Conferences, Lectures, & Seminars
Events for January
-
There and Back Again: Linking Regional and Global Air Quality and Climate
Wed, Jan 16, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Kevin Bowman Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA Recent advances in observational capabilities, global chemistry transport models, and data assimilation techniques have the potential to revolutionize our understanding of global and regional air quality. This progress has been enabled in part by the Tropospheric Emission Spectrometer (TES), a high-resolution Fourier Transform spectrometer aboard the NASA Aura spacecraft that provides the global distribution of vertical ozone and carbon monoxide profiles. These observations can characterize how pollutants such as ozone can be transformed and transported on global scales with important implications for both regional air quality and climate. In particular, I will discuss research in understanding the contribution of surface emissions to tropical tropospheric ozone and the impact of summertime ozone over North America on chemistry climate coupling. In addition, we will explore a new effort to develop "sensor webs" that can combine ground, aircraft, and satellite observations with adjoint modeling techniques to design optimal sampling strategies that maximize the information available in these observations. I will show how sensor webs could be used in intensive air quality campaigns and their potential role for global environmental monitoring.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited
Contact: Evangeline Reyes
-
Uncertainty and Bayesian inference in inverse problems
Thu, Jan 17, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Youssef Marzouk, Massachusetts Institute of Technology"Uncertainty and Bayesian inference in inverse problems"Predictive simulation rests on validated models, which often must be conditioned on indirect observations. In this context, Bayesian statistics provides a foundation for inference from noisy and limited data, a natural mechanism for regularization in the form of prior information, and a quantitative assessment of uncertainty in the objects of inference. Inverse problemsâ"representing indirect estimation of model parameters, inputs, or structural componentsâ"can be fruitfully cast in this framework. Complex and computationally intensive forward models arising in physical applications, however, can render a Bayesian approach prohibitive. This difficulty is compounded by high dimensionality, as when the unknown is a spatial field.We present new algorithmic developments for Bayesian inference in this context, showing strong connections with the forward propagation of uncertainty. In particular, we introduce a stochastic spectral formulation that accelerates the Bayesian solution of inverse problems via rapid evaluation of a surrogate posterior. We also pursue dimensionality reduction for the inference of spatiotemporal fields, using truncated Karhunen-Loève representations of Gaussian process priors. These approaches are demonstrated on scalar transport problems arising in contaminant source inversion and in the inference of inhomogeneous transport properties.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited
Contact: Evangeline Reyes
-
Modeling Critical Infrastructures with Networked Agent-based Approaches
Wed, Jan 23, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars
Speaker:
Dr. Robert J. Glass,
Distinguished Member of Technical Staff
Complex Adaptive Infrastructures and Behavioral Systems
National Infrastructure Simulation and Analysis Center
Sandia National LaboratoryCritical Infrastructures are formed by large numbers of components that interact within complex networks. As a rule, infrastructures contain strong feedbacks either explicitly through the action of hardware/software control, or implicitly through the action/reaction of people. Individual infrastructures influence others and grow, adapt, and thus evolve in response to their multifaceted physical, economic, cultural, and political environments. Simply put, critical infrastructures are complex adaptive systems.Our general approach to modeling such systems distills the system of interest to a network (or multiple networks) of nodes and connections with a set of tailored interaction rules (static to adaptive) for each. Combined with drives and dissipations we can evaluate how general features, such as network connectivity and interaction rules, or specific perturbations such as a hurricane, can influence system failure (often cascading failure) and the choice of mitigation strategy once a cascade begins. Examples will be drawn from recent work that applies our general approach to areas as diverse as community mitigation for pandemic influenza (e.g., bird flu), congestion and cascades in coupled large value payment systems (e.g., foreign exchange coupled US and Euro systems, trillions of $ per day), and hurricane induced perturbations to US petrochemical supply chains.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited
Contact: Evangeline Reyes
-
A Multiscale Characterization and Analysis Methodology for Ductile Fracture in Heterogeneous Metalli
Thu, Jan 24, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Somnath Ghosh,Department of Mechanical Engineering, Ohio State UniversityHeterogeneous metallic materials e.g. cast aluminum alloys or metal matrix composites are widely used in automotive, aerospace, nuclear and other engineering systems. The presence of precipitates and particulates in the microstructure often affect their failure properties like fracture toughness or ductility in an adverse manner. Important micromechanical damage modes that are responsible for deterring the overall properties include particulate fragmentation, debonding at interfaces and ductile matrix failure due to void initiation, growth and coalescence, culminating in local ductile failure. The complex interaction between competing damage modes in the presence of multiple phases makes failure and ductility prediction for these materials quite challenging. While phenomenological and straightforward micromechanics models have predicted stress-strain behavior and strength of multi-phase materials with reasonable accuracy, their competence in predicting ductility and strain-to-failure, which depends on the extreme values of distribution, is far from mature. To address the needs of a robust methodology for ductility, the work will discuss a comprehensive multi-scale characterization based domain