Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “machine learning optimization”

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 1,333 records · Page 74
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Lithium-Ion Battery Diagnostics Using Electrochemical Impedance via Machine-Learning

Diagnosing battery states such as health, state-of-charge, or temperature is crucial for ensuring the safety and reliability of electrochemical energy storage systems. While some states, such as temperature, may be measured using cheap sensors, accurate diagnosis of battery health metrics usually requires time-consuming performance measurements, making them infeasible for use in real-world operation. These health metrics can be measured during lab-testing and then estimated on-line using predictive life models or via state observer algorithms such as Kalman filters, but these predictive methods should be supplemented by actual measurement of battery health whenever possible to ensure reliability. Rapid measurement of battery health may be done by various types of fast diagnostic techniques such as electrochemical impedance spectroscopy (EIS), which can be performed in only a few minutes and require only a fraction of the energy and power needed for a full charge and discharge measurement. But there is a substantial challenge for estimating battery health using EIS data, as EIS is sensitive to cell temperature, state-of-charge, current, and resting time in addition to health. Thus, utilizing EIS data to predict battery capacity requires correcting for all these additional variables, a task that is extremely difficult to handle analytically. This talk utilizes machine-learning methods to estimate the effectiveness of battery capacity prediction from EIS data, leveraging a data set of hundreds of EIS measurements recorded at varying temperature and state-of-charge throughout a 500-day aging study of 32 commercial, large-format NMC-Graphite lithium-ion batteries. Using EIS as input to machine-learning models is complicated by the nonlinear response of impedance to battery health, temperature, and state-of-charge, as well as the collinearity between the impedance response at neighboring frequencies, which can easily lead to overfit models. To train robust models, features from EIS data need to be extracted from the data or some subset of critical frequencies selected. Many approaches for extracting and selecting features from EIS data from electrochemical analysis and machine-learning fields were identified for analysis: using the entire raw spectra; selection of one, two, or many frequencies from the entire spectra; selecting interesting points from the EIS measurement using domain knowledge; fitting EIS with an equivalent-circuit model; calculating statistics on the raw impedance values; and reducing the dimensionality of the data using unsupervised linear (principal component analysis) and non-linear (uniform manifold approximation and projection) methods. These approaches were rigorously compared using a machine-learning pipeline approach, training linear, Gaussian process, and random forest regression models and quantifying performance using cross-validation as well as a held-out test set. An artificial neural network model trained on the raw spectra was also tested. Promising pipelines were fine-tuned via Bayesian hyperparameter optimization using cross-validation loss and training with class-specific weights to counter data set imbalance. The most reliable method for utilizing impedance in this work was the selection of two optimal frequencies through an exhaustive search, resulting in about 2% mean absolute error on test data for both Gaussian process and random forest model architectures. Interrogation of a variety of models reveals critical frequencies of 100 Hz and 103 Hz for this data set, though the optimal set of frequencies is not necessarily intuitive, i.e., the best performing models are not simply those that use impedance at frequencies that have the highest correlation to the relative discharge capacity. The best performing model is an ensemble model, which is able to predict battery capacity with 1.9% mean absolute error for unseen cells using impedance recorded at a variety of temperatures and states-of-charge.

battery↗
DOE OSTI2023

Optimization with Neural Network Feasibility Surrogates: Formulations and Application to Security-Constrained Optimal Power Flow

In many areas of constrained optimization, representing all possible constraints that give rise to an accurate feasible region can be difficult and computationally prohibitive for online use. Satisfying feasibility constraints becomes more challenging in high-dimensional, non-convex regimes which are common in engineering applications. A prominent example that is explored in the manuscript is the security-constrained optimal power flow (SCOPF) problem, which minimizes power generation costs, while enforcing system feasibility under contingency failures in the transmission network. In its full form, this problem has been modeled as a nonlinear two-stage stochastic programming problem. In this work, we propose a hybrid structure that incorporates and takes advantage of both a high-fidelity physical model and fast machine learning surrogates. Neural network (NN) models have been shown to classify highly non-linear functions and can be trained offline but require large training sets. In this work, we present how model-guided sampling can efficiently create datasets that are highly informative to a NN classifier for non-convex functions. We show how the resultant NN surrogates can be integrated into a non-linear program as smooth, continuous functions to simultaneously optimize the objective function and enforce feasibility using existing non-linear solvers. Overall, this allows us to optimize instances of the SCOPF problem with an order of magnitude CPU improvement over existing methods.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2020

Phase behavior of continuous-space systems: A supervised machine learning approach

