Multiple-distributed generation planning under load uncertainty and different penetration levels

Ugranli F. , KARATEPE E.

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS, vol.46, pp.132-144, 2013 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 46
  • Publication Date: 2013
  • Doi Number: 10.1016/j.ijepes.2012.10.043
  • Page Numbers: pp.132-144
  • Keywords: Distributed generation, Optimal size and location, Artificial neural network, Power loss, Load uncertainty, Penetration level, PARTICLE SWARM OPTIMIZATION, DISTRIBUTION NETWORKS, DG ALLOCATION, DISTRIBUTION-SYSTEMS, GENETIC ALGORITHM, PLACEMENT, RELIABILITY, LOSSES, IMPACT


The penetration of distributed generation (DG) in power system is continually increasing. Hence, there is a need to investigate the potential benefits and drawbacks of DGs when integrating DG units in existing networks. The challenge of identifying the optimal locations and sizes has triggered research interest and many studies have been presented in this purpose. Different analytical techniques have been developed to minimize power losses for single-DG unit integration. If DG units are integrated at nonoptimal locations, the power losses increase, resulting in increased cost of energy. The novelty of this paper lies in studying the optimal placement of multiple-DG units in order to minimize power losses. In this study, an optimality criterion is investigated to minimize losses by including load uncertainty, different DG penetration levels and reactive power of multiple- DG concept. The simulation results show that it is not possible to form an analytical equation for optimum planning of DG in terms of load distribution, penetration level and reactive power. Due to the complexity of the multiple-DG concept, artificial neural network based optimal DG placement and size method is developed. The proposed method is implemented to the IEEE-30 bus test network and the results are presented and discussed. The results show that the proposed method can be applied to a power network for all possible scenarios. (C) 2012 Elsevier Ltd. All rights reserved.