基于改进粒子群算法的横波速度预测方法研究

发布时间：2018-03-15 02:00

本文选题：粒子群算法　切入点：横波速度预测　出处：《吉林大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着全球经济的回暖,我国社会经济逐年稳步提高,能源是经济发展的动力。人民的环境保护意识不断提高,新型能源不断被开发与推广,但是新型能源的应用仍然具有局限性,传统的能源特别是对油气资源的需求量更为凸显,这也导致了对油气藏的进一步勘探开发。随着勘探开发不断增加,常规储层储量逐年降低,这也使得开发的方向逐步趋于岩性油气藏、复杂型储层,使得勘探难度大大提升,勘探方法进一步的改进。正因为储层勘探的进一步深入,对于地震储层预测的精度要求也有了新的提升。不仅需要从预测储集层的发育程度出发,更要求了对预测储集层所包含的流体性质、岩石的弹性性质等的进一步了解,从而在实际生产中提高油气藏勘探开发的可实现性,大大减低开发风险。岩石物理学的研究目的就是探究岩石弹性性质与岩石参数之间的对应关系,便于我们进一步的研究地质地层结构与岩石孔隙流体对地震响应的影响,因此岩石物理学研究是建立在地震响应与测井资料之间的一道桥梁,更是油气藏地震勘探的重要物理基础。在勘探开发中,地震反演作为一个重要的环节,是储层预测与流体识别的主流方法。横波速度信息不仅是叠前反演所需要的重要信息也是叠前地震数值模型不可缺少的参数信息。往往在实际生产中,由于横波速度信息提出成本高、初至拾取难度大,应用不为广泛,故如果找到一种有效的横波速度预测方法对有着至关重要的意义。本文将从岩石物理模型的角度出发,详细介绍粒子群算法并提出改进思路,将改进粒子群算法引入横波速度预测中来,利用传统遍历搜索方法和改进粒子群算法两种方法计算最佳孔隙纵横比,使理论纵波速度与实际纵波速度的误差最小,进行横波速度预测,将预测结果与实测横波速度对比,验证了引入粒子群算法的横波速度预测方法的有效性与准确性。同时,通过改变岩石物理模型对横波速度预测做进一步的研究,文中选取了SCA模型、SCA+Gasmann模型、Hudson模型、Chapman模型,这四种模型进行横波速度预测分析。其中基于SCA+Gasmann模型的横波预测结果,最接近实际资料,预测曲线与实际横波曲线拟合度最高;基于Hudson模型的横波速度预测结果,对比基于其他三种岩石物理模型横波速度预测结果,其误差最高,曲线拟合度最差。
[Abstract]:With the warming of the global economy, China's social economy is steadily improving year by year, energy is the driving force of economic development, people's awareness of environmental protection is constantly improved, and new types of energy are constantly being developed and popularized. However, the application of new energy is still limited, and the demand for traditional energy, especially for oil and gas resources, is more prominent, which also leads to further exploration and development of oil and gas reservoirs. The reserves of conventional reservoirs are decreasing year by year, which also makes the direction of development gradually tend to lithologic reservoirs, complex reservoirs, which makes exploration much more difficult and exploration methods further improved. Precisely because of the further development of reservoir exploration, The accuracy requirement of seismic reservoir prediction has also been improved. It is necessary not only to predict the development of reservoir, but also to have a better understanding of the fluid property and elastic property of rock, etc. In order to improve the realizability of exploration and development of oil and gas reservoirs in actual production and greatly reduce the risk of development, the purpose of the study of rock physics is to explore the corresponding relationship between the elastic properties of rocks and the parameters of rocks. Therefore, the study of rock physics is a bridge between seismic response and logging data. It is also an important physical basis of seismic exploration in oil and gas reservoirs. Seismic inversion is an important link in exploration and development. S-wave velocity information is not only the important information needed for pre-stack inversion, but also the indispensable parameter information for pre-stack seismic numerical model. It is very important to find an effective prediction method for shear wave velocity because of its high cost, difficulty in picking up the first arrival, and not widely used. In this paper, from the point of view of rock physical model, it is very important to find an effective prediction method of shear wave velocity. The particle swarm optimization algorithm is introduced in detail, and the improved particle swarm optimization algorithm is introduced into the prediction of shear wave velocity. Two methods, the traditional traversal search method and the improved particle swarm optimization algorithm, are used to calculate the optimum aspect ratio. The error between theoretical and actual P-wave velocities is minimized, and the prediction results are compared with the measured S-wave velocities to verify the validity and accuracy of the S-wave velocity prediction method, which is based on particle swarm optimization (PSO). By changing the rock physical model to further study the prediction of shear wave velocity, this paper selects SCA model and Gasmann model and Hudson model and Chapman model to predict and analyze the shear wave velocity. Among them, the prediction results of shear wave based on SCA Gasmann model are analyzed. Compared with the prediction results of the other three rock physical models, the prediction results of shear wave velocities based on Hudson model are the highest and the fitting degree of the curves is the worst.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P618.13;P631.4

【相似文献】