Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Generative Neural Networks”

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.

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

At least 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Generative Thermodynamic Computing

Here, we introduce a generative modeling framework for thermodynamic computing, in which structured data are synthesized from noise by the natural time evolution of a physical system governed by Langevin dynamics. While conventional diffusion models use neural networks to perform denoising, here the information needed to generate structure from noise is encoded by the dynamics of a thermodynamic system. Training proceeds by maximizing the probability with which the computer generates the reverse of a noising trajectory, which ensures that the computer generates data with minimal heat emission. We demonstrate this framework within a digital simulation of a thermodynamic computer. If realized in analog hardware, such a system would function as a generative model that produces structured samples without the need for artificially injected noise or active control of denoising.

Whitelam, Stephen [Lawrence Berkeley National Labo↗
DOE OSTI2024

Sub-millisecond keyhole pore detection in laser powder bed fusion using sound and light sensors and machine learning

Laser powder bed fusion is a mainstream additive manufacturing technology widely used to manufacture complex parts in prominent sectors, including aerospace, biomedical, and automotive industries. However, during the printing process, the presence of an unstable vapor depression can lead to a type of defect called keyhole porosity, which is detrimental to the part quality. In this study, we developed an effective approach to locally detect the generation of keyhole pores during the printing process by leveraging machine learning and a suite of optical and acoustic sensors. Simultaneous synchrotron x-ray imaging allows the direct visualization of pore generation events inside the sample, offering high-fidelity ground truth. A neural network model adopting SqueezeNet architecture using single-sensor data was developed to evaluate the fidelity of each sensor for capturing keyhole pore generation events. Our comparative study shows that the near infrared images gave the highest prediction accuracy, followed by 100 kHz and 20 kHz microphones, and the photodiode sensitive to processing laser wavelength had the lowest accuracy. Using a single sensor, over 90% prediction accuracy can be achieved with a temporal resolution as short as 0.1 ms. A data fusion scheme was also developed with features extracted using SqueezeNet neural network architecture and classification using different machine learning algorithms. Our work demonstrates the correlation between the characteristic optical and acoustic emissions and the keyhole oscillation behavior, and thereby provides strong physics support for the machine learning approach.

36 MATERIALS SCIENCE↗
NASA NTRS1992

Inversion of parameters for semiarid regions by a neural network

Microwave brightness temperatures obtained from a passive radiative transfer model are inverted through use of a neural network. The model is applicable to semiarid regions and produces dual-polarized brightness temperatures for 6.6-, 10.7-, and 37-GHz frequencies. A range of temperatures is generated by varying three geophysical parameters over acceptable ranges: soil moisture, vegetation moisture, and soil temperature. A multilayered perceptron (MLP) neural network is trained with a subset of the generated temperatures, and the remaining temperatures are inverted using a backpropagation method. Several synthetic terrains are devised and inverted by the network under local constraints. All the inversions show good agreement with the original geophysical parameters, falling within 5 percent of the actual value of the parameter range.

Zurk, Lisa M.↗
DOE OSTI2021

MACHINE LEARNING BASED CHEMICAL EXPLOSIVE MODE ANALYSIS (ML-CEMA)

The software consists of a Machine Learning based Chemical Explosive Mode Analysis (ML-CEMA) tool for advanced computational flame diagnostics. CEMA, originally based on eigen-analysis of the local thermochemical system, is capable of identifying reaction fronts and limit phenomena such as auto-ignition and extinction in practical combustion systems, such as internal combustion engines and gas turbine combustors. However, the original CEMA is computationally expensive for large reaction mechanisms that are typically needed to describe fuel chemistry of practical large-hydrocarbon fuels. This novel ML-CEMA tool employs a ML technique to accelerate the eigen-analysis of the basic CEMA approach by orders of magnitude, thus making it suitable for practical fuels. In ML-CEMA,zero-dimensional (0D) reactors and one-dimensional (1D) premixed flames are first used to generate a large number of data points for neural network based ML training. The trained ML model is then used to perform CEMA prediction. This ML-CEMA tool has been demonstrated in canonical 0D and 1D configurations as well as highly-transient three-dimensional spray flames exhibiting multi-mode turbulent combustion, showing promising results. ML-CEMA, as a standalone tool, can be used for computationally-efficient diagnostics of massive datasets generated from both experiments and simulations. For example, based on spatially resolved measurements of a small set of reactive scalars(such as temperature, hydroxyl radical and formaldehyde), ML-CEMA can effectively identify flame fronts and rare events. ML-CEMA also provides a robust online or offline flame feature detection tool. When used for on-the-fly simulations, ML-CEMA further enables zone-adaptive combustion modeling, in which the predicted eigenvalue is used as a robust mode indicator for judicious assignment of locally-valid combustion models. This ML-CEMA based zone-adaptive model can lead to substantial computational cost savings when used for large-scale simulations of multi-mode combustion systems. Third Party Code Web Page to Download Code Web Page Location of Third Party License

