An Adaptive Recursive Wavelet Based Algorithm for Real-Time Measurement of Power System Variables during Off-nominal Frequency Conditions

Abstract

An adaptive wavelet-based method of phasor and frequency estimations and its application for on-line monitoring, control and protective equipment is presented in this paper. The method is recursive with a low computational burden, suited for real-time implementations, and does not need any pre-processing stages. It computes the frequency of the analyzing signal accurately during nominal and off-nominal frequency conditions. To reduce computational errors of the phase angle and amplitude estimations that occur during off-nominal frequency conditions, the proposed method adapts itself according to the frequency drift. The proposed recursive adaptive method along with three other approaches based on Recursive Wavelet Transform (RWT), Discrete Fourier Transform (DFT) and Recursive DFT (RDFT) are implemented and compared using Matlab, considering dynamic and static conditions. The methods are also employed for real-time computation of the phasor and frequency of the recorded data of the South West Interconnected System (SWIS) in Western Australia, and their efficiencies are assessed and compared. In addition, a real 33.3 MVA, 132/23.3 kV transformer located in SWIS is modeled in the EMTP software, and the recorded signals have been employed for performance evaluation of the above-mentioned methods. The method is also implemented on a computer-based system in the laboratory and its performance is appraised accordingly. Extensive Matlab simulations, laboratory implementation, field data and EMTP results show that the proposed adaptive method has better static and dynamic performances in comparison to the other approaches.