岩石物理模型约束下声波测井数据评价及处理*

声波测井数据是石油天然气勘探开发中最为重要的基础资料，其数据质量受井壁环境、频散等因素影响较大，且部分钻井并未采集声波。这对后续开展速度分析、储层识别和烃类检测都带来较大影响。因此需要对声波测井数据进行环境校正、补齐、预测等处理。目前，主要依据研究区域内已钻井的统计规律对声波测井数据进行评估和预测。而在深水区，因钻井距离远、地质条件差异较大等原因，不同地区的声波变化较大，依据统计方法评估数据存在较大的多解性。本文提出采用岩石物理模型来评估声波质量和预测声波数据。基于Xu&White岩石物理模型，采用密度曲线反演计算泥岩孔隙度，改善了岩石物理模拟在储层段和非储层段的拟合效果，且在多口井的适用性更好。基于新方法，本文建立了可在区域内稳定适用的岩石物理模型，可计算多口井在不同岩性、不同深度下的速度，为在不同地区和不同深度的声波质量评估和预测提供依据，指导后期储层预测和含气性检测。

Evaluation and correction of logging acoustic waves under rock physical model constraint

Acoustic wave data is the most important information in oil exploration and development. The quality of acoustic wave data may be greatly affected by borehole environment, frequency dispersion, and so on, which results in data errors. In addition, some drilling wells did not collect wave data. So data errors and the lack of wave data will have a negative impact on subsequently analysis such as reservoir identification and hydrocarbon detection. Therefore, it is necessary to optimize acoustic wave data such as environmental correction, supplement or wave prediction.Currently, the evaluation and prediction of acoustic wave data are mainly based on the statistical results of drilled wells in the region. However, in deep water area, due to the long distance and significant geological differences among the drilled wells, the statistical methods have huge uncertainties. This article proposes a rock-physical model to evaluate data quality and even predict acoustic data. Based on the classical Xu&White rock-physical model, this article optimizes porosity interpretation by using density to calculate the porosity of mud-stone and getting better fit between modelling wave data and logging data. And this method can be applied to other wells. Finally, we can obtain velocities of different wells with different lithology in different depth by using a single model in a region, which provide a stable reference for the wave data quality evaluation and prediction, and guiding further reservoir prediction and gas detection.