Energy Management of an Active Distribution Network Considering Correlation Between Uncertain Input Variables

Abstract

This paper deals with the probabilistic energy management of an active distribution network in the presence of plug-in hybrid electric vehicle loads and power electronic devices like the soft open point and the smart transformer. The energy management scheme aims to simultaneously reduce the average voltage deviation, improve voltage stability, and maximize the daily profit. The proposed energy management measures involve simultaneous optimal scheduling of the soft open point, the smart transformer, the battery energy storage system, and an optimal power procurement from renewable sources. A probabilistic method models input uncertainties (load, renewable generation, plug-in hybrid electric vehicle charging load, and grid energy price). The correlation between uncertain input variables is incorporated by modifying the “Hong’s 2m point estimate method”. The multi-objective optimization problem is solved using a normalized weighted average technique. Simulation studies on a 33-bus active distribution network elucidate the benefits of the proposed approach. The expected daily profit improves by ∼ 2.95 % , the voltage stability by ∼ 17.94 % , and the average voltage deviation reduces by ∼ 75 % with the energy management approach presented in this paper. To validate the proposed method of solving a multi-objective problem, the combined objective has also been solved using the ubiquitous “controlled elitist genetic algorithm” of the MATLAB toolbox. The optimal solution yielded by the weighted sum method lies on the Pareto front obtained by the controlled elitist genetic algorithm. © 2022, King Fahd University of Petroleum & Minerals.