Combined Monte Carlo and Gradient Procedures Used to Optimized Solution of the Comprehensive Interpretation of Well Logs

Abstract

The comprehensive interpretation of well logs is a mathematical task on a solution of a set of equations. The number of equations is determined by the number of the logs and conditions determining boundary and geological requirements. The common methods (e.g. gradient method, GRAD) are limited to cases when one global minimum occurs in the examined range of variables. Only Monte Carlo method, MC, which consists on numerous solutions obtained by random sampling of possible variables can find the global minimum of the function when other local minima are present. The MC method is time consuming, especially, in application to the well logging interpretation when enormous number of points along the depth profile should be calculated. We propose a complex method of solution combining both MC and GRAD methods. The method has been applied to the Interlog application in the GeoWin system for processing and interpretation of well logs and the field example is presented in the paper. The results confirm that the combined MC-GRAD solutions can give more information than the simple GRAD solution. Presented approach can be treated as an additional tool, especially dedicated for log analysts, for solving a set of logs difficult to interpretation.