Optimization system for valve control in intelligent wells under uncertainties

Almeida, Luciana Faletti; Vellasco, Marley; Pacheco, Marco Aurélio Cavalcanti

doi:10.1016/j.petrol.2010.05.013

Cited by 56 publications

(33 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An important advance was achieved by employing evolutionary computation methods such as genetic algorithms (Alghareeb et al, 2009;Almeida et al, 2010), an efficient method of global optimization in sweeping for the optimal solution, or very close to it, in a feasible computational time but still dependent on the complexity of the reservoir model, the number of variables used (number of valves), and the computational power available. Despite being an efficient global optimization method in scanning for solutions, genetic algorithms are not efficient in finding the local maximum.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Another problem is an unfair comparison between two options, due to fact that IW and CW are not optimized in the best way before being compared. The CW is not optimized in terms of completion and the IW is not optimized in terms of number and placement of valves (Silva and Schiozer, 2009;Almeida et al, 2010). This is due to the fact that an evaluation of valve operation in the early stage of the field, in the phase of development planning, involves a large number of variables when representing future operations, which in turn leads to greater difficulty in solving the problem through classical optimization methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Assisted optimization method for comparison between conventional and intelligent producers considering uncertainties

Sampaio

Barreto

Schiozer

2015

Journal of Petroleum Science and Engineering

View full text Add to dashboard Cite

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Assisted optimization method for comparison between conventional and intelligent producers considering uncertainties

Sampaio

Barreto

Schiozer

2015

Journal of Petroleum Science and Engineering

View full text Add to dashboard Cite

“…Recently, Almeida et al (2010) used a genetic algorithm (GA) to select the best operation of the valves in the control of IW. A representation of chromosome was employed to allow the formulation of a strategy to control all valves, presenting an alternative, for a certain time.…”

Section: Literature Reviewmentioning

confidence: 99%

Optimization of Proactive Control Valves of Producer and Injector Intelligent Wells under Economic Uncertainty

Sampaio

Barreto

Schiozer

2012

SPE Europec/Eage Annual Conference

View full text Add to dashboard Cite

Intelligent wells can improve oil recovery, mitigate risks and avoid unnecessary well intervention in petroleum fields. However, there is no consolidated methodology to evaluate the applicability of intelligent wells and to represent them in commercial simulators, which complicates the comparison with conventional wells. Moreover, there are two main modes of operation of intelligent well valves, reactive and proactive; each one can provide different benefits. In general, proactive control seeks maximum oil recovery, but it requires larger computational effort and greater knowledge of the reservoir than the reactive control. This paper presents a comparison between different configurations of intelligent wells with proactive control and mode operation on/off: (1) five-spot configuration with conventional wells (producer and injectors), (2) one intelligent producer and four conventional injectors, (3) one conventional producer and four intelligent injectors and (4) one intelligent producer and four intelligent injectors, in order to compare the different behaviors. The objective of this study is to evaluate the potential of proactive operation for each type of configuration and the benefits of the intelligent injectors and producer acting separately or together, considering the effects on production and costs of intelligent completion. For this, a genetic algorithm was coupled to a commercial simulator to optimize the proactive control and to search the maximum net present value (NPV), determining the optimum operation control for each valve. The case study consists of one heterogeneous reservoir model, light oil and three economic scenarios (pessimistic, probable and optimistic). Results show that the use of intelligent injector and producer wells together, in this case study, can increase of oil production and decrease of water production, although it may not be the most advantageous alternative because of the higher investment. On the other hand, the configuration using only an intelligent producer well (lower investment) is capable of increasing oil recovery sufficiently, therefore making the best investment with intelligent completion, in this case study.

show abstract

“…Thus, the exact estimation of reservoir/formation properties will stay unknown and accordingly, optimization just based on a single and uncertain geological model would not be reliable. Applying various realizations of uncertain parameters with equal probability is an alternative way to rely on just one deterministic/stochastic value (van Essen et al, 2009;Alhuthali et al, 2010;Almeida et al, 2010;Chen and Hoo, 2012). This method is called Monte Carlo method and is a common way to deal with uncertain parameter values (Bayer et al, 2008;Carrera et al, 2005;Mantoglou and Kourakos, 2007;Baú, 2012).…”

Section: Introductionmentioning

confidence: 99%