Pattern recognition and time location of power quality disturbances using TT-transform is proposed in this thesis. TT-transform is performed to the original signal and the result is a complex matrix. The absolute value of the complex matrix called modulus matrix. Unit cylinder is got by every row vector of modulus matrix as the mother curve and the cylinders are superposed. A 3D figure is got and it is used as the pattern to recognize power quality disturbances. The maxima of the modulus matrix for every column are used to form a vector called modulus vector. The prominent values in the modulus vector are the transition point and the time location is got. Simulation results indicate that the patterns are unique for power quality disturbances and the time location is accurate under noisy circumstance.
Published in:
Intelligent System Design and Engineering Application (ISDEA), 2010 International Conference on
(Volume:2
)
Date of Conference: 13-14 Oct. 2010