The Department of Civil Engineering at the University at Buffalo and NCEER jointly sponsored the "Structural and Geotechnical Alumni Seminar Series" as part of the University at Buffalo's Engineering School's 50th anniversary celebration. The seminar series featured alumni from the Civil Engineering Department, many of whom have participated in NCEER research projects. The seminars were held weekly during the Spring 1997 semester on Friday afternoons on the University at Buffalo campus, and were coordinated by Professors Tsu T. Soong and Andrei Reinhorn. Seminar reviews were written by graduate students in the school of engineering. Reviews from the first six seminars were included in the April issue of the NCEER Bulletin, Vol. 11, No. 2. The final four seminars are reviewed herein.
Vulnerability of Bridges to Vessel Collision
Application of Boundary Element Methods in Dynamic Analysis and Frame Mechanics
Semi-active Hybrid Control Systems for Seismic Protection of Structures
Phenomenological Constitutive Models and Causality
On May 2, 1997, Dr. Zolan Prucz, an associate at Modjeski and Master Inc. (New Orleans, LA), made a presentation on the vulnerability of bridges to vessel collision. Over 50 faculty and graduate students attended the seminar.
Dr. Prucz, who is a 1984 Ph.D. alumnus of the department of civil engineering at the University at Buffalo, started with a brief history of vessel-bridge collisions and the recent rise in the number of these accidents. He cited these accidents as a result of a three-way interaction between the vessel, waterway and bridge. Some of the attributable causes are the increase in the number of vessels in the world fleet, the increase in vessel sizes, difficulties in navigation through channels passing under two side-by-side bridges, weather changes, human error, and equipment failure.
Case histories were used to demonstrate some of the causal effects. It became obvious that the problem is world-wide. In many cases, the bridge structures sustained more damage than the vessels involved in the accidents. Notable among these is the Sunshine Skyway accident of May 1980, in which 35 lives were lost. Dr. Prucz highlighted this as an example of accidents caused by navigational difficulties in channels passing under two similar bridges, located side by side. In this instance, the vessel Summit Venture strayed outside of the navigation channel and hit the pier of the bridge, causing collapse of the span.
As a result of the Sunshine Skyway bridge accident, actions have been taken in the U.S. to develop computational tools and technology to address this problem. The audience's attention was drawn to the important lessons derived from these case histories, which for new bridges are: the importance of initial planning, collision risk analysis, consideration of input forces as a function of ship/barge sizes, and the incorporation of redundancy in structural engineering designs. Dr. Prucz also emphasized the need for more studies to better establish collision forces. With regard to existing bridges, he mentioned the need to use protective systems. Some of these systems are fenders (timber or rubber), "dolphins," protective islands, and floating shear booms. In a brief comparative discussion of these protective systems, the benefits of using systems which allow parts replacement after an accident was apparent. With regard to economic considerations for new bridge projects, he stated that studies have indicated that including protection systems in the original cost of a new bridge is far less than the total costs resulting from vessel collisions.
During the discussion session following the presentation, it was noted that damping systems and active control are also being explored as collision mitigating systems. In conclusion, Dr. Prucz also described efforts on the part of vessel owners to use sophisticated electronic systems to reduce the occurrence of such accidents.
Back to top
Application of Boundary Element Methods in Dynamic Analysis and Fracture Mechanics was presented on April 18, 1997 by Dr. Gary F. Dargush. Dr. Dargush is an assistant professor of civil engineering at the University at Buffalo. He received his Ph.D. from the University at Buffalo in 1987 and joined the faculty in September 1996. Prior to joining the University at Buffalo, Dr. Dargush worked as a structural analyst and design engineer at Ford Motor Company and General Motors Corporation.
The boundary element method, as an analytical tool, is proposed for use particularly in applications involving stress discontinuities or singularities. The former occur, for example, in problems of wave propagation, while the latter are associated with cracks or other abrupt geometric features. In both cases, boundary element methods (BEM) are able to capture these discontinuities or singular response by making direct use of infinite space fundamental solutions of the governing differential equations.
Dr. Dargush began the presentation with an overview of boundary element formulations suitable for elasto-dynamics and fracture analysis, and then concentrated on several recent applications of the method. Use of BEM to study the effectiveness of nondestructive evaluation via the impact echo method was shown. Included was discussion on the use of BEM to study seismic wave propagation in layered media. Some results for poroelastic response of a circular foundation were given.
One of the most popular areas of BEM use is in fracture mechanics. The effectiveness of this method for evaluating turbomachinery components such as turbine blade was shown. Stress intensity factors for a cracked blade were calculated. Behavior of the stress intensity factor was studied by BEM and compared with some analytical solution available in the literature.
