| Abstract |
Recent events such as the collapse of the Morandi Bridge in Italy, and the breach of Sanford Dam in North Carolina (USA), have brought
to the table the relevance of infrastructure maintenance. In this context, aspects related to risk analysis are gaining relevance in dam
engineering, since allow making informed decisions in terms of prioritisation of investments.
Although there is a well-defined framework, both nationally and internationally, for risk analysis of dams, the existing guides also highlight
aspects with a clear margin for improvement. This is the challenge that will be addressed with this project, by means of the development of
new computational tools, based on machine learning and numerical methods, to improve the processes involved in the quantification of
uncertainty in dam safety studies. The benefits will be translated into a more detailed analysis of failure modes that are currently
addressed with very simplified methods and the application of advanced numerical modelling in other failure modes whose probability is
currently estimated by other means.
This objective requires the use of technologies that have been developed and are applied in other fields, but that must be adapted to the
particular case of the analysis of dam behaviour: numerical modelling, risk analysis, and machine learning. The project team includes
experts with strong expertise in each of these three fields, all in relation to dam engineering.
The project is aligned with:
a) Priority II within Societal Challenge 5, which deals with risk analysis, safety and security of infrastructures.
b) A COST Action promoted by the PI, with 21 recognized experts on dam engineering from Europe and the USA, whose objectives
partially overlap and whose activities are complementary to those funded within this project.
c) The priorities of the European Club of ICOLD. It supports the posed objectives with the participation of its president in the COST Action.
d) The ExaQUte European Research Project, coordinated by CIMNE, whose objective is constructing a framework to enable Uncertainty
Quantification (UQ) and Optimization Under Uncertainties (OUU) in complex engineering problems using computational simulations.
The members of the project team have strong international connections with researchers from relevant Universities in Europe and the
USA, as well as with International Institutions such as the US Army Corps of Engineers, the World Bank or the Inter-American
Development Bank. |