Short-term Production Optimization by Automated Adaptive Modeling and Control

Abstract

Abstract Short-term production optimization relying on model-based predictions over a short period (weeks to months) requires the use a near-borehole reservoir model. Such a model is usually developed and validated through standard well testing. Standard well testing has to be repeated periodically, with related loss of production. Production losses may be reduced by prolonging the interval between tests, but that may compromise the quality of information about reservoir properties, such as skin (or productivity index), which would ultimately compromise production as well. Therefore, a need exists for a methodology that maximizes both reservoir information and production simultaneously. Because these two tasks are inherently contradictory, a compromise has to be found. In this work we propose a methodology that combines well testing and production in an optimal way, resulting in overall production optimization. This methodology relies on a short-term moving-horizon optimization of an objective function that includes terms referring both to the quality of reservoir information and to production net present value. Reservoir information is captured by empirical (proxy) models that are built adaptively on-line as a result of optimal perturbations of production rates and recording of dynamic responses of related bottomhole pressures. Besides, the entire workflow can be automated. Simulations are presented that illustrate the mechanics and value of the proposed methodology. Introduction The oil and gas industry is facing remarkable challenges to maximize profitability in a dynamic and uncertain environment while satisfying a variety of constraints. Current practices of production optimization involve combining mathematical models, field data and experience to make decisions about optimal production scenarios. In recent literature, a number of proxy modeling techniques [1–10] have been proposed where the output variables (oil recovery factor, multiphase flow rates etc.) are modeled as a function of the input variables. However, most of these methods focus on data-driven approaches such as response surface techniques based on regression, interpolation, neural network etc. These methods are relatively easy to setup and capture the nonlinear effects in the training data set. However, reservoir phenomena unseen in the past (e.g., water breakthrough) or operating regimes that lie outside the range of training data set are not adequately predicted by such models. Further, most proxy modeling approaches used in production optimization actually model the reservoir simulator outputs and are seldom validated against real field data. The authors of this paper have developed a parametric modeling methodology for Real-Time Production Optimization (RTPO) strategy [11,12]. Since the parametric model structure is derived from reservoir physics, it is expected that the model will be suitable to extrapolate outside the training data set. A feasible approach to continuous model updating and short-term forecasting using this approach was presented in [13].