运筹offer:一周学界招聘总结(2023.7.10-2023.7.14)

We have an open fixed-term position for a lecturer (3 years""akademischer Rat a.Z."") in Numerical Analysis at the University of Jena (Germany) starting from 1st October 2023. We support the development of an independent research line and will equip the position with a budget for travel and guests. Teaching in German will be expected.

15. July 2023

Climate change has many consequences for natural gravity hazards in mountain environments, particularly by increasing their intensity and frequency.

The aim of this exploratory project is to apply digital sciences to the prediction and mitigation of these risks.

Numerical simulation of gravity flows (falling boulders, rock flows, debris flows, ice, etc.) has certainly reached a certain level of maturity, but its use in prediction and prevention is still in its infancy.

As part of a GRANIER exploratory project, so-called \data-driven modelling\ methods will be explored for gravity flows and protective structures. The aim will be to make the most of laboratory and observatory data in order to build and calibrate models, assess their sensitivity, improve their predictive nature, i.e. control and take account of uncertainties, using variational, statistical and AI methods. In turn, it is hoped that this will improve data generation and sustainability. These methods, which have already been tried and tested in the context of climate modelling, have hardly been used for gravity flows and complex rheologies of the frictional cohesive type, which are intrinsically non-smooth. This is the highly exploratory nature of this research.

The TRIPOP team is a joint research team of Inria Grenoble Rhône-Alpes and the Laboratoire Jean Kuntzmann (LJK), following on from the BIPOP team (2003-2017).

The team is mainly interested in the modelling, mathematical analysis, simulation and control of non-smooth dynamic systems. Non-smooth dynamics concerns the study of the temporal evolution of systems that are not smooth in the mathematical sense of the term, i.e. systems that are characterised by a lack of differentiability, either of the mappings in their formulations, or of their solutions with respect to time.The team is one of the few in the world to have brought together researchers in applied mathematics, control theory, computational mechanics and scientific computing in the field of non-smooth dynamics. In mechanics, the main examples of non-smooth dynamic systems are multi-body systems with one-sided Signorini contact, fixed-value friction (of the Coulomb type) and impacts, or plasticity.

The aim of this post-doc is to work on the joint use of data from different sources to evaluate and improve the predictive capacity of gravity hazard models.

Key words: Data-driven modelling, uncertainties, calibration, data assimilation, reduced-order models, substitution models.

The post-doc will develop the following points:

1) Statistical models integrating data of various kinds and the hazard models developed. The identification of hazard model parameters, in particular using Bayesian approaches, will make it possible to calibrate and quantify the uncertainties associated with the models.

2) Model reduction approaches (POD, PGD, etc.) or construction of substitution models (Sparse Polynomial Chaos, Gaussian Processes, etc.) to build simplified models that can be used in this context.

3) Application of various data assimilation techniques (particle or variational filters) to the models described in the first section and to reduced-order models.

The calibrated models will be integrated into an overall approach aimed at developing quantitative risk analysis methods.

Targeted applications: All the gravity hazards mentioned above, starting with the simplest and those for which data is available, such as block falls.

The candidate must hold a PhD in Mathematics applied to Mechanics.

Skills in uncertainty quantification, data assimilation and model reduction are essential, as is a taste for scientific computing (implementation and use of calculation codes). Knowledge of mechanical systems or geophysical flows would be an additional advantage.