The modeling and risk management of fatigue crack growth is a problem of critical importance in any mechanical system. The work presented in this study demonstrates two efficient methods for the modeling of non-planar fatigue crack growth and also the quantification of uncertainty in the model prediction. Discretization errors, both temporal and spatial, are considered for each model. The model parameters for crack growth models typically are obtained via experimental data. A methodology is presented here to calibrate those parameters when experimental errors are considered. The consideration of experimental errors leads to a more realistic approximation of the model parameters. A methodology for model calibration and validation using multiple sources of information is accomplished through a Bayesian network approach. Finally, natural variability, data uncertainty and model uncertainty are considered to provide an informed framework for risk management decision making with respect to inspection scheduling and fidelity.

Design optimization of large structural systems

Transmission of pollutants

Transportation system management techniques

In order to protect public health and the natural environment, regulatory agencies such as the U.S. Environmental Protection Agency (USEPA) and the Ontario Ministry of the Environment (MOE) have developed a variety of clean-up programs to remediate contaminated sites. Before a remedial strategy can be implemented, physical, chemical and biological information of the site must be characterized so that an appropriate remedial approach can be selected. Problems exist with common characterization applications such as non-representative sample collection, time-consuming and costly testing, and loss of sample integrity due to transport. The success of the dipole flow test (DFT) and dipole flow tracer test (DFTT) indicate the potential for identifying a wide range of physical aquifer characteristics through in situ techniques; this has led to the development of the dipole flow reactive tracer test (DFRTT), which can be used to identify sorption and degradation characteristics in addition to those characteristics already identifiable by the DFT and DFTT. The DFRTT contains numerous advantages including the ability to inject a variety of reactive tracers using pulse or continuous injection, it can be used at multiple locations and depths across a site, and it is cost-effective. The focus of this thesis was to develop and validate an efficient numerical framework for the reactive-transport component of the DFRTT interpretation model so that it could be used to simulate a variety of reactive tracers. The model will then be calibrated to match the resulting DFRTT field breakthrough curves (BTCs) so that certain aquifer property values can be obtained. The comprehensive-reactive transport model (CRTM) was developed to solve the advective-dispersive-reactive equation (ADRE) using a streamline-oriented control volume (CV) mesh, to minimize the introduction of numerical dispersion (due to the dipole’s rapidly converging and diverging flow field). In order to reduce computational effort, the ADRE was decoupled using operator-splitting techniques (OS) and the resulting partial differential equation for the transport component was solved using a finite volume approach, while the fourth-order form of Runge-Kutta was used to solve the resulting ordinary differential equation of the reactive component. Four OS techniques were implemented to decouple the ADRE; two of the techniques were iterative, while the other two were non-iterative. The use of iterative OS techniques enabled the introduction of a flux corrected transport (FCT) scheme (in addition to the commonly used central differencing scheme (CDS) and upwind differencing scheme (UDS)), for solving the transport portion of the ADRE. The CRTM was validated against a current “off-the-shelf” model, MODFLOW/MT3DMS. The resulting simulations indicated that the CRTM and MODFLOW/MT3DMS BTCs compared well for a conservative, decaying (first-order), and sorbing (linear sorption isotherm) tracer under specific conditions. However, MODFLOW/MT3DMS illustrated signs of failure when a large flow rate or a small longitudinal dispersivity coefficient was employed. It was concluded that MODFLOW/MT3DMS was unable to handle the rapidly converging and diverging dipole flow field, and that it was necessary to develop/utilize a dipole specific model for modelling the DFRTT application. Analysis of the three advective schemes indicated that CDS was an inappropriate method for the dipole configuration used in the investigation (due its non-monotone solutions), and that the UDS, although unconditionally monotone, produced excessive numerical dispersion. The FCT scheme had the benefits of both CDS and UDS; however, it was computationally slower than the UDS due to its iterative nature. Examination of the OS techniques indicated that in most cases there were no significant differences between any of the OS methods; however if a prominent sink term was utilized, the iterative techniques were deemed superior over the non-iterative techniques because of their ability to correct mass depletion. It was also determined that because of FCTs second-order accuracy, the OS techniques employing FCT had lower errors than those which did not. The CRTM was used to design a DFRTT biodegradation experiment under Canadian Forces Base (CFB) Borden conditions. Dipole parameters were altered so that a dipole configuration could be recommended for field testing. It was determined that a decent zone of influence aquifer volume (~ 65 m3) with a detectable BTC tail occuring between 2 and 3 days required a dipole length = 0.8 m, and a half-chamber length = 0.15 to 0.30 m. The oxidation of toluene by aerobic bacteria present within the Borden aquifer was also simulated. The purpose of these simulations was to obtain a detectable substrate BTC so that experimental parameters could be recommended for conducting a field experiment. To achieve a detectable substrate BTC the following recommendations were made: (i) sampling events should occur once every 1 to 2 hours (if possible) so that the changes in effluent concentration can be documented accurately; (ii) dipole parameters from Section 5.1 with a standard 2” well should be employed; (iii) a large substrate concentration should be utilized (> 4.0 mg/L); (iv) a flow rate of 1x10-4 m3/s should be used; and (v) an injection duration of 2 hours should be applied. As long as the field substrate BTC is measurable then Monod kinetics parameters can be estimated after a field and CRTM BTC comparison; this enables us to understand the aquifer’s biodegradation potential (for remedial design purposes).

