Frequency-domain AVA inversion using sparse Bayesian learning
Yongzhen Ji, Shangxu Wang, Sanyi Yuan, Fengfan Yuan, Junzhou Liu · 2015
Summary AVA inversion method utilizing angle information to obtain elastic parameters of subsurface rock is vital to reservoir characterization. Under the assumption of blocky layered media, a novel AVA inversion algorithm combining spectral reflectivity inversion with sparse Bayesian learning is presented in this paper. The method retrieves sparse P-wave, S-wave and density reflectivity series by sequentially adding, deleting or re-estimating hyper-parameters, without pre-setting the number of nonzero reflectivity spikes. Moreover, the method can highlight the stratigraphic boundaries and improve vertical resolution of inverted parameters, which is helpful for the subsequent interpretation. Model data and real data examples illustrate the effectiveness of the method.