2018/03/17 by Deepjyoti Deka, Michael Chertkov, Deka, Deepjyoti +3 · 1 citation
Computer Science · #Blind Source Separation Techniques #FOS: Computer and information sciences #FOS: Electrical engineering #FOS: Mathematics #Image and Signal Denoising Methods #Machine Learning (stat.ML) #Neural Networks and Applications #Optimization and Control (math.OC) #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1803.06531
openalex publication_date 2018/03/17 · openalex created_date 2022/08/08 · openalex updated_date 2026/07/28
Distribution grid is the medium and low voltage part of a large power system.\nStructurally, the majority of distribution networks operate radially, such that\nenergized lines form a collection of trees, i.e. forest, with a substation\nbeing at the root of any tree. The operational topology/forest may change from\ntime to time, however tracking these changes, even though important for the\ndistribution grid operation and control, is hindered by limited real-time\nmonitoring. This paper develops a learning framework to reconstruct radial\noperational structure of the distribution grid from synchronized voltage\nmeasurements in the grid subject to the exogenous fluctuations in nodal power\nconsumption. To detect operational lines our learning algorithm uses\nconditional independence tests for continuous random variables that is\napplicable to a wide class of probability distributions of the nodal\nconsumption and Gaussian injections in particular. Moreover, our algorithm\napplies to the practical case of unbalanced three-phase power flow. Algorithm\nperformance is validated on AC power flow simulations over IEEE distribution\ngrid test cases.\n