Adjoint Method used in History Matching and Optimization Workflows
The focus of this R&D project is the development of a new integrated workflow using state of the art optimization techniques combined with the adjoint method for history matching. The integration concept is aimed at developing a robust workflow supporting decision processes for evaluating alternative production scenarios of oil and gas fields. For this purpose, SenEx, which is an advanced sensitivity computation tool capable of modifying rock properties on grid-block level, has been linked to the existing optimization framework MEPO. The Hotlink thus developed has been tested, reviewed and enhanced. Within the framework of this re-search project, the developed workflows have been applied to a spectrum of history match-ing problems with varying complexities.
In general, multiple geological realizations were used to capture existing uncertainty in an extended optimization workflow, including history matching and production optimization. Us-ing this method, history match is typically achieved with significantly less number of iterations compared to existing methods. The workflow applied is useful for evaluating alternative re-development scenarios, quantifying uncertainties and generating realistic history matched models. Benchmarking work performed provided better insights into benefits and limitations of the adjoint method.