Sujith Mangalathu

Parameterized Seismic Fragility Curves for Curved Multi-frame Concrete Box-Girder Bridges Using Bayesian Parameter Estimation

ABSTRACT This paper addresses the application of a Bayesian parameter estimation method to a regional seismic risk assessment of curved concrete bridges. For this purpose, numerical models of case-study bridges are simulated to generate multiparameter demand models of components, consisting of various uncertainty parameters and an intensity measure (IM). The demand models are constructed using a Bayesian parameter estimation method and combined with limit states to derive the parameterized fragility curves. These fragility curves are used to develop bridge-specific and bridge-class fragility curves. Moreover, a stepwise removal process in the Bayesian parameter estimation is performed to identify significant parameters affecting component demands.

Predicting the dissolution kinetics of silicate glasses using machine learning

Abstract Predicting the dissolution rates of silicate glasses in aqueous conditions is a complex task as the underlying mechanism(s) remain poorly understood and the dissolution kinetics can depend on a large number of intrinsic and extrinsic factors. Here, we assess the potential of data-driven models based on machine learning to predict the dissolution rates of various aluminosilicate glasses exposed to a wide range of solution pH values, from acidic to caustic conditions. Four classes of machine learning methods are investigated, namely, linear regression, support vector machine regression, random forest, and artificial neural network. We observe that, although linear methods all fail to describe the dissolution kinetics, the artificial neural network approach offers excellent predictions, even for untrained data, thanks to its inherent ability to handle non-linear data. We further note that the predictive ability of simpler methods, such as linear regression, could be improved using additional physics-based constraints. Such methods, called as physics-informed machine learning can be used to extrapolate the behavior of untrained compositions as well. Overall, we suggest that a more extensive use of machine learning approaches could significantly accelerate the design of novel glasses with tailored properties.

Application of Bayesian Methods to Probabilistic Seismic Demand Analyses of Concrete Box-Girder Bridges

This paper proposes a set of probabilistic demand models of bridge components using the Bayesian parameter estimation method. To develop probabilistic demand models of individual bridge components, the material, structural, and geometric properties used in the bridge models serve as independent variables and the response data of engineering demand parameters for individual components monitored from the analyses serve as dependent response variables. To illustrate the proposed methodology, a typical reinforced two span, three frame curved concrete box-girder bridge in California is selected as a case study. Probabilistic numerical bridge models are developed, and then nonlinear time history analyses are performed using a set of ground motions representative of the seismic hazard. The significant input parameters are identified through a stepwise removal process in the Bayesian approach. The demand models generated in the Bayesian approach provides a more reliable estimation of the seismic demand compared to the traditional approach in which the ground motion intensity measure is the only input parameter.

Adjustment Factors to Account for the Effect of Bridge Deck Horizontal Curvature on the Seismic Response of Concrete Box-Girder Bridges in California

This research suggests adjustment factors to account for the effect of bridge deck horizontal curvature on the probabilistic seismic demand model (PSDMs) and fragility curves of concrete box-girder bridges in California. For this purpose, typical configurations of horizontally curved bridges in California are selected to create detailed three-dimensional (3-D) probabilistic bridge models with different levels of bridge deck horizontal curvature. Simulation results from the nonlinear time history analysis (NLTHA) of bridges are used to compare the PSDM of individual bridge components using a statistical technique called analysis of covariance (ANCOVA). Comparison results are used to group bridge classes and to suggest adjustment factors. Grouping results indicate that the PSDMs of unseating and bearing displacement are statistically significant for bridges with different levels of deck horizontal curvature. The effect of deck curvature and the use of the modification factors are demonstrated in this paper through the generation of fragility curves.