scholarly journals Application of nature-inspired algorithms and artificial neural network in waterflooding well control optimization

2021 ◽  
Author(s):  
Cuthbert Shang Wui Ng ◽  
Ashkan Jahanbani Ghahfarokhi ◽  
Menad Nait Amar
Keyword(s):  
Case Studies ◽  
Net Present Value ◽  
Computational Cost ◽  
Data Driven ◽  
Coefficient Of Determination ◽  
Reservoir Model ◽  
Well Control ◽  
Optimization Task ◽  
Proxy Models ◽  
Artificial Neural

AbstractWith the aid of machine learning method, namely artificial neural networks, we established data-driven proxy models that could be utilized to maximize the net present value of a waterflooding process by adjusting the well control injection rates over a production period. These data-driven proxies were maneuvered on two different case studies, which included a synthetic 2D reservoir model and a 3D reservoir model (the Egg Model). Regarding the algorithms, we applied two different nature-inspired metaheuristic algorithms, i.e., particle swarm optimization and grey wolf optimization, to perform the optimization task. Pertaining to the development of the proxy models, we demonstrated that the training and blind validation results were excellent (with coefficient of determination, R2 being about 0.99). For both case studies and the optimization algorithms employed, the optimization results obtained using the proxy models were all within 5% error (satisfied level of accuracy) compared with reservoir simulator. These results confirm the usefulness of the methodology in developing the proxy models. Besides that, the computational cost of optimization was significantly reduced using the proxies. This further highlights the significant benefits of employing the proxy models for practical use despite being subject to a few constraints.

Fast History Matching and Optimization Using a Novel Physics-Based Data-Driven Model: An Application to a Diatomite Reservoir

2021 ◽  
Author(s):  
Zhenzhen Wang ◽  
Jincong He ◽  
William J. Milliken ◽  
Xian-Huan Wen
Keyword(s):  
Capillary Pressure ◽  
Network Model ◽  
History Matching ◽  
Net Present Value ◽  
Analytic Solutions ◽  
Data Driven ◽  
Production Data ◽  
Well Control ◽  
Compositional Flow ◽  
Control Optimization

Abstract Full-physics models in history matching and optimization can be computationally expensive since these problems usually require hundreds of simulations or more. We have previously implemented a physics-based data-driven network model with a commercial simulator that serves as a surrogate without the need to build the 3-D geological model. In this paper, we reconstruct the network model to account for complex reservoir conditions of mature fields and successfully apply it to a diatomite reservoir in the San Joaquin Valley (SJV) for rapid history matching and optimization. The reservoir is simplified into a network of 1-D connections between well perforations. These connections are discretized into grid blocks and the grid properties are calibrated to historical production data. Elevation change, saturation distribution, capillary pressure, and relative permeability are accounted for to best represent the mature field conditions. To simulate this physics-based network model through a commercial simulator, an equivalent 2-D Cartesian model is designed where rows correspond to the above-mentioned connections. Thereafter, the history matching can be performed with the Ensemble Smoother with Multiple Data Assimilation (ESMDA) algorithm under a sequential iterative process. A representative model after history matching is then employed for well control optimization. The network model methodology has been successfully applied to the waterflood optimization for a 56-well sector model of a diatomite reservoir in the SJV. History matching result shows that the network model honors field-level production history and gives reasonable matches for most of the wells, including pressure and flow rate. The calibrated ensemble from the last iteration of history matching yields a satisfactory production prediction, which is verified by the remaining historical data. For well control optimization, we select the P50 model to maximize the Net Present Value (NPV) in 5 years under provided well/field constraints. This confirms that the calibrated network model is accurate enough for production forecasts and optimization. The use of a commercial simulator in the network model provided flexibility to account for complex physics, such as elevation difference between wells, saturation non-equilibrium, and strong capillary pressure. Unlike traditional big-loop workflow that relies on a detailed characterization of geological models, the proposed network model only requires production data and can be built and updated rapidly. The model also runs much faster (tens of seconds) than a full-physics model due to the employment of much fewer grid blocks. To our knowledge, this is the first time this physics-based data-driven network model is applied with a commercial simulator on a field waterflood case. Unlike approaches developed with analytic solutions, the use of commercial simulator makes it feasible to be further extended for complex processes, e.g., thermal or compositional flow. It serves as an useful surrogate model for both fast and reliable decision-making in reservoir management.

