Model-based lifecycle optimization of well locations and production settings in petroleum reservoirs

摘要: The coming years there is a need to increase production from petroleum reservoirs, and an enormous potential do so by increasing the recovery factor. This possible making better use of recent technological developments, such as horizontal wells, downhole valves sensors. However, actually these improved capabilities difficult because many open problems in reservoir management operations processes. Consequently, significant scope factor oil gas fields tailoring tools systems control community efficiently perform dynamic optimization wells (e.g. number, locations) their settings bottom-hole pressures, flow rates, valve settings) based on uncertain models, sense that they lead good decisions while requiring limited time user. thesis aims at developing tools, main contributions are follows. Many setting can be written optimal linear control. If only constraints upper lower bounds control, expected have pure bang-bang solutions. adjoint method derive gradients cost function with respect combined robust compute against uncertainty models. used directions which iteratively improve upon initial well configuration surrounding to-be-placed pseudo (i.e. operate negligible rate). controllability observability properties single-phase model analyzed. It shown pressures near we rate or pressure controllable, whereas measure observable. Finally, new regularization history matching presented, this analysis.