Estimation of the steady states parameters in open-loop distribution networks based on feedforward neural networks

¹ Tashkent State Technical University named after Islam Karimov, Department of Power Plants, Networks and Systems. 100095, 2 University Street, Tashkent, Uzbekistan
² Department of Power Supply and Renewable Energy Sources, “Tashkent Institute of Irrigation and Agricultural Mechanization Engineers” National Research University, 100000 Tashkent, Uzbekistan
³ Karshi Engineering Economic Institute, Faculty of Engineering, 180101, Qashqadaryo region, Qarshi, 225 Mustaqillik shoh Street, Karshi, Uzbekistan

^* Corresponding author: muzaffar_hb@mail.ru

Abstract

Power flow calculations play major role during the operational stage of any distribution networks for its control, as well as during the design stage. Moreover, the main purpose of any power flow calculations is to compute precise steady-state voltages of nodes, the real and reactive power flows on each branches, under the assumption of known loads. That is, one of the main results of the calculation is steady-state voltages of nodes. As a rule, iterative methods are used to calculate load flows in distribution networks. This places high demands on calculations in terms of speed and reliability of obtaining results in any operating conditions. Given this in the article presents models for node voltages estimation in distribution networks based on feedforward artificial neural networks. Their use makes it possible to increase the speed of the power flow calculations in distribution networks. We examined the effectiveness of the models on the example of real schemes of 6-10 kV open-loop distribution networks.