微粒群算法在地震波阻抗反演中的应用

Application of Particle Swarm Optimization in Seismic Impedance Inversion

地震波阻抗反演是油藏描述和储层预测中的关键技术,其本质属于多参数的非线性组合优化问题,诸如人工神经网络、模拟退火、遗传算法等非线性反演方法在地震波阻抗反演中已经得到了运用。起源于生物社会学研究和生物行为学模拟的微粒群算法,在多参数、非线性、多极值函数优化问题中具有较强的优越性。通过分析微粒群算法的原理,本文用该非线性算法实现了地震波阻抗反演,并且在理论模型的实验中,证明了算法的可行性。

Seismic impedance inversion is the key technique in the reservoir description and estimation. It is essentially a method of optimizing the multi--parameter nonlinear combination. Nonlinear inversion methods, such as artificial neural network, simulated annealing, genetic algorithm etc. , have been applied into the seismic impedance inversion. Particle Swarm Optimization, originated from the study of biologic sociology and the simulation of biologic behavior, enjoys great superiority in the optimization of multi--parameter, nonlinear and multi--maximum function. By analyzing the principle of Particle Swarm Optimization, this paper adopts this nonlinear algorithm to realize the seismic impedance inversion and it is proved that the algorithm is feasible in the theoretical model experiment.