Loading...

Design of a Smart Algorithm Based on Two Dimensional Wavelet Transformation for Detection and Classification of Power Quality Disturbances

Mollayi, Nader | 2010

410 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 40440 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Mokhtari, Hossein
  7. Abstract:
  8. Power Quality can be simply defined as the quality of voltage at electrical loads. Detection and classification of voltage and current disturbances is of high importance in power system protection and monitoring. This procedure cannot be implemented by operators because of the high volume of the data which must be processed. So, it is needed to automate this procedure. Systems designed for this purpose usually contain three main parts: feature generation, feature selection and classifier design. The algorithms used for feature generation for power quality disturbances are mainly based on discrete Fourier transformation or discrete wavelet transformation. These approaches have shown some shortcomings in their operation. The main disadvantages of algorithms based on discrete Fourier transformation are the opposition between time and frequency domains and sensitivity to change in system base frequency and in case of discrete wavelet transformation they are not being able to detect disturbances with slow variations and steady state disturbances. Here, a new approach based on two dimensional representation of power quality data is proposed which uses two dimensional wavelet transformation in its structure as a vital part. In this approach, a two dimensional matrix is formed based on 32 cycles of voltage signal, so that the samples of voltage in each cycle will form one row of the matrix. This matrix can be considered a 32 by 256 pixel image. The resulted image is decomposed into approximation and details by two dimensional wavelet transformation. Details contain the useful information. By processing the details, special patterns associated with each type of disturbance can be detected. Two new approaches are also proposed for feature extraction. Features are classified based on the nearest neighbor classifier system. The main advantages of our proposed system were the ability of detection of disturbances with slow variations, detection of a wide variety of disturbances from transients to steady state disturbances with one processing algorithm and robustness of system in presence of power system frequency variations for high frequency disturbances. Testing the operation of system with a database which contained real data obtained from power system beside simulated data was another advantage of our project
  9. Keywords:
  10. Power Quality ; Image Processing ; Character ; Classifier ; Wavelet Transform ; Two Dimentional Transform ; Power Quality Disturbances

 Digital Object List

 Bookmark