OBSERVABILITY ANALYSIS AND OPTIMAL METER PLACEMENT IN ACTIVE DISTRIBUTION NETWORKS WITH DISTRIBUTED GENERATION

摘要

The increasing penetration of Distributed Generation (DG) is transforming the distribution networks from passive to active conditions. In this paper, combined with the characteristics of the active distribution network (ADN) with DG, the observability analysis method based on meter placement was employed. Each consumer is classified, and the DG is considered as a special negative load with constant power, which contains two types: photovoltaic (PV) power and wind turbine generation (WTG). The load variation is represented using typical load patterns, and the DG output uncertainty is handled by Latin hypercube sampling (LHS) technique. An efficient graph partitioning method is proposed to partition the weighted tree in the performing meter placement to achieve the observability. In addition, an optimal meter placement algorithm is presented for improving state estimation (SE) accuracy and minimizing the number of measurements, in which analytic hierarchy process (AHP) is employed to determine the weight of observability degree and economy. Finally, test results based on the 14 node system and IEEE 33-bus distribution network verify the efficiency and feasibility of the proposal.