SGER: IDENTIFICATION OF STRUCTURAL DAMAGE IN TANKS AND INDUSTRIAL FACILITIES DUE TO HURRICANE KATRINA

University Of Puerto Rico Mayaguez, Mayaguez PR

Investigators

Abstract

SGER: IDENTIFICATION OF STRUCTURAL DAMAGE IN TANKS AND INDUSTRIAL FACILITIES DUE TO HURRICANE KATRINA PI: Luis A. Godoy Department of Civil Engineering and Surveying, University of Puerto Rico at Mayaguez, PR 00681-9041 Abstract. The assessments of the stability of industrial structures and oil storage tanks is mainly carried out at present using computational techniques. However, their reliability should be validated using information from real events affecting real structures. Such information is only available when a high consequence disaster occurs, as is the case of hurricane Katrina. The site visit to the areas affected by Hurricane Katrina should greatly expand our existing inventories and allow a closer examination of more contentious conclusions being formulated. For the case of buckling of tanks and other storage facilities, we will observe buckling modes in shells, which can be accurately recorded by means of photographs. With the mode and the level of wind speeds recorded in the area, we will later perform computations using our finite element codes to estimate the wind speed at buckling (buckling loads) and the structural shape at buckling (buckling modes). Comparisons between observations and predictions will allow having a more complete understanding of the mechanics of failure due to a real hurricane, so that the specific sequence of events can be reconstructed to explain the observed failures. The observations carried out will provide evidence to confirm the lower bound approach which has been recently proposed by the PI and Prof. J. Croll. We will also observe and document the performance of tanks founded on soft alluvial soils, which are common in the area affected by hurricane Katrina. Structural damage due to poor foundation performance, such as differential settlements, will be recorded. For the case of industrial buildings affected by hurricane Katrina, we expect fatigue modes of failure to be dominant in the majority of cases. The pull-over rate should be a function of location, both in relations to the panel as part of the system and the system as part of the neighborhood, and those features will be observed during the visits. The information gathered during the site visits will be classified and interpreted according to the state of the art in the field, and will be made available to the engineering community within two months following the site visit. Broader impact: The field data collected during our site visit will lead to refinements in the current dynamic protocols, in terms of the minimum size of future specimens, failure modes not well understood at present, and the cyclic load protocol itself. It may also help strengthen the arguments to adopt rational dynamic system rating standards. This research will also impact the graduate program in Civil Engineering at the University of Puerto Rico at Mayaguez. UPRM has a growing Ph.D. program in which the topic of behavior of structures subjected to natural hazards is an integral part. More than 95% of the students at UPRM are Hispanic. Intellectual merit: The PI and Prof. J. Croll have recently advanced a lower bound buckling theory for wind loaded structures based on reduced energy/reduced stiffness concepts. The observations carried out will provide evidence to confirm the lower bound approach. Such definite confirmation can only be achieved now, that the approach is mature and there are specific questions that need to be answered, and it is expected that the reconstruction of the mechanisms of failure in tanks will shed light on the interpretation of the damage observed in the field.