基于频变AVO反演的深层储层含气性识别方法

地震波在地下含油气介质储层中传播时会发生地震振幅的衰减与弹性特征频散现象,造成深层含油气储层地震流体识别困难。为此,基于Chapman裂隙—孔隙微结构衰减理论模型,分析多种频变弹性参数的流体敏感程度,优选出Gassmann流体项的频散程度作为深层储层含气性预测的识别因子;联合连续小波变换时频谱分解方法对部分角度叠加地震数据展开频谱分析确定参考频率;在此基础上,研究基于贝叶斯Cauchy约束准则的叠前地震频变Gassmann流体项反演优化方法,利用频变Gassmann流体项反演结果指导储层的流体检测,并在我国近海某盆地P探区对该方法进行了深层储层含气性预测验证。研究结果表明,该方法可以实现基于叠前地震资料的频变Gassmann流体参数的可靠提取,由其所得到的深层储层流体识别结果与实际测井解释结果吻合度较高。结论认为,频变Gassmann流体项能够有效识别深层储层的流体类型,为深层天然气层的识别提供了新的技术思路和方法。

A gas-bearing property identification method for deep reservoirs based on frequency-dependent AVO inversion

During the propagation of seismic wave in underground hydrocarbon bearing reservoirs, the phenomena of seismic amplitude attenuation and elastic characteristic dispersion happen, which makes it difficult to identify the fluids in deep hydrocarbon bearing reservoirs based on seismic data. In this paper, the fluid sensitivity degrees of a variety of frequency-dependent elastic parameters were analyzed based on the Chapman theoretical model of fractured–porous microstructure attenuation. And accordingly, the dispersion degree of Gassmann fluid term was selected as an identification factor for the gas-bearing prediction of deep reservoirs. Then, combined with the frequency spectrum decomposition method which is used for continuous wavelet conversion, spectrum analysis was carried out on some seismic data stacked with angle to determine the reference frequency. Based on this, the inversion optimization method of prestack seismic frequency-dependent Gassmann fluid term based on the Bayes Cauchy constraint criterion was researched, and the inversion result of frequency-dependent Gassmann fluid term was used to guide reservoir fluid detection. Finally, this method was applied in P exploration area in one offshore basin of China to verify its role in gas-bearing prediction of deep reservoirs. And it is indicated that by virtue of this method, the frequency-dependent Gassmann fluid parameters based on prestack seismic data can be extracted reliably, and correspondingly the identification results of deep reservoir fluid are better consistent with the actual logging interpretation results. In conclusion, the frequency-dependent Gassmann fluid term is conducive to identifying deep reservoirs effectively and provides a new idea and method for the identification of deep gas layers.