Xu, Chao↗
DOE OSTI2021

Enhanced Data Efficiency Using Deep Neural Networks and Gaussian Processes for Aerodynamic Design Optimization

Adjoint-based optimization methods are attractive for aerodynamic shape design primarily due to their computational costs being independent of the dimensionality of the input space and their ability to generate high-fidelity gradients that can then be used in a gradient-based optimizer. This makes them very well suited for high-fidelity simulation based aerodynamic shape optimization of highly parametrized geometries such as aircraft wings. However, the development of adjoint-based solvers involve careful mathematical treatment and their implementation require detailed software development. Furthermore, they can become prohibitively expensive when multiple optimization problems are being solved, each requiring multiple restarts to circumvent local optima. In this work, we propose a machine learning enabled, surrogate-based framework that replaces the expensive adjoint solver, without compromising on predicting predictive accuracy. Specifically, we first train a deep neural network (DNN) from training data generated from evaluating the high-fidelity simulation model on a model-agnostic design of experiments on the geometry shape parameters. The optimum shape may then be computed by using a gradient-based optimizer coupled with the trained DNN. Subsequently, we also perform a gradient-free Bayesian optimization, where the trained DNN is used as the prior mean. We observe that the latter framework (DNN-BO) improves upon the DNN-only based optimization strategy for the same computational cost. Overall, this framework predicts the true optimum with very high accuracy, while requiring far fewer high-fidelity function calls compared to the adjoint-based method. Furthermore, we show that multiple optimization problems can be solved with the same machine learning model with high accuracy, to amortize the offline costs associated with constructing our models. Our methodology finds applications in the early stages of aerospace design. (C) 2021 Published by Elsevier Masson SAS.

Renganathan, S. Ashwin↗
DOE OSTI2024

Neuro-Spark: A Submicrosecond Spiking Neural Networks Architecture for In-Sensor Filtering

Neuro-Spark, which is a new neuromorphic architecture with a field-programmable gate array (FPGA) implementation for ultrafast spiking neural network (SNN) inference at the edge, facilitates smart-pixel in-sensor filtering for high-energy physics experiments at the Large Hadron Collider (LHC). Utilizing the evolutionary optimization for neuromorphic systems (EONS) training method, we generate compact SNN models with 91% signal efficiency, akin to convolutional neural networks but with half the parameters. However, deploying near the detector poses a challenge because the SNN must handle a sustained input data rate exceeding 1013 GB/s. To overcome this, we propose a novel hardware architecture that uses high-level synthesis to construct a tuned architecture for the EONS-trained SNN. In addition to the analysis and validation with an AMD Xilinx Artix-A7 FPGA, our solution consumes only ç24% of FPGA LUT and flipflops. We also introduce an innovative quantization method that reduces FPGA resource utilization by ç15% without compromising accuracy. Our FPGA implementation achieves computing latency of ç10 ns for smart-pixel application inference on an edge FPGA.

Miniskar, Narasinga Rao↗
DOE OSTI2024

Evaluating the Use of Foundational Chemical Language Models in Multimodal Graph Fusion

Rapid and accurate prediction of the physicochemical properties of molecules given their structures remains a key challenge in cheminformatics. Machine learning approaches offer high-throughput options, but the optimality of inductive biases and data representations are up for debate. For example, BERT-based masked language models (MLMs) can be trained in a self-supervised way on hundreds of millions to billions of readily available SMILES strings. Another option is graph neural networks (GNNs), which can operate directly on molecular structures. Yet, generating accurate molecular geometry is computationally expensive, leading to a relative scarcity in data compared to SMILES strings. It is attractive to combine these two paradigms by pre-training an LM on a large corpus of SMILES strings and embedding these representation into a geometric graph neural network. Despite the promise of such an approach, and contrary to previous studies, we find mixed results with the combination of the LMs and GNNs on several molecule datasets. In particular, we found evidence for improvement on the FreeSolv and QM7 benchmarks, but degraded performance on the ESOL, LIPO and QM9 datasets compared to a GNN baseline.

