Abstract:The paper studied the character detection of sample-based seismic attribution,established non-linear mapping relation between time difference of sonic log(Δt) and seismic attribution by using BP neural networks.The inversion data for all well of studied area have been extrapolated by distant-weighted averaging.The wavelet suitable for seismic signal decomposition and reconstruction has been studied and analyzed in detail in order to improve the resolution and the correctness of results.The paper also studied the changing relations of logging delta time and seismic impedance vs formation interface.Finally,the paper made correction of interface influence for inverted time difference and calculated the porosity para;neter with higher resolution and as consistent with real situation.The calculated results are consistent with real situation and have satisfied effects after proving by known wells and new drillings.
宋维琪, 王小马, 杜玉民, 付金华. 综合应用地震属性、测井数据反演储层参数[J]. 石油地球物理勘探, 2002, 37(5): 491-494.
Song Weiqi, Wang Xiaoma, Du Yumin, Fu Jinhua. Inversion of reservoir parameters by integrative application of seismic attribution and logging data. OGP, 2002, 37(5): 491-494.