An anisotropic critical state model for monotonic and cyclic loading of Ko-consolidated clay is presented, based on extension of an existing model. First, the performance of two existing critical state models is evaluated for anisotropic and cyclic loading. The first model is based on isotropic hardening, which incorporates a reduction of the yielding surface during elastic unloading in order to simulate a real soil behavior under cyclic loading. The second critical state model combines isotropic hardening with kinematic hardening, and is based on a two-surface concept. The proposed model adopts the two-surface concept, consisting of the consolidation surface and the inner yield surface, and the combined isotropic and kinematic hardening laws. To account for the effects of the anisotropy induced by the initial consolidation stage, the proposed model introduces a rotation of the consolidation and inner yield surface. The behavior of the new model is illustrated through numerical parametric investigations and comparisons with a limited number of existing test data.

Significant reduction in vibration can be obtained by using semi-active control in structures subjected to pulse type excitations. This can be achieved by (a) independently variable stiffness control; (b) independently variable damping control or (c) combined variable stiffness and damping control. This thesis presents nonlinear control algorithms developed to vary stiffness and damping in single and multidegree of freedom structures subjected to pulse type excitations to reduce vibrations. The efficiency of the controllers is studied both analytically and experimentally. Nonlinear response spectra and time-histories are presented to show the reduction in response obtained by using the control algorithms. The study concludes that the variable stiffness and variable damping systems lead to significant response reduction over a broad period range and can be implemented in practice.

This dissertation consists of four parts. The first part deals with the response of liquid storage tanks to a vertical component of ground motion. Galerkin's method is used to solve this problem approximately, and its accuracy is established by comparing the results with those obtained by the exact modal superposition method. The effects of soil structure interaction are examined, and a simple practical design procedure is proposed for providing for these effects. The second part deals with the response of liquid storage tanks to a horizontal component of ground shaking. The problem is analyzed by application of the Rayleigh-Ritz procedure in combination with Lagrange's equation. The details of the method of analysis are presented along with comprehensive numerical data which may be used readily in design applications. The third part of the dissertation deals with the response of liquid storage tanks, both rigid and flexible, to a rocking component of ground shaking. Emphasis is placed on understanding the behavior of the tank-liquid system and establishing the interrelationship of the responses of the system to rocking and to a lateral, translational motion. The fourth part deals with the harmonic response of massless ring foundations supported on a homogeneous elastic halfspace and subjected either to a vertical force or an overturning moment. The theory of elastic wave propagation is used to obtain the impedance functions. The method of analysis takes due account of the mixed boundary conditions at the surface of the halfspace. The effect of foundation mass on the response is also studied for the vertically excited system. The reported data are believed to be of high accuracy.

Existence of structural damage in civil structures may greatly deteriorate the overall performance of the system or even lead to disastrous consequences. Therefore, structural damage detection and health monitoring are very important from life-safety and economic viewpoints. In this research, three new fault detection filter-based damage detection methods, including the equivalent static output feedback controller design method, eigenstructure assignment method and Hinfinity/H_ robust detection filter design method, are developed. Meanwhile, a new flexibility-based algorithm is proposed to estimate the extent of structural damage. Firstly, the problem of finding a stable detection filter is converted to the equivalent problem of finding a decentralized static output feedback controller. Then, the iterative Linear Matrix Inequalities (LMI) technique and Genetic Algorithm (GA) are, respectively, applied to find the stable detection filter gain. Although the above equivalent problem, iterative LMI technique and GA algorithm are not new, the idea of combing them together for structural damage detection is novel and has great potentials in solving robust damage detection filter gains. Secondly, a novel eigenstructure assignment method is developed and combined with LMI technique for robust structural damage detection. The obtained robust detection filter can reduce the noise effect on the isolated output residuals, which improves structural damage detection results. Thirdly, a novel damage detection method using H infinity/H_ concept and iterative LMI technique is proposed to find a stabilizing fault detection and isolation filter which not only bounds the Hinfinity norm of the transfer function from disturbances to the output residual, but also does not degrade the component of the output residual due to the fault. Fourthly, a two degree-of-freedom (DOF) system with semi-active independently variable stiffness (SAIVS) device is tested to verify the above proposed structural damage detection methods. The experiment results show that the proposed methods are effective and efficient. Finally, this thesis also develops a new flexibility-based numerical algorithm to detect structural damage. Based on the characteristics of flexibility matrix which is easy to obtain with enough accuracy experimentally, Gauss-Newton optimization method is used to find the optimal structural parameters. Comparing them with the healthy parameters, structural elements with and extent of damage can be determined.

The strain sensing ability of single-walled carbon nanotubes (SWNTs) in buckypaper are explored using Raman spectroscopy. This sensing ability is also examined based on electrical properties using SWNTs in buckypaper, a nanocomposite (2% wt.) from National Aeronautics and Space Administration (NASA), and a PVDF-DWNT (polyvinylidene-fluoride double-walled) composite (0.1% wt.). The voltage and resistance change is measured by a four-point probe setup; with the voltage or resistance calculated using Ohm's law. Compressive and tensile forces are applied using an MTS servohydraulic testing machine. The effect of temperature on the NASA specimen is also studied. Results will show an approximately linear resistance change in the buckpaper and NASA specimen when subjected to tension and compression forces, while the PVDF sample will not show this because the % wt. is below the percolation value (0.19%) necessary to achieve conductivity. A linear relationship between temperature and change in resistance in NASA specimen is shown.

