Research Abstract

Many societies want to achieve an energy production without any fossil fuels. However, we are not yet able to incorporate all the potential of renewable energies and thus still heavily rely on conventional energy producers.

There are two ways to optimise energy production. On the one hand, we can tune the production schedule to follow demand better. On the other hand, we can change the consumption behaviour to support an optimal supply strategy. Traditionally, demand forecasts are the basis for production schedules. However, with an increasing and changing amount of variables influencing the system, perfect predictions seem unrealistic. This thesis considers a flexible demand side as an opportunity to optimise power production for an energy grid.

Research interests

Data Mining
Bayesian Statistics
Feature Selection
Forecasting

Publications

How much demand side flexibility do we need? - Analyzing where to exploit flexibility in industrial processes [in press].
Barth, L.; Hagenmeyer, V.; Ludwig, N.; Wagner, D.
2018. 9th ACM International Conference on Future Energy Systems (ACM e-Energy), 12th - 15th June 2018, Karlsruhe, Germany, ACM, New York

Concept and benchmark results for Big Data energy forecasting based on Apache Spark.
González Ordiano, J. Á.; Bartschat, A.; Ludwig, N.; Braun, E.; Waczowicz, S.; Renkamp, N.; Peter, N.; Düpmeier, C.; Mikut, R.; Hagenmeyer, V.
2018. Journal of Big Data, 5 (1), Art.Nr. 11. doi:10.1186/s40537-018-0119-6

Mining Flexibility Patterns in Energy Time - Series from Industrial Processes.
Ludwig, N.; Waczowicz, S.; Mikut, R.; Hagenmeyer, V.
2017. Proceedings. 27. Workshop Computational Intelligence, Dortmund, 23. - 24. November 2017. Hrsg.: F. Hoffmann, 13-32, KIT Scientific Publishing, Karlsruhe

A comprehensive modelling framework for demand side flexibility in smart grids.
Barth, L.; Ludwig, N.; Mengelkamp, E.; Staudt, P.
2018. Computer science - research and development, 33 (1-2), 13-23. doi:10.1007/s00450-017-0343-x

Towards coding strategies for forecasting-based scheduling in smart grids and the energy lab 2.0.
Jakob, W.; Gonzalez-Ordiano, J. Á.; Ludwig, N.; Mikut, R.; Hagenmeyer, V.
2017. GECCO '17 : Proceedings of the Genetic and Evolutionary Computation Conference Companion, Berlin, Germany, 15th - 19th July 2017, 1271-1278, Association for Computing Machinery, New York (NY). doi:10.1145/3067695.3082481

Lastprognosen für das Fernwärmenetz der Stadt Karlsruhe.
Mikut, R.; Renkamp, N.; Gonzalez Ordiano, J. A.; Ludwig, N.; Schlagel, T.; Waczowicz, S.; Hagenmeyer, V.; Rink, M.; Iser, A.; Hitzel, J.
2016. Umwelt - Energie - Stadt : 4.Workshop der Stadt Karlsruhe mit den KIT-Zentren 'Klima und Umwelt' sowie 'Energie', Karlsruhe, 29.November 2016

Time Series Analysis for Big Data: Evaluating Bayesian Structural Time Series Using Electricity Prices.
Ludwig, N.; Feuerriegel, S.; Neumann, D.
2016). In Volker Nissen, Dirk Stelzer, Steffen Straßburger, and Daniel Fischer, editors, Multikonferenz Wirtschaftsinformatik (MKWI) 2016, volume III, pages 1569--1580. Universitätsverlag Ilmenau, Ilmenau, 2016.

Putting Big Data analytics to work: Feature selection for forecasting electricity prices using the LASSO and random forests.
Ludwig, N.; Feuerriegel, S.; Neumann, D.
2015. Journal of Decision Systems, 24(1):19--36, 2015. [DOI ]