SEARCH · Search NASA

Results for “optimal design parameters”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 181 records · Page 10

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Machine learning-based optimization of air-cooled heat sinks

Machine learning-based models using Artificial Neural Network (ANN) and greedy search algorithm are used to optimize air-cooled parallel plate-finned heat sinks (PPFHSs) subjected to laminar flow over an extensive range of design parameters. Here, the thermal and hydraulic performances of PPFHSs are represented by heat transfer coefficient (h) and pressure drop (ΔP), respectively. Optimization objectives for PPFHS designs can vary from industry to industry depending on their design priorities. The present study proposes a novel and generalized optimization method that defines practical optimization objectives and provides an accurate optimization process to design effective PPFHSs for a wide range of industrial applications with different design requirements. Three optimization objectives are presented in this study: (i) the largest h PΔ, (ii) the largest h within a specified maximum allowed flow rate, and (iii) the lowest weight that maximizes h for operation within the maximum allowed flow rate. While the shortcoming of the first objective is demonstrated, the other two objectives are found to be suitable for designing effective heat sinks (HSs) across different applications. Results suggest a promising trend from the third objective to develop HSs with ~ 37-68% lower weight, 80-85% reduced ΔP, and negligible penalty in h compared with optimized HSs obtained from the second objective. However, since the third objective leads to HSs with thinner fins, structural analysis should be performed to ensure reliable operation of the HSs.

42 ENGINEERING↗

DOE OSTI2020

Extreme Fast Charging Lithium-Ion Batteries (Final Technical Report)

The objective of this research project is to develop, design, fabricate, and demonstrate lithium ion battery cells that optimize energy density, reduce cost, and demonstrate eXtreme Fast Charging (XFC) capabilities. The XFC capability of the cell will be determined by the final specific energy of the cell after completing 500 6C charge /1C discharge cycles. Over the first 18 months of the project, over 800 cells were produced in groups of 20 cells to determine the optimum configuration of following parameters to optimized XFC performance: (1) Graphite anode material optimization (2) Electrolyte and additive optimization (3) Electrode and cell design optimization. In the final 6 months of the project, over 200 optimized cells were produced that featured optimized parameters from each of the independent studies. The final optimized cells exhibited a final specific energy of 150 Wh/kg, meeting the DOE project goal, but failing to meet the initial specific energy requirement of 180 Wh/kg.

25 ENERGY STORAGE↗

DOE OSTI2022

Throughput Optimization of Molybdenum Carbide Nanoparticle Catalysts in a Continuous Flow Reactor Using Design of Experiments

Transition metal carbides (TMCs) have attracted significant attention because of their applications toward a wide range of catalytic transformations. However, the practicality of their synthesis is still limited because of the harsh conditions in which most TMCs are prepared. Recently, a solution-phase synthesis of phase-pure a-MoC1-x nanoparticles was presented. While this synthetic route yielded nanoparticles with exceptional catalytic performance, the reaction parameter space was not explored, and catalyst throughput was not optimized for scale-up. Continuous flow platforms coupled with statistical design of experiments (DoE) can provide a powerful method for understanding the reaction parameter space for optimizations. Here, we demonstrate the use of statistical DoE in tandem with response surface methodology for a parametric screening analysis to optimize the throughput of a MoC1-x nanoparticle synthesis utilizing a millifluidic flow reactor. A full factorial design was implemented to evaluate four input variables (reaction temperature, flow rate, solvent fraction of oleylamine, and precursor concentration) that carry statistically significant effects on three responses (throughput, residence time, and isolated yield). A Doehlert matrix was implemented to investigate each significant variable at a higher number of levels to optimize throughput. Our results give a nonintuitive set of experimental conditions that resulted in an optimized throughput of 2.2 g h-1. This translates to a 50-fold increase in throughput compared to the previously reported batch method. The catalytic performance of the MoC1-x nanoparticles produced under optimized throughput was demonstrated in the CO2 hydrogenation reaction. This DoE screening analysis and throughput optimization of MoC1-x synthesis open the door to an increased feasibility for scale-up.

