Many of today?s steel bridges are being constructed with longer spans and higher skew. As a result, the North Carolina Department of Transportation (NCDOT) has experienced numerous problems in predicting the dead load deflections of steel plate girder bridges. In response to these problems, the NCDOT has funded this research project (Project Number 2004-14). Common dead load deflection prediction methods, which traditionally utilize single girder line (SGL) analysis, have been shown to over predict the dead load deflections; the inaccuracy can result in various costly construction delays and maintenance and safety issues. The primary objective of this research is to develop a simplified procedure to predict dead load deflections of skewed and non-skewed steel plate girder bridges. In developing the simplified procedure, ten steel plate girder bridges were monitored during placement of the concrete deck to observe the deflection of the girders. Detailed three-dimensional finite element models of the bridge structures were generated in the commercially available finite element analysis program ANSYS, and correlations were made between the simulated deflections and the field measured deflections. With confidence in the ability of the developed finite element models to capture bridge deflection behavior, a preprocessor program was written to automate the finite element generation. Subsequently, a parametric study was conducted to investigate the effect of skew angle, girder spacing, span length, cross frame stiffness, number of girders within the span, and exterior to interior girder load ratio on the girder deflection behavior. The results from the parametric were used to develop a simplified procedure, which modifies traditional SGL predictions with empirical equations to account for skew angle, girder spacing, span length, and exterior to interior girder load ratio. Predictions of the deflections from the simplified procedure and from SGL analyses were compared to the deflections predicted from finite element models (ANSYS) and the field measured deflections to validate the procedure. It was concluded that the simplified procedure may be utilized to predict dead load deflections for simple span, steel plate girder bridges. Additionally, an alternative prediction method has been proposed to predict deflections in continuous span, steel plate girder bridges with equal exterior girder loads, and supplementary comparisons were made to validate this method.

Accidental/intentional contamination continues to be a major concern for the security management in water distribution systems. Once a contaminant has been initially detected, an effective algorithm is required to recover the characteristics of the contaminantâs source based on dynamically varying streams of sensor observations. This dissertation focuses on the development and demonstration of a new algorithm to characterize a contaminant source quickly, accurately, and robustly. An evolutionary algorithm (EA)-based adaptive dynamic optimization technique (ADOPT) is proposed, potentially providing a real-time response. In addition to offering adaptive capacity in a dynamic environment, this algorithm is able to assess the degree of non-uniqueness in the solution through multi-population scheme. This approach, however, requires a large number of time-consuming simulation runs to evaluate possible solutions, and it may be difficult to converge on the best solution or a set of alternative solutions within a reasonable computational time. For this reason, it is desirable to appropriately reduce the decision space over which the optimization procedure must search to reduce the computational burden and to produce faster convergence. A logistic regression-based prescreening technique is investigated in order to reduce the decision space by estimating the probability of a node being a contaminant source location. When a small set of potential source nodes are identified, applying the local search procedure to this set of locations is computationally efficient and potentially good at identifying the best solution. The EA-based ADOPT is then integrated with a logistic regression analysis and a local improvement method to expedite the convergence and to solve the problem potentially faster. The effectiveness of the proposed methods is demonstrated for contamination source identification problems in two illustrative water distribution networks.

Collapse of unreinforced masonry (URM) structures, including infill walls, is a leading cause of property damage and loss of life during extreme loading events. Many existing structures are in need of retrofit to bring them in compliance with modern design code provisions. Conventional strengthening techniques are often time-consuming, costly, and add significant weight to the structure. These limitations have driven the development of alternatives such as externally bonded (EB) glass fiber reinforced polymer (GFRP) strengthening systems, which are not only lightweight, but can be rapidly applied and do not require prolonged evacuation of the structure. The objective of this research program was to evaluate the effectiveness of strengthening infill masonry walls with externally bonded GFRP sheets to increase their out-of-plane resistance to loading. The experimental program comprises fourteen full-scale specimens, including four un-strengthened (control) specimens and ten strengthened specimens. All specimens consisted of a reinforced concrete (RC) frame (which simulates the supporting RC elements of a building superstructure) that was in-filled with solid concrete brick masonry. The specimens were loaded by out-of-plane uniformly distributed pressure in cycles up to failure. Parameters investigated include the aspect ratio, the strengthening ratio, the number of wythes, and the type of FRP anchorage used. The type of FRP anchorage was found to greatly influence the failure mode. Un-strengthened specimens failed in flexure. However, strengthened specimens without overlap of the FRP onto the RC frame failed due to sliding shear along the bed joints which allowed the walls to push out from the RC frames in a rigid body fashion. In the case where GFRP sheets were overlapped onto the RC frames, the aforementioned sliding shear caused delamination of the GFRP sheets from the RC frames. Use of steel angles anchored along the perimeter of the walls as shear restraints allowed these walls to achieve three times the service load without any visible signs of distress. GFRP strengthening of infill masonry walls was found to be effective, provided that proper anchorage of the FRP laminate was assured.

