Abstract
Pre-stack AVO inversion is a key technology for obtaining underground elastic parameters. Traditional methods usually construct inversion problems based on the forward relationship between elastic parameters and reflection coefficients. Due to the lack of directly solver inversion operators, regularization is often relied upon to ensure the stability of inversion. Therefore, this paper proposes a stable AVO inversion operator construction mode, which can realize the prediction of elastic parameters directly from the reflection coefficient. By introducing the Sparse Incremental Selection and Shrinkage Operator (SISSO) regression algorithm, the bidirectional and explicit mathematical expression between the reflection coefficient and the elastic parameter is directly constructed, so as to realize the establishment of the forward and inversion operators synchronously. The established forward operator is consistent with the form and accuracy of classical Aki-Richards approximation, which verifies its theoretical reliability. The inversion operator can directly predict the elastic parameters from the reflection coefficient, reducing the reliance on prior constraints such as regularization, while clarifying the dependency relationship between the elastic parameters and the reflection coefficient with the incident Angle varies. Experimental results based on single-well synthetic data, the Marmousi model, and field seismic records demonstrate that, compared with conventional methods, the proposed approach enables more direct and stable estimation of elastic parameters under weak prior constraints. This offers a new technical pathway for enhancing both the stability and reliability of pre-stack AVO inversion.
Paper Information:
Peng Zhen., Cao Danping. A stable AVO inversion operator: Direct prediction of elastic parameters from reflection coefficient[J]. Petroleum Science, 2026. Doi: https://doi.org/10.1016/j.petsci.2026.09.036