Road traffic noise is becoming a major public concern. Many transportation agencies are looking for practical and economical means to reduce traffic noise generation and propagation. In 2003, the University of Waterloo’s Centre for Pavement and Transportation Technologies (CPATT) and the Regional Municipality of Waterloo embarked on a partnership to design quiet pavement test sections and to conduct controlled sound level measurement on four different types of asphalt surface courses. Four different surface courses, two Rubberized Open Graded Friction Course Asphalt Pavements (rOFC and rOGC), Stone Mastic Asphalt Pavement (SMA), and a control mix Hot-Laid 3 (HL-3), were placed in lengths of 600 m. The overall 2.4 km test area was closed to traffic and test vehicles were driven through the test area at the prescribed control speeds with sound level meters recording sound levels both at the tire/pavement interface as well as at the monitoring stations off the roadway. Impedance Tube Method and Reverberation Time Method were performed to determine the sound absorption coefficients of the pavement mixes. In order to evaluate the sound attenuation ability of the mixes, the results from rOFC, rOGC, and SMA were used to compare with the result from the control mix HL-3. Statistical analysis of measurement results was performed to see whether the differences between mixes are significant at a 95% confidence interval. Life cycle cost analysis was also performed in order to determine the cost effectiveness of each asphalt mix. Results indicate that traffic sound level increases as vehicle speed and size increase regardless of asphalt types. rOFC and rOGC perform significantly better than HL-3, but the performance slightly deteriorate after one year because of the clogging problem. SMA does not attenuate sound as effectively when compare to HL-3 at the early age. However, sound attenuation ability improves after one year of service. Overall result indicates that rOGC performs the best among all mixes in terms of the sound attenuation ability. Life cycle cost analysis shows that HL-3 is the most economical mix but it is the worst mix in terms of sound attenuation ability. It is recommended to conduct additional sound level and skid resistance measurements in the future to monitor the long-term pavement performance. Also investigation of the relationship between the sound level and sound absorption coefficient measurements is beneficial for the future acoustical evaluation for the asphalt mix.

Peak base displacement is one of the most important quantities in the design of base-isolated buildings. During the preliminary stages of design, a nonlinear time-history analysis is often not possible or too expensive, and hence reliable measures for predicting peak base displacement must be obtained through other means. In this study, regression models are developed in order to predict the peak displacement using a series of intensity measures (IMs) as model inputs. This thesis utilizes two methods for this purpose, Principal Component Regression (PCR) and a newly proposed method known as Sorted-Input Independent Component Regression (SI-ICR). In the framework of PCR and SI-ICR, the problem that exists due to correlation of IMs is addressed, which allows the transformation of correlated components into uncorrelated ones. This step is followed by dimensionality reduction of the components that do not contribute significantly to the explained variance of the original data set. A regression model using only one IM, peak ground velocity (PGV), is also developed to compare the advantages of using multiple IMs as opposed to one. Prediction results are presented and compared to simulation results for building models with increasing degree of complexity, starting with a two degree of freedom uniaxial case to a twelve degree of freedom biaxial model. It is concluded that PCR and SI-ICR significantly outperform the PGV model with PCR slightly outperforming SI-ICR. PCR is regarded as a more suitable and practical regression method for predicting the responses of nonlinear base isolated structures.

