Machine learning based demand side management in residential buildings

↵

Background

In order to react to the fluctuating electricity generation by renewable energy sources, flexible electrical loads are necessary. Especially, heat pumps and electric vehicles can react to the green energy of photovoltaic systems and wind turbines in residential buildings. Thus they can help to decarbonize the buildings sector while stabilizing the electricity grid. To use flexible loads intelligent control strategies are essential. Neural networks can learn such complex control strategies based on synthetically generated data and apply the learned strategies in real world settings. The objective of this master thesis is the development of machine learning based approaches for optimal heating system operation.

Tasks of the thesis

• Become familiar with an existing building energy system simulation environment (Python)

• Design of the experimental procedure

• Development of supervised machine learning models for optimal heating technology control strategies

• Multi-criteria optimal operation w.r.t. costs, CO2-emissions and system impact

Requirements

• Interest in the field of machine learning

• Interest in demand side management and smart grids

• Programming skills (first experience with Python is recommended)

• Responsible and motivated working attitude

• Good English or German language skills

Formal aspects

• Begin: from now on or as you wish (duration: 6 months)

• Language: English or German

Application

• Transcript of records of your study programs and a CV

Contact

M. Sc. Max Kleinebrahm or Dr.-Ing. Thomas Dengiz
E-Mail: max.kleinebrahm∂kit.edu or thomas.dengiz∂kit.edu

Upon successful completion of the Master's thesis, there is an opportunity to further refine the developed algorithms as part of a PhD.