论文部分内容阅读
目前电阻率法三维反演方法大多是基于最小构造或最平缓模型约束的反演,这些反演算法能稳定收敛,但有时反演结果分辨率较差,不利于地质解释.本文在分析Zhdanov(2002)提出的基于最小支撑泛函聚焦反演方法的基础上,将最小支撑泛函引入到电阻率法三维反演的目标函数中,然后采用高斯牛顿法求解反演目标函数最优化问题,同时结合预条件共轭梯度法得到电阻率法三维聚焦反演结果.通过对几个典型模型的试算,并与传统光滑模型约束反演结果进行比较,表明本文反演方法结果与实际模型吻合的更好,分辨率更高,模型更聚焦,而且算法收敛速度较快.
At present, the three-dimensional inversion method of resistivity method is mostly based on the inversion of minimum structure or the most gentle model constraint, and these inversion algorithms can stably converge, but sometimes the inversion result has poor resolution and is not conducive to geological interpretation.In this paper, Zhdanov ( 2002), the minimum support functional is introduced into the objective function of the three-dimensional inversion of the resistivity method, and then Gauss Newton method is used to solve the optimization problem of the objective function of the inversion. Simultaneously, The results of three-dimensional inversion of the resistivity method are obtained by using the conjugate gradient method of preconditioner.Through comparison of some typical models and comparison with the results of traditional smooth model constrained inversion, the result of the inversion method in this paper is in good agreement with the actual model Better, higher resolution, more focused model, and faster convergence of the algorithm.