Francel, Collin [University of Alabama]↗
DOE OSTI2020

Fast predictions of liquid-phase acid-catalyzed reaction rates using molecular dynamics simulations and convolutional neural networks

The rates of liquid-phase, acid-catalyzed reactions relevant to the upgrading of biomass into high-value chemicals are highly sensitive to solvent composition and identifying suitable solvent mixtures is theoretically and experimentally challenging. We show that the complex atomistic configurations of reactant–solvent environments generated by classical molecular dynamics simulations can be exploited by 3D convolutional neural networks to enable accurate predictions of Brønsted acid-catalyzed reaction rates for model biomass compounds. We develop a 3D convolutional neural network, which we call SolventNet, and train it to predict acid-catalyzed reaction rates using experimental reaction data and corresponding molecular dynamics simulation data for seven biomass-derived oxygenates in water–cosolvent mixtures. We show that SolventNet can predict reaction rates for additional reactants and solvent systems an order of magnitude faster than prior simulation methods. This combination of machine learning with molecular dynamics enables the rapid, high-throughput screening of solvent systems and identification of improved biomass conversion conditions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

MPEX AI Digital Twins

All magnetically confined plasma fusion power plant concepts (Tokamak, Spherical Tokamak, Stellarator, Mirror, ...) must exhaust the heat and plasma from the core confinement region to the material walls. The primary channel for this exhaust is through a plasma divertor which directs plasma along open magnetic field lines to a material target. The Material Plasma Exposure eXperiment (MPEX) illustrated in Figure 1, is a high-power, steady-state linear plasma device designed to produce the plasma material interaction (PMI) conditions of the divertor of future magnetic confinement fusion power plants: energy flux 20MW/m 2 , ion fluence 1031/m 2 , pulse duration 106 sec. These goals of plasma exposure in MPEX are well beyond those achieved in magnetic fusion experimental devices. Successfully achieving these high power steady state conditions for long pulses requires operational control of the heating and particle sources and the plasma flux to the walls and target. The MPEX AI Hot Spot Controller, proposed in this project, will help achieve the operational milestones of MPEX. The MPEX device will begin commissioning at the end of FY26. A smaller proto-MPEX was operated for 14,666 plasma discharges and will resume operation in September of 2025 as proto-MPEX-lite, with reduced capability, to test a new window for the Helicon plasma source. The proto-MPEX data has undergone surrogate modeling with machine learning methods (R. Archibald, 2022 IEEE International Conference on Big Data). This proto-MPEX data will be used to begin development of the AI digital twins described in this white paper. The scientific mission of MPEX is to qualify materials of different composition for use in the high energy and plasma flux conditions of a fusion power plant. The materials exposed in MPEX will in some cases be exposed to high neutron fluxes at other ORNL facilities to measure the changes to their PMI properties. The targets exposed in MPEX will be transported under vacuum to a Surface Analysis Station (SAS). The SAS will be equipped with the following diagnostics: Focused Ion Beam (FIB) for trench milling, 100-400 angstrom resolution scanning electron microscope (SEM), surface mapping x-ray spectrometer, high resolution camera, and a future upgrade to a laser induced breakdown spectroscopy quadruple mass spectrometer (LIBS-QMS). The MPEX experiments will generate diverse pre- and post-exposure measurement data of detailed material properties down to the crystal grain level in 3D for post-exposure assessment of PMI damage (e.g. cracking, melting, erosion and redeposition of the material). Physics models for the PMI, and how the material composition and manufacturing impact its performance under high energy plasma exposure, need to be validated with MPEX data to guide the selection of new candidate materials. Our vision for the MPEX AI Digital Twins project is to supply experimental and physics model simulation data to train Artificial Intelligence (AI) models for data processing, analysis, operational control, PMI and materials simulation to maximize the scientific output of the MPEX device. Ultimately, an AI digital twin of MPEX material assessment metrics for tested and synthetic material types with simulated PMI will be trained by the AI Modeling Teams on the experimental and physics simulation data submitted to the American Science Cloud by this project. A purely empirical search for the best material is inefficient given the finite number of samples that can be tested on MPEX. In order to expand the material properties database for training the MPEX Material Assessment AI Digital Twin, and to gain physics understanding of the PMI processes, physics models of the material properties and PMI processes are required. The physics simulations provide detailed simulation data, like impact angles for plasma ions, sputtering yields, transport of the ionized sputtered target material in the plasma, and redeposition locations. This simulation data expands the measurement data for deeper physics understanding. The experimental data is essential to validate the PMI and material structure simulation models. The validated models can then be used to generate new simulation data of MPEX material assessments for synthetic material compositions that have not been exposed in MPEX. These predictive simulations, plus the whole experimental dataset, will be used to train the MPEX Material Assessment AI Digital Twin allowing a rapid generative AI search for new materials with reduced PMI damage by interpolating the domain of the training set. These new optimum materials can be simulated with the physics codes and/or tested in MPEX. The ability of AI neural networks to interpolate multi-dimensional parameter spaces and generate virtual data is exploited for a more efficient search for optimum materials. The advent of the Transformational AI Models Consortium (TAIMC) is an opportunity to engage with state of the art private and public AI developers to achieve the goals of the AI digital twins and AI accelerated physics models proposed in this project. Our partners at ORNL from the Advance Scientific Computing Research (ASCR) organization will collaborate in accelerating the integrated plasma material interaction simulation framework. This simulation framework will provide a platform for generating simulation data across a range of physical fidelities, including hybrid methods that produce multi-fidelity results. This data will be leveraged for AI model development, both for generation of surrogates and the automation of simulation campaigns. A part of the research below will include collaborative efforts with the TAIMC to (i) adapt data storage approaches to ensure AI-readiness, (ii) provide a protypical exemplar to inform and exercise constructed workflows, and (iii) generate and share data, using the TAIMC unified AI data standard, for foundational models that will be trained from multiple sources across the DOE complex. We will also collaborate with the TAIMC, as well as the planned AI modeling teams, to develop approaches for reducing the cost of data generation. These include tailored multi-fidelity approaches as well as fine-tuning strategies to augment general, large-scale foundational models.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRSDate not supplied