Information on season-ahead streamflow forecasts is beneficial for the operation and management of water supply systems. Developing such long-lead (3-12 months) stream flow forecasts typically depend on exogenous climatic conditions particularly sea surface temperature (SSTs) conditions in tropical oceans. Identification of such conditions that influence the moisture transport into water resources regions is important to develop low-dimensional statistical models and to analyze climatic forecasts from General Circulation Models (GCMs). The main purpose of this study is to develop probabilistic streamflow forecasts for the Falls Lake Reservoir, NC, for the summer season that is critical for developing water management strategies so that the City of Raleigh?s water demands could be met through water conservation measures. The study develops two low-dimensional statistical models based on SSTs in the tropical Pacific, tropical Atlantic and over the NC Coast. Given that prediction from any model is bound to have unavoidable error/model uncertainty, the study intends to combine the forecasts from individual models to develop an improved multi-model forecast. For this purpose, the study develops an algorithm for combining forecasts from individual forecasts by evaluating the performance of individual forecasts contingent on climatic (predictor) conditions. The methodology is demonstrated through the development of multi-model ensembles of streamflow forecasts for the Falls Lake reservoir by combining probabilistic streamflow forecasts from two low dimensional statistical models. Using Rank Probability Score (RPS) for evaluating each year?s streamflow forecasts for the summer months (July-August-September) from the two low dimensional models, the methodology proportionately gives higher representation by drawing increased ensembles for a model that has better predictability under similar predictor conditions. The performance of the multi-model forecasts is compared with the individual model?s performance using various performance evaluation measures. By developing multi-model ensembles based on leave-one out cross validation and split sampling, the study shows that evaluating the model?s performance based on the predictor state provides a better alternative in developing multi-model ensembles instead of combining the models purely based on their long-term predictability. The method is also extended to combine various GCMs to get improved winter (December-January-February) precipitation forecast for the entire US. Finally the study shows the utility of the multi model precipitation ensembles to develop improved streamflow forecasts for the Falls Lake through statistical downscaling.

Lab scale reactors were used to measure the emissions of non-methane organic compounds (NMOCs)and hazardous air pollutants (HAPs) from decomposing refuse. Reactors containing municipal solid waste (MSW) as well as reactors containing individual waste components (paper waste, yard waste, food waste, household hazardous waste) were operated under anaerobic conditions with leachate neutralization and recycle. Decomposition of MSW under aerobic and nitrate-reducing conditions was also studied. NMOC yields ranged from 0.016 mg-C/dry gm (paper) to 0.347 mg-C/dry gm (food). The data suggest that volatilization, or air stripping, is the primary mechanism of NMOC release, although the decomposition process also contributes.

Unreinforced masonry (URM) structures comprise a considerable proportion of the building stock worldwide. However, these structures generally do not behave well under extreme wind or earthquake loading. As part of on-going research, methods of repairing or strengthening URM walls subject to out-of-plane loading using fiber-reinforced polymers (FRP) are being investigated. For several reasons, one method showing particular promise is the use of near-surface mounted (NSM) Carbon FRP strips. Research to-date has made significant progress in quantifying the fundamental behavior of the bonded FRP-to-masonry interface and the behavior of URM walls repaired with vertical FRP strips subject to out-of-plane loading. This thesis presents the experimental results of five large-scale clay brick masonry walls strengthened using NSM CFRP strips and loaded out-of-plane statically to failure using an airbag system. Vertical and Horizontal NSM configurations were tested separately as well as orthogonal grid configurations constructed using two different techniques. The experimentally observed failure mechanisms are described in detail for each wall.

Laboratory experiences, i.e. visualization of material covered in class and hands-on use of equipment, are especially advantageous to engineering classes such as structural mechanics. Unfortunately it is sometimes difficult for on-campus students to be taken to a lab setting and impossible for those who are off campus due to work, disabilities, or other complexities and taking class through distance education. This project describes a shake table experiment that is being converted to a distance-learning environment. This will include remote access, control, and protection from misuse. An aspect of the project that differentiates it from simple remote viewing of a lecture or experiment is the need to control the experiment and to protect against the possibility of damage occurring to this particular setup if left unmonitored. This last point necessitates the inclusion of sufficient safety protocols. The environment must allow remote controlling of the system, multi-user viewing, data saving, and download capabilities. The technology selected for use in this project is the LabVIEW programming environment in conjunction with its real time counterpart, LabVIEW RT. By using this language, practical and intuitive control panels coupled with easy to follow data flow block diagrams are made possible. The LabVIEW code likewise handles the data acquisition. The information sent and received through the DAQ card is processed by LabVIEW RT code embedded in the real time processor. The information is then sent to a host computer for saving, visualization, and distribution to remote clients. This visualization includes an oscilloscope for displaying the accelerations from both the table and the structure residing on the table. Further visualization is given by way of a video camera. The code must be made safe from unauthorized usage in addition to allowing for the university network to remain protected. This research outlines in detail the setup required and programs needed to implement such a system and presents the information in a manner that can be helpful regardless of the programming language chosen.

