8 March 2019 from 15:00 to 17:00

DSCC Central Topic Seminar #9: Machine Learning Applications in Forecasting

The Data Science & Complexity Centre (DSCC) Central Topic Seminars are a series of seminars co-organized by the Utrecht Applied Data Science, the Utrecht Bioinformatics Center, and the Centre for Complex Systems Studies. It will consist of tutorials, excursions, software training and specialist lectures. We aim to expose the central topic "Machine Learning" for the broad community within and outside Utrecht University and bring the researchers from different backgrounds together.

In the first hour, Dr. Markus Abel, the CEO of the 4cast (dedicated to generate power production forecasts at the highest level of precision imaginable), will give a specialist lecture on Machine Learning Applications in Forecasting titled "Building a machine learning engine for the forecast of wind energy production".

Abstract: The prognosis of energy production from renewable sources is important to sustain the energy supply to commercial and private consumers. The most prominent stakeholders are TSOs (transmission system operators) and traders of energy. Most often, the forecasts depend on weather prediction and production data. In this talk, I will present the setup and algorithms used in our 4Cast system for the prediction of wind power. Details on the underlying IT will be given together with a discussion of algorithms. I will sketch the system components explain differences in physical, data-based and mixed approaches. Eventually, an example how to build a model is shown and shortcomings as well as success is discussed.

In the second hour, Dr. Remco Verzijlbergh, the Founder & Director of Operations of the Whiffle (produces fine-scale weather forecasts and weather simulations by using cutting-edge computing technology), will give a specialist lecture on Machine Learning Applications in Forecasting titled "Machine learning applications for renewable energy forecasting".

Abstract: The integration of weather driven renewable energy sources (RES) in the energy system poses great societal and scientific challenges. At the core of these challenges lies the very nature of renewable sources: their output is variable and difficult to predict. Forecasting of renewable energy production is therefore essential for a secure and cost-efficient operation of the power system. In this talk we discuss the role of machine learning in the wide spectrum of approaches that is being used to tackle the RES forecasting challenge. We touch upon the fundamentals of the numerical weather prediction problem and the role of machine learning in this context. We contrast the data based approaches with the ones based on physical modeling. Day-to-day examples of wind energy forecasting will be used to illustrate the challenges.

Both students and staff are welcome.

DSCC Central Topic Seminars:

Some of the Seminars are available to watch via the CCSS YouTube channel.

Venue: Room 2.01, Minnaert Building, Leuvenlaan 4, De Uithof, Utrecht.

Please register before Thursday 7 March 2019.

Start date and time
8 March 2019 15:00
End date and time
8 March 2019 17:00