基于混沌蚁群算法的弹性阻抗反演

弹性阻抗反演属于叠前反演,能够反映振幅随偏移距或入射角的变化,可以获得纵波速度、横波速度、密度、泊松比以及拉梅系数等多种岩性参数。目前,弹性阻抗反演的常规算法为广义线性反演,该反演方法强烈依赖初始模型,容易陷入局部极小。为此,尝试将一种优秀的非线性算法——混沌蚁群算法应用于弹性阻抗反演中,该算法具有能达到全局最小,不依赖初始模型等优点。模型测试证明了混沌蚁群算法的稳定性和可靠性。将基于混沌蚁群的弹性阻抗反演方法应用于胜利油田某储层,其反演结果与实际钻探及测井解释相一致,验证了该反演方法的可行性和有效性。

 As a pre-stack inversion,elastic impedance inversion not only characterizes amplitude variation with offset or incident angle,but also offers more lithologic parameters,such as P-wave velocity,S-wave velocity,density,Poisson’s ratio,Lame coefficient etc.However,the common elastic impedance inversion algorithms are generalized linear methods,which are strongly dependent on initial model and easily fall into local minimum.A good nonlinear algorithm,chaos ant colony algorithm,is introduced to elastic impedance inversion in this paper.The chaos ant colony algorithm is model-independent and able to reach global minimum.Model test has demonstrated the reliability and stability of the algorithm.The inversion results of a reservoir in Shengli Oilfield are consistent with the drilling data,which proves the feasibility and effectiveness of the method.

国家自然科学基金项目(41004046和41230318)资助。