High-quality ground motion records are required for engineering applications including response history analysis, seismic hazard development, and validation of physics-based ground motion simulations. However, the determination of whether a ground motion record is high-quality is poorly handled by automation with mathematical functions and can become prohibitive if done manually. Machine learning applications are well-suited to this problem, and a previous feed-forward neural network was developed (Bellagamba et al. 2019) to determine high-quality records from small crustal events in the Canterbury and Wellington regions for simulation validation. This prior work was however limited by the omission of moderate-to-large magnitude events and those from other tectonic environments, as well as a lack of explicit determination of the minimum usable frequency of the ground motion. To address these shortcomings, an updated neural network was developed to predict the quality of ground motion records for all magnitudes and all tectonic sources—active shallow crustal, subduction intraslab, and subduction interface—in New Zealand. The predictive performance of the previous feed-forward neural network was matched by the neural network in the domain of small crustal records, and this level of predictive performance is now extended to all source magnitudes and types in New Zealand making the neural network applicable to global ground motion databases. Furthermore, the neural network provides quality and minimum usable frequency predictions for each of the three orthogonal components of a record which may then be mapped into a binary quality decision or otherwise applied as desired. This framework provides flexibility for the end user to predict high-quality records with various acceptability thresholds allowing for this neural network to be used in a range of applications.
Timber-based hybrid structures provide a prospective solution for utilizing environmentally friendly timber material in the construction of mid-rise or high-rise structures. This study mainly focuses on structural damage evaluation for a type of timber-steel hybrid structures, which incorporate prefabricated light wood frame shear walls into steel moment-resisting frames (SMRFs). The structural damage of such a hybrid structure was evaluated through shake table tests on a four-story large-scale timber-steel hybrid structure. Four ground motion records (i.e., Wenchuan earthquake, Canterbury earthquake, El-Centro earthquake, and Kobe earthquake) were chosen for the tests, with the consideration of three different probability levels (i.e., minor, moderate and major earthquakes) for each record. During the shake table tests, the hybrid structure performed quite well with visual damage only to wood shear walls. No visual damage in SMRF and the frame-to-wall connections was observed. The correlation of visual damage to seismic intensity, modal-based damage index and inter-story drift was discussed. The reported work provided a basis of knowledge for performance-based seismic design (PBSD) for such timber-based hybrid structures.
Earthquakes cause significant damage to buildings due to strong vibration of the ground. Levitating houses using magnets and electromagnets would provide a complete isolation of ground motion for protecting buildings from seismic damage. Two types of initial configuration for the electromagnet system were proposed with the same air gap (10mm) between the electromagnet and reluctance plate. Both active and passive controller are modelled to investigate the feasibility of using a vibration control system for stabilizing the magnetic system within the designed air gap (10mm) in the vertical direction. A nonlinear model for the magnetic system is derived to implement numerical simulation of structural response under the earthquake record in Christchurch Botanic Gardens on 21 February 2011. The performance of the uncontrolled and the controlled systems are compared and the optimal combination of control gains are determined for the PID active controller. Simulation results show both active PID controller with constant and nonlinear attracting force are able to provide an effective displacement control within the required air gap (+/-5mm). The maximum control force demand for the PID controller in the presence of nonlinear attracting force is 4.1kN, while the attracting force in equilibrium position is 10kN provided by the electromagnet. These results show the feasibility of levitating a house using the current electromagnet and PID controller. Finally, initial results of passive control using two permanent magnets or dampers show the structural responses can be effectively reduced and centralized to +/-1mm using a nonlinear centring barrier function.
Rising disaster losses, growth in global migration, migrant labour trends, and increasingly diverse populations have serious implications for disaster resilience around the world. These issues are of particular concern in New Zealand, which is highly exposed to disaster risk and has the highest proportion of migrant workers to national population in the OECD. Since there has been no research conducted into this issue in New Zealand to date, greater understanding of the social capital used by migrant workers in specific New Zealand contexts is needed to inform more targeted and inclusive disaster risk management approaches. A New Zealand case study is used to investigate the extent and types of social capital and levels of disaster risk awareness reported by members of three Filipino migrant workers organisations catering to dairy farm, construction and aged care workers in different urban and rural Canterbury districts. Findings from (3) semi-structured interviews and (3) focus groups include consistently high reliance on bonding capital and low levels of bridging capital across all three organisations and industry sectors, and in both urban and rural contexts. The transitory, precarious residential status conveyed by temporary work visas, and the difficulty of building bridging capital with host communities has contributed to this heavy reliance on bonding capital. Social media was essential to connect workers with family and friends in other countries, while Filipino migrant workers organisations provided members with valuable access to industry and district-specific networks of other Filipino migrant workers. Linking capital varied between the three organisations, with members of the organisation set up to advocate for dairy farm workers reporting the highest levels of linking capital. Factors influencing the capacity of workers organisations to develop linking capital appeared to include motivation (establishment objectives), length of time since establishment, support from government and industry groups, urban-rural context, income levels and gender. Although aware of publicity around earthquake and tsunami risk in the Canterbury region, participants were less aware of flood risk, and expressed fatalistic attitudes to disaster risk. Workers organisations offer a valuable potential interface between CDEM Group activities and migrant worker communities, since organisation leaders were interested in accessing government support to participate (with and on behalf of members) in disaster risk planning at district and regional level. With the potential to increase disaster resilience among these vulnerable, hard to reach communities, such participation could also help to build capacity across workers organisations (within Canterbury and across the country) to develop linking capital at national, as well as regional level. However, these links will also depend on greater government and industry commitment to providing more targeted and appropriate support for migrant workers, including consideration of the cultural qualifications of staff tasked with liaising with this community.
