高分辨率的储层弹性与物性参数同步反演研究
发布时间：2018-07-20 10:54
【摘要】：石油和天然气素有工业血液之称,是一个国家的重要战略性资源。近年随着未勘探的易开发区域逐渐减少,地球物理研究人员已将油气勘探的重点目标转向对技术要求更高的隐藏性油气藏和复杂构造性油气藏。目前单纯的储层弹性或物性参数反演已不能完全满足油气勘探的需要,而利用叠前地震资料开展高分辨率的储层弹性与物性参数同步反演已成为学术界和工业界共同关注的热点问题。本文在回顾地震资料高分辨率处理和地震反演的研究背景及意义的基础上,总结归纳出了目前叠前地震反演中存在的两个问题:1)如何获得高分辨率的地震资料;2)如何实现高分辨率的叠前地震同步反演的确定性优化方法。为解决以上两个问题,本文分别做了如下创新工作:1、针对问题1),本文提出了一种基于BP人工神经网络的地震资料高分辨率处理方法。该方法利用BP人工神经网络建立井旁地震道记录振幅谱与补偿系数之间的非线性映射关系,进而利用该关系计算出其它待补偿地震记录振幅谱的补偿系数,接着对补偿系数进行空间加权平滑和自适应补偿位置选择处理,最后将其作用于振幅谱,得到补偿后的高分辨率地震记录。该方法相较于其它方法,同时采用了测井资料信息和地震资料信息,尽量避免了补偿不足或补偿过多的现象,增强了补偿依据。2、针对问题2),本文提出了一种基于双参数弹性速度模型的储层弹性与物性参数叠前地震同步反演的确定性优化方法。该方法利用双参数弹性速度模型建立储层弹性与物性参数之间的联系,在贝叶斯反演框架下以储层弹性与物性参数联合的后验概率为目标函数,同时利用地震资料高分辨处理方法提升同步反演初值的分辨率,最后用自适应变步长优化方法求解得到分辨率更高的储层弹性和物性参数。一方面本方法采用双参数弹性速度模型,与其它岩石物理模型相比,该模型能够更好的建立弹性参数与孔隙形状参数之间的联系,有助于认识孔隙形状对储层弹性性质的影响;另一方面本方法采用确定性优化方法构建反演框架和求解,与随机优化方法相比,反演速度更快、精度更高。本文提出的方法均运用实际工区数据进行了验证,从验证效果来看,本文提出的地震资料高分辨率处理方法在提升地震资料分辨率上有明显的效果;本文提出的同步反演的确定性优化方法具有很好的收敛速度和稳态效果,井曲线投上去后吻合度好,满足叠前反演要求。
[Abstract]:Oil and natural gas have been known as industrial blood, is an important strategic resources of a country. In recent years, with the decrease of unexplored areas, geophysical researchers have turned the key targets of oil and gas exploration to hidden reservoirs and complex structural reservoirs with higher technical requirements. At present, the simple inversion of reservoir elastic or physical parameters can no longer fully meet the needs of oil and gas exploration. The simultaneous inversion of reservoir elastic and physical parameters with high resolution using prestack seismic data has become a hot issue in both academia and industry. On the basis of reviewing the research background and significance of high-resolution processing and seismic inversion of seismic data, This paper summarizes two problems existing in prestack seismic inversion: 1) how to obtain high resolution seismic data 2) how to realize the deterministic optimization method of high resolution prestack seismic synchronous inversion. In order to solve the above two problems, this paper proposes a high resolution processing method of seismic data based on BP artificial neural network. In this method, BP artificial neural network is used to establish the nonlinear mapping relationship between amplitude spectrum and compensation coefficient of seismic track records, and then the compensation coefficient of amplitude spectrum of other seismic records to be compensated is calculated. Then the compensation coefficients are processed by spatial weighting smoothing and adaptive compensation position selection. Finally, the compensated high resolution seismic records are obtained by applying them to the amplitude spectrum. Compared with other methods, this method uses logging and seismic information to avoid the phenomenon of insufficient compensation or too much compensation. This paper presents a deterministic optimization method for simultaneous inversion of reservoir elastic and physical parameters based on two-parameter elastic velocity model. In this method, the relationship between reservoir elasticity and physical parameters is established by using a two-parameter elastic velocity model. In the framework of Bayesian inversion, the posterior probability of the combination of reservoir elasticity and physical parameters is taken as the objective function. At the same time, the high resolution processing method of seismic data is used to improve the resolution of the initial value of synchronous inversion, and the adaptive variable step size optimization method is used to solve the reservoir elastic and physical parameters with higher resolution. On the one hand, the two-parameter elastic velocity model is used in this method. Compared with other rock physical models, this model can better establish the relationship between elastic parameters and pore shape parameters, and help to understand the effect of pore shape on the elastic properties of reservoir. On the other hand, the deterministic optimization method is used to construct the inversion framework and solve it. Compared with the stochastic optimization method, the inversion speed is faster and the precision is higher. The methods proposed in this paper are verified by using the actual work area data. From the result of verification, the high resolution processing method of seismic data presented in this paper has obvious effect on improving the resolution of seismic data. The deterministic optimization method for synchronous inversion presented in this paper has good convergence rate and steady state effect, and the well curve has good coincidence after being put into the well, which meets the requirements of prestack inversion.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P618.13;P631.4
