With rapid increase of the grid-connected equipment based on power electronic converters(such as renewable energy generation

energy storages and electric vehicles)

the electrical signals of distribution networks are becoming more and more complex. A wideband signal model of power systems is established

which is composed of deterministic components and stochastic noise components. Based on this signal model

a decomposition scheme for wideband signals is proposed. Firstly

the robust local regression smoothing method is used to extract and filter the stochastic noise components. A determination method for the adaptive threshold based on the estimation of mean and standard deviation is proposed

which is used to decompose the stochastic noise components. Then an adaptive threshold based on the spectrum of interharmonic sub-groups is built to detect the interharmonics. Finally

the Taylor-Fourier transform algorithm is used to estimate the parameters of deterministic components based on the independent sub-signals decomposed by the infinite impulse response filter bank. Simulation results show that the proposed scheme can achieve time-varying interharmonics detection and precise decomposition of wideband signals under the conditions of low signal to noise ratio and frequency dynamics

and the field voltage measurement is decomposed by using the proposed decomposition method.