Spectral Decomposition for Hydrocarbon Detection Based on VMD and Teager–Kaiser Energy

摘要

Hydrocarbons can cause anomalies in the energy density of seismic signals when seismic waves pass through them. Teager–Kaiser energy (TKE) is an important attribute that can be utilized to depict the energy density of a seismic signal and the energy distribution of a seismic wavefield. In this letter, a novel spectral decomposition-based approach for hydrocarbon detection is proposed that applies the variational mode decomposition (VMD) associated with TKE to seismic data, which is called the VMDTKE algorithm. The proposed method not only possesses the better performance of TKE in focusing instantaneous energy, but also inherits the merit of high time–frequency resolution from VMD. The Marmousi2 example is used to demonstrate that the VMDTKE approach is capable of depicting the location and extent of strong anomalies which correlate to hydrocarbons more clearly. We compare the spectral decomposition results with that from the conventional VMD-based method. Application on field data further confirms the potential of the VMDTKE algorithm in delineating strong amplitude anomalies that are associated with hydrocarbon reservoirs.