Abstract: Due to deep burial and complex diagenesis, tight sandstone reservoirs are characterized by low porosity, low permeability, and extreme heterogeneity. Accurately identifying and predicting high-quality reservoirs is a critical foundation for the efficient development of tight gas. Existing methods for predicting high-quality reservoirs, which relies on geological theoretical models and geophysical theories, fail to meet the degrees of accuracy, effectiveness, and applicability required for hydrocarbon exploration and development. This paper is a case study of the Oligocene Huagang Formation tight sandstone reservoir in Block H of Xihu sag, East China Sea shelf basin. Using core, logging, and seismic data, a seismic prediction method based on the coupling of sedimentary microfacies, lithofacies, and diagenetic facies is proposed for high-quality reservoirs. This method achieves accurate prediction of high-quality tight sandstone reservoirs. The research results show as follows. (1)The combined characterization of core, logging, and seismic data clearly defines the distribution of sedimentary microfacies. The integration of logging cross-plot analysis and stepwise discriminant analysis significantly enhances the accuracy of lithology identification and interpretation. The high-quality reservoirs of Huagang Formation in Block H of Xihu sag develop in the medium sandstone facies belt of the underwater distributary channel. (2)The diagenetic facies are quantitatively classified based on the flow zonation index, and machine learning is applied to achieve intelligent logging interpretation and macroscopic prediction of diagenetic facies. Under the dual constraints of sedimentary microfacies and lithofacies, high-quality reservoirs are precisely located. (3)The three-phase coupling model indicates that high-quality reservoirs are distributed in the overlapping zone of the underwater distributary channel sedimentary microfacies, medium sandstone lithofacies, and strong dissolution diagenetic facies. The analysis effectively reduces ambiguity and significantly improves prediction accuracy, providing a new approach for predicting high-quality tight sandstone reservoirs.

Key words: East China Sea shelf basin, Xihu sag, Oligocene, Huagang Formation, tight sandstone, sedimentary microfacies, lithofacies, diagenetic facies

摘要： 致密砂岩储层受埋深大和成岩作用复杂的影响,具有低孔低渗和非均质性极强的特征,准确识别及预测优质储层是实现致密气高效开发的重要基础。现有的基于地质理论模式和地球物理理论等预测优质储层的方法,其预测精度、效果和适用性不能满足勘探开发需求。以东海陆架盆地西湖凹陷H区块渐新统花港组致密砂岩储层为例,基于岩心、测井和地震资料,提出一种基于沉积微相-岩石相-成岩相耦合的优质储层地震预测方法,实现了致密砂岩优质储层的精准预测。研究结果表明:①岩心-测井-地震联合表征明确了沉积微相的展布,测井交会图解与逐步判别分析结合使得岩性的识别与解释精度显著提升,西湖凹陷H区块花港组优质储层发育于水下分流河道的中砂岩相带。②基于 流动分层指数定量划分成岩相,机器学习实现了成岩相的智能测井解释与宏观预测,在沉积微相与岩石相双重约束条件下精准定位优质储层。③三相耦合模式表明,优质储层分布于"水下分流河道沉积微相-中砂岩岩石相-强溶蚀成岩相"叠合区。分析结果有效降低了多解性,显著提高了预测精度,为致密砂岩优质储层预测提供了新思路。

关键词: 东海陆架盆地, 西湖凹陷, 渐新统, 花港组, 致密砂岩, 沉积微相, 岩石相, 成岩相