Spectral Variation-Based Compression of Three-Phase Signals

This work adapts a spectral analysis-based compression algorithm, originally designed for single-phase leakage current signals, for compression of three-phase electrical signals with real disturbances. A PCA-based preprocessing step is introduced to reduce dimensionality and improve compression efficiency. The method is evaluated using real signals containing representative power quality events. The results show that the proposed approach achieves higher compression ratios, for instance, in some scenarios exceeding 118:1 and 76:1 for voltage and current signal, respectively. The reconstruction quality was comparable to the original algorithm, indicating its suitability for efficient storage and transmission in smart grid applications.