The phase behavior of complex fluids is a challenging problem for molecular simulations. Supervised machine learning (ML) methods have shown potential for identifying the phase boundaries of lattice models. In this work, we extend these ML methods to continuous-space systems. We propose a convolutional neural network model that utilizes grid-interpolated coordinates of molecules as input data of ML and optimizes the search for phase transitions with different filter sizes. We test the method for the phase diagram of two off-lattice models, namely, the Widom–Rowlinson model and a symmetric freely jointed polymer blend, for which results are available from standard molecular simulations techniques. The ML results show good agreement with results of previous simulation studies with the added advantage that there is no critical slowing down. We find that understanding intermediate structures near a phase transition and including them in the training set is important to obtain the phase boundary near the critical point. The method is quite general and easy to implement and could find wide application to study the phase behavior of complex fluids.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

The information content of jet quenching and machine learning assisted observable design

Jets produced in high-energy heavy-ion collisions are modified compared to those in proton-proton collisions due to their interaction with the deconfined, strongly-coupled quark-gluon plasma (QGP). In this work, we employ machine learning techniques to identify important features that distinguish jets produced in heavy-ion collisions from jets produced in proton-proton collisions. We formulate the problem using binary classification and focus on leveraging machine learning in ways that inform theoretical calculations of jet modification: (i) we quantify the information content in terms of Infrared Collinear (IRC)-safety and in terms of hard vs. soft emissions, (ii) we identify optimally discriminating observables that are in principle calculable in perturbative QCD, and (iii) we assess the information loss due to the heavy-ion underlying event and background subtraction algorithms. We illustrate our methodology using Monte Carlo event generators, where we find that important information about jet quenching is contained not only in hard splittings but also in soft emissions and IRC-unsafe physics inside the jet. This information appears to be significantly reduced by the presence of the underlying event. We discuss the implications of this for the prospect of using jet quenching to extract properties of the QGP. Since the training labels are exactly known, this methodology can be used directly on experimental data without reliance on modeling. We outline a proposal for how such an experimental analysis can be carried out, and how it can guide future measurements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Imaging and spatially resolved mass spectrometry applications in nephrology

The application of spatially resolved mass spectrometry (MS) and MS imaging approaches for studying biomolecular processes in the kidney is rapidly growing. These powerful methods, which enable label-free and multiplexed detection of many molecular classes across omics domains (including metabolites, drugs, proteins and protein post-translational modifications), are beginning to reveal new molecular insights related to kidney health and disease. Further, the complexity of the kidney often necessitates multiple scales of analysis for interrogating biofluids, whole organs, functional tissue units, single cells and subcellular compartments. Various MS methods can generate omics data across these spatial domains and facilitate both basic science and pathological assessment of the kidney. Optimal processes related to sample preparation and handling for different MS applications are rapidly evolving. Emerging technology and methods, improvement of spatial resolution, broader molecular characterization, multimodal and multiomics approaches and the use of machine learning and artificial intelligence approaches promise to make these applications even more valuable in the field of nephology. Overall, spatially resolved MS and MS imaging methods have the potential to fill much of the omics gap in systems biology analysis of the kidney and provide functional outputs that cannot be obtained using genomics and transcriptomic methods.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Design of lightweight BCC multi-principal element alloys with enhanced hydrogen storage using a machine learning-driven genetic algorithm

Body-centered cubic (BCC) based multi-principal element alloy (MPEA) hydrides have demonstrated significant potential for compact and efficient hydrogen storage. In this work, we first leverage machine learning (ML) models to predict the hydrogen affinity, storage capacity and phase stability of BCC MPEAs, creating a unique hydrogen-to-metal (H/M) predictor for materials with unprecedented performance. We developed a metaheuristic optimizer high-throughput framework by interfacing ML models with a genetic algorithm for the accelerated search of {Mg, Al, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Nb, Mo} based lightweight BCC MPEAs with improved hydrogen storage characteristics. We report five new MPEAs with a predicted gravimetric hydrogen storage capacity of around 3.5 wt% or more, including Cr 0.09 Mg 0.73 Ti 0.18 (4.25 wt% H) and Cr 0.21 Nb 0.11 Ti 0.35 V 0.33 (3.5 wt% H). The electronic structure of the top-performing composition, Cr 0.09 Mg 0.73 Ti 0.18 , was analyzed using density functional theory (DFT) to understand the reasons for its improved hydrogen storage properties compared to TiFe (1.90 wt% H), LaNi 5 (1.37 wt% H) or BCC MPEAs like TiVNbCr (3.70 wt% H). Temperature-dependent molecular dynamics (MD) studies were further performed on optimized BCC MPEAs to qualitatively study hydrogen mobility and analyze the effect of different elemental composition on bulk hydrogen diffusion. Our findings demonstrate how a ML assisted genetic algorithm framework can be used for efficient search of stable, lightweight and cost-effective MPEAs while minimizing the need for expensive ab initio calculations.

