Track / Overview

Transforming our energy systems to make them free of carbon emissions poses technological and scientific challenges. These hinge on aspects of production of electricity and its delivery to end users, consumption patterns and future scenarios, and electrification of transportation and heating. As well as socio-economic and environmental issues that need to be addressed for a rapid and sustainable transition. Many important problems in these areas are currently being studied across technical, social and environmental sciences. The application of machine learning (ML) and artificial intelligence (AI) spans these disciplines, offering powerful tools and methods.

Relevant topics include the demonstration of smart grid technologies by means of application cases and experiments that illustrate current consumption patterns and their spatial distribution. Modelling of production and consumption with future energy pricing concepts. Closer to the power plant and electric grid operators, themes such as the modelling of energy production at different time scales, and the optimization of system operation and life cycle. All of these topics are currently experiencing a large flow of data from smart meters, environmental monitoring systems, renewal energy power plants, electricity market, electrical grid monitoring systems, as well as physics based models. Thus, novel applications of state-of-the-art methods for supervised and unsupervised classification, time series modelling and forecasting, computer vision, optimization with feedback from data are expected. For the sake of strengthening analyses and results from new methods, comparisons are encouraged. Similarly, aspects of data and model selection, and hyperparameter tuning.

Transforming the way we produce and consume energy to reduce our environmental footprint, requires also social and economic aspects to be taken into account. Such aspects of the energy transition also benefit from a large influx of data. Thus, themes such as environmental benefits of electrification, management of energy needs at the city, cantonal, national or even inter-country level, and analyses and modelling of environmental data are important in this track. Because they can yield new valuable feedback to policy makers. They may also include a diversity of heterogeneous data, including time series, images, and unstructured data. 

Track / Schedule

Bringing intelligence in system health management of power systems: From system health monitoring to decision support

With Olga Fink

Data analytics and machine learning applications for a sustainable energy future

With Angelos Selviaridis

Monitoring, control, and digitalization of electrical distribution grids using automated and data-driven solutions

With Omid Mousavi

From Big Data to a Virtual Power Plant

With Florian Hochstrasser & Sabine Vincent

Digital Transformation in Renewables: Challenges and Opportunities

With Pramod Bangalore

ML for Energy: why academic models fail in the real world

With Emanuele Fabbiani

Poster Pitches

Break

Long-term heat load forecasts using hierarchical archetype modelling and hourly smart meter data

With Martin H. Kristensen

A Model Selection Approach for Time-Series Based Local Day-Ahead Electric Load Forecasting for Industrial Customers

With Gwendolin Wilke

A fast inference Machine Learning model to assess the rooftop solar photovoltaic diffusion

With Roberto Castello

Semi-deep learning with user feedback improving energy prediction models and improving user experience

With Jiufeng Shi

PID autotuning via two-stage safe Bayesian optimization for heat pump control: Simulation and experimental results

With Bratislav Svetozarevic

Wind Farm Dynamic Yield Optimization using Reinforcement Learning

With Giorgio Cortiana

Final Wrap Up

With Philipp Schütz

Track / Speakers

Philipp Schütz

Professor, HSLU

Omid Mousavi

R&D Director, DEPsys

Pramod Bangalore

Head of Research, Greenbyte AB

Angelos Selviaridis

Energy Systems Engineer, EKZ

Giorgio Cortiana

Head of Advanced Analytics, E.ON Digital Technology

Roberto Castello

Scientific Collaborator, EPFL

Jiufeng Shi

Data Scientist, Discovergy GmbH

Bratislav Svetozarevic

Postdoctoral Researcher, Empa

Emanuele Fabbiani

Head of Data Science @ xtream

Sabine Vincent

Data Scientist, tiko Energy Solutions AG

Martin H. Kristensen

Postdoctoral Researcher, Aarhus Municipality

Gwendolin Wilke

Lecturer, Project Manager, FHNW

Olga Fink

Professor, EPFL

Florian Hochstrasser

Data Scientist, tiko

Track / Co-organizers

Braulio Barahona

Researcher

Philipp Schütz

Professor, HSLU

Andreas Melillo

Research Associate, HSLU – Lucerne University of Applied Sciences and Arts

Patrick Meyer

Researcher, HSLU

AMLD EPFL 2020 / Tracks & talks

AI & Nutrition

Marinka Zitnik, Marcel Salathé, Fabio Mainardi, Tome Eftimov, Barbara Koroušić Seljak, Nives Ogrinc, Aleksandra Kovachev

13:30-17:00 January 282A

AI & Policy

Joanna Bryson, Sofia Olhede, Emanuele Baldacci, Sabrina Kirrane, Bruno Lepri, Dennis Diefenbach, Ioannis Kaloskampis, Benoît Otjacques, Steve MacFeely, Christina Corbane

13:30-17:30 January 272A

Challenge Track

Danny Lange, Sunil Mallya, Marcel Salathé, Florian Laurent, Erik Nygren, Sharada Mohanty, Parth Kothari, Navid Rekabsaz, Wilhelmina Welsch, Ewan Oglethorpe, Nicholas Jones, Gokula Krishnan, Jeremy Watson, Andrew Melnik

13:30-17:00 January 284A

AMLD / Global partners