【摘要】 岩石的脆性是非常规油气藏勘探开发中需考虑的关键岩石力学参数之一。岩石脆,压裂过程中易产生缝网,从而有效改造储层,达到提高单井产量的目的。基于岩石弹性参数的归一化杨氏模量和泊松比方法,是当前广泛采用的可用于地震预测的岩石脆性评价技术,但其结果受归一化参数取值的影响较大,适用性受到影响。利用杨氏模量和泊松比的商作为脆性评价指标,并用大量岩石力学实验刻度,大大提高了岩石脆性判别的可信度。利用地震叠前弹性参数反演,开展吉木萨尔凹陷二叠系芦草沟组致密油岩石脆性预测,指导水平井部署。为保证更好的储层改造效果,在水平井轨迹设计过程中不但考虑了储层物性,还充分考虑了岩石脆性。在优选的有利区钻探水平井,并实施了体积压裂,微地震监测结果显示压裂产生了缝状网,储层改造效果理想。该水平井试油获得了稳定高产工业油流。更多还原

【Abstract】 Brittleness of rocks is one of the key parameters of rock mechanics which should be considered in exploration and development of unconventional oil and gas reservoirs. When the rock is brittle,fracture networks tend to form during hydraulic fracturing,thus improving the reservoir properties effectively,and enhancing the single-well productivity. Based on the current technologies,evaluating the brittleness of rocks can be applied to seismic prediction using normalized Young’s modulus and Poisson ratio. Although it is widely used,the evaluation results are highly influenced by the normalization parameters,so this approach is not perfect. This work used the quotient of Young’s modulus and Poisson ratio( EPR) as the value of brittleness,which is calibrated with rock mechanic experiment results. It has significantly increased the in credibility of rock brittleness identification. Meanwhile,by using prestack elastic parameters inversion,the rocks brittleness of Permian Lucaogou Formation in the Jimusar sag was predicted,and the result was used to deploy a horizontal well. In order to ensure good fracturing,not only rock reservoir physical properties,but more important,rock brittleness is fully considered in the design of horizontal well trajectory. A horizontal well was drilled in the favorable area that has better brittleness as shown by seismic prediction.Multi-stage hydraulic fracturing was carried out to this horizontal well. Microseismic detection shows that fracture networks were generated during fracturing in this well,which indicated that good stimulation has been achieved. High-yield oil flow was obtained from this horizontal well after fracturing.更多还原