，
本文编号：2133296
[Abstract]:Oil and natural gas have been known as industrial blood, is an important strategic resources of a country. In recent years, with the decrease of unexplored areas, geophysical researchers have turned the key targets of oil and gas exploration to hidden reservoirs and complex structural reservoirs with higher technical requirements. At present, the simple inversion of reservoir elastic or physical parameters can no longer fully meet the needs of oil and gas exploration. The simultaneous inversion of reservoir elastic and physical parameters with high resolution using prestack seismic data has become a hot issue in both academia and industry. On the basis of reviewing the research background and significance of high-resolution processing and seismic inversion of seismic data, This paper summarizes two problems existing in prestack seismic inversion: 1) how to obtain high resolution seismic data 2) how to realize the deterministic optimization method of high resolution prestack seismic synchronous inversion. In order to solve the above two problems, this paper proposes a high resolution processing method of seismic data based on BP artificial neural network. In this method, BP artificial neural network is used to establish the nonlinear mapping relationship between amplitude spectrum and compensation coefficient of seismic track records, and then the compensation coefficient of amplitude spectrum of other seismic records to be compensated is calculated. Then the compensation coefficients are processed by spatial weighting smoothing and adaptive compensation position selection. Finally, the compensated high resolution seismic records are obtained by applying them to the amplitude spectrum. Compared with other methods, this method uses logging and seismic information to avoid the phenomenon of insufficient compensation or too much compensation. This paper presents a deterministic optimization method for simultaneous inversion of reservoir elastic and physical parameters based on two-parameter elastic velocity model. In this method, the relationship between reservoir elasticity and physical parameters is established by using a two-parameter elastic velocity model. In the framework of Bayesian inversion, the posterior probability of the combination of reservoir elasticity and physical parameters is taken as the objective function. At the same time, the high resolution processing method of seismic data is used to improve the resolution of the initial value of synchronous inversion, and the adaptive variable step size optimization method is used to solve the reservoir elastic and physical parameters with higher resolution. On the one hand, the two-parameter elastic velocity model is used in this method. Compared with other rock physical models, this model can better establish the relationship between elastic parameters and pore shape parameters, and help to understand the effect of pore shape on the elastic properties of reservoir. On the other hand, the deterministic optimization method is used to construct the inversion framework and solve it. Compared with the stochastic optimization method, the inversion speed is faster and the precision is higher. The methods proposed in this paper are verified by using the actual work area data. From the result of verification, the high resolution processing method of seismic data presented in this paper has obvious effect on improving the resolution of seismic data. The deterministic optimization method for synchronous inversion presented in this paper has good convergence rate and steady state effect, and the well curve has good coincidence after being put into the well, which meets the requirements of prestack inversion.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P618.13;P631.4
，
本文编号：2133296
本文链接：https://www.wllwen.com/kejilunwen/shiyounenyuanlunwen/2133296.html