DFT↗
DOE OSTI2022

Adaptive Generation of Training Data for ML Reduced Model Creation

Machine learning proxy models are often used to speed up or completely replace complex computational models. The greatly reduced and deterministic computational costs enable new use cases such as digital twin control systems and global optimization. The challenge of building these proxy models is generating the training data. A naive uniform sampling of the input space can result in a non-uniform sampling of the output space of a model. This can cause gaps in the training data coverage that can miss finer scale details resulting in poor accuracy. While larger and larger data sets could eventually fill in these gaps, the computational burden of full-scale simulation codes can make this prohibitive. In this paper, we present an adaptive data generation method that utilizes uncertainty estimation to identify regions where training data should be augmented. By targeting data generation to areas of need, representative data sets can be generated efficiently. The effectiveness of this method will be demonstrated on a simple one-dimensional function and a complex multidimensional physics model.

Cianciosa, Mark↗
DOE OSTI2024

AXI4MLIR: User-Driven Automatic Host Code Generation for Custom AXI-Based Accelerators

Tensor algebra operations represent an important class of algorithms used across many applications, including machine learning, scientific computing, and data analytics. As a result, the efficient generation of custom accelerators for tensor operations has received increased attention. Previous efforts have produced automated tools enabling users to prototype and explore optimized accelerators. However, little effort has been focused on the host-accelerator interaction in these tools. Efficient use of hardware accelerators requires knowledge about the accelerator's capabilities (operations, data formats, and opcode support), the host CPU microarchitecture (e.g., memory hierarchy), the host-accelerator interface, and the application's features (which code regions should be mapped onto an accelerator). Manually rewriting the original applications to facilitate improved custom accelerator mapping is an error-prone and time-consuming endeavor. To cope with this, we propose AXI4MLIR, a new framework to automatically generate and optimize the communication between the host CPU and arbitrary accelerators that implement linear algebra algorithms. AXI4MLIR extends the MLIR compiler framework to automatically generate efficient host-accelerator driver code for accelerators with AXI-based interfaces. Our compiler extensions enable automatic driver code generation while carefully considering the host's memory hierarchy and target accelerator features. To demonstrate the flexibility and utility of AXI4MLIR, we test it with diverse use cases that include different types of accelerators, tiling scenarios, and dataflow schemes. We compare our experimental results to manual implementations of host-accelerator driver code and find that our approach can reduce CPU cache references by 56% and deliver up to a 1.65x speedup.

Bohm Agostini, Nicolas↗
DOE OSTI2025

SymProp: Scaling Sparse Symmetric Tucker Decomposition via Symmetry Propagation

Sparse symmetric tensors are an important class of tensors, and their decompositions serve as powerful tools for revealing low-rank structures. This paper introduces SymProp, a novel approach for scaling sparse symmetric Tucker decomposition by propagating symmetry through intermediate computations. SymProp optimizes two key computational kernels: Sparse Symmetric Tensor Times Same Matrix chain (S3 TTMc) for Higher-Order Orthogonal Iteration (HOOI) and Sparse Symmetric Tensor Times Same Matrix chain Times Core (S3 TTMcTC) for Higher-Order QR Iteration (HOQRI). Our method employs a metaprogramming-based index iteration approach to efficiently handle the upper triangular parts of intermediate dense symmetric tensors. SymProp achieves up to 50.9× speedup over SPLATT and up to 360.8× over Compressed Sparse Symmetric (CSS) format on the S3 TTMc operation. Moreover, our S3 TTMc and S3 TTMcTC implementations support tensor orders four levels higher than state-of-the-art methods. Our HOQRI demonstrates superior scalability and up to a 33.6× speedup over optimized HOOI. By enabling more scalable Tucker decompositions for higher orders, decomposition ranks, and dimension sizes, SymProp opens new possibilities for analyzing complex hypergraph structures in fields such as network science, data mining, and machine learning.

Li, Zecheng [North Carolina State University]↗
DOE OSTI2026

PyJMAK: An Open-Source Python Toolkit for Modeling Solid-State Metallurgical Phase Transformations