scholarly journals Stochastic optimization strategies applied to the OLYMPUS benchmark

2019 ◽  
Vol 24 (6) ◽  
pp. 1943-1958 ◽  
Author(s):  
V. L. S. Silva ◽  
M. A. Cardoso ◽  
D. F. B. Oliveira ◽  
R. J. de Moraes
Keyword(s):  
Stochastic Optimization ◽  
Net Present Value ◽  
Computational Cost ◽  
Real Life ◽  
Sequential Optimization ◽  
Optimization Approach ◽  
Well Placement ◽  
Well Control ◽  
Field Development ◽  
Control Optimization

AbstractIn this work, we discuss the application of stochastic optimization approaches to the OLYMPUS case, a benchmark challenge which seeks the evaluation of different techniques applied to well control and field development optimization. For that matter, three exercises have been proposed, namely, (i) well control optimization; (ii) field development optimization; and (iii) joint optimization. All applications were performed considering the so-called OLYMPUS case, a synthetic reservoir model with geological uncertainty provided by TNO (Fonseca 2018). Firstly, in the well control exercise, we successfully applied an ensemble-based approximate gradient method in a robust optimization formulation. Secondly, we solve the field development exercise using a genetic algorithm framework designed with special features for the problem of interest. Finally, in order to evaluate further gains, a sequential optimization approach was employed, in which we run one more well control optimization based on the optimal well locations. Even though we utilize relatively well-known techniques in our studies, we describe the necessary adaptations to the algorithms that enable their successful applications to real-life scenarios. Significant gains in the expected net present value are obtained: in exercise (i) a gain of 7% with respect to reactive control; for exercise (ii) a gain of 660% with respect to a initial well placement based on an engineering approach; and for (iii) an extra gain of 3% due to an additional well control optimization after the well placement optimization. All these gains are obtained with an affordable computational cost via the extensive utilization of high-performance computing (HPC) infrastructure. We also apply a scenario reduction technique to exercise (i), with similar gains obtained in the full ensemble optimization, however, with substantially inferior computational cost. In conclusion, we demonstrate how the state-of-the-art optimization technology available in the model-based reservoir management literature can be successfully applied to field development optimization via the conscious utilization of HPC facilities.

TLBO-FLN: Teaching-Learning Based Optimization of Functional Link Neural Networks for Stock Closing Price Prediction

2020 ◽  
Vol 10 (4) ◽  
pp. 522-532 ◽  
Author(s):  
Sarat Chandra Nayak ◽  
Subhranginee Das ◽  
Mohammad Dilsad Ansari
Keyword(s):  
Neural Networks ◽  
Computational Cost ◽  
Optimization Techniques ◽  
Fine Tuning ◽  
Functional Link ◽  
Price Prediction ◽  
Closing Price ◽  
Teaching Learning Based Optimization ◽  
Artificial Neural ◽  
Teaching Learning

Background and Objective: Stock closing price prediction is enormously complicated. Artificial Neural Networks (ANN) are excellent approximation algorithms applied to this area. Several nature-inspired evolutionary optimization techniques are proposed and used in the literature to search the optimum parameters of ANN based forecasting models. However, most of them need fine-tuning of several control parameters as well as algorithm specific parameters to achieve optimal performance. Improper tuning of such parameters either leads toward additional computational cost or local optima. Methods: Teaching Learning Based Optimization (TLBO) is a newly proposed algorithm which does not necessitate any parameters specific to it. The intrinsic capability of Functional Link Artificial Neural Network (FLANN) to recognize the multifaceted nonlinear relationship present in the historical stock data made it popular and got wide applications in the stock market prediction. This article presents a hybrid model termed as Teaching Learning Based Optimization of Functional Neural Networks (TLBO-FLN) by combining the advantages of both TLBO and FLANN. Results and Conclusion: The model is evaluated by predicting the short, medium, and long-term closing prices of four emerging stock markets. The performance of the TLBO-FLN model is measured through Mean Absolute Percentage of Error (MAPE), Average Relative Variance (ARV), and coefficient of determination (R2); compared with that of few other state-of-the-art models similarly trained and found superior.