Improving the energy efficiency of the existing building stock is beneficial in terms of reducing both operational cost and environmental impact. Solid, load bearing masonry buildings comprise an appreciable part of the existing building stock, are valued for their durability and are often of historical significance, however their thermal performance is often poor. Thermal upgrades with interior insulation are usually considered risky as this lowers the drying potential of the wall enclosure which increases the moisture content of the masonry, possibly to damaging levels during freeze-thaw cycles. The current approach to assessing the suitability of brick for exterior use is based on acceptance criteria or a severe freeze-thaw test as prescribed by ASTM and CSA. The first method is based on an incomplete understanding of freeze-thaw physics. The second method subjects the bricks to a set of conditions that cannot encompass the highly variable range of in-service exposures. Past research indicates that rigid porous materials, such as brick and concrete, experience frost damage only if saturated beyond a fixed, critical degree of saturation. A more useful approach to assessing the risk of damage due to freeze-thaw would be to determine the critical degree of saturation of a material and then compare that to anticipated moisture loads under service conditions using computer modelling software such as WUFI. A test method was developed to determine the critical degree of saturation of a material requiring less than two weeks of testing by means of frost dilatometry. Representative brick were first sliced into 10 mm thick specimens. Material properties of the brick specimens were determined, including dry weight, dry density, porosity (determined by either boiling or vacuum saturation), water uptake coefficient (A-value), and initial length, to allow for computer modeling and analysis of results. The specimens were wetted to various degrees of saturation (based on either boil or vacuum saturation), sealed in a manner to ensure minimal moisture loss, and then subjected to at least six freeze-thaw cycles. The change in length of the specimens following freeze thaw testing was measured and the expansion, which in this study was expressed in terms of microstrain, plotted against degree of saturation. Any specimen that experienced expansion greater than that attributable to instrument error was considered damaged by frost. Determining the total open porosity of a material, necessary for measuring degree of saturation, is difficult due to close voids and trapped air in dead end pores. Boiling saturation and vacuum saturation are two methods commonly used to determine open porosity. For the size of specimens used in the test, almost all water absorption when using the boil method occurred during the first hour of boiling and first hour of soaking. The repeatability of boiling to determine total open porosity is low, with approximately 10% variability experienced over three rounds of boiling. Specimens absorbed a greater or equal amount of moisture using the vacuum saturation method than the boiling saturation method, indicating that vacuum saturation is more effective at determining total open porosity. The repeatability of vacuum saturation is good with very little variability observed over 3 rounds. The benefit of boiling saturation is that its equipment requirements and procedure are simpler than that of vacuum saturation. Freeze-thaw cycling was carried out by immersing the sealed specimens in a liquid bath. The change in length was determined by measuring pins attached to the specimens with an outside digital micrometer with a ratchet stop. A combination of experimentation and computer modeling showed that the samples could be wetted to target moisture contents with excellent accuracy and that moisture redistributed evenly in less than 24 hours for most samples. The minimum required temperature of the freeze cycle was chosen to be less than -12°C, in order to freeze all water contained in pores with a radius of 10nm and greater. It was determined, for the small specimen sizes used, that the cooling and warming phases should be at least one hour in duration to ensure that the specimens entirely freeze and thaw. Three sets of brick were subjected to the test to determine their critical degree of saturation: modern extruded brick, pressed brick from the 1950s, and historic brick dating to the 1870s. The older bricks experienced frost dilatation above 0.25 and 0.30 vacuum saturation respectively, with no damage evident below these thresholds. The modern extruded brick only experienced damage at 0.87 of vacuum saturation and greater. Knowledge of the critical degree of saturation of a brick allows architects and engineers to pursue retrofit strategies that increase thermal performance without jeopardizing the durability of the material by creating the conditions that allow the moisture content to increase to dangerous levels.

Fatigue durability, damage tolerance and strength evaluations of cracked structural components require accurate determination of stress intensity factors (SIF). Most practical crack configurations are embedded and surface breaking planar cracks subjected to complex two-dimensional stress fields. The only cracked body configuration which has been studied analytically for all types of applied stress fields is a circular crack in an infinite elastic solid. However, this model is suitable only for a narrow class of practical applications. Much wider class of practical problems can be solved using the model of an elliptical crack. The exact analytical SIF solutions for an elliptical crack were obtained only for some particular cases of polynomial applied stress fields. In the present work the exact analytical SIF solution has been obtained for an elliptical crack embedded in an infinite elastic body and subjected to an arbitrary applied normal stress field (Mode I). The most effective method of evaluating the stress intensity factor induced by an applied stress field is by using the weight function for a given cracked body. The weight function represents the SIF induced by a unit concentrated load. The only exact analytical weight function for a planar crack was obtained for a circular one. In the present research the exact analytical weight function has been derived for an elliptical crack embedded in an infinite elastic solid. The weight function for an elliptical crack was subsequently employed in the alternating method to obtain the unique SIF solution for a surface breaking semi-elliptical crack in a semi-infinite body subjected to an arbitrary applied stress field. The solutions obtained in the present work can be used in various practical applications, such as cracks in pressure vessels, welded structures and mechanical engineering components subjected to cyclic loading.

