




Price of privacy in smartgrid 
Student(en):

Betreuer:
Christos Dimitrakakis, Maryam Kamgarpour 
Beschreibung:
Introduction
A premise of smartgird is a more secure and efficient power system. An enabling factor is realtime pricing, where the consumers adjust their consumption based on (near) realtime electricity prices. Since electricity prices reflect the balance of supply and demand, realtime pricing is hoped to help in stabilizing grid, while saving francs on consumers’ bills. In this project, we aim to design pricing and consumption mechanisms that ensure consumers' data remain private while achieving benefits of realtime pricing.
Project Description
We consider a scenario in which an aggregation of consumers communicates with a central utility. Each individual consumer shares its realtime consumption patterns with the central utility. The central utility broadcasts the price of electricity in a given planning horizon, according to a price model. Within this setup we address the following questions. First, by observing the public variable, the price, what can an intruder learn about an individual consumer’s private data such as occupancy and sleep pattern, given partial information on patterns of consumption. Second, how can the central utility modify the pricing scheme to ensure the private data such as occupancy is preserved, while ensuring the benefits of realtime pricing. Depending on the progress, we address the question of how could individual consumers use thermal storage or batteries to hide their private data, such as occupancy patterns.
Required background
courses in probability theory, linear dynamical systems, stochastic control, dynamic programming.
Acquired Skills
application of probability theory, Markov chains, information theory and dynamical system theory to a relevant problem in smart grid.
The project requires strong background and grades in mathematics. If you are interested in the project, make sure to apply and include your uptodate transcript of grades for Bachelor's and Master's studies.
Weitere Informationen

Professor:
Maryam Kamgarpour

Projektcharakteristik:
Typ:
Art der Arbeit: Masters/Semester
Voraussetzungen: strong background in mathematics, control, optimization and probability.

Anzahl StudentInnen:
Status: open

Projektstart: Semester: 
