This work describes a new methodology for integrated decision analysis in the development and management of petroleum fields considering reservoir simulation, risk analysis, history matching, ...uncertainty reduction, representative models, and production strategy selection under uncertainty. Based on the concept of closed-loop reservoir management, we establish 12 steps to assist engineers in model updating and production optimization under uncertainty. The methodology is applied to UNISIM-I-D, a benchmark case based on the Namorado field in the Campos Basin, Brazil. The results show that the method is suitable for use in practical applications of complex reservoirs in different field stages (development and management). First, uncertainty is characterized in detail and then scenarios are generated using an efficient sampling technique, which reduces the number of evaluations and is suitable for use with numerical reservoir simulation. We then perform multi-objective history-matching procedures, integrating static data (geostatistical realizations generated using reservoir information) and dynamic data (well production and pressure) to reduce uncertainty and thus provide a set of matched models for production forecasts. We select a small set of Representative Models (RMs) for decision risk analysis, integrating reservoir, economic and other uncertainties to base decisions on risk-return techniques. We optimize the production strategies for (1) each individual RM to obtain different specialized solutions for field development and (2) all RMs simultaneously in a probabilistic procedure to obtain a robust strategy. While the second approach ensures the best performance under uncertainty, the first provides valuable insights for the expected value of information and flexibility analyses. Finally, we integrate reservoir and production systems to ensure realistic production forecasts. This methodology uses reservoir simulations, not proxy models, to reliably predict field performance. The proposed methodology is efficient, easy-to-use and compatible with real-time operations, even in complex cases where the computational time is restrictive.
This paper presents a novel few-shot proxy modeling approach for the oil and gas industry to reduce reliance on numerical simulators for reservoir analysis. The strategy introduces a regression ...framework leveraging deep hierarchical self-distillation to construct a meta-model based on ensemble learning. The proposed method employs a cascade training scheme, which uses predictions derived from a superior hierarchical level to distill knowledge into the next predictor. The pivotal idea of self-distillation is to generate “soft targets” rather than hard ones. Soft targets represent probability distributions over potential output curves rather than a correct answer. This smoothing information provides additional guidance to the model during training, helping it generalize better. This architecture utilizes three-dimensional (3D) maps in proxy modeling to forecast cumulative fluid production and generate risk curves indispensable for effective field decisions. The study employed data acquired from a complex oil field characterized by a substantial degree of uncertainty. The hierarchical self-distillation technique outperforms alternative methods, achieving a symmetric mean absolute percentage error (SMAPE) below 2%. It reduces computational overhead by 84% for a probabilistic model with 900 simulations and 62% for a model with 200 simulations. The developed proxy model delivers valuable insights for decision-making in oil field management, offering the potential to decrease expenses and enhance efficiency in field exploration endeavors.
Managing geological uncertainties in reservoir engineering involves significant challenges mainly due to the prohibitive computational costs of traditional simulation methods. These simulations, ...essential for generating geological models, often require extensive computational resources and can take days or weeks to complete. The computational load limits the number of viable models while the high dimensionality of properties compounds the challenge, and the abundance of producing wells, each associated with different objective functions, further complicates the problem. Current methodologies have focused on training an Artificial Neural Network (ANN) for each producer well to create a proxy model. Instead, we propose the development of a single ANN designed to simultaneously predict the behavior of multiple objective functions. This work proposes a novel end-to-end deep neural network that can handle 3D geological uncertainties and replicate the simulator’s response for several cumulative production curves. This approach leverages 3D convolutions to process spatial and depth dimensions within heterogeneous reservoirs, including diverse geostatistical realizations. The end-to-end solution was successfully implemented in a Brazilian offshore field and accurately replicated the simulator’s behavior and yielded results that outperformed state-of-the-art methods. The results indicate correlations that exceeded 0.95 between the proxy response and the numerical simulator. Additionally, our approach demonstrated robustness even when trained with a limited number of simulation models, and it notably reduced computational costs. The findings highlight our architecture’s capacity to integrate dimensional reduction and regression analysis within a unified framework, effectively predicting different fluid behaviors in the reservoir and showcasing robustness against high dimensionality and sparse data.
•End-to-end proxy capable of managing 3D geological uncertainties.•Dealing with high dimensionality to reproduce the simulator response quickly.•We only train one network for all fluids, unlike most approaches.•Time efficient, requiring 9 min for training, testing and prediction.•Significantly cuts training time and resources, enhancing computational efficiency.
The use of time-lapse seismic data to improve reservoir characterization is becoming a common practice in the oil industry. Nevertheless, the integration of datasets with different characteristics, ...such as flow simulation and seismic data, is still a challenge. One of the possible ways to perform the integration is the use of extracted pressure and saturation from 4D seismic in the history matching process. However, the quantitative use of pressure and saturation difference maps in the objective function needs more accurate estimation of these dynamic properties. Thus, this work proposes a methodology to use multiple simulation model realizations, generated through the combination of uncertain reservoir attributes, to guide an inversion process that evaluates pressure and saturation from 4D seismic, in order to provide estimations that are more reliable. The application of the methodology in a synthetic dataset showed promising results. The main contribution of this work is to show that it is possible to use available knowledge from flow simulation and reservoir characterization to constrain time-lapse data and extract from it more reliable information.
