Nowadays, the false data injection attack (FDIA), which can bring inestimable losses to smart grids, has become one of the most threatening cyber attacks in cyber physical systems. Previous studies for false data detection focused on state estimation, which require a huge computational overhead at the control center. In this paper, we propose a confidence-aware collaborative detection mechanism for false data attacks, which is a fast and lightweight scheme. Firstly, we propose a trust-based compromised PMU identification method, in order to identify malicious PMUs by monitoring behaviors of PMUs in a cycle. Secondly, we propose a voting-based detection method based on physical rules, in order to detect FDIA collaboratively. This method improves the detection rate while reducing the computational cost at control center. We also make extensive experiments on real-time data that are collected from the PowerWorld simulator. The experimental results show the efficiency and effectiveness of our proposed mechanism and methods.

如今，错误数据注入攻击（FDIA）可能给智能电网带来不可估量的损失，已成为网络物理系统中最具威胁性的网络攻击之一。以前的错误数据检测研究集中在状态估计上，这需要控制中心的大量计算开销。在本文中，我们提出了一种用于虚假数据攻击的置信度感知协作检测机制，该机制是一种快速且轻量级的方案。首先，我们提出了一种基于信任的受损PMU识别方法，以通过监视一个周期内的PMU行为来识别恶意PMU。其次，提出一种基于物理规则的基于投票的检测方法，以协同检测FDIA。该方法提高了检测率，同时降低了控制中心的计算成本。我们还对从PowerWorld模拟器收集的实时数据进行了广泛的实验。实验结果表明了我们提出的机制和方法的有效性和有效性。