Abstract—To improve the operating performance of a distribution network, on line monitoring is required. For this purpose, sensors (metering devices) are installed. To reduce the number of sensors, state estimation approach can be used to estimate the voltage of buses which do not have sensors. This paper proposes online state estimator for three phase active distribution networks using Neural Network and displayed the results on Geographic Information System (GIS). Neural Network based state estimation is used to estimate the bus voltages by using learning approach from power flow patterns. K-matrix three phase distribution power flow is used in this method as an analytical tool. The K-matrix approach is combined with Particle Swarm Optimization (PSO) in handling a Distributed Generation (DG) which is operated as a voltage controlled (PV) bus. The test results show that the proposed method can reduce the number of sensors significantly (almost 50%).
Index Terms—State estimator, neural network, K-matrrix, PSO and GIS.
Dimas Fajar U. P., Indri Suryawati, Ontoseno Penangsang, and Adi Soeprijanto are with the Electrical Engineering Department Institut Teknologi Sepuluh Nopember, Indonesia (e-mail: dimasfup@ee.its.ac.id, indrisuryawati@gmail.com, zenno_379@yahoo.com, adisup@ee.its.ac.id).
Matt Syai’in is with the Department of Marine Electrical Engineering Surabaya Shipbuilding State Polytechnic Surabaya, Indonesia (matt.syaiin@gmail.com).
[PDF]
Cite:Dimas Fajar U. P., Indri Suryawati, Ontoseno Penangsang, Adi Suprijanto, and Mat Syai’in, "Online State Estimator for Three Phase Active Distribution Networks Displayed on Geographic Information System," Journal of Clean Energy Technologies vol. 2, no. 4, pp. 357-362, 2014.