design of experiments↗

DOE OSTI2024

Multiphysics Design Optimization and Additive Manufacturing of Nuclear Components (Final CRADA Report - Executive Summary)

Westinghouse Electric Company (WEC) actively participated in the advancement of the nuclear fuel and reactor design space and requested the help of Oak Ridge National Laboratory (ORNL) in the creation of a new design tool set. This report details the creation of a collection of software tool sets that are linked together to collectively assist WEC design engineers in developing novel ideas outside the normal scope of traditional nuclear fuel and reactor design formulas. Specifically, Siemens HEEDS, a design space exploration and parametric optimization software, monitored and changed parameters in a collection of softwares to meet the team’s objective. The HEEDS parametric optimization method, SHERPA, was developed to control the Siemens NX CAD platform to adjust the native CAD of a hexahedral spacer grid. This new geometry can be used to execute a topological design optimization by the NX Topology software add-in. The resulting geometry is additively manufacturable. This topological optimization occurred twice—once on the spacer grid’s spring, and once on the dimple geometry. These new geometries were imported by Siemens’ STAR-CCM+, a multiphysics structural and fluid dynamic computational solver in which the spring geometry is deflected to match the rod insertion configuration. Along with the dimple geometry, this new deflected spring was used to complete a hydraulic assessment of a single-unit cell comprising one rod, one spring, and two dimples. The HEEDS SHERPA algorithm ranks the design based on the final mass of the unit cell and the hydraulic pressure drop performance. The ORNL team demonstrated the ability to use this software and provided engineering judgement to apply modern aerospace aerodynamic design. The effort has been focused on thinking outside the conventional design space and redesigning a spacer grid to perform beyond the WEC set objectives. Furthermore, the ORNL team also demonstrated that the HEEDS optimization routine can independently develop a design that meets the WEC design goals. Although these designs were at a low technology readiness level, their demonstration confirmed the team’s capability to create novel advanced nuclear concepts.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2020

LCOE Design Optimization Using Genetic Algorithm with Improved Component Models for Medium-Voltage Transformerless PV Inverters

For real-world installations of photovoltaic and other renewable energy resources, the critical design metric is the levelized cost of energy (LCOE); however, many power electronics design optimizations are performed with efficiency and power density as the primary design goals. Recent work has shown that a new LCOE-focused optimization approach can yield improved system designs balancing cost and energy generation. This paper expands the LCOE optimization approach by considering comprehensive optimization parameters, adding new modeling of inductor cost, extending the semiconductor model to include effects of losses on housing cost, and implementing a genetic algorithm to improve computation efficiency.

41 EE - Solar Energy Technologies Office (EE-4S)↗

DOE OSTI2024

Interlaced Characterization and Calibration (ICC) for Improved Computational Simulation Credibility

Accurate material characterization and model calibration are pivotal for simulations used for high-consequence engineering decisions. Current characterization and calibration methods (1) use simplified test specimen geometries and global data, (2) cannot guarantee that sufficient characterization data is collected for a specific model of interest, (3) provide only mean parameter values with no uncertainty quantification, and (4) are sequential, inflexible, and time-consuming. This work developed a new paradigm—coined Interlaced Characterization and Calibration (ICC)—which drives forward the state-of-the-art in model calibration by bringing together recent advancements into one improved workflow. The ICC paradigm (1) employs tools to efficiently use full-field data to calibrate high-fidelity material models, (2) aligns the data needed with the data collected by adopting an optimal experimental design protocol, (3) provides uncertainty metrics on the calibrated model parameters, and (4) incorporates these advances into a quasi real-time feedback loop. The ICC framework was validated synthetically with both low-fidelity and high-fidelity simulations paired with several different elastoplastic material models, and was also demonstrated experimentally with an aluminum 6061 cruciform exemplar specimen. Results showed that the ICC framework—in which Bayesian optimal experimental design actively guided the experiment— resulted in calibrations with similar or better accuracy than predetermined experiments based on subject matter expertise. Moreover, the ICC framework produced a complete model calibration— with quantified uncertainties on model parameters—in 1 week, a 5 - 10× increase in efficiency over traditional approaches. Thus, the ICC paradigm improves both the calibration process and quality, by (1) improving efficiency, which increases agility of solid mechanics modeling and enables utilization of computational simulation (CompSim) at earlier stages of the design cycle and (2) providing quantified, and in some cases reduced, parameter uncertainties, which increases confidence in model predictions and supports credible decision making.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Computation of Direct Sensitivities of Spatial Multibody Systems With Joint Friction