A two-stage design optimization framework for multifield origami-inspired structures

2021 ◽  
pp. 1045389X2110116
Author(s):  
Wei Zhang ◽  
Saad Ahmed ◽  
Jonathan Hong ◽  
Zoubeida Ounaies ◽  
Mary Frecker
Keyword(s):  
Case Studies ◽  
Computing Time ◽  
Computational Cost ◽  
High Fidelity ◽  
Fe Model ◽  
Two Stage ◽  
Baseline Design ◽  
Optimization Framework ◽  
Highly Nonlinear ◽  
Stage 1

Different types of active materials have been used to actuate origami-inspired self-folding structures. To model the highly nonlinear deformation and material responses, as well as the coupled field equations and boundary conditions of such structures, high-fidelity models such as finite element (FE) models are needed but usually computationally expensive, which makes optimization intractable. In this paper, a computationally efficient two-stage optimization framework is developed as a systematic method for the multi-objective designs of such multifield self-folding structures where the deformations are concentrated in crease-like areas, active and passive materials are assumed to behave linearly, and low- and high-fidelity models of the structures can be developed. In Stage 1, low-fidelity models are used to determine the topology of the structure. At the end of Stage 1, a distance measure [Formula: see text] is applied as the metric to determine the best design, which then serves as the baseline design in Stage 2. In Stage 2, designs are further optimized from the baseline design with greatly reduced computing time compared to a full FEA-based topology optimization. The design framework is first described in a general formulation. To demonstrate its efficacy, this framework is implemented in two case studies, namely, a three-finger soft gripper actuated using a PVDF-based terpolymer, and a 3D multifield example actuated using both the terpolymer and a magneto-active elastomer, where the key steps are elaborated in detail, including the variable filter, metrics to select the best design, determination of design domains, and material conversion methods from low- to high-fidelity models. In this paper, analytical models and rigid body dynamic models are developed as the low-fidelity models for the terpolymer- and MAE-based actuations, respectively, and the FE model of the MAE-based actuation is generalized from previous work. Additional generalizable techniques to further reduce the computational cost are elaborated. As a result, designs with better overall performance than the baseline design were achieved at the end of Stage 2 with computing times of 15 days for the gripper and 9 days for the multifield example, which would rather be over 3 and 2 months for full FEA-based optimizations, respectively. Tradeoffs between the competing design objectives were achieved. In both case studies, the efficacy and computational efficiency of the two-stage optimization framework are successfully demonstrated.

Feed-forward artificial neural network provides data-driven inference of functional connectivity

2019 ◽  
Vol 29 (9) ◽  
pp. 091101 ◽  
Author(s):  
Nikita Frolov ◽  
Vladimir Maksimenko ◽  
Annika Lüttjohann ◽  
Alexey Koronovskii ◽  
Alexander Hramov
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Functional Connectivity ◽  
Data Driven ◽  
Feed Forward ◽  
Artificial Neural

Multidisciplinary Optimization of Airfoil Cooling Structure

2008 ◽  
Author(s):  
Grzegorz Nowak
Keyword(s):  
Cooling System ◽  
Computational Cost ◽  
System Optimization ◽  
Multidisciplinary Optimization ◽  
Internal Cooling ◽  
Technical Problems ◽  
Thermal Boundary Conditions ◽  
Optimization Task ◽  
External Conditions ◽  
Cooling Air

This paper discusses the problem of cooling system optimization within a gas turbine airfoil regarding to thermo-mechanical behavior of the component, as well as some economical aspects of turbine operation. The main goal of this paper is to show the possibilities of evolutionary approach application to the cooling system optimization. This method, despite its relatively high computational cost, seems to be a valuable tool to such technical problems. The analysis involves the optimization of location and size of internal cooling passages within an airfoil. Initially cooling is provided with circular passages and heat is transported by convection. During the optimization the number of channels can vary. The task is approached in 3D configuration. Each passage is fed with cooling air of constant parameters at the inlet. Also a constant pressure drop is assumed along the passage length. The thermal boundary conditions in passages vary with diameter and local vane temperature (passage wall temperature). The analysis is performed by means of the genetic algorithm for the optimization task and FEM for the heat transfer predictions within the component. In the present study the airfoil profile is taken as aerodynamically optimal and the objective of the search procedure is to find cooling structure variant that at given external conditions provides lower stresses, material temperature and indirectly coolant usage.

scholarly journals Over-parameterisation, a major obstacle to the use of artificial neural networks in hydrology?