Machine Learning Methods for Estimating Propeller Source Noise Spheres

In this work, several neural network function approximations are compared for inter- polating, storing, and sampling acoustic source spheres with applications to propeller noise estimation. These methods are compared using an acoustic model of the three bladed GL-10 propeller at different flight conditions, with training data generated using NASA’s ANOPP-PAS module. The source spheres used to train the networks capture the tonal propeller noise due to both the blade thickness and loading. This tonal noise prediction method allows the vehicle noise to be estimated for auralization and acoustic control. Three radial basis function neural network architectures are compared in this work. The first two networks directly estimate the parameters of the source sphere at different flight conditions but differ in the number of layers used. The third network estimates the parameters of the source sphere using a weighted combination of spherical basis functions. These networks are trained on numerically generated source spheres, with operating points given in terms of the propeller rotation rate, freestream speed, and propeller angle of attack. The performance of the neural network is determined using a validation dataset of withheld data points. This performance is quantified in terms of the approximation error, training time, and sample time. The third network, which estimates the weights of the spherical basis functions, performs the best in both average and maximum approximation errors in all cases. This network’s worst case performance is 5.6 % relative dif- ference of a model parameter associated with acoustic pressure. The direct estimation network with a single layer has the worst approximation error in all cases. Additionally, the spherically defined network has the slowest sample time at 0.05 seconds per thousand points. Both direct estimation methods produce a thousand sample points in approximately 0.001 seconds.

Acoustics↗
DOE OSTI2025

Large-scale experimental validation of thermochemical water-splitting oxides discovered by defect graph neural networks

Thermochemical water-splitting (TCH) based on 2-step thermal redox cycles in metal oxides is a promising approach to generating H 2 , but state-of-the-art (SOTA) CeO 2 has several practical limitations, which has motivated continued materials discovery efforts in this field. Here, in this study, we improve upon a SOTA defect graph neural network (dGNN) surrogate model's oxygen vacancy predictions and combine them with materials project phase diagrams to down-select and discover structurally diverse, experimentally known metal oxides whose TCH performance was previously unknown. Amongst twelve candidates selected based on our high-throughput screening and down-selection criteria, we achieved ∼80% accuracy in identifying materials with stable redox cycling and hydrogen production in stagnation flow reactor water-splitting experiments. Closer to 100% accuracy can be achieved if higher-accuracy, hybrid DFT-predicted vacancy formation energies were computed and used in lieu of the most uncertain dGNN-based screening predictions, as they correct false positives to true negatives. Notably, two discovered candidates, Sr 3 PrMn 2 O 8 and Ba 2 Fe 2 O 5 , display hydrogen yields greater than CeO 2 under specific redox conditions. In conclusion, these results demonstrate our ability to computationally predict and experimentally validate promising candidate TCH materials that have the potential to compete with CeO 2 .