Abstract Friction exists in most mechanical systems and may have a major influence on the dynamic performance of the system. The incorporation of friction in dynamic systems has been a subject of active research for several years owing to its high nonlinearity and its dependence on several parameters. Consequently, optimization of dynamic systems with friction becomes a challenging task. Gradient-based optimization of dynamical systems is a prominent technique for optimal design and requires the computation of model sensitivities with respect to the design parameters. The novel contribution of this paper is the derivation of the analytical methodology for the computation of direct sensitivities for smooth multibody systems with joint friction using the Lagrangian index-1 formulation. System dynamics have been computed using two different friction models; the Brown and McPhee, and the Gonthier et al. model. The methodology proposed to obtain model sensitivities has also been validated using the complex finite difference method. A case study has been conducted on a spatial multibody system to observe the effect of friction on the dynamics and model sensitivities, compare sensitivities with respect to different parameters and demonstrate the numerical and validation aspects. Since design parameters can have very different magnitudes and units, the sensitivities have been scaled with the parameters for comparison. Finally, a discussion has been presented on the interpretation of the case study results. Due to the incorporation of joint friction, ‘jumps’ or discontinuities are observed in the model sensitivities akin to those observed for hybrid dynamical systems.

Engineering↗

DOE OSTI2021

Application of machine learning to estimate fireball characteristics and their uncertainty from infrared spectral data

Experiments or events involving high explosives (HE) can be monitored remotely by infrared (IR) sensors to gather information about the configuration or materials involved in the device. Researchers at the Air Force Institute of Technology (AFIT) developed a phenomenological model for HE fireball spectra in the IR range that allows for parameters to be extracted from Fourier transform infrared (FTIR) data. This model includes parameters tied to physical characteristics of the fireball: temperature, size, soot, and gas concentrations. Previous works have sought to recover these parameters by the fitting of either whole spectra or select wavenumber bands to this phenomenological model. Difficulties arise due to the complex relationships between the parameters to be fit. Uncertainty quantification of the estimated fireball parameters is also problematic since HE experiments do not have any ground truth information on the parameters. It is suggested that artificial neural network (ANN) based approaches may be well suited to this problem, because of their ability to capture complex and highly nonlinear relationships. As such, this work seeks to explore the efficacy of deep artificial neural networks (DNNs) for this problem of parameter recovery from spectra and to also investigate the uncertainty of recovering the fireball parameters from FTIR data. Networks are designed using the hyperparameter optimization tool Hyperopt and trained/tested on artificial data generated using the phenomenological model developed by AFIT. The results of applying the network to the artificial data set are compared to a physics-based band approach that uses a selected number of bands based on their physical properties. Information on the uncertainty of estimating parameters from remotely sensed experimental data is obtained by treating the accuracy of the DNN model on artificial data as an upper bound and by examining the impact of emissivity due to soot on parameter estimation error; the results for artificial data are likely to be optimistic as compared to recovering parameters from experimental data.

