Data compression technique in recording electric arc furnace voltage and current waveforms for tracking power quality

A data compression technique is used to enhance the storage capability of electric power quality instruments in recording electric arc furnace voltage and currents waveforms, where voltage flicker and harmonic current distortion are major power quality disturbances. The compression approaches by multi-resolution analysis with threshold coding and vector quantization coding are compared. The digital filters are designed by the discrete wavelet transform. The results from spectrum analysis show high compression ratios while keeping low information loss. From the calculation results, the reconstructed voltage waveforms using threshold coding almost keep the same voltage flicker values. The vector quantization coding is better for current waveforms to keep harmonic distortion values. From the simulation results of the measurement data, the memory requirement is reduced to 20.4% while power quality characteristics are almost kept.