Page 251 - Intelligent Digital Oil And Gas Fields

Integrated Asset Management and Optimization Workflows 201

Recent developments of optimization methods can be mainly divided into
deterministic and heuristic approaches. Deterministic approaches leverage
analytical properties of the problem and generate a sequence of points that
converge to a global optimal solution. Heuristic (often referred to as prob-
abilistic or stochastic) approaches are perceived to be more flexible and
efficient than deterministic methods; however, they tend to be more
computationally demanding most often since the probability of finding
the global solution decreases when the problem size increases.
Prevalent deterministic optimization approaches are linear program-
ming, mixed-integer linear programming (MILP), nonlinear programming,
and mixed-integer nonlinear programming (MINLP). The application of
linear programming in the oil and gas industry is to a large extent a matter
of the past with publications dating back to the late 1990s (Eeg and Herring,
1997), because they are most effective in solving (rather scarce) linear opti-
mization problems. One of the most comprehensive IAM planning solutions
developed using MILP was by Iyer et al. (1998), which proposes planning
and scheduling of investment and operation in offshore oil-field facilities by
the rigorous incorporation of nonlinear reservoir performance, surface pres-
sure constraints, and drilling rig resource constraints. Iyer et al. (1998) intro-
duce a tractable MILP model with several thousand binary variables and
sequential model decomposition strategy by the (dis)aggregation of time
periods and wells.
However, nonlinear programming and particularly MINLP can provide
general tools for solving optimization problems to obtain a global or an
approximately global optimum and are still actively pursued, for example,
in well-spacing optimization by maximizing NPV (John and Onjekonwu,
2010) and for generalized field development optimization in terms of
well-drilling locations and corresponding (time-varying) controls (Isebor
et al., 2014).
The rest of this section briefly reviews the mainstream optimization
approaches in exploration and production (E&P). For a comprehensive
overview of different optimization methodologies for decision making in
intelligent DOF, see Echeverria Ciaurri et al. (2012). For a technology sur-
vey of real-time optimization of offshore oil and gas production systems, see
Bieker et al. (2006). Temizel et al. (2014) outline the most prevalent advan-
tages and drawbacks of optimization techniques in real-time production
optimization of intelligent fields. Finally, for a comprehensive general
review of optimization techniques we recommend the encyclopedia of
optimization (Floudas and Pardalos, 2009) as an excellent resource.