Privacy-preserving and secure market-based operation of distributed energy systems

Abstract

Due to significant shifts in energy generation and demand patterns, contemporary power systems encounter formidable power imbalance issues. Addressing ramifications stemming from uncertain energy generation and demand, market-based mechanisms such as real-time peer-to-peer (P2P) energy trading are proposed. Nevertheless, the escalating coupling of power systems with communication networks presents substantial hurdles for the market-based operation of distributed energy systems. Within the market, multiple agents are geographically dispersed and interconnected via power lines. Ensuring the normal operation of the market necessitates the guarantee of physical network security. Additionally, safeguarding the personal interests of diverse agents requires privacy preservation. Moreover, given the potentially extensive data shared in the market, prioritizing data security becomes imperative. This thesis investigates privacy-preserving and secure market-based operation strategies to balance highly uncertain energy generation and demand.

1) First, we formulate the real-time P2P energy trading problem as a spatial-temporally constrained stochastic optimization problem by considering ESS and the spatial network constraints to ensure the physical network security. To deal with the uncertainties online, a tailored Lyapunov optimization method is proposed to approximately reformulate the stochastic optimization problem into an online one by relaxing the time-coupling constraints, underpinned by theoretical performance guarantees. This decomposition enables closed-form solutions for the sub-problems, contributing to a substantial increase in computational efficiency.

2) Second, we endeavor to further mitigate privacy leakage issues that emerge in the realistic fully distributed P2P energy trading. We developed a secure multi-party computation mechanism consisting of offline and online phases to ensure the security of shared data by leveraging the tailored secret sharing method. In addition, the customized Paillier encryption method is proposed for both the secure two-party computation and the offline phase of the multi-party computation. The random encryption coefficient is designed to enhance the security of the two-party computation and simultaneously guarantee the convergence of the distributed optimization. The feasible range for the encryption coefficient is derived with a strict mathematical proof.

3) Third, we dedicate to bolster the data security within the distributed P2P energy trading. To investigate malicious impacts of Byzantine agents, we formulate the distributed P2P energy trading problem by considering high-fidelity physical network constraints. Subsequently, to thwart Byzantine faults, an online spatial-temporal anomaly detection approach with closed-form solutions for updating model parameters is proposed by leveraging the domain knowledge-informed tensor learning. Theoretical conditions for the distributed optimization with online anomaly detection mechanisms are obtained to ensure the convergence and optimality.

4) Finally, we aim to address privacy leakage concerns from the non-convex synergy problem of hierarchical data center penetrated power networks in the day-ahead market. The synergy problem is formulated as a mixed integer quadratically constrained quadratic programming considering both communication and energy conservation. To mitigate impacts of the highly non-convex nature, the normalized multi-parametric disaggregation technique is leveraged to reformulate the problem into a mixed integer non-linear programming. To further overcome non-smoothness of the reformulated problem, we propose the customized $\ell_1-$surrogate Lagrangian relaxation method with convergence guarantees to solve the problem in a distributed privacy-preserving manner.