Flag Shape hysteresis to idealize modal cyclic pushover curves. However, Conventional approach, found to provide inaccurate demand prediction with UMRHA when a structure is subjected to strong ground shaking. That is, results become less accurate once structure undergo significant yielding. The current study is an attempt to improve the inelastic demand prediction by introducing a Modified Flag Shape hysteretic idealization with UMRHA. The accuracy of UMRHA with both idealization approach is examined with a 40 story case study building for ground motions from four group of earthquakes. Results show that it is possible to predict nonlinear seismic demands accurately even beyond structure yielding with newly developed Modified Flag Shape idealization. Results provide good match with exact responses from Nonlinear Response History Analysis (NLRHA) for both response time histories and response envelops. Overall, the new theory is performing well for the adopted case study building and expected to work well for similar kind of buildings. The second part of the current thesis focus on developing Simplified Analysis Schemes to represent nonlinear response of inelastic Single Degree of Freedom System (SDOF) with Modified Flag Shape hysteresis. Assuming Modified Flag Shape hysteretic properties are known, two different simplified analysis procedures are developed. They are Equivalent Linearization (EL) and Response Reduction Procedure (RRP). Both Equivalent Linearization and Response Reduction Procedure are derived for different earthquake groups and their prediction accuracy is checked. Results show that Response Reduction Procedure exhibit better prediction compared to Equivalent linearization. In fact, prediction is more stable and reliable for long initial periods. Also an example is demonstrated to assess nonlinear response of inelastic building using Response Reduction Procedure with the aid of 40 story case study building. The concept of RRP is further tested for the influence of Modified Flag Shape hysteretic properties. The critical properties that effect RRF are identified.

Thesis (M. Eng.) - Asian Institute of Technology, 2018