City Research Online

Abstract

The growing demand for electricity and the increasing complexity of energy systems
have necessitated innovative approaches to efficient, secure, and sustainable energy management. Energy systems are undergoing a transformative shift driven by Smart Grid technologies that integrate renewable energy sources and distributed energy sources, as well as advanced data-driven technologies. These innovations aim to enhance energy management, reduce environmental impact, and empower consumers as active participants in energy markets. However, traditional energy systems face challenges such as inefficiencies in pricing and energy trading, centralization risks, and data privacy concerns. Recent research highlights the limitations of centralized systems, emphasizing the need for secure, scalable, and user-centered approaches that preserve data privacy while enabling efficient energy management. Emerging technologies, such as blockchain, differential privacy, and federated learning, offer promising solutions to address these challenges. This work focuses on innovative pricing models for energy trading, advanced privacy-preserving techniques for data-sharing, and secure collaborative frameworks for energy demand forecasting to enhance the functionality, security, and equity of modern energy systems. To
this end, several contributions are presented.

The first contribution is a novel dynamic pricing model tailored for a microgrid of prosumers with photovoltaic panels. The proposed model introduces mathematical frameworks for determining equilibrium prices based on supply-demand ratios and incorporates mechanisms for calculating energy usage costs, profits, and penalties for participants who deviate from predicted energy profiles. The effectiveness of the model is validated using real-world energy profiles, showcasing its potential to reduce energy costs.

The second contribution is a blockchain-based data-aggregation scheme for a microgrid of prosumers. This scheme ensures prosumers’ privacy by concealing their real energy usage data, thereby mitigating risks like eavesdropping and man-in-the-middle cyber attacks. The proposed scheme utilizes on-chain and off-chain techniques to protect user privacy while efficiently reducing the blockchain size. The results show that the proposed technique achieves high accuracy in data aggregation while ensuring user privacy.

The third contribution is a user-centric data-sharing scheme that leverages local differential privacy techniques to preserve the privacy of end-users. The proposed scheme provides the end-user with control over the utility of their data, with the level of privacy being calculated from individual utility preferences. The results show that the proposed scheme allows keeping the utility within the boundaries defined by the end-user, while providing the maximum possible level of privacy.

Lastly, a privacy-preserving federated learning framework incorporating local differential
privacy is designed. This framework enables several parties to collaboratively train a
central machine learning model without sharing their private datasets. Compared to
the state-of-the-art schemes that propose a fixed privacy setting based on a number of
simulations, in the proposed scheme, the privacy configuration changes over time to
match the pattern of the desired level of accuracy of the central model.

Publication Type:	Thesis (Doctoral)
Subjects:	T Technology T Technology > TA Engineering (General). Civil engineering (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering
Departments:	School of Science & Technology > Department of Engineering School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads