Predictive var management of distributed generators

M. Z.C. Wanik, I. Erlich, A. Mohamed, H. Shareef

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)


This paper presents and describes a smart predictive technique for managing reactive power from a numbers of distributed generation (DG) units connected to low voltage (LV) buses in a distribution network. The technique applies an optimization process in the first stage and in the second stage the procedure is generalized using artificial neural network (ANN). The ANN is trained to replace the role of optimization process which is repetitive in nature and time consuming. The technique can speed up the time while scarifying a little accuracy. The objective is to develop an intelligent management tool that can be used to manage reactive power from a group of DG units for online management. This technique predicts the optimal reactive power fro the next time step that needs to be supplied by each DG unit with the objective of minimizing active power losses and keeping the voltage profile within the required limit. The effectiveness of the method is tested by predicting reactive power from twelve DG units simultaneously and the result is promising. Intelligent management technique presented in this paper is suitable to be integrated into online management scheme under Smart Grid concept.

Original languageEnglish
Title of host publication2010 9th International Power and Energy Conference, IPEC 2010
Number of pages6
Publication statusPublished - 2010
Externally publishedYes
Event2010 9th International Power and Energy Conference, IPEC 2010 - Singapore, Singapore
Duration: Oct 27 2010Oct 29 2010

Publication series

Name2010 9th International Power and Energy Conference, IPEC 2010


Conference2010 9th International Power and Energy Conference, IPEC 2010


  • Distributed generation
  • Neural networks
  • Online management
  • Optimal reactive power
  • Smart grid

ASJC Scopus subject areas

  • Energy Engineering and Power Technology


Dive into the research topics of 'Predictive var management of distributed generators'. Together they form a unique fingerprint.

Cite this