This study is concerned with the performance of reinforced concrete flat-slab buildings during the Loma Prieta Earthquake. Three flat-slab buildings ranging from six to twenty-four stories were selected for investigation. Each building is analyzed using a number of different 3-D analytical models with a finite element program. Various models account for the change of stiffness of the slabs, columns, and shearwalls during the earthquake. The characteristic parameters determined from the analytical models are then compared with the identified characteristic parameters using the system identification techniques. Based on the comparison, a most appropriate model is selected which is then used in analyzing each building for lateral drift and base shear using the code defined lateral loads and also using the recorded ground acceleration. Based on the comparison of the recorded and calculated response, an estimate of the stiffness of the flat-slab is suggested which gives a better predication of the drift response under seismic loading. The comparison of the computed base shear from the dynamic analysis and the UBC seismic code shows the validity of the assumed distribution of the shear forces.

Nonlinear dynamic analysis of non-ductile flat-plate buildings requires a suitable hysteretic model and an equivalent frame approach which appropriately models the lateral stiffness of the buildings. Based on the test results of non-ductile individual interior and exterior connections and two-bay slab-column subassemblies, hysteretic parameters related to stiffness degradation, strength deterioration and pinching of hysteretic loops are identified. The hysteretic model thus obtained is then incorporated into an equivalent frame analysis approach. The proposed equivalent frame approach is based on the equivalent beam concept which is calibrated to match the lateral stiffness and the moment-transfer strength of slab-column connections. Simulation of the load-drift response of the test subassemblies with the proposed analytical approach has good agreement with the measured response. Several slab-column buildings are designed as non-ductile systems and their response to typical earthquake loading is studied.

Recent earthquakes have shown the vulnerability of flat-slab structures to severe ground motion. The failure in such structures typically initiates at the slab-column connections in the form of a punching failure. This investigation was carried out to evaluate the adequacy of current procedures for the design of slab-column connections. The variables studied included the intensity of gravity loading, slab shear reinforcement at the column line, slab overhang and stiff edge beam at exterior connections, and the indeterminacy of the connection subassembly. Nine half scale slab-column subassemblies were tested under simulated earthquake loading. Seven of the subassemblies simulated a single floor of a two-bay flat-plate structure. Each subassembly consisted of one interior and two exterior slab-column connections. The remaining two specimens were individual interior and exterior connections. All specimens were subjected to the same predefined displacement routine which consisted of twenty cycles of incremental displacements increasing to a maximum of seven percent drift. Increased slab gravity load significantly reduced the drift capacity of both interior and exterior connections. To achieve a lateral drift level of 1.5 percent prior to failure, the ultimate direct shear on flat-plate connections must be limited to $V\sb{u} \leq C\sb{d}\sqrt{f\sb{c}\sp\prime}b\sb{o}d$, where $C\sb{d}$ = 2.0 for exterior connections and $C\sb{d}$ = 1.4 for interior connections. The ACI code design approach for interior connections was unconservative for higher gravity load levels. For the range of shear stress levels studied in these tests, the ACI Committee 352 recommendations that moment and shear be treated independently for design of exterior connections appears reasonable. The ultimate direct shear capacity of the exterior connections in this study was $V\sb{u} = 2\sqrt{f\sb{c}\sp\prime}b\sb{o}d$. Closed hoop stirrups enclosing the slab flexural reinforcement passing through the interior connection prevented punching shear failure and increased the ductility of the connection. A stiff edge beam or slab overhang at the exterior connections increased both the strength and ductility of these connections. The behavior of the combined specimens was similar to a summation of the individual connections especially at drift levels less than 1.5 percent.

The stochastic response of a yielding 2DF system subjected to a stationary Gaussian white noise excitation is investigated. The particular model considered represents a structure with yielding girders, so is more realistic than the common rigid floor model. The responses considered are the stationary root mean square (RMS) distortions of columns and rotations of girders. Simulation results are presented, as well as an equivalent linearization to approximate the response levels. A numerical integration scheme is presented to obtain simulation data on the stationary RMS distortions and rotations. The stationary RMS response is obtained by averaging a long sample time history (assuming ergodocity) of the response after the response has reached stationarity. The equivalent linearization approach is based on a substitute structure concept. Values of linear parameters are determined from simulation data for the SDF system and are functions of ductility. The Lyapunov equation is then written to calculate the stationary RMS response of the linearized system. (Abstract shortened with permission of author.)

One can safely claim that not one single surface roughness parameter, in terms of completely characterizing a surface, can be exhaustive. This study is both an experimental and theoretical study of characterizing surface finish in relation to fatigue; and the role of this finish in fatigue initiation. A total of forty-two samples, comprising three different groups that exhibited different surface finish properties, were fatigued using constant amplitude loading. The S-N curves for these groups were established. A correlation between surface finish parameters and fatigue initiation was established. The parameters of importance were expected to be the deepest scratch and the rms valley curvature (sharpness) of the measured surfaces. The product of both parameters showed a reasonable degree of correlation with fatigue initiation at different stress levels. This product became a basis of establishing the functional relationships between surface finish parameters and fatigue failure. A number of specimens were prepared having a single scratch that varied in depth from 0.004 inches to 0.02 inches. The scratches had a fixed radius of curvature of 0.003 inches. The specimens were fatigued and the resulting S-N curves were grouped into one spectrum. The single scratch spectrum utilizing the deepest scratch and sharpness was used as a basis to predict fatigue failure. First-passage criteria were used to predict the deepest scratch in longer sampling lengths. The product of the predicted deepest scratches and curvature were used with the single scratch spectra to predict fatigue failure. The predicted life was within the 95% confidence bands of the S-N curves generated from fatiguing the specimens.