A novel approach for the classification of power quality disturbance using combined adaptive decomposition structure and neural network

  • Authors:

  • N. P. Subramaniam;S. Jayashree;K. Bhoopathy Bagan

  • Affiliations:

  • Dept. Electronics Engg., Anna University, Chennai, India;Dept. Electronics Engg., Anna University, Chennai, India;Dept. Electronics Engg., Anna University, Chennai, India

  • Venue:
  • AEE'05 Proceedings of the 4th WSEAS international conference on Applications of electrical engineering
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Power quality disturbance (PQD) are normally subjected to transients and non periodic components which present a problem to the overall performance of the system. The use of traditional wavelet transform to extract the fundamental frequency components from the disturbed signal is inappropriate to different faults with single mother wavelet. A novel adaptive decomposition structure based technique for PQD has been presented in this paper, which has the ability to perform the statistical analysis using histogram analysis block of adaptive decomposition signals and neural network for classification of fault. The proposed method is described in detail and implemented using MATLAB and SIMULINK. Finally, the effect of different parameters on the algorithm is examined in order to highlight its performance. It is found that the adaptive decomposition is an excellent tool for disturbance detection and classification.