基于分区数据协商与异常共识机制的配电网去中心化状态估计
Decentralized State Estimation in Distribution Networks Based on Multiarea Data Negotiation and Anomaly Consensus Mechanism
-
摘要: 针对配电网规模不断扩大和多种量测装置并存的现状, 提出一种更加灵活的配电网去中心化状态估计方法。该方法主要分为两阶段。第一阶段为私密数据协商:底层(物理层)将配电网划分为子区域和联络线区域, 对来自不同量测设备或不同用户的本地数据进行处理从而草拟合同。第二阶段为边界合同制定:利用异常共识机制检测异常并判断合同是否生效, 同时采用投票机制来验证合同的有效性并防止数据篡改。无异常的合同将形成区块并上链。经过校验的边界估计值将作为新的伪量测数据用来提升子区状态估计的效果。仿真结果表明, 所提算法在保证分区隐私条件下, 能够提高状态估计精度, 有效抑制坏数据影响, 并且适用于具有不同采样周期的混合量测配置场景。Abstract: This paper proposed a flexible decentralized distribution network state estimation method in view of the increasing scale of the distribution network and the coexistence of hybrid measurement devices. The method was divided into two stages. The first stage was the confidential data negotiation: in the underlying layer (physical layer), the distribution network was divided into subareas and tie-line areas and then local data from different measurement equipment or users was processed to draft contracts. The second stage was the boundary contract settlement, which utilizes anomaly consensus mechanism to determine whether the contract is effective. The voting mechanism was adopted to verify the new block and prevent data tampering. An effective contract would generate block and be chained in existing blockchain. The verified boundary estimations were used as new pseudo- measurements to enhance the estimation accuracy of subareas. Simulation results show that the proposed method improves the estimation accuracy with ensuring subarea privacy, effectively suppress the influence of bad data, and can be applied to hybrid measurements with different measurement rates.