叠前道集优化处理技术及其应用

Pre-stack gather optimization processing technology and its application

叠前地震反演是复杂油气储层预测及流体检测的重要手段之一.反演结果的精度与可靠性一方面取决于反演算法本身,另一方面与叠前道集品质密切相关.叠前道集优化处理能够消除或压制非岩性、流体信息引起的地震信息异常,恢复能够客观反映岩性及流体变化的地震信息,揭示地下介质真实的AVO特征.本文提出一种基于AVO特征分析的叠前道集优化处理技术,从实际井数据出发,通过建立井旁道AVO合成道集,对比分析原始道集与合成道集的AVO特征差异,从振幅、相关性、剩余时差以及最大频率四个方面分析评价原始道集存在的问题,针对性制定合适的处理流程,并基于井旁道测试确定最佳处理参数来进行道集优化处理.每步处理均采用AVO特征、叠后频谱和井震标定进行有效质控,确保道集优化处理的合理性,最终为叠前反演提供较高质量的数据基础.该技术成功应用于XJ某区块,相较于原始道集,优化后道集叠前反演的精度明显提升.

Pre-stack seismic inversion is one of the important methods for complex oil and gas reservoir prediction and fluid detection. The accuracy and reliability of the inversion result depends on the inversion algorithm itself on the one hand, and is closely related to the quality of the pre-stack gather on the other hand. Pre-stack gather optimization processing can eliminate or suppress seismic information anomalies caused by non-lithological and fluid information, restore seismic information that can objectively reflect changes in lithology and fluid, and reveal the true AVO characteristics of the underground medium. This paper proposes a pre-stack gather optimization processing technology based on AVO feature analysis. Starting from actual well data, through the establishment of well sideway AVO synthetic gathers, the difference in AVO features between original and synthetic gathers is compared and analyzed. Analyze and evaluate the existing problems of the original gather from the four aspects of amplitude, correlation, residual time difference and maximum frequency, formulate a suitable processing flow, and determine the best processing parameters based on the well side channel test to optimize the gather. Each step of processing uses AVO features, post-stack spectrum and well seismic calibration for effective quality control to ensure the rationality of gather optimization processing, and ultimately provide a higher-quality data basis for pre-stack inversion. This technology has been successfully applied to a block of XJ. Compared with the original gather, the accuracy of the pre-stack inversion of the optimized gather is significantly improved.