These weeks' work has been devoted to the following activities:
- Reading Stefano Moret's thesis
- ECOS 2015 paper extension
- Application for Finnish Academy PostDoc funding
- MSc project for control optimization of hybrid ferry
- Osmose test function
Reading Stefano Moret's thesis
Stefano Moret is a PhD student at our lab (IPESE) at EPFL. He is going to graduate in September on the subject of dealing with uncertainty in the strategic planning of large energy systems.
In short, he focused on the energy system of the whole Switzerland, trying to answer the question "How can we make a strategic plan for the development of the Swiss energy system in the future?"
Methodologically, this turned out to be strongly connected to the question of how to deal with uncertainties when optimizing an energy system. This is true for any energy system, but particularly for the case of ones (like a whole country's energy system) where the uncertainties (on future prices of commodities, interest rates, etc.) are huge.
So, Stefano focused on the following aspects:
- Uncertainty analysis: how does each uncertain parameter affect the uncertainty of the final optimal solution?
- Sensitivity analysis: how sensitive is the final optimal solution to the uncertainty on each individual parameter? This is very useful, as it allows to eliminate from the analysis (and, hence, simplify the otherwise huge task) the parameters that are the least relevant
- Robust optimization: How can we optimize the system, being aware of the uncertainty, protecting ourselves by the worst outcomes while not compromising the efficiency of the system?
This was a very interesting reading!
ECOS 2015 paper extension
We got to a turning point: the calculations are basically over.
So we had a Skype meeting, pointing the way through. The manuscript is starting to take shape, and although there is still much work to be done on the definition and verification of the calculations and the assumptions, I can finally see the light at the end of the tunnel.
TV (Tuong-Van Nguyen) took responsibiltiy for the critical part of clustering the resulting energy demand into representative days/hours/trips, while Fredrik and I will help in the practical implementation.
We are almost there!
And finally, thanks to Francesca (Belfiore, also PhD student at EPFL) I managed to have a correct implementation of the calculation of the bypass air in the main engines!!
Application for Finnish Academy PostDoc funding
Finally I have got an idea, and it is taking shape. I will not go much far into the details, but it is about autonomous vessels, wind propulsion, machine learning, gray-box modelling and condition monitoring. All, of course, for energy efficiency.
Keep posted, it's going to be awesome!
MSc project for control optimization of hybrid ferry
Here there has been an unexpected turn of events: the responsible person at Damen Shipyards quit.
However, this turned out not to be too bad: he put us in contact with two new people (Erik-Jan and Peter) who seem to be still quite interested in the project, although they asked to shift the interest a bit.
The idea now is that:
- We will focus on a ferry, rather than on a tug. Still hybrid, so who cares
- We will have a particular focus on adaptive control: the controller should be able to adapt automatically to new routes without external inputs (or, with as few external inputs as possible). Challenge!
Osmose test function
The question of the Osmose test function came up again, so I took finally action.
Now the test function also works for Mac, and is implemented in the master version of Osmose on Github. So, anyone who has access to Osmose can simply download it and use the tests, as well as add his/her own ones...yey!