MA: Renewable-Aware Edge, Fog and Cloud Computing
This topic area covers a variety of different thesis topics in the realm of green computing, especially renewable-aware edge, fog and cloud computing. Please contact us for concrete information on current research topics and ideas for your thesis.
Two major topic areas for potential bachelor and master theses are:
1. Carbon-aware load shifting
Since the carbon intensity (emissions per energy) of electricity varies at different times and locations depending on demand and energy sources, so does the environmental impact of consuming this electricity. For example, the German carbon intensity is usually above 400 gCO2/kWh but can drop below 100 gCO2/kWh when there is a lot of sun and wind. Meanwhile carbon intensity values and fluctuation in countries such as Norway is a lot lower, since it relies mainly on hydropower. Get an intuition here .
The idea behind carbon-aware load shifting is to move computational load towards times and locations such that we avoid consuming dirty energy. To do so students can use a variety of methods such as formulating and solving of optimization problems, time series forecasting for demand and supply of green energy, etc. Created methods are expected to be evaluated experimentally, e.g. using simulation .
2. Synergies between distributed computing and distributed power generation
Future application domains such as the Internet of Things (IoT) demand highly distributed data processing close to the edge of the network. At the same time, energy generation is becoming more and more distributed too, due to the adoption of renewable sources such as roof installations of photovoltaic panels. Can we identify and exploit synergies in these two areas in order to reduce their carbon footprint and stabilize the power grid?
- Can we use weather and electricity demand forecasts to manage resources IoT, fog, and edge computing environments in a way that maximizes their usage of renewable energy?
- In which scenarios is it feasible to place computation tasks close to renewable energy sources in order to consume energy right when and where it is produced and, hence, reduce variability in the power grid?
- Under which circumstances is it ecologically and economically sensible to directly equip IoT devices or edge computing nodes with on-site power generation such as PV panels?
- What is the role of energy storage in such systems? To which level does it make sense to strive for energy-autonomous nodes, considering battery aging caused by frequent charge/discharge cycles? Can we schedule load in a way to prolong the lifespan of such batteries?
- How feasible and reliable are such systems under real-life conditions?
Prerequisites for working on these topics are advanced programming and data analysis skills for implementing and evaluating the methods and scenarios. We are mainly working with simulation and co-simulation, so previous knowledge with these technologies is an advantage. Further skills depend on your concrete topic and can be adjusted individually.
Thesis can be written in either German or English language.