Transforming Decision Making in Carbonate Matrix Acidization Domain

Abstract

Abstract Acid stimulation treatments are typically designed using a combination of experience, available field/lab testing data and a stimulation design software that ties it all together. The final design, given the large degrees of uncertainty, requires multiple iterations to arrive at the (near) optimum stimulation design. To account for the large number of variables and their uncertainty, a framework and associated tools have been developed for automating and optimizing the design process that allow to significantly reduce the optimization time. The adopted approach leverages modern web applications for overcoming the limitations associated with the legacy stimulation design tools. Python Dash is used for creating an interactive web application, allowing users to load a file containing all necessary inputs for the calculation and specify the ranges for different design parameters. The web app then generates a parameter space and spawns multiple parallel runs of the calculation engine. An optimization module utilizing existing python optimization libraries is developed within the web app framework, which guides the selection of the next sample point in the parameter space after each iteration until the optimum point is reached. After implementing this framework across multiple workflows, considerable improvements have been realized with time savings ranging from hours to days. A detailed case study is presented to demonstrate the application of the framework for designing a Limited Entry Liner (LELs) completion, where the nozzle distribution in different reservoir sections (compartments) is varied to achieve a desired acid placement into the formation and improve stimulation efficiency.