Data assimilation (also known as history matching) and uncertainty assessment is the process of conditioning reservoir models to dynamic data to improve its production forecast capacity. One of the ...main challenges of the process is the representation and updating of spatial properties in a geologically consistent way. The process is even more challenging for complex geological systems such as highly channeling reservoirs, fractured systems and super-K layered reservoirs. Therefore, mainly for highly heterogeneous reservoirs, a proper parameterization scheme is crucial to ensure an effective and consistent process. This paper presents a new approach based on cumulative distribution function (CDF) for parameterization of complex geological models focused on layered reservoir with the presence of high permeability zones (super-K). The main innovative aspect of this work is focused on a new sampling procedure based on a cut-off frequency. The proposed method is simple to implement and, at the same time, very robust. It is able to properly represent super-K distribution along the reservoir during the data assimilation process, obtaining good data matches and reducing the uncertainty in the production forecast. The new method, which preserves the prior characteristics of the model, was tested in a complex carbonate reservoir model (UNISIM-II-H benchmark case) built based on a combination of Brazilian Pre-salt characteristics and Ghawar field information available in the literature. Promising results, which indicate the robustness of the method, are shown.
•A new parameterization method for data assimilation and uncertainty assessment was proposed.•The method is suitable for complex geological models with the presence of high permeability zones (super-K).•The proposed method is simple to implement and, at the same time, very robust.•The new method preserves the prior characteristics of the model.•The proposed method was validated in a complex carbonate reservoir model.
We present an efficient and risk-informed closed-loop field development (CLFD) workflow for recurrently revising the field development plan (FDP) using the accrued information. To make the process ...practical, we integrated multiple concepts of machine learning, an intelligent selection process to discard the worst FDP options and a growing set of representative reservoir models. These concepts were combined and used with a cluster-based learning and evolution optimizer to efficiently explore the search space of decision variables. Unlike previous studies, we also added the execution time of the CLFD workflow and worked with more realistic timelines to confirm the utility of a CLFD workflow. To appreciate the importance of data assimilation and new well-logs in a CLFD workflow, we carried out researches at rigorous conditions without a reduction in uncertainty attributes. The proposed CLFD workflow was implemented on a benchmark analogous to a giant field with extensively time-consuming simulation models. The results underscore that an ensemble with as few as 100 scenarios was sufficient to gauge the geological uncertainty, despite working with a giant field with highly heterogeneous characteristics. It is demonstrated that the CLFD workflow can improve the efficiency by over 85% compared to the previously validated workflow. Finally, we present some acute insights and problems related to data assimilation for the practical application of a CLFD workflow.
This paper introduces a new methodology, combining a Genetic Algorithm (GA) with multi-start simulated annealing to integrate Geostatistical Realizations (GR) in data assimilation and uncertainty ...reduction process. The proposed approach, named Genetic Algorithm with Multi-Start Simulated Annealing (GAMSSA), comprises two parts. The first part consists of running a GA several times, starting with certain number of geostatistical realizations, and the second part consists of running the Multi-Start Simulated Annealing with Geostatistical Realizations (MSSAGR). After each execution of GA, the best individuals of each generation are selected and used as starting point to the MSSAGR. To preserve the diversity of the geostatistical realizations, a rule is imposed to guarantee that a given realization is not repeated among the selected individuals from the GA. This ensures that each Simulated Annealing (SA) process starts from a different GR. Each SA process is responsible for local improvement of the best individuals by performing local perturbation in other reservoir properties such as relative permeability, water-oil contact, etc. The proposed methodology was applied to a complex benchmark case (UNISIM-I-H) based on the
Namorado Field
, located in the Campos Basin, Brazil, with 500 geostatistical realizations and other 22 attributes comprising relative permeability, oil-water contact, and rock compressibility. Comparisons with a conventional GA algorithm are also shown. The proposed method was able to find multiple solutions while preserving the diversity of the geostatistical realizations and the variability of the other attributes. The matched models found by the GAMSSA method provided more reliable forecasts when compared with the matched models found by the GA.
Reservoir management decisions are often based on simulation models and probabilistic approaches. Thus, the response of the model must be sufficiently accurate to base sound decisions on and fast ...enough to be practical for methodologies requiring many simulation runs. However, simulation models often forecast production rates different to real production rates for various reasons. Two possible causes of these deviations are (1) upscaling (a technique to reduce the computational time of simulation models by reducing the number of grid blocks) and (2) uncertainties (the values established to attributes are different from real values caused by lack of knowledge of real reservoir). Morosov and Schiozer (2016) applied a closed-loop technique in a benchmark case where decisions taken using the simulation models are applied to a reference case. The optimized production strategy, using simulations models, increased the expected monetary value of the project by about 29%, but the Net Present Value (NPV), calculated using a reference case, decreased by 2%. The real NPV was outside the expected range and revealed that the set of models did not fully represent the real field, even for high-quality history-matched models. The objective of this study is to identify the causes of these discrepancies. To reach this goal, we investigate and analyze both the impact of the upscaling and the uncertainty on production and economic indicators. We use a set of representative models of benchmark UNISIM-I (Avansi and Schiozer, 2015) to consider the effects of uncertainty and upscaling. Our main concern was the uncertainties in the distribution of petrophysical properties that strongly influence the productivity and injectivity of wells, noted by Morosov and Schiozer (2016) as being the main cause for differences among models. Furthermore, to verify the isolated effects of the possible causes of deviation, we use a single model to show only the effects of upscaling, and another set of models showing only the uncertainty. The results showed that the impact of the uncertainties was higher than the upscaling for the studied case. The upscaling generated an optimistic bias for production and economic indicators, but well-correlated with the reference case. The uncertainties significantly affected the production forecasts for this study. This happened because the response of the wells is highly dependent on the petrophysical properties of the model, which varies widely between the different models representing uncertainties and was not adequately depicted by the representative models.