Between 1920 and 1950 an oil gasification plant operated on a property adjacent to Kettle Creek about 0.2 km from the mouth of Port Stanley harbour on Lake Erie, Ontario, Canada. Oil tar wastes from the gasification plant were stored on the site until it was eventually abandoned in 1987. At that time the Ontario Ministry of the Environment (MOE) determined that the site was contaminated with heavy metals and polycyclic aromatic hydrocarbons (PAHs) and that some of this waste had been flowing into Kettle Creek through the George Street drain in the village of Port Stanley for an undetermined period of time. The site was completely remediated in 1995 and the flow of contaminated water from the drain ceased. However, sediment sampling revealed the presence of heavy metals and PAHs in Kettle Creek, the inner and outer harbours, and in Lake Erie. From a drinking water source protection perspective, there was an interest in identifying the oil tar contaminants and assessing contaminated sediment transport within the Elgin Area Water Treatment Plant intake protection zones (IPZs). The effectiveness of conventional treatment processes currently available within Elgin Area Water Treatment Plant (WTP) in removing these contaminants was also evaluated. According to historical monitoring data from various compartments including soil, sediment, groundwater and surface water, three types of contaminants were identified, including heavy metals (Sb, As, Cd, Cr, Cu, Fe, Pb, Ni, Se, V and Zn), PAHs and volatile organics (benzene, toluene, phenols). Due to extremely low toxicity and exposure probability, some unregulated contaminants (iron, vanadium, zinc, phenol and some PAHs) were removed from the final contaminant list and were not discussed from the perspective of treatment. A technique developed by the USEPA to characterize and track contaminant plumes in water, the fingerprint analysis of leachate contaminants (FALCON), was for the first time investigated for its suitability as a tool to assist with the interpretation of contaminated sediment transport in surface water originating from a former oil/coal gasification plant and its potential to help assess drinking water intake protection zones. A source fingerprint based on 4 heavy metals (As, Cr, Pb, Ni) and 6 polycyclic aromatic hydrocarbons (PAHs) from 12 sampling sites in a contaminant-impacted harbour was generated. This source fingerprint of the contaminated harbour sediments was then compared to 48 fingerprints generated at other sites in the vicinity of two intake protection zones of a drinking water treatment plant. The source fingerprint did not match fingerprints of sites upstream from the contaminant input source in the creek which fed the small harbour and other potential contamination sources to the east and west in the lake. However, the source fingerprint did match most sites in an outer harbour and some outside the harbour break walls, including sediments collected from within the drinking water intake pipe ~3 km to the east of the harbour. A high correlation between water intake sediments and the source fingerprint demonstrated that contaminated sediments have reached water intake. However, no exceedances of the target contaminants were reported in intake surface water in the period from 1990 – 2010. It was also found that the correlation between the source fingerprint and those in the intake has been decreasing over the period for which data are available, confirming the success of remediation efforts. Surface water monitoring has demonstrated that PAH concentrations are lower than detection limits and only iron (Fe) exceeds the Ontario Drinking Water Quality Standards (ODWQS) aesthetic objective. The concentrations of oil tar contaminants in treated water were all below the MOE regulated concentrations, indicating that the current Elgin Area Water Treatment Plant configuration is effectively removing any oil tar contaminants present in raw water. Critical raw water concentrations (CRWCs), which represent maximum raw water concentrations that can reliably be removed by the Elgin WTP, were predicted for each oil tar contaminant. The probability of each contaminant exceeding the CRWC was then estimated using a Log Pearson Type III distribution. Copper was found to be the contaminant with the highest exceedance probability. A point system was designed to evaluate the cost-effectiveness of other treatment alternatives and to select the most appropriate of these to improve the robustness of the WTP. Granular activated carbon (GAC) was determined to be the most cost-effective compared to other techniques and hence is considered as the most suitable technique to be implemented in the plant in order to improve its robustness as it relates to dissolved heavy metal species and PAHs.