The main aim is to present the key ideas behind application of probabilistic entropy, and also probabilistic distance in civil and mechanical engineering problems subjected to various sources of uncertainty. A motivation for this research is to propose some alternative approach to the traditional moments-based methods available in probabilistic mechanics, where single universal uncertainty measure is determined instead of various statistics. This idea is implemented using Shannon entropy and further, an application of the probabilistic distance in structural reliability assessment is proposed, where Bhattacharyya distance is applied and contrasted with the traditional First Order Reliability Method (FORM). Numerical methods implemented for the solution of the static and dynamical problems are in turn – Monte-Carlo simulation, iterative generalized stochastic perturbation technique, and also semi-analytical approach. Their programming in the computer algebra system MAPLE is displayed. Engineering computations are delivered in conjunction with the Finite Element Method (FEM) systems ABAQUS and ROBOT. The test usage of the Artificial Neural Network (ANN) implemented in Python environment is also documented. The case studies under consideration include but are not restricted to dynamic response of selected engineering structures, homogenization of some composite materials, and also some multi-physics problems.
March 3, 2025, 3:30pm-5:30pm
The seminar will take place in blended mode: in presence at the Aula Seminari Capannone C, Faculty of Architecture, via Gramsci 53, Rome and online via Zoom.
For the seminars supplied by the PhD Program of the DISG credits are not provided. Also certificates of attendance will not be produced for seminars.