Multiscale Formation Resistivity Imaging

摘要

A conventional interpretation technique for array resistivityrnlog data is based on inversion. Log inversion methodsrncommonly use a single formation model for the entirerninterpretation process. This single model approach has strongrnlimitations when executing either joint or sequentialrninversions. Here we present a new multiscale array resistivityrndata inversion method that utilizes a set of multiscale, logresolution-rndependent formation models determined for eachrnsubarray log involved in the interpretation process.rnOur method consists of two main interpretation runs. Forrnthe first run, we interpret separate subarray logs using a set ofrnmultiscale models. For the second, we generate the formationrnresistivity model or its image by inverting the results from thernfirst. Application of multiscale models leads to a rapid andrnstable reconstruction of the deep formation structure consistentrnwith all subarray logs.rnThe method is applicable to any array log data acquired inrnvertical, deviated, and horizontal wells. To evaluate thernconcept of our method, we performed a series of 2-Drnbenchmark modeling and inversion tests using array inductionrndata.