.png){kind=icon}
.png){kind=icon}
Return
Day-Ahead Pricing Strategy for Virtual Power Plants Considering Privacy Protection of User Data
.png){kind=icon}
OA
AI
B
P
W
G
DOI：10.35833/mpce.2025.000971.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
As an advanced aggregation and control technology for multi-type source-load resources, virtual power plant (VPP) enables intelligent interconnection and coordinated optimization of distributed load resources. This paper proposes a day-ahead pricing strategy for VPPs considering privacy protection of user data. First, a day-ahead pricing model for VPP is constructed, which considers historical demand response (DR) contributions of users to enhance fairness in the pricing process. The asynchronous advantage actor-critic (A3C) algorithm is adopted to solve this model and optimize pricing decisions. Furthermore, an encrypted horizontal federated learning (HFL) method is proposed for model training, which is based on improved weighted federated averaging algorithm and aims to mitigate data silos and protect privacy of user data across different regions. Instead of directly transmitting user data during training, partially homomorphic encryption (PHE) scheme is applied to encrypt exchanged model parameters, thereby reducing the risk of privacy leakage. Case studies validate that the proposed pricing strategy can enhance the fairness of multi-user regulation decisions and improve the performance of VPPs in day-ahead DR while ensuring the security of user data.
Keywords:
Virtual power plant (VPP)
demand response (DR)
day-ahead pricing
privacy protection
horizontal federated learning (HFL)
partially homomorphic encryption (PHE)
asynchronous advantage actor-critic (A3C)
Journal
IF:
6.1
Papers:
1.6K
Citations:
6.0K
Organization
Cited Papers