Search

found 2 results

Research papers, University of Canterbury Library

The development of cheap, whilst effective and relatively non-invasive structural retrofit techniques for existing non-ductile reinforced concrete (RC) structures still remains the most challenging issue for a wide implementation on a macro scale. Seismic retrofit is too often being confused as purely structural strengthening. As part of a six-years national project on “Seismic retrofit solutions for NZ multi-storey building”, focus has been given at the University of Canterbury on the development of a counter-intuitive retrofit strategy for earthquake vulnerable existing rc frame, based on a “selective weakening” (SW) approach. After an overview of the SW concept, this paper presents the experimental and numerical validation of a SW retrofit strategy for earthquake vulnerable existing RC frame with particular focus on the exterior beam-column (b-c) joints. The exterior b-c joint is a critically vulnerable region in many existing pre-1970s RC frames. By selectively weakening the beam by cutting the bottom longitudinal reinforcements and/or adding external pre-stressing to the b-c joint, a more desirable inelastic mechanism can be attained, leading to improved global seismic performance. The so-called SW retrofit is implemented on four 2/3-scaled exterior RC b-c joint subassemblies, tested under quasi-static cyclic loading at the University of Canterbury. Complemented by refined 3D Finite Element (FE) models and dynamic time-history analyses results, the experimental results have shown the potential of a simple and cost-effective yet structurally efficient structural rehabilitation technique. The research also demonstrated the potential of advanced 3D fracture-mechanics-based microplane concrete modelling for refined FE analysis of non-ductile RC b-c joints.

Research Papers, Lincoln University

Artificial Neural Networks (ANN) as a tool offers opportunities for modeling the inherent complexity and uncertainty associated with socio-environmental systems. This study draws on New Zealand ski fields (multiple locations) as socio- environmental systems while considering their perceived resilience to low probability but potential high consequences catastrophic natural events (specifically earthquakes). We gathered data at several ski fields using a mixed methodology including: geomorphic assessment, qualitative interviews, and an adaptation of Ozesmi and Ozesmi’s (2003) multi-step fuzzy cognitive mapping (FCM) approach. The data gathered from FCM are qualitatively condensed, and aggregated to three different participant social groups. The social groups include ski fields users, ski industry workers, and ski field managers. Both quantitative and qualitative indices are used to analyze social cognitive maps to identify critical nodes for ANN simulations. The simulations experiment with auto-associative neural networks for developing adaptive preparation, response and recovery strategies. Moreover, simulations attempt to identify key priorities for preparation, response, and recovery for improving resilience to earthquakes in these complex and dynamic environments. The novel mixed methodology is presented as a means of linking physical and social sciences in high complexity, high uncertainty socio-environmental systems. Simulation results indicate that participants perceived that increases in Social Preparation Action, Social Preparation Resources, Social Response Action and Social Response Resources have a positive benefit in improving the resilience to earthquakes of ski fields’ stakeholders.