Well-Test Optimization and Automation

Cramer, Ron; Jakeman, Simon Victor John; Berendschot, Leo

doi:10.2523/99971-ms

Cited by 9 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Historically, a variety of schemes for model parameters estimation have been developed [9, 10, 11, 12? ]. Similarly various techniques for experimental design could be adopted [13,14,15] depending on the objectives of measurements and constraints on the experimental setup. For instance, sampling based methods for experimental design allow one to investigate all possible combinations of model parameters and investigate their interactions.…”

Section: Introductionmentioning

confidence: 99%

Optimal Bayesian experimental design for subsurface flow problems

Тараканов

Elsheikh

2020

Computer Methods in Applied Mechanics and Engineering

View full text Add to dashboard Cite

Optimal Bayesian design techniques provide an estimate for the best parameters of an experiment in order to maximize the value of measurements prior to the actual collection of data. In other words, these techniques explore the space of possible observations and determine an experimental setup that produces maximum information about the system parameters on average. Generally, optimal Bayesian design formulations result in multiple high-dimensional integrals that are difficult to evaluate without incurring significant computational costs as each integration point corresponds to solving a coupled system of partial differential equations. In the present work, we propose a novel approach for development of polynomial chaos expansion (PCE) surrogate model for the design utility function. In particular, we demonstrate how the orthogonality of PCE basis polynomials can be utilized in order to replace the expensive integration over the space of possible observations by direct construction of PCE approximation for the expected information gain. This novel technique enables the derivation of a reasonable quality response surface for the targeted objective function with a computational budget comparable to several singlepoint evaluations. Therefore, the proposed technique reduces dramatically the overall cost of optimal Bayesian experimental design. We evaluate this alternative formulation utilizing PCE on few numerical test cases with various levels of complexity to illustrate the computational advantages of the proposed approach. * The current manuscipt is a preprint of the paper published in Computer Methods in Applied Mechanics and Engineering.

show abstract

Section: Introductionmentioning

confidence: 99%

Optimal Bayesian experimental design for subsurface flow problems

Тараканов

Elsheikh

2020

Computer Methods in Applied Mechanics and Engineering

View full text Add to dashboard Cite

show abstract

Advanced Production Monitoring

Hooimeijer

Azmi

2006

International Oil &Amp; Gas Conference and Exhibition in China

View full text Add to dashboard Cite

An effective and comprehensive integrated production-system monitoring framework plays an important role in the Smart Fields philosophy to turn data into actionable information and therefore value. This has been demonstrated by implementing some of the principal requirements for such a framework in the application called Data Broker in Shell. The Data Broker infrastructure has allowed Shell Engineers to perform exception-based surveillance. Used together with other Smart Fields applications such as FieldWare Production Universe, FieldWare Well Test, IPM and Energy Components, automated event recognition and workflows has freed up engineer's time to do more value added work, such as technical analysis and decision-making - time that would otherwise be spent looking for and validating data from many different sources. The system allows for a more efficient handling of data in terms of efficiency, quality and timeliness in reporting to end-users. The advent of real-time algorithms, monitoring tools, better integration software, event and workflow application tools allows the above infrastructure to be realised. Introduction There is a drive in the industry towards more advanced ways of managing hydrocarbon fields and many companies have initiated programs to implement new innovative ways of working [1]. These programs have been called various names: in Shell the program is called Smart Fields. These programs typically involve extending access to, and improving the quality of, production and metering data in the field so operators and engineers are aware of what the field is doing in real time. As fields are getting equipped with more and more gauges, large volumes of data have become available, both in the process control domain, and in the office domain. The industry is faced with the challenge to turn this data into value. The prize contained with the data is reduced deferment, optimised production, accurate reservoir characterisation and better security of supply. If this concept is combined with a right degree of automation it also has the potential to increase the efficiency of staff as they can focus on the real problem areas rather that spending a lot of time trying to analyse the data and identify the problems. The term that has been coined for the latter is " exception based surveillance". Exception based surveillance focuses on the exceptions that require a response from engineering staff. This means staff can spend their time working on real problems, and do not have to spend a lot of time looking for them. Surveys have shown that Engineers spend as much as 60% of their time to gather, QC, format, and convert data. At the heart of the Smart Fields concept are value loops (see Figure 1), which can be limited in scope and have short cycle times (such as control loops in a Process Control system) or cover a large scope and have a long cycle time (such as asset optimisation). In any case the loop needs to be closed, i.e. conclusion needs to be derived from information and action needs to be taken to improve the overall asset performance. If the loop is not closed, then there is a lot of wasted effort and investments are not capitalized on. For example, if you have invested in measuring down hole pressure, but data are not analyzed continuously and events (such as shut-ins) are not detected by staff in a timely fashion, then a lot of deferment can ensue. The loop from measured data to integrated modelling, event detection and follow up actions was not closed in that case. Deferment could have been reduced by a timely response on the basis of the data, and the gauge could have returned the value of purchase, installations and maintenance many times over. This paper first describes principal requirements for a comprehensive monitoring framework. In Shell an application called Data Broker has been developed that has implemented several of these principal requirements. In the description and application section, some of the implementation details of Data Broker are used to illustrate how one can realise a comprehensive monitoring framework. Practical examples are shown in the "Examples" section.

show abstract

Optimal Well-Testing Strategy for Production Optimization: A Monte Carlo Simulation Approach

Bieker¹,

Slupphaug²,

Johansen³

2006

SPE Eastern Regional Meeting

View full text Add to dashboard Cite

TX 75083-3836 U.S.A., fax 01-972-952-9435. AbstractWell testing may be performed to support many decisions including ones related to production optimization of an oil production system. In production optimization information such as gas oil ratio and water cut/water oil ratio is used to decide on, for example, which wells to prioritize for choking back/opening to avoid over-/underutilization of the production capacity. Since the reservoir properties change with time, the uncertainties of their estimated values increase with time, and eventually a new well test will be required. As the uncertainty in the estimates grows, so does the risk for prioritizing the wrong well(s) giving lower oil production rate than possible. A computer program is developed to decide which well to test based on historical test data. The program uses a Monte Carlo approach for identifying which well test is likely to lead to highest increase in the total oil production rate after the well test information is utilized to optimize the oil production. The computer program is applied to field data quantifying the benefits when applied to this specific field.

show abstract

Well-Test Optimization and Automation

Cited by 9 publications

References 0 publications

Optimal Bayesian experimental design for subsurface flow problems

Optimal Bayesian experimental design for subsurface flow problems

Advanced Production Monitoring

Optimal Well-Testing Strategy for Production Optimization: A Monte Carlo Simulation Approach

Contact Info

Product

Resources

About