42 ENGINEERING↗

DOE OSTI2023

Optimizing Perovskite Thin‐Film Parameter Spaces with Machine Learning‐Guided Robotic Platform for High‐Performance Perovskite Solar Cells

Abstract Simultaneously optimizing the processing parameters of functional thin films remains a challenge. The design and utilization of a fully automated platform called SPINBOT is presented for the engineering of solution‐processed functional thin films. The SPINBOT is capable of performing experiments with high sampling variability through the unsupervised processing of hundreds of substrates with exceptional experimental control. Through the iterative optimization process enabled by the Bayesian optimization (BO) algorithm, the SPINBOT explores an intricate parameter space, continuously improving the quality and reproducibility of the produced thin films. This machine learning (ML)‐guided reliable SPINBOT platform enables the acceleration of the optimization process of perovskite solar cells via a simple photoluminescence characterization of films. As a result, this study arrives at an optimal film that, when processed into a solar cell in an ambient atmosphere, immediately yields a champion power conversion efficiency (PCE) of 21.6% with satisfactory performance reproducibility. The unsealed devices retain 90% of their initial efficiency after 1100 h of continuous operation at 60–65 °C under metal‐halide lamps. It is anticipated that the integration of robotic platforms with the intelligent algorithm will facilitate the widespread adoption of effective autonomous experimentation to address the evolving needs and constraints within the materials science research community.

14 SOLAR ENERGY↗

DOE OSTI2022

Parametric dynamic mode decomposition for reduced order modeling

Dynamic Mode Decomposition (DMD) is a model-order reduction approach, whereby spatial modes of fixed temporal frequencies are extracted from numerical or experimental data sets. The DMD low-rank or reduced operator is typically obtained by singular value decomposition of the temporal data sets. For parameter-dependent models, as found in many multi-query applications such as uncertainty quantification or design optimization, the only parametric DMD technique developed was a stacked approach, with data sets at multiple parameter values were aggregated together, increasing the computational work needed to devise low-rank dynamical reduced-order models. Here in this paper, we present two novel approach to carry out parametric DMD: one based on the interpolation of the reduced-order DMD eigen-pair and the other based on the interpolation of the reduced DMD (Koopman) operator. Numerical results are presented for diffusion-dominated nonlinear dynamical problems, including a multiphysics radiative transfer example. All three parametric DMD approaches are compared.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2021

Variational Self-Consistent Theory for Beam-Loaded Cavities

A new variational theory is presented for beam loading in microwave cavities. The beam{field interaction is formulated as a dynamical interaction whose stationarity according to Hamilton’s principle will naturally lead to steady-state solutions that indicate how a cavity’s resonant frequency, Q and optimal coupling coefficient will detune as a result of the beam loading. A driven cavity Lagrangian is derived from first principles, including the effects of cavity wall losses, input power and beam interaction. The general formulation is applied to a typical klystron input cavity to predict the appropriate detuning parameters required to maximize the gain (or modulation depth) in the average Lorentz factor boost, (Δ γ). Numerical examples are presented, showing agreement with the general detuning trends previously observed in the literature. The developed formulation carries several advantages for beam-loaded cavity structures. It provides a self-consistent model for the dynamical (nonlinear) beam{field interaction, a procedure for maximizing gain under beam-loading conditions, and a useful set of parameters to guide cavity-shape optimization during the design of beam-loaded systems. Enhanced clarity of the physical picture underlying the problem seems to be gained using this approach, allowing straightforward inclusion or exclusion of different field configurations in the calculation and expressing the final results in terms of measurable quantities. Two field configurations are discussed for the klystron input cavity, using finite magnetic confinement or no confinement at all. Formulating the problem in a language that is directly accessible to the powerful techniques found in Hamiltonian dynamics and canonical transformations may potentially carry an additional advantage in terms of analytical computational gains, under suitable conditions

