Lightweight Defense Against Data Consistency Attacks in Distributed DC Optimal Power Flow

Distributed DC optimal power flow (DC-OPF) is increasingly adopted in multi-region power systems to scale power-flow optimization across regions while preserving local data privacy. However, this multi-agent architecture also creates new attack surfaces, where Byzantine or data-consistency adversaries can tamper with exchanged messages and quietly steer the system toward unfair and inefficient operating points. Fully distributed DC-OPF mitigates such threats with blockchain, but its peer-to-peer (P2P) messaging incurs $O(n<^>{2})$ communication complexity and prohibitive consensus latency, making it unsuitable for iterative optimization. Coordinator-based DC-OPF avoids blockchain but typically relies on P2P cross-verification for data consistency, which again results in quadratic overhead. To ensure data integrity, freshness, and consistency in distributed DC-OPF while retaining $O(n)$ communication complexity for normal operations, this paper proposes a lightweight security framework that can perfectly detect Byzantine manipulation without any P2P cross-checking. The method integrates a modified Schnorr protocol over elliptic curve cryptography into a coordinator-based architecture, enabling secure and efficient consensus at each iteration. Simulations on the IEEE 118-bus system show that, for a 10-region power system, the proposed secure DC-OPF converges in only 45 iterations and 21 s, whereas the blockchain-based alternative requires 439 iterations and 36.6 min. By combining robust security with low computational and bandwidth overhead, the framework provides a scalable and practical solution for real-time secure distributed DC-OPF. Note to Practitioners-Modern cyber-physical power systems increasingly rely on distributed DC optimal power flow (DC-OPF) to coordinate multiple control areas and clear electricity markets in real time. In these settings, economic efficiency and fairness depend on the integrity of the iterative messages that determine generator dispatch and locational marginal prices (LMPs). A compromised insider or man-in-the-middle adversary can quietly send inconsistent signals to different regions, pushing the system away from cost-minimizing dispatch, distorting LMPs, and shifting profits between market participants, even though the optimization appears to converge and all physical constraints are met. This work proposes a lightweight security layer that can be integrated on top of an existing coordinator-based DC-OPF or market-clearing system. Using digital signature aggregation, agents collectively verify at each iteration that they have received the same global update and that messages are fresh and authentic. When a discrepancy is detected, the framework triggers a simple recovery phase based on majority voting over received updates. From a market economics perspective, this helps preserve socially optimal dispatch and prices, mitigates unfair profit reallocation, and reduces the risk that subtle cyber manipulation leads to inefficient global variables. The protocol is implemented as middleware between the optimization engine and the communication layer, requiring no changes to the underlying DC-OPF formulation or pricing rules. One limitation of the proposed framework is its reliance on a single coordinator; future work will explore designs with multiple coordinators.