2018/06/15 by Roel Dobbe, Dobbe, Roel, Werner van Westering +9 · 1 citation
Computer Science · Engineering · #Data Analysis #Energy Load and Power Forecasting #FOS: Electrical engineering #FOS: Mathematics #FOS: Physical sciences #Image and Signal Denoising Methods #Optimization and Control (math.OC) #Statistics and Probability (physics.data-an) #Systems and Control (eess.SY) #Traffic Prediction and Management Techniques #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1806.06024
openalex publication_date 2018/06/15 · openalex created_date 2022/09/18 · openalex updated_date 2026/07/28
Implementing state estimation in low and medium voltage power distribution is\nstill challenging given the scale of many networks and the reliance of\ntraditional methods on a large number of measurements. This paper proposes a\nmethod to improve voltage predictions in real-time by leveraging a limited set\nof real-time measurements. The method relies on Bayesian estimation formulated\nas a linear least squares estimation problem, which resembles the classical\nweighted least-squares (WLS) approach for scenarios where full network\nobservability is not available. We build on recently developed linear\napproximations for unbalanced three-phase power flow to construct voltage\npredictions as a linear mapping of load predictions constructed with Gaussian\nprocesses. The estimation step to update the voltage forecasts in real-time is\na linear computation allowing fast high-resolution state estimate updates. The\nuncertainty in forecasts can be determined a priori and smoothed a posteriori,\nmaking the method useful for both planning, operation and post-hoc analysis.\nThe method outperforms conventional WLS and is applied to different test\nfeeders and validated on a real test feeder with the utility Alliander in The\nNetherlands.\n