About the Institute: The Department of Computer Science of the University of Pisa is the oldest of its kind in Italy, and the Computer Science Ph.D. School has the same distinction. The Doctorate accepts applications for all areas of Computer Science, Optimization and Numerical Analysis.
Duration: Three years. The deadline for the application is October 15, 2021. The selection process is expected to be performed swiftly thereafter and the activity to start before year's end.
Compensation: Standard Italian Ph.D. student fare, about 1150 Euro/month net. A contribution to increased living expenses will be available to support the student during the foreseen stay abroad (at least 6 months).
Requirements for the PhD position: basically any STEM Master degree (Engineering, Mathematics, Computer Science, ...), decent English level and very good academic record. Knowledge of optimization techniques and software is a plus, as is resilience and a taste for discovery.
Project abstract: In the "Fit for 55" climate package the EU has set the world's most ambitious targets for the decarbonification of the energy system (ES). Achieving those objectives requires all the stakeholders to forecast the response of the ES to the possible changes over long timescales. This requires building mathematical models that take into account uncertainties, be multi-level and multi-scale, and properly represent the many combinatorial and nonlinear aspects of the ES. The corresponding monolithic optimization problems are too large to tackle even on the best HPC systems. The only approaches that can possibly tackle problems of this scale and complexity without resorting to dramatic over-simplifications, as customary in current strategic ES models, are decomposition ones, since they are capable of exploiting all the several forms of structure (scenarios, geographical and temporal scales É) they possess. However, standard modelling systems do not support implement multi-level, heterogeneous decomposition approaches.
We build over previous international research projects that produced the Structured Modeling System++ (SMS++), a C++ software framework whose unique design choices already allow to tackle highly challenging mid-term models with approaches combining several forms of decomposition (Stochastic Dual Dynamic Programming, Lagrangian relaxation). We will develop the remaining algorithmic components that are needed for solving models of the ES spanning multi-decade time horizons with resolution down to the hourly one. Achieving this result, which will require exploiting the native parallel capabilities of SMS++ to run the approaches on HPC architectures, will be a very significant scientific achievement resulting on a seismic change in the perception of what is possible for long-term stochastic models. This will allow all stakeholders of energy systems to accurately plan for pathways towards the "Fit for 55" objectives ensuring economic and societal sustainability along all the transition period. The work will require the expertise of the industrial partner EdF, that has a long-standing collaboration with the University of Pisa and is already invested in the development and use of SMS++. SMS++ is open source and all the newly developed models will be so, which will make the results available to all the interested parties.
Further information can be found at
This is an University-wide call for many bourses, the one of interest here is "Modellazione della Transizione Verso un Sistema Energetico Decarbonizzato" at p. 41. The instruction to participate to the call can be found at Articolo 3, p. 3 of the document. Unfortunately the document is Italian-only for now (an English translation may appear, but this is not certain). Basically, the candidates need to upload their CV and a copy of the certificate that they have the requisite Master degree, indicating which of the bourses (s)he applies to. A Research Project can also be uploaded. Further instructions will be available when logging in the system as per the instructions on the document.
For any enquiries on the research project or assistance with signing up in the selection process (in particular for non-Italian candidates) do not hesitate to contact Antonio Frangioni (frangio@di.unipi.it).
----------------------------------------------------------------------
**********************************************************
*
* Contributions to be spread via DMANET are submitted to
*
* DMANET@zpr.uni-koeln.de
*
* Replies to a message carried on DMANET should NOT be
* addressed to DMANET but to the original sender. The
* original sender, however, is invited to prepare an
* update of the replies received and to communicate it
* via DMANET.
*
* DISCRETE MATHEMATICS AND ALGORITHMS NETWORK (DMANET)
* http://www.zaik.uni-koeln.de/AFS/publications/dmanet/
*
**********************************************************
Posts
