Integrating Building Thermal Flexibility Into Distribution System: A Privacy-Preserved Dispatch Approach
Shuai Lu, Zeyin Hou, Wei Gu, Yijun Xu
TL;DR
The paper addresses privacy concerns in incorporating building thermal flexibility into distribution-system dispatch by proposing a privacy-preserved centralized optimization framework. It models BLAs with an aggregate linear dynamic model (ATDM) and derives a compact, matrix-form dispatch formulation that couples BLA controls with grid variables. A novel privacy-preserved algorithm is introduced, combining transformation-based encryption (TE-I/TE-II) with constraint relaxation and extension (CRT and CET) to securely mask private parameters and states while preserving optimality, with theoretical privacy guarantees against semi-honest adversaries and eavesdroppers. Numerical tests on IEEE test systems demonstrate that the PPCC method maintains solution accuracy and cost performance while achieving strong privacy protection, with only modest increases in computation due to additional slack and constraint handling. The work highlights the practical viability of privacy-preserving, building-aware dispatch and outlines future directions in privacy-aware allocation, cybersecurity, and occupant-behavior uncertainty handling.
Abstract
The inherent thermal storage capacity of buildings brings considerable thermal flexibility to the heating/cooling loads, which are promising demand response resources for power systems. It is widely believed that integrating the thermal flexibility of buildings into the distribution system can improve the operating economy and reliability of the system. However, the private information of the buildings needs to be transferred to the distribution system operator (DSO) to achieve a coordinated optimization, bringing serious privacy concerns to users. Given this issue, we propose a novel privacy-preserved optimal dispatch approach for the distribution system incorporating buildings. Using it, the DSO can exploit the thermal flexibility of buildings without accessing their private information, such as model parameters and indoor temperature profiles. Specifically, we first develop an optimal dispatch model for the distribution system integrating buildings, which can be extended to other storage-like flexibility resources. Second, we reveal that the privacy-preserved integration of buildings is a joint privacy preservation problem for both parameters and state variables and then design a privacy-preserved algorithm based on transformation-based encryption, constraint relaxation, and constraint extension techniques. Besides, we implement a detailed privacy analysis for the proposed method, considering both semi-honest adversaries and external eavesdroppers. Case studies demonstrate the accuracy, privacy-preserved performance, and computational efficiency of the proposed method.