By using the harmonic filter bank(HFB)technique to separate the harmonic components from the transient ones in the power quality(PQ)disturbance signals

a data compression method for PQ disturbance based on HFB is proposed. The harmonic component parameters can be estimated with high precision by adopting the interpolated discrete Fourier transform(IpDFT)algorithm. The harmonic components are then compressed by parameter quantization

and the residue of disturbance is compressed by wavelet transform. The real PQ disturbance data are finally used to test the proposed method. The results show that the method is effective for various PQ disturbance data

and has even better performance than that of the standard wavelet compression method with the same compression ratio. This work is supported by National Natural Science Foundation of China(No. 51477124).