No description available.

No description available.

In this thesis I develop a mathematical model for an apartment building during an earthquake. The movement of the building is restricted to a plane and twisting motions have been assumed negligible. A control system for the building is developed using optimal control techniques. For a quadratic objective functional, the existence of an optimal control is determined and numerical results are generated that show that the controller significantly lowers the chaotic oscillations in the building. The numerical work was done with the Miser3 package for Matlab. Relaxation of different constraints are considered, including multiple controls, varying stories, and different objective functionals.

acase@tulane.edu / 1 / Qiaoyun Zhang

Thesis (Ph.D.)--Oregon State University, 1999. / Includes bibliographical references.

Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

The 1946 M7.3 Ancash earthquake triggered a large number of landslides in an epicentral area along the continental divide in the Andes of Peru. Reconnaissance reports by Silgado (1947, 1951), Heim (1949) and Rüegg (1950) published shortly after the earthquake describe some of the landslides, but lack any detailed documentation. A review of literature, field traverses, aerial photograph interpretation and GIS mapping, identified 45 landslides inferred to have been triggered by the event. The cumulative volume-frequency plot shows a strong power law relation, similar in form to that derived for landslides triggered by the 1994 Northridge earthquake (Harp and Jibson, 1995). Data for the Ancash earthquake plots near the regression line calculated by Keefer (1994), and modified by Malamud et al. (2004), for earthquake moment magnitude vs. total volume moved by landslides during the earthquake. A denudation (average surface lowering) of 76 mm and a denudation rate of 0.00585 mm/yr was estimated for the area over which the earthquake-triggered landslides occurred. Rock avalanches constituted 83% of the landslides which involved Cretaceous limestones interbedded with shales. Rock slope failures were common on slopes with a southwest aspect, an orientation corresponding to the regional dip direction of major planar structures in the epicentral area. In valleys oriented transverse to the NW-SE structural trend of the epicentral area, south-westerly dipping bedding planes combined with orthogonal joint sets to form numerous wedge failures. Three rock avalanches were analysed using conventional wedge failure criteria. Two of the rock avalanches which did not satisfy conventional criteria were utilized to develop a new criteria, i.e. stepped wedge criteria. Stepped wedge failure involves wedge movement along a stepped line of intersection formed by a nearly-vertical discontinuity, striking roughly parallel to the slope, and the conventional line of intersection formed by the bedding and orthogonal joint set. The nearly-vertical discontinuity forms the essential backscarp needed to facilitate stepped wedge failure. Additional characteristics of stepped wedge failure were also distinguished based on the wedge failures studied in the epicentral area of the 1946 Ancash earthquake.

The 1946 M7.3 Ancash earthquake triggered a large number of landslides in an epicentral area along the continental divide in the Andes of Peru. Reconnaissance reports by Silgado (1947, 1951), Heim (1949) and Rüegg (1950) published shortly after the earthquake describe some of the landslides, but lack any detailed documentation. A review of literature, field traverses, aerial photograph interpretation and GIS mapping, identified 45 landslides inferred to have been triggered by the event. The cumulative volume-frequency plot shows a strong power law relation, similar in form to that derived for landslides triggered by the 1994 Northridge earthquake (Harp and Jibson, 1995). Data for the Ancash earthquake plots near the regression line calculated by Keefer (1994), and modified by Malamud et al. (2004), for earthquake moment magnitude vs. total volume moved by landslides during the earthquake. A denudation (average surface lowering) of 76 mm and a denudation rate of 0.00585 mm/yr was estimated for the area over which the earthquake-triggered landslides occurred. Rock avalanches constituted 83% of the landslides which involved Cretaceous limestones interbedded with shales. Rock slope failures were common on slopes with a southwest aspect, an orientation corresponding to the regional dip direction of major planar structures in the epicentral area. In valleys oriented transverse to the NW-SE structural trend of the epicentral area, south-westerly dipping bedding planes combined with orthogonal joint sets to form numerous wedge failures. Three rock avalanches were analysed using conventional wedge failure criteria. Two of the rock avalanches which did not satisfy conventional criteria were utilized to develop a new criteria, i.e. stepped wedge criteria. Stepped wedge failure involves wedge movement along a stepped line of intersection formed by a nearly-vertical discontinuity, striking roughly parallel to the slope, and the conventional line of intersection formed by the bedding and orthogonal joint set. The nearly-vertical discontinuity forms the essential backscarp needed to facilitate stepped wedge failure. Additional characteristics of stepped wedge failure were also distinguished based on the wedge failures studied in the epicentral area of the 1946 Ancash earthquake.

Whether it is for pre-event prevention and preparedness or for post-event response and recovery of a catastrophic earthquake, estimates of damage, death and downtime (3d) losses are needed by engineers, owners, and policy makers. In this research, a quantitative "scenario-based" risk analysis approach as developed to investigate the 3d losses for buildings. The "Redbook Building" is taken as the typical New Zealand construction exemplar and analyzed for the 22 February 2011 Christchurch Earthquake. Losses are presented in the form of attenuation curves that also include the associated uncertainties. The spatial distribution of 3d damages over the height of buildings is also considered. It is thus shown that it is possible to discriminate between losses that lead to building replacement versus less severe losses that require structures to be repaired. The 3d loss results show that within the Christchurch city (17 km radial distance from the earthquake epicenter): (a) the expected physical damage loss ratio is about 50% of the property value; (b) the expected probability that someone is killed or seriously injured is about 4%; and (c) the expected downtime for the building being out of service is about 24 weeks. However, when considering various uncertainties, one can have 90% confidence that these loss estimations will be as high as: (a) complete loss (100% physical damage), implying structure has a great chance of collapse; (b) 8% possibility of fatality, implying deaths and significant injuries are likely; and (c) 1-year downtime due to post-event reconstruction demand surge. These informative results demonstrate that even though structures, such as the "Redbook Building", may have been well designed and constructed to contemporary standards, significant damage can still be expected and the downtime loss is particularly large. In order to solve this problem, new building structures should ideally be built stronger, include recentering attributes, and use Damage Avoidance Design (DAD) armoring connection details.

Thesis (M. Phil.)--University of Hong Kong, 2002. / Includes bibliographical references (leaves 206-217).