Development of a Multi-Solution Framework for Simultaneous Well Placement, Completion, and Control Optimization

Optimal field development and control aim to maximize the economic profit of oil and gas production. This,however, results in a complex optimization problem with a large number of correlated control variables atdifferent levels (e.g. well locations, completions and controls) and a computationally expensive objectivefunction (i.e. a simulated reservoir model). The typical limitations of the existing optimization frameworksare: (1) single-level optimization at a time (i.e. ignoring correlations among control variables at differentlevels); and (2) providing a single solution only whereas operational problems often add unexpectedconstraints likely to reduce the -optimal-, inflexible solution to a sub-optimal scenario. The developed framework in this paper is based on sequential iterative optimization of control variablesat different levels. An ensemble of close-To-optimum solutions is selected from each level (e.g. for welllocation) and transferred to the next level of optimization (e.g.To control settings), and this loop continuesuntil no significant improvement is observed in the objective value. Fit-for-purpose clustering techniques aredeveloped to systematically select an ensemble of solutions, with maximum differences in control variablesbut close-To-optimum objective values, at each level of optimization. The framework also considers pre-defined constraints such as the minimum well spacing, irregular reservoir boundaries, and production/injection rate limits. The proposed framework has been tested on a benchmark case study, known as the Brugge field, to findthe optimal well placement and control in two development scenarios: with conventional (surface controlonly) and intelligent wells (with additional zonal control using Interval Control Valves). Multiple solutionsare obtained in both development scenarios, with different well locations and control settings but close-To-optimum objective values. We also show that suboptimal solutions from an early optimization level canapproach and even outdo the optimal one at the higher-level optimization, highlighting the value of the here-developed multi-solution framework in exploring the search space as compared to the traditional single-solution approaches. The development scenario with intelligent completion installed at the optimal welllocation and optimally controlled during the production period achieved the maximum added value. Ourresults demonstrate the advantage of the developed multi-solution optimization framework in providing themuch-needed operational flexibility to field operators.