The discussion following the seminar by the 50 faculty members and students who attended centered on the accuracy of BEM and the ability to use it in plastic fracture mechanics.
Back to top
Semi-active hybrid control was the subject of the seminar presented by Dr. Michael Symans on April 4, 1997. Dr. Symans earned his Ph.D., M.S. and B.S. in civil engineering at the University at Buffalo in 1995, 1993 and 1990, respectively. He also has a B.S. in physics from SUNY/College at Fredonia (1990). His graduate studies concentrated on the development of passive and semi-active fluid dampers for seismic response control of structures. He is currently assistant professor in the department of civil and environmental engineering at Washington State University. He is a recent recipient of the NSF Career Award and is a member of the ASCE Committee on Structural Control.
This seminar focused on a current research project involving the application of variable fluid dampers to the control of base-isolated structures.
First, Dr. Symans briefly talked about performance requirements of different codes and also gave a historical review of different control systems. He described the strong points of different control systems, such as passive control, active control and semi-active control, and pointed out each systems' weak points, which are mainly induced by the unpredictable characteristic of input force. In order to achieve a better control effect, Dr. Symans suggested that a hybrid control system, which combines passive control and semi-active control, is more efficient than other control systems.
Dr. Symans discussed the semi-active hybrid control system used in his research project. The most unique part of this control system is that fuzzy logic theory was utilized for determining the control force. Fuzzy control has two important features when compared to conventional control. The first is verbose statement consistent with human reasoning, and the second is that I/O relation is transparent although complex. Meanwhile, conventional control involves complicated mathematical expressions that are less intuitive. As stated by Dr. Symans, fuzzy control has the ability to work with imprecise data and uncertainty and is well-suited to earthquake motions since each event is unique and unpredictable.
Finally, Dr. Symans discussed the results of intensive numerical studies about fuzzy control, which were recently completed. These studies were carefully conducted so as to permit comparisons between the hybrid control system and a well-designed passive control system. One of the significant results of the numerical analysis is that the hybrid base isolation system can be used to limit the force transferred through the isolation system to an acceptable value while simultaneously absorbing the maximum amount of energy possible under the force-limiting constraint. This may have significant implications for seismic retrofitting of weak structures which cannot be easily strengthened.
Back to top
Phenomenological Constitutive Models and Causality by Dr. Nikos Makris was the topic of the eighth seminar on structural and geotechnical earthquake engineering held on March 28, 1997. Over 40 faculty and graduate students attended the seminar. Dr. Makris is an assistant professor of civil and environmental engineering at the University of California at Berkeley. He received his civil engineering diploma from the National Technical University of Athens and his M.S. and Ph.D. degrees from the University at Buffalo in 1990 and 1992, respectively. Dr. Makris' research interests include seismic isolation, energy dissipation, passive and active control, and soil-structure interaction.
The proposed causal hysteretic model is appropriate for the analysis of the nonlinear response of structures equipped with different protective systems and the analysis of structures subjected to short duration ground motions (impulsive type).
The basic transfer functions and time response of linear phenomenological models were first reviewed (dynamic stiffness, memory function, impedance and relaxation stiffness of the linear spring element - Hook's model).
The relation between the analyticity of the transfer function (frequency response function) and the causality of the corresponding time response function was extended for the case of generalized transfer functions.
Using the properties of the Hilbert transform and the associated Kramers-Kroning relations, Dr. Makris demonstrated that transfer functions that have a singularity at w = 0 in their imaginary part should be corrected by adding a Delta function in their real part. In this way the noncausality characteristic of the function is removed and the function becomes analytic. This operation ensures that the resulting time response function is causal and is consistent with the theory of generalized functions. New formulas for basic transfer functions and time response functions for different standard constitutive models were proposed.
The causal hysteric element with a frequency independent loss stiffness is then constructed and analyzed (dynamic stiffness, memory function, impedance and relaxation stiffness of the complex spring element). The element is a physically realizable mechanical model at finite frequencies, whereas w = 0 is not defined. The dynamic stiffness of the proposed hysteretic model has the same imaginary part as the "ideal" hysteretic damper, but has the appropriate real part that makes the model causal. The proposed model is constructed by requiring that the real and imaginary parts of its transfer functions satisfy the Kramer-Kroning relations. This condition ensures that the corresponding time response functions of the proposed model are zero at negative times.
The causal hysteretic element is the limiting case of a linear viscoelastic model (model exhibiting linear hysteretic damping) with nearly frequency independent loss stiffness, as proposed by M. Biot in 1958. This demonstrates that there is a continuous transition from a linear viscoelastic model to a model that is ideally hysteretic.
The discussion following the seminar centered on the areas of application and the importance of the proposed model in structural analysis.
Back to top
Click here to return to the Table of Contents