EES-UETP description upcoming courses previous courses templates about

2018-04: Advanced Data Analytics for Energy Systems
September 3rd-5th, 2018


INESC Technology and Science (INESC TEC)

Local Organizer Committee

Ricardo Bessa received his Licenciado (five-year) degree from the Faculty of Engineering of the University of Porto, Portugal (FEUP) in 2006 in Electrical and Computer Engineering. In 2008, he received the M.Sc. degree in Data Analysis and Decision Support Systems on the Faculty of Economics of the University of Porto (FEP). He obtained his Ph.D. degree in the Doctoral Program in Sustainable Energy Systems (MIT Portugal) at FEUP in 2013. Currently, he is a Senior Researcher and Area Manager at INESC TEC in its Center for Power and Energy Systems and technical coordinator of the Horizon 2020 project InteGrid. He worked in several international projects such as the European Projects FP6, FP7 SuSTAINABLE, FP7 EvolvDSO, Horizon 2020 SENSIBLE and UPGRID and an international collaboration with Argonne National Laboratory for the U.S. Department of Energy. He is co-author of 27 journal papers, 63 conference papers and 4 book chapters about renewable energy forecasting, power system operation and electricity markets.

Description of the course

The technological revolution in the electric power system sector is producing large volumes of data with relevant impact in the business and functional processes of system operators, energy utilities and grid users. This course aims to cover different theoretical and practical aspects of data analytics in energy systems, according to the following viewpoint:
(1) The future generation of big data functions will combine spatial-temporal information and distributed learning techniques that exploit recent advances in high performance and distributed computing.
(2) The output should be probabilistic (uncertainty) information and with high value for integration in decision-aid methods under risk.
(3) Deep learning techniques represent an added value for automatic feature extraction and reduction, but manual feature engineering using domain (expert) knowledge cannot be abandoned.
(4) Machine learning algorithms can be used to control grid assets, for instance embedded in reinforcement learning techniques or to create proxy models for complex physical systems.
(5) Creation of new business models for knowledge extraction from data is also expected in a near future. Some examples are data analytics for consumer engagement in demand response, big data pre-processing from grid sensors, electricity markets modelling and predictive maintenance of electrical assets.

Contents and Schedule

Day 1 (3rd September 2018)

(9h15-9h30) Course Opening (Ricardo Bessa, INESC TEC)
(9h30-10h30) Data streams and online learning (João Gama, INESC TEC/FEP)
(10h30-11h00) Coffee-break
(11h00-12h30) Data streams and online learning (João Gama, INESC TEC/FEP)
(12h30-14h00) Lunch
(14h00-16h30) Statistical learning for uncertainty forecast (Jethro Browell, University of Strathclyde)
(16h30-17h00) Coffee-break
(17h00-18h00) Feature engineering to improve time series forecasting (Ricardo Bessa, INESC TEC)

Day 2 (4th September 2018)

(9h30-10h30) Reinforcement learning for data-driven optimization (Damien Ernst, University of Liège)
(10h30-11h00) Coffee-break
(11h00-12h00) Reinforcement learning for data-driven optimization (Damien Ernst, University of Liège)
(12h00-13h15) Lunch
(13h15-14h15) Introduction to deep learning (Stefan Leijnen, Amsterdam University of Applied Sciences)
(14h15-15h15) Implementation of deep learning with TensorFlow (Stefan Leijnen, Amsterdam University of Applied Sciences)
(15h15-16h15) Decision-making under risk (Manuel Matos, INESC TEC/FEUP)
(16h15-16h30) Coffee-break
(16h30-18h00) Data analytics for asset management (Bruce Stephen, University of Strathclyde)

Day 3 (5th September 2018)

(9h00-10h30) Data analytics in transmission system operators (Miguel Moreira da Silva, REN)
(10h30-11h00) Coffee-break
(11h00-12h30) Big data analytics for electrical utilities (Pedro Ferreira, EDP Inovação)
(12h30-13h30) Lunch
(13h30-15h00) Data mining for modelling electricity markets (José Villar Collado, INESC TEC/University of Comillas)
(15h00-16h30) Consumer engagement with big data techniques (Vassilis Nikolopoulos, Intelen)


Bruce Stephen, University of Strathclyde, Scotland
Damien Ernst, University of Liège, Belgium
Jethro Browell, University of Strathclyde, Scotland
João Gama, INESC TEC/FEP, Portugal
José Villar Collado, INESC TEC/University of Comillas, Portugal/Spain
Manuel Matos, INESC TEC/FEUP, Portugal
Miguel Moreira da Silva, REN, Portugal
Pedro Ferreira, EDP Inovação, Portugal
Ricardo Bessa, INESC TEC, Portugal
Stefan Leijnen, Amsterdam University of Applied Sciences, Netherlands
Vassilis Nikolopoulos, Intelen, U.S.A.


The course will take place at INESC TEC buildings, situated in the Campus of the Faculty of Engineering of the University of Porto, in Porto, Portugal


We will issue EES-UETP verified certificates for participation and the course corresponds to a workload of 2 ECTS credits.


Course fees Course fees will include lectures, course aids (lectures on pen drive, leaflets, brochures,), coffee breaks, three lunches and a course dinner in a restaurant.
An invoice will be given to each registered participant during the Course.
Payments are requested before the beginning of the Course.

To register, please fill in the Form.

Accommodation and further information

Please see the Leaflet