Numerical Evaluation Of Different Retrofitting Techniques Of New Orleans Floodwalls And Implementation Of Advanced Constitutive Model In Flac3D

The objective of this study was to re-examine some but the crucial established failure mechanism, and also to establish other failure mechanisms of the floodwalls due to hurricanes in New Orleans so that the necessary retrofitting techniques could be sought out. Depending on the failure mechanisms, two of the proposed retrofitting techniques such as structural cap to maintain integrity by tying adjacent panels of I-type floodwalls, and bentonite apron to self seal gaps that occur between the floodwall and foundation soil were chosen, for which their performances had to be evaluated. This dissertation documents all the numerical analysis conducted to address those tasks. At first, this study utilized the field monitoring results from a full scale loading test in London Ave. Canal in New Orleans, LA to calibrate/back-calculate soil moduli. From this study, it turned out that the moduli of the field soils are mostly higher (1 to 12 times) than predicted magnitudes from laboratory tests or empirical relations, and quantitatively confirmed that the gap formation and strength reduction in the water glazed marsh layer greatly decreased the stability of floodwall. Using the calibrated parameters and the failure mechanism established, effectiveness of the structural cap was evaluated and it was found that the structural cap is effective in restricting the relative displacement of the adjacent panels maintaining integrity. In order to justify the application of retrofitting techniques design to seal these gaps (Bentonite Apron), this study conducted numerical and analytical evaluations to investigate the in-depth mechanism of gap formation using an effective stress approach. It was found that question of whether gap is the cause or the result of failure depends on soil properties and site conditions. But improving the existing foundation soil for the case where gap is the result of failure tends to cause the formation of gap which necessitates the implementation of gap sealing mechanism in any circumstances. The study also shothat swelling potential of the Bentonite Apron was such that it can seal the gap effectively, and at the same time it is not that large to cause the excessive displacement of the wall and the peripheral soil. Incorporation of anisotropy in numerical analysis is very important since most of the soils are anisotropic in nature. So far, none of the researchers of New Orleans flood protection system have incorporated this behavior in their numerical analysis as they are not readily available in commercial codes. So, to fill this gap, as a second objective, this study implemented the new user defined constitutive model (Anisotropic Modified Cam Clay Model) in FLAC3D, by taking the advantage of user defined model capability. The implemented Model was well verified and validated with the mathematical model and the experimental results respectively. When applied to the New Orleans floodwall sections, the results shothat isotropic Mohr Coulomb might be good for evaluating overall behavior in moderate deformation problems, while Anisotropic Modified Cam Clay Model was good even for large strain problems where more accurate evaluation of yielding is needed