2003 ◽  
Vol 7 (5) ◽  
pp. 693-706 ◽  
Author(s):  
E. Gaume ◽  
R. Gosset
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Conceptual Model ◽  
Linear Model ◽  
Case Studies ◽  
Stream Flow ◽  
River Flow ◽  
Prediction Models ◽  
Rainfall Runoff ◽  
Artificial Neural

Abstract. Recently Feed-Forward Artificial Neural Networks (FNN) have been gaining popularity for stream flow forecasting. However, despite the promising results presented in recent papers, their use is questionable. In theory, their “universal approximator‿ property guarantees that, if a sufficient number of neurons is selected, good performance of the models for interpolation purposes can be achieved. But the choice of a more complex model does not ensure a better prediction. Models with many parameters have a high capacity to fit the noise and the particularities of the calibration dataset, at the cost of diminishing their generalisation capacity. In support of the principle of model parsimony, a model selection method based on the validation performance of the models, "traditionally" used in the context of conceptual rainfall-runoff modelling, was adapted to the choice of a FFN structure. This method was applied to two different case studies: river flow prediction based on knowledge of upstream flows, and rainfall-runoff modelling. The predictive powers of the neural networks selected are compared to the results obtained with a linear model and a conceptual model (GR4j). In both case studies, the method leads to the selection of neural network structures with a limited number of neurons in the hidden layer (two or three). Moreover, the validation results of the selected FNN and of the linear model are very close. The conceptual model, specifically dedicated to rainfall-runoff modelling, appears to outperform the other two approaches. These conclusions, drawn on specific case studies using a particular evaluation method, add to the debate on the usefulness of Artificial Neural Networks in hydrology. Keywords: forecasting; stream-flow; rainfall-runoff; Artificial Neural Networks

A Proxy Peng-Robinson EOS for Efficient Modeling of Phase Behavior

2021 ◽  
Author(s):  
Mark Zhao ◽  
Ryosuke Okuno
Keyword(s):  
Phase Behavior ◽  
Computational Cost ◽  
Training Data ◽  
Petroleum Engineering ◽  
Fugacity Coefficient ◽  
Fugacity Model ◽  
Proxy Model ◽  
Reservoir Conditions ◽  
Fugacity Coefficients ◽  
Proxy Models

Abstract Equation-of-state (EOS) compositional simulation is commonly used to model the interplay between phase behavior and fluid flow for various reservoir and surface processes. Because of its computational cost, however, there is a critical need for efficient phase-behavior calculations using an EOS. The objective of this research was to develop a proxy model for fugacity coefficient based on the Peng-Robinson EOS for rapid multiphase flash in compositional flow simulation. The proxy model as implemented in this research is to bypass the calculations of fugacity coefficients when the Peng-Robinson EOS has only one root, which is often the case at reservoir conditions. The proxy fugacity model was trained by artificial neural networks (ANN) with over 30 million fugacity coefficients based on the Peng-Robinson EOS. It accurately predicts the Peng- Robinson fugacity coefficient by using four parameters: Am, Bm, Bi, and ΣxiAij. Since these scalar parameters are general, not specific to particular compositions, pressures, and temperatures, the proxy model is applicable to petroleum engineering applications as equally as the original Peng-Robinson EOS. The proxy model is applied to multiphase flash calculations (phase-split and stability), where the cubic equation solutions and fugacity coefficient calculations are bypassed when the Peng-Robinson EOS has one root. The original fugacity coefficient is analytically calculated when the EOS has more than one root, but this occurs only occasionally at reservoir conditions. A case study shows the proxy fugacity model gave a speed-up factor of 3.4% in comparison to the conventional EOS calculation. Case studies also demonstrate accurate multiphase flash results (stability and phase split) and interchangeable proxy models for different fluid cases with different (numbers of) components. This is possible because it predicts the Peng-Robinson fugacity in the variable space that is not specific to composition, temperature, and pressure. For the same reason, non-zero binary iteration parameters do not impair the applicability, accuracy, robustness, and efficiency of the model. As the proxy models are specific to individual components, a combination of proxy models can be used to model for any mixture of components. Tuning of training hyperparameters and training data sampling method helped reduce the mean absolute percent error to less than 0.1% in the ANN modeling. To the best of our knowledge, this is the first generalized proxy model of the Peng-Robinson fugacity that is applicable to any mixture. The proposed model retains the conventional flash iteration, the convergence robustness, and the option of manual parameter tuning for fluid characterization.

Sign in / Sign up

Export Citation Format

Share Document