43 PARTICLE ACCELERATORS↗

DOE OSTI2020

Thermophysical Property Measurements on Salt Mixture

Thermal conductivity, specific heat, viscosity, and density are critical properties for understanding thermal hydraulics of heat transfer fluids (HTFs) like molten salts. These properties are important for design and safe operation of any HTF loop. These characterizations are required to benchmark computational chemical models that are used to predict properties, but existent data are limited. These measurements are extremely difficult to make due to the operating temperature range and corrosive properties of the salt.Density measurements are needed to predict volumetric behavior. Thermal expansion mismatches can cause mechanical failure of structural materials and important components that are in contact with the molten salt. All of these properties are related to heat transfer in HTFs. The heat generated in a reactor must be high enough to derive power; however it should not be so hot that it cannot be safely removed via the HTF at an appropriate rate. It is important to understand the thermal hydraulics required to operate at the desired power generation level and to avoid situations in which thermal shifts can damage components,cause failure of structural materials, and/or lead to accidents. Understanding these thermal properties can aid in component and process design and can contribute to optimized operational and safety parameters.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2021

Development of an in situ dual-channel thermal desorption gas chromatography instrument for consistent quantification of volatile, intermediate-volatility and semivolatile organic compounds

Abstract. Aerosols are a source of great uncertainty in radiative forcing predictions and have poorly understood health impacts. Most aerosol mass is formed in the atmosphere from reactive gas-phase organic precursors, forming secondary organic aerosol (SOA). Semivolatile organic compounds (SVOCs) (effective saturation concentration, C*, of 10−1–103 µg m−3) comprise a large fraction of organic aerosol, while intermediate-volatility organic compounds (IVOCs) (C* of 103–106 µg m−3) and volatile organic compounds (VOCs) (C* ≥ 106 µg m−3) are gas-phase precursors to SOA and ozone. The Comprehensive Thermal Desorption Aerosol Gas Chromatograph (cTAG) is the first single instrument simultaneously quantitative for a broad range of compound-specific VOCs, IVOCs and SVOCs. cTAG is a two-channel instrument which measures concentrations of C5–C16 alkane-equivalent-volatility VOCs and IVOCs on one channel and C14–C32 SVOCs on the other coupled to a single high-resolution time-of-flight mass spectrometer, achieving consistent quantification across 15 orders of magnitude of vapor pressure. cTAG obtains concentrations hourly and gas–particle partitioning for SVOCs every other hour, enabling observation of the evolution of these species through oxidation and partitioning into the particle phase. Online derivatization for the SVOC channel enables detection of more polar and oxidized species. In this work we present design details and data evaluating key parameters of instrument performance such as I/VOC collector design optimization, linearity and reproducibility of calibration curves obtained using a custom liquid evaporation system for I/VOCs and the effect of an ozone removal filter on instrument performance. Example timelines of precursors with secondary products are shown, and analysis of a subset of compounds detectable by cTAG demonstrates some of the analytical possibilities with this instrument.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

MODELLING OF ADDITIVE MANUFACTURING – COMPRESSION MOLDING PROCESS USING COMPUTATIONAL FLUID DYNAMICS

A computational fluid dynamics model has been developed to predict the behavior of a printed strands during a novel material extrusion additive manufacturing and compression molding process. While the traditional additive manufacturing process enables control over the fiber orientation within the part, it would also result in high void content. On the other hand, compression molding produces parts with low porosity levels at rapid processing cycle time but lacks control over the microstructure. The novel additive manufacturing – compression molding (AM-CM) integrating both these processes offers control over both the microstructure and porosity to manufacturing high performance composite parts. The numerical model developed here enables to analyze the effect of processing parameters on the behaviour of printed layer that is subsequently compressed and to determine the optimal printing parameters for high-performance composite part design.

