Research On Distributed State Estimation In Distribution Systems With Distributed Generation

The distribution network is directly connected with customers,which is the key link to ensure the quality of power supply and of user service,as well as improve the operation efficiency of the power system.However,the distribution network has the inherent properties of large scale,massive nodes,and three-phase imbalance.Due to economic cost and other limits,it will be in a situation of insufficient observability with heterogeneous measurements as well as uncertainties in many aspects for a long time.Moreover,with the accelerating development of distribution networks,the importance and urgency of distribution network state estimation are becoming increasingly obvious.Therefore,accurate situational awareness through state estimation is of great significance to the safe,reliable and economic operation of distribution network with high penetration of renewable energies.This paper focuses on distributed state estimation in distribution systems with distributed generation.The specific research work is as follows.1)It introduces the research background and significance of the subject.It summarizes the current research in the multi-area distribution system state estimation,the power system partition methods,and the interval state estimation and then analyzes the shortcomings of existing methods,providing a basis for the following work in this paper.2)Based on the electrical coupling strength and size-uniformity of subareas,the nonoverlapping and node-overlapping partition methods for large-scale three-phase unbalanced distribution networks are proposed respectively.On the first,the equivalent electrical distance is proposed as a quantitative index of the electrical coupling relationship.On the second,the distribution network is partitioned into several high-inter-cohesive sub-areas with similar scales based on the point/edge division community discovery theory,greedy algorithm and tabu search algorithm.3)Consider the coordinate tensions and time synchronization compatibility problem under a mixed measurement environment,and a distributed state estimation framework in three-phase unbalanced distribution systems is proposed on the basis of convergence consistency,including SCADA measurement delay estimation and distributed state estimation algorithms two parts.The two data compatibility problems under mixed measurement are alleviated by measurement transformation and the introduction of time-delay error in the measurement equation.4)Considering the uncertainties of distributed generator output and measurement errors,a distributed interval state estimation model in distribution systems is constructed based on the unknown but bounded principle.The interval measurement transformation and the introduction of time-delay errors are applied to mitigate compatibility issues under mixed measurements.The proposed model includes two parts: local interval state estimation based on interval algorithms and system coordination by communication of neighboring subareas.