08 HYDROGEN↗
DOE OSTI2025

Graph Neural Networks for Surrogate Modeling of Offshore Floating Platforms

Floating offshore wind turbines (FOWTs) present an significant opportunity to increase renewable energy generation. However, significant challenges remain before FOWTs can be widely commercialized and deployed. In particular, hydrodynamic loading on the platforms can stress the overall structure, damage the mooring systems, and impact power generation. Studying these loads is difficult and often relies on computationally expensive models or experiments. In this work, we explore the use of graph neural networks (GNNs) to construct flexible, data-driven surrogates for hydrodynamic loads on platforms. We leverage the natural graph-like structure of offshore wind platform designs to enable the GNN model to learn to approximate the loads for different wave conditions and structural designs. We demonstrate potential uses for the surrogate by performing parameter sweeps and ridge analysis on the trained model to identify the impacts of different wave and structural features on the loads.

floating offshore wind turbines↗
DOE OSTI2020

Numerical Investigation of a Central Fuel Property Hypothesis Under Boosted Spark-Ignition Conditions

In the present work, a central fuel property hypothesis (CFPH), which states that fuel properties are sufficient to provide an indication of a fuel’s performance irrespective of its chemical composition, was numerically investigated. In particular, the objective of the study was to determine whether Research Octane Number (RON) and Motor Octane Number (MON), as fuel properties, are sufficient to describe a fuel’s knock-limited performance under boosted spark-ignition (SI) conditions within the framework of CFPH. To this end, four TPRF-bioblendstock surrogates having different compositions but matched RON (=98) and MON (=90), were first generated using a non-linear regression model based on artificial neural network (ANN). Additionally, three unconventional bioblendstocks were included in the analysis: di-isobutylene (DIB), isobutanol, and Anisole. Skeletal reaction mechanisms were generated for the TPRF-DIB, TPRF-isobutanol, and TPRF-anisole blends from a detailed kinetic mechanism. Thereafter, numerical simulations were performed for the fuel surrogates using the skeletal mechanisms and a virtual cooperative fuel research (CFR) engine model, under a representative boosted operating condition. In the computational fluid dynamics (CFD) model, the G-equation approach was employed to track the turbulent flame front and the well-stirred reactor model combined with the multi-zone binning strategy was used to capture auto-ignition in the end-gas. In addition, laminar flame speed (LFS) was tabulated for each blend as a function of pressure, temperature, and equivalence ratio a priori, and the lookup tables were used to prescribe laminar flame speed as an input to the G-equation model. Parametric spark timing sweeps were performed for each fuel blend to determine the corresponding knock-limited spark advance (KLSA) and 50% burn point (CA50) at the respective KLSA timing. It was observed that despite same RON, MON, and engine operating conditions, the TPRF-anisole blend exhibited markedly different knock-limited performance from the other three blends. This deviation from the octane index (OI) expectation was shown to be caused by differences in laminar flame speed. However, it was found that relatively large fuel-specific differences in LFS (>20%) would have to be present to cause any appreciable deviation from the OI framework. Otherwise, RON and MON would still be robust enough to predict a fuel’s knock-limited performance.

42 ENGINEERING↗
DOE OSTI2024

Upsampling Monte Carlo Reactor Simulation Tallies in Depleted Sodium-Cooled Fast Reactor Assemblies Using a Convolutional Neural Network

The computational demand of neutron Monte Carlo transport simulations can increase rapidly with the spatial and energy resolution of tallied physical quantities. Convolutional neural networks have been used to increase the resolution of Monte Carlo simulations of light water reactor assemblies while preserving accuracy with negligible additional computational cost. Here, we show that a convolutional neural network can also be used to upsample tally results from Monte Carlo simulations of sodium-cooled fast reactor assemblies, thereby extending the applicability beyond thermal systems. The convolutional neural network model is trained using neutron flux tallies from 300 procedurally generated nuclear reactor assemblies simulated using OpenMC. Validation and test datasets included 16 simulations of procedurally generated assemblies, and a realistic simulation of a European sodium-cooled fast reactor assembly was included in the test dataset. We show the residuals between the high-resolution flux tallies predicted by the neural network and high-resolution Monte Carlo tallies on relative and absolute bases. The network can upsample tallies from simulations of fast reactor assemblies with diverse and heterogeneous materials and geometries by a factor of two in each spatial and energy dimension. The network’s predictions are within the statistical uncertainty of the Monte Carlo tallies in almost all cases. This includes test assemblies for which burnup values and geometric parameters were well outside the ranges of those in assemblies used to train the network.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
NASA NTRS2017