Seta, Berin↗

DOE OSTI2025

A geometric framework for momentum-based optimizers for low-rank training

Low-rank pre-training and fine-tuning have recently emerged as promising techniques for reducing the computational and storage costs of large neural networks. Training low-rank parameterizations typically relies on conventional optimizers such as heavy ball momentum methods or Adam. In this work, we identify and analyze potential difficulties that these training methods encounter when used to train low-rank parameterizations of weights. In particular, we show that classical momentum methods can struggle to converge to a local optimum due to the geometry of the underlying optimization landscape. To address this, we introduce novel training strategies derived from dynamical low-rank approximation, which explicitly account for the underlying geometric structure. Our approach leverages and combines tools from dynamical low-rank approximation and momentum-based optimization to design optimizers that respect the intrinsic geometry of the parameter space. We validate our methods through numerical experiments, demonstrating faster convergence, and stronger validation metrics at given parameter budgets.

Schotthoefer, Steffen [ORNL] (ORCID:00000002156965↗

DOE OSTI2025

Structural evolution and magnetic hardness of (Sm,Zr)(Fe,Co,Ti) 12 alloy particles via reduction-diffusion

The quest for achieving high coercivity in Sm(Fe,Co,Ti) 12 alloys, despite their inherent strong magnetocrystalline anisotropy, has posed significant challenges. Recently, (Sm,Zr)(Fe,Co,Ti) 12 monocrystalline particles have exhibited coercivity μ 0 H c > 1.2 T, showcasing promising prospects and significant potential for both manufacturing and research endeavors. This study delves into the structural evolution of (Sm,Zr)(Fe,Co,Ti) 12 (1:12) alloy particles made via the calciothermic reduction-diffusion synthesis process as influenced by the molar ratios of Ca atoms to O 2- ions (Ca/O), annealing time and annealing temperature. Critical insight that informs conditions to optimize the magnetic response is gained via systematic experimentation and advanced electron microscopy. Complex structural features, including core-shell morphologies and intricate multiphase compositions within individual particles, are unveiled. An optimal Ca/O ratio of 1.30 produces particles with a coercivity up to μ 0 H c = 1.63 T, while higher Ca/O ratios induce the formation of a Sm-rich TbCu 7 -type (1: 7 ) phase, which only partially transforms into the desired 1:12 phase during annealing. Persistent remnants of the 1:7 phase locally impact atomic structure, particle morphology, and coercivity. Furthermore, these findings underscore the complex interplay between synthesis parameters, resulting structures, and magnetic properties, informing the design and optimization of high-performance permanent magnets comprised of the (1:12) compound.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Microreactor Optimization Using Simulation And Economics (mouse)

Microreactor Optimization Using Simulation and Economics (MOUSE) is a tool that integrates both nuclear microreactor design and reactor economics to provide comprehensive evaluations and optimizations. This tool enables stakeholders to explore the interplay between technical and economic variables, guiding them towards effective and competitive microreactor solutions. For the reactor core simulations, MOUSE leverages the OpenMC Monte Carlo Particle Transport Code to perform detailed core simulations for various microreactor designs. The included OpenMC models are 2D core designs of a Liquid Metal Thermal Microreactor (LMTR), a Gas-Cooled TRISO-Fueled Microreactor (GCMR), and a Heat Pipe Microreactor. Beyond core design, MOUSE includes simplified calculations for: - Calculating the masses of heat exchangers within the system. - Mechanical power of pumps. - Estimating the area occupied by various buildings within the nuclear plant. For the economic analysis, MOUSE provides detailed bottom-up cost estimates, encompassing a wide range of costs including preconstruction costs, direct costs, indirect costs, training costs, financial costs, operation & maintenance (O&M) costs, and fuel costs. These cost estimations are developed using data from the MARVEL project and additional literature sources, enabling the calculation of total capital costs and levelized cost of energy for both first-of-a-kind and nth-of-a-kind microreactors. MOUSE also enables analysis of the cost drivers and competitiveness in the electricity market. MOUSE allows users to modify a wide array of technical and economic parameters to evaluate different scenarios and their impacts. Examples of these parameters include: Fuels, coolants, or reflector materials Enrichment levels Control drum materials and geometry Fuel pin geometry and materials Moderator pin geometry and materials Reactor core and reflector dimensions Packing factor for the TRISO particles Nuclear reactor power and reactor burnup Number of sensors Shielding thickness Reactor vessel and guard vessel dimensions Operational staff requirements Number of emergency shutdowns Levelization period Interest rate Construction duration Since MOUSE is powered by the WATTS toolkit, it supports optimization studies, parametric analyses, and uncertainty calculations/propagation. The optimization techniques enable users to identify optimal design and economic configurations. The parametric analysis tools allow users to explore the sensitivity of various parameters, while uncertainty propagation helps quantify the impact of uncertainties on overall performance and cost. User Interface and Workflow: Currently, MOUSE is a command-line-based tool. Users can input various reactor design or economic parameters, modify the designs, run simulations, and visualize results through comprehensive data visualization and reporting capabilities. The typical workflow involves setting up the reactor model, defining economic parameters, running simulations, and analyzing the results to make informed decisions. By combining advanced design calculations with detailed economic modeling, MOUSE provides a robust framework for optimizing nuclear microreactor technologies, enhancing their competitiveness, and guiding stakeholders towards innovative and cost-effective solutions.

