A Decentralized State Estimation Method for Distribution Network

Abstract

A distribution network decentralized state estimation method, including the first stage, private data negotiation process: the physical layer divides the distribution network into sub-regions and tie-line regions, and processes local data from different measurement devices or users to draft Contract; the second stage, boundary contract formulation: use the abnormal consensus mechanism to detect abnormalities to determine whether the contract is valid, and use the voting mechanism to verify and prevent data tampering. The invention solves the contradiction between the observability of the distribution network and the privacy of user data. And while protecting privacy, only exchange and store boundary values, reducing transmission and storage requirements; at the same time, it is compatible with different measurement sampling rates, and is suitable for distribution networks equipped with multi-sampling cycle mixed measurement and communication equipment, achieving more flexibility.

Description

technical field [0001] The invention relates to the technical field of power grid state estimation, in particular to a decentralized state estimation method for a distribution network based on partition data negotiation and abnormal consensus mechanism. Background technique [0002] With the increasing scale of distribution network, the traditional centralized state estimation (State Estimation, SE) encountered increasing complexity and computational burden. At the same time, the measurement devices in the distribution network are limited, and the redundancy of real-time measurement data is low, often relying on key pseudo-measurement data. In recent years, the development of advanced measurement systems and new measurement devices such as synchrophasor measurement has been rapid, and various measurement cycle data will coexist for a long time. At present, most of the literature focuses on the processing of centralized SE for different measurement cycle data, which cannot e...

AI Technical Summary

Problems solved by technology

At present, most of the literature focuses on the processing of centralized SE for different measurement cycle data, which cannot effectively deal with the problem of different measurement cycle data in different sub-areas.

Moreover, the centralized SE needs to collect every piece of measurement data from the global measurement device, so this poses a challenge to real-time performance

In addition to the complexity and computational burden faced by large-scale power systems with more nodes, traditional centralized SE encounters bad data, which will lead to global estimation errors or convergence failures

Moreover, all current state estimation methods assume that the data of the power company is public, and do not consider data transactions or exchanges under privacy protection.

Images