The purpose of traffic signs is to provide information for the orderly movement that guides all road users as to direction, regulations and warnings. A driver may be able to view sign sheetings from a long distance with no sight obstruction. However, there are many cases in which the driver¡¯s line of sight is blocked by obstructions, such as trees, hills, curves, and other signs. These obstructions may reduce the effectiveness of the sheetings and therefore the frequency of safe response by drivers. The objectives of this project are to determine the distribution of the visibility distance for obstructed and unobstructed signs and to identify reasons for being obscure using videologs. The videologs contain images that can be used for this visibility distance study with the benefits of less cost, less potential danger, and less time than data obtained from a manual field survey. Most signs were obstructed. A little more than half of the obstructions were curves and hills, and trees were the most prevalent obstruction in the urban area that was biggest problem among the road types. About 25.4% of obstructed signs had less than 400 feet of visibility distance. In order to avoid the possible reduction of the effectiveness of traffic signs, sign placement should be considered during geometric highway design, and the impact of trees that may be growing should be taken into consideration.

Performance modeling of asphalt concrete pavements is one of the most difficult, but important tasks facing pavement engineers. Experiences at North Carolina State University suggest that this task is best accomplished by utilizing two separate models; one to account for the material behavior and another to account for boundary conditions, such as tire-pavement interaction, temperature gradient along the layer thickness, pavement structural design, etc. The material characterization model should focus on the material irrespective of geometry, i.e., fundamental properties. The structural model should be robust enough to account for the range of conditions experienced by pavements in service. Two peer-reviewed and published papers are presented here which deal with the development of a constitutive material model for asphalt concrete. In the first, the viscoelastoplastic continuum damage model in tension is applied to materials from the Federal Highway Administration?s Accelerated Load Facility study on modified mixture performance. It is shown that the material model is capable of describing the behavior of the tested mixtures over a range of conditions from primarily viscoelastic to primarily viscoplastic. Further, the model shows sensitivity to changes in asphalt binder and the ability to predict the behavior of asphalt concrete mixtures containing polymer modified binder. The second paper presents results from an experimental study of anisotropy in asphalt concrete. Anisotropy occurs due to the preferential orientation of aggregate particles in the mixture and is found to have varying levels of significance depending on both the mode of loading and the levels of deformation applied. In the linear viscoelastic range, anisotropy is found to have little effect on the material behavior, whereas under monotonic compressive loading until failure, it is found to contribute significantly. Further, it is found that temperature and rate affect the significance of anisotropy in asphalt concrete. Conclusions and plans for future work are also presented.

Filamentous bulking in activated sludge treatment plants is a worldwide problem. Understanding the growth requirements of specific filamentous organisms will allow the development of better control strategies for bulking. In this study, the short term effects of eight carbon sources and three inoculum sources on the growth of filamentous bacteria were tested. Three lab scale sequencing batch reactors (SBR) were operated. Microscopic (Gram and Neisser staining) and molecular methods (Denaturing Gradient Gel Electrophoresis [DGGE], Fluorescent in Situ Hybridization [FISH]) were used to track the microbial population changes in the reactors. Sludge volume index (SVI) measurements were used to monitor bulking in the reactors. DGGE and sequencing results indicated the presence of the filamentous bacteria Sphaerotilus natans and Thiothrix. S. natans grew in glucose-, acetate-, and sucrose-fed reactors, regardless of the inoculum source. It also grew in propionate- and pyruvate-fed reactors inoculated with the sludge from the Neuse River Wastewater Treatment Plant (WWTP). Thiothrix was detected in propionate- and pyruvate-fed reactors inoculated with sludge from the South Cary WWTP, and in glucose- and acetate-fed inoculated with the sludge from the Neuse River WWTP inoculated reactors. In addition to these two filaments, Gram and Neisser staining indicated the presence of Nostocoida limicola in Neuse River WWTP inoculated reactors. The presence of S. natans and T. nivea was confirmed with FISH. SVI measurements were consistent with the level of bulking, showing an increase as the number of filaments in the reactors increased. This study confirmed that readily biodegradable substrates favored the growth of S. natans, T. nivea and, N. limicola in activated sludge. The simultaneous use of microscopic and molecular tools provided the information above with one method compensating for the other method?s biases.