Hanna, Botros [Idaho National Laboratory (INL), Id↗

DOE OSTI2022

Optimization of Water-Alternating-CO2 Injection Field Operations Using a Machine-Learning-Assisted Workflow

Summary This paper will present a robust workflow to address multiobjective optimization (MOO) of carbon dioxide (CO2)-enhanced oil recovery (EOR)-sequestration projects with a large number of operational control parameters. Farnsworth unit (FWU) field, a mature oil reservoir undergoing CO2 alternating water injection (CO2-WAG) EOR, will be used as a field case to validate the proposed optimization protocol. The expected outcome of this work would be a repository of Pareto-optimal solutions of multiple objective functions, including oil recovery, carbon storage volume, and project economics. FWU’s numerical model is used to demonstrate the proposed optimization workflow. Because using MOO requires computationally intensive procedures, machine-learning-based proxies are introduced to substitute for the high-fidelity model, thus reducing the total computation overhead. The vector machine regression combined with the Gaussian kernel (Gaussian-SVR) is used to construct proxies. An iterative self-adjusting process prepares the training knowledge base to develop robust proxies and minimizes computational time. The proxies’ hyperparameters will be optimally designed using Bayesian optimization to achieve better generalization performance. Trained proxies will be coupled with multiobjective particle swarm Optimization (MOPSO) protocol to construct the Pareto-front solution repository. The outcomes of this workflow will be a repository containing Pareto-optimal solutions of multiple objectives considered in the CO2-WAG project. The proposed optimization workflow will be compared with another established methodology using a multilayer neural network (MLNN) to validate its feasibility in handling MOO with a large number of parameters to control. Optimization parameters used include operational variables that might be used to control the CO2-WAG process, such as the duration of the water/gas injection period, producer bottomhole pressure (BHP) control, and water injection rate of each well included in the numerical model. It is proved that the workflow coupling Gaussian-SVR proxies and the iterative self-adjusting protocol is more computationally efficient. The MOO process is made more rapid by squeezing the size of the required training knowledge base while maintaining the high accuracy of the optimized results. The outcomes of the optimization study show promising results in successfully establishing the solution repository considering multiple objective functions. Results are also verified by validating the Pareto fronts with simulation results using obtained optimized control parameters. The outcome from this work could provide field operators an opportunity to design a CO2-WAG project using as many inputs as possible from the reservoir models. The proposed work introduces a novel concept that couples Gaussian-SVR proxies with a self-adjusting protocol to increase the computational efficiency of the proposed workflow and to guarantee the high accuracy of the obtained optimized results. More importantly, the workflow can optimize a large number of control parameters used in a complex CO2-WAG process, which greatly extends its utility in solving large-scale MOO problems in various projects with similar desired outcomes.

Energy & Fuels↗