Skip to main content
placeholder image

Enhancement of load balancing in electrical distribution networks using Artificial Neural Networks

Conference Paper


Abstract


  • In this paper, problem related to the network reconfiguration for load balancing in distribution systems has been addressed. For obtaining efficient and real-time solution for the above problem, a generalised Artificial Neural Network (ANN) model has been developed based on the Daily Load Curves (DLCs) to predict the switching status of the dynamic switches for optimal configuration under varying load conditions. To show the performance of the proposed ANN model, a 16-bus test system is used and the model is trained by applying the input vectors generated for the test system, using Conjugate Gradient Descent Backpropagation Algorithm. The trained ANN model is tested with 20 actual input-sets taken arbitrarily from the DLCs and the test results are found to be the same as that obtained by off-line simulation.

Publication Date


  • 2003

Citation


  • Kashem, M. A., Ganapathy, V., & Negnevitsky, M. (2003). Enhancement of load balancing in electrical distribution networks using Artificial Neural Networks. In IPEC 2003 - 6th International Power Engineering Conference (pp. 907-912).

Scopus Eid


  • 2-s2.0-12744253828

Web Of Science Accession Number


Start Page


  • 907

End Page


  • 912

Abstract


  • In this paper, problem related to the network reconfiguration for load balancing in distribution systems has been addressed. For obtaining efficient and real-time solution for the above problem, a generalised Artificial Neural Network (ANN) model has been developed based on the Daily Load Curves (DLCs) to predict the switching status of the dynamic switches for optimal configuration under varying load conditions. To show the performance of the proposed ANN model, a 16-bus test system is used and the model is trained by applying the input vectors generated for the test system, using Conjugate Gradient Descent Backpropagation Algorithm. The trained ANN model is tested with 20 actual input-sets taken arbitrarily from the DLCs and the test results are found to be the same as that obtained by off-line simulation.

Publication Date


  • 2003

Citation


  • Kashem, M. A., Ganapathy, V., & Negnevitsky, M. (2003). Enhancement of load balancing in electrical distribution networks using Artificial Neural Networks. In IPEC 2003 - 6th International Power Engineering Conference (pp. 907-912).

Scopus Eid


  • 2-s2.0-12744253828

Web Of Science Accession Number


Start Page


  • 907

End Page


  • 912