ART artificial neural networks based adaptive phase selector

Author(s): Tai, N.L. | Yang, Y. | Yu, W.Y. |

Year: 2005

Citation: ELECTRIC POWER SYSTEMS RESEARCH Volume: 76 Issue: 1-3 Pages: 115-120

Abstract: This paper introduces a new phase selector based on adaptive resonance theory (ART). Because conventional phase selector cannot adapt dynamically to the power system operating conditions, it presents different characters under different power system conditions. To overcome the disadvantage, an adaptive phase selector, which utilizes artificial neural network based on ART, is designed. ART based neural network (ARTNN) has some advantages such as no local extremum, quickly convergence and so on. Therefore, the proposed ARTNN based phase selector has better performances compared with other neural networks based phase selector, and the new selector can adapt dynamically to the varying power system operation conditions. Furthermore, the phase selector can be trained and learned on-line. A lot of EMTP simulations and experimental field data tests have illustrated the phase selector s correctness and effectiveness.

Topics: Machine Learning, Applications: Industrial Control, Models: ART 2-A,

PDF download



Cross References


  1. ART 2-A: An adaptive resonance algorithm for rapid category learning and recognition
    This article introduces Adaptive Resonance Theory 2-A (ART 2-A), an efficient algorithm that emulates the self-organizing pattern recognition and hypothesis testing properties of the ART 2 neural network architecture, but at ... Article Details


Back to Articles