UAV Trajectory Modeling Using Neural Networks

Massive small unmanned aerial vehicles are envisioned to operate in the near future. While there are lots of research problems need to be addressed before dense operations can happen, trajectory modeling remains as one of the keys to understand and develop policies, regulations, and requirements for safe and efficient unmanned aerial vehicle operations. The fidelity requirement of a small unmanned vehicle trajectory model is high because these vehicles are sensitive to winds due to their small size and low operational altitude. Both vehicle control systems and dynamic models are needed for trajectory modeling, which makes the modeling a great challenge, especially considering the fact that manufactures are not willing to share their control systems. This work proposed to use a neural network approach for modelling small unmanned vehicle's trajectory without knowing its control system and bypassing exhaustive efforts for aerodynamic parameter identification. As a proof of concept, instead of collecting data from flight tests, this work used the trajectory data generated by a mathematical vehicle model for training and testing the neural network. The results showed great promise because the trained neural network can predict 4D trajectories accurately, and prediction errors were less than 2:0 meters in both temporal and spatial dimensions.

Neural Networks↗
NASA NTRS1993

Parametric motion control of robotic arms: A biologically based approach using neural networks

A neural network based system is presented which is able to generate point-to-point movements of robotic manipulators. The foundation of this approach is the use of prototypical control torque signals which are defined by a set of parameters. The parameter set is used for scaling and shaping of these prototypical torque signals to effect a desired outcome of the system. This approach is based on neurophysiological findings that the central nervous system stores generalized cognitive representations of movements called synergies, schemas, or motor programs. It has been proposed that these motor programs may be stored as torque-time functions in central pattern generators which can be scaled with appropriate time and magnitude parameters. The central pattern generators use these parameters to generate stereotypical torque-time profiles, which are then sent to the joint actuators. Hence, only a small number of parameters need to be determined for each point-to-point movement instead of the entire torque-time trajectory. This same principle is implemented for controlling the joint torques of robotic manipulators where a neural network is used to identify the relationship between the task requirements and the torque parameters. Movements are specified by the initial robot position in joint coordinates and the desired final end-effector position in Cartesian coordinates. This information is provided to the neural network which calculates six torque parameters for a two-link system. The prototypical torque profiles (one per joint) are then scaled by those parameters. After appropriate training of the network, our parametric control design allowed the reproduction of a trained set of movements with relatively high accuracy, and the production of previously untrained movements with comparable accuracy. We conclude that our approach was successful in discriminating between trained movements and in generalizing to untrained movements.

Bock, O.↗
DOE OSTI2023

Efficient Optimization of Energy Recovery From Geothermal Reservoirs With Recurrent Neural Network Predictive Models

Improving the long-term energy production performance of geothermal reservoirs can be accomplished by optimizing field development and management plans. Reliable prediction models, however, are needed to evaluate and optimize the performance of the underlying reservoirs under various operation and development strategies. In traditional frameworks, physics-based simulation models are used to predict the energy production performance of geothermal reservoirs. However, detailed simulation models are not trivial to construct, require a reliable description of the reservoir conditions and properties, and entail high computational complexity. Data-driven predictive models can offer an efficient alternative for use in optimization workflows. This paper presents an optimization framework for net power generation in geothermal reservoirs using a variant of the recurrent neural network (RNN) as a data-driven predictive model. The RNN architecture is developed and trained to replace the simulation model for computationally efficient prediction of the objective function and its gradients with respect to the well control variables. The net power generation performance of the field is optimized by automatically adjusting the mass flow rate of production and injection wells over 12 years, using a gradient-based local search algorithm. Two field-scale examples are presented to investigate the performance of the developed data-driven prediction and optimization framework. Furthermore, the prediction and optimization results from the RNN model are evaluated through comparison with the results obtained by using a numerical simulation model of a real geothermal reservoir.

15 GEOTHERMAL ENERGY↗