A novel method for removal of DC component in nonsinusoidal signals of power systems

A method for removal of DC component in nonsinusoidal signals of power systems is presented in this paper. In digital power metering and digital protection relaying, some unwanted signals, such as DC components and other nonperiod signals with exponential decay, must be filtered out so as to obtain much higher measurement accuracy of voltages and currents. This paper aims at proposing a algorithm for estimating the amplitudes and phase angles of nonsinusoidal voltage signals by removing DC components, employing central numerical differentiation with 11 points and discrete Fourier transform (DFT). Discrete wavelet transform is used to estimate the time constant of nonperiod signals, and numerical differentiation is used to remove the DC components and to estimate the amplitudes and phase angles of nonsinusoidal signals of power systems. The simulation results of a study example carried out in Matlab show better application to digital relay and meter in accuracy and time.