decomposition method followed by a multi-scale model for deformation and ductile failure. Adaptive multi-scale models are developed for quantitative predictions at critical length scales, establishing functional links between microstructure and response, and following the path of failure from initiation to rupture. The work is divided into three modules. (i) Multi-scale morphology based domain partitioning to develop a pre-processor for multiscale modeling, (ii) Enriched Voronoi Cell FEM for particle and matrix cracking leading to ductile fracture and (iii) Macroscopic homogenization continuum damage model for ductile fracture. Finally a robust framework for two-way multi-scale analysis module is the coupling of different with different inter-scale transfer operators and interfaces is developed.Bio-SketchDr. Somnath Ghosh is the John B. Nordholt Professor in the Department of Mechanical Engineering at the Ohio State University. He received his M.S. in Theoretical and Applied Mechanics from Cornell University and PhD from the University of Michigan. His research interests include multiple scale modeling in spatial and temporal domains, failure and fatigue modeling in heterogeneous materials and structures, Computational nanotechnology, etc. He is a fellow of American Association for the Advancement of Science (2008), US Association of Computational Mechanics (2007), ASM International (2006) and ASME (2002). In 2007, the Ohio State University awarded him the University Distinguished Scholar award. He got the National Science Foundation Young Investigator award of NSF in 1994. He was an elected member of the executive council of the US Association of Computational Mechanics (USACM) from 2002-2006 and is the Chair of USACMâs Materials Modeling committee. The US Association of Computational Mechanics has chosen him to be the organizer of the 10th US National Congress of Computational Mechanics in 2009.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited
Contact: Evangeline Reyes
-
Torsional Effects on the Inelastic Seismic Response of Structures
Tue, Jan 29, 2008 @ 01:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars
Oral Defense by: Mehran Mansuri,
Sonny Astani Department of Civil and Environmental EngineeringABSTRACT:
To evaluate inelastic torsional response of buildings due to different parameters such as unsymmetrical distribution of mass or lateral load resisting elements in the plan of the structure or yielding and inelastic behavior of resisting elements and loss of the resistance of such an element during an earthquake, a full three-dimensional nonlinear dynamic analysis is a powerful tool to evaluate such a nonlinear response.
The results of nonlinear dynamic analyses of two actual steel moment frame buildings that were damaged during the 1994 Northridge earthquake subjected to couple of different recorded ground motions from Northridge and Loma Prieta earthquakes are presented and the importance of different parameters such as discontinuity of lateral resisting elements, unsymmetrical distribution of mass or resistance in the plan of structure, intensity and frequency content of earthquake ground motions, accidental eccentricity as prescribed by code and the effect of geometric nonlinearity (P-Delta) on the inelastic lateral-torsional response of structures is discussed. Response parameters considered include lateral story displacement, Interstory drift index, plastic hinge rotation demand and torsional rotation of each floor.
The analysis procedures use three-dimensional nonlinear dynamic analytical models developed for the PERFORM 3-D computer program.
Study of the results for different models with different eccentricities clearly shows the effect of inelastic torsion in comparison with elastic torsion on the response of structures. The torsional rotation of floors considered as a main parameter of torsional response of the building has an average increase of 30 to 60 percent for material nonlinearity. By adding geometric nonlinearity (P-Delta), this increases 70 to 100 percent of elastic torsional rotation. This clearly shows the inelastic torsional response of structures may be significantly underestimated by a linear dynamic analysis, especially for large value of mass or stiffness eccentricity and intensity of the ground motion.
Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited
Contact: Evangeline Reyes
-
Equation-Free Modeling and Computation for Complex/Multiscale Systems
Wed, Jan 30, 2008 @ 02:00 PM - 03:00 PM
Sonny Astani Department of Civil and Environmental Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Yannis Kevrekidis,
Princeton University"Equation-Free Modeling and Computation for Complex/Multiscale Systems"In current modeling practice for complex/multiscale systems, the best available descriptions often come at a fine level (atomistic, stochastic, microscopic, individual-based) while the questions asked and the tasks required by the modeler (prediction, parametric analysis, optimization and control) are at a much coarser, averaged, macroscopic level. Traditional modeling approaches start by first deriving macroscopic evolution equations from the microscopic models, and then bringing our arsenal of mathematical and algorithmic tools to bear on these macroscopic descriptions. Over the last few years, and with several collaborators, we have developed and validated a mathematically inspired, computational enabling technology that allows the modeler to perform macroscopic tasks acting on the microscopic models directly. We call this the "equation-free" approach, since it circumvents the
step of obtaining accurate macroscopic descriptions. We will argue that the backbone of this approach is the design of (computational) experiments. Traditional continuum numerical algorithms can thus viewed as protocols for experimental design (where "experiment" means a computational experiment set up and performed with a model at a different level of description). Ultimately, what makes it all possible is the ability to initialize computational experiments at will. Short bursts of appropriately initialized computational experimentation through matrix-free numerical analysis and systems theory tools like variance reduction and estimation- bridge microscopic simulation with macroscopic modeling. I will also discuss some recent developments in data mining algorithms, exploring large complex data sets to find good "reduction coordinates".Location: Kaprielian Hall (KAP) - 209
Audiences: Everyone Is Invited
Contact: Evangeline Reyes