Accurate prediction of metallurgical phase transformations is an essential basis for autonomous optimization and rapid part qualification. Several methods can be used to estimate the evolution of phase fractions such as JMAK kinetics-based models, phase-field models, thermodynamic models, and data-driven machine learning models. Thermodynamic and phase-field-based methodologies solve multiphysics equations requiring numerous calibration parameters and significant computational resources. As a result, the computation domain is limited to a point or on order of micron-meters. The data-driven models rely on large datasets from experiments and simulations. While the JMAK model only provides information about phase fraction evolution, it can predict this evolution in near real-time using thermal history and thermodynamic data without restriction on the domain. JMAK models have been popularly used by researchers to model phase transformations occuring during additive manufacturing or over arbitrary temperature profiles. Commercial proprietary software such as Abaqus and Ansys or closed-source in-house implementations offer the ability to model JMAK based kinetics to predict phase transformation. However, these software packages are not open-source or freely available for use and development in conjunction with manufacturing machines, sensors, and machine learning algorithms. In addition, the use of the model is restricted by a license token. In contrast, given temperature profiles at multiple points in the domain, this Python-based PyJMAK model can compute phase evolution in parallel due to its stand-alone modular, voxel-based structure, and it can be executed on high-performance computing resources without any license restrictions.

Prabhune, Bhagya [Oak Ridge National Laboratory (O↗
DOE OSTI2025

Developing a Prototype Methodology to Rank CO2-EOR Wells and Assess Their Reuse Potential for Geologic Carbon Storage

This paper presents a prototype methodology to assess the possible transition of Class II carbon dioxide-enhanced oil recovery (CO2-EOR) wells to Class VI wells. The focus is on wellbore construction materials—casing, cement, tubing, and the packer—and includes comprehensive workflows to evaluate these materials, with primary emphasis on compliance with Environmental Protection Agency (EPA) Class VI well construction and conversion guidelines. These workflows systematically assess material properties and performance criteria to ensure regulatory compliance and optimize long-term wellbore integrity and functionality. Utilizing Python scripts and JavaScript Object Notation (JSON) representations, the study automates checks on digitized Texas Railroad Commission (TRRC) data to rank wells based on workflow criteria. By emphasizing critical factors such as casing integrity, cementing techniques, tubing compatibility, and packer selection, the methodology helps well owners and operators prioritize wells for potential reuse as CO2 injection wells. Given limitations in digitized data, manual user verification is required in some sections. Future improvements include integrating non-digitized data through web scraping and machine learning techniques. This research serves as a practical guide for stakeholders, supporting environmental compliance and sustainable well operations.

geologic carbon sequestration↗
DOE OSTI2026

Investigation of the Effect of Framework Flexibility on Adsorption in SIFSIX-3-Cu using a Machine-Learned Force Field

Metal-organic frameworks (MOFs) are a promising class of adsorbents. The performance of MOF sorbents relies on high selectivity and low regeneration energy. This work focuses on the use of machine learned force fields (MLFFs) to model adsorption in a flexible MOF, SIFSIX-3-Cu. A DeePMD-based MLFF was trained to reproduce DFT (PBE+D3) energies, forces, and stresses, using an iterative sampling scheme combining sampling based on molecular dynamics, Monte Carlo, and geometry optimization to capture both attractive and repulsive regions of the potential energy surface. Flexibility of the MOF was explicitly included in this model. Hybrid Monte Carlo/molecular dynamics (MC/MD) simulations using the MLFF predicted adsorption isotherms in good agreement with experimental data for a range of pressures (40 Pa – 104 Pa) in contrast to rigid models, which overpredict CO2 adsorption at low pressures. The improvement was the result of a description of the variability of fluorine-fluorine diagonal distances at adsorption sites. This detailed description of flexibility afforded by the MLFF resulted in more accurate predictions adsorption isotherms when compared to the experimentally measured values. These results underscore the importance of including framework flexibility when modeling adsorption phenomena in MOFs, particularly for low pressure applications and provide a robust procedure for training MLFF models for MOFs.

Atomistic Simulation↗
DOE OSTI2020

Using Machine Learning Tools to Predict Compressor Stall

Clean energy has become an increasingly important consideration in today’s power systems. As the push for clean energy continues, many coal-fired power plants are being decommissioned in favor of renewable power sources such as wind and solar. However, the intermittent nature of renewables means that dynamic load following traditional power systems is crucial to grid stability. With high flexibility and fast response at a wide range of operating conditions, gas turbine systems are poised to become the main load following component in the power grid. Yet, rapid changes in load can lead to fluid flow instabilities in gas turbine power systems. These instabilities often lead to compressor surge and stall, which are some of the most critical problems facing the safe and efficient operation of compressors in turbomachinery today. Although the topic of compressor surge and stall has been extensively researched, no methods for early prediction have been proven effective. This study explores the utilization of machine learning tools to predict compressor stall. The long short-term memory (LSTM) model, a form of recurrent neural network (RNN), was trained using real compressor stall datasets from a 100 kW recuperated gas turbine power system designed for hybrid configuration. Two variations of the LSTM model, classification and regression, were tested to determine optimal performance. The regression scheme was determined to be the most accurate approach, and a tool for predicting compressor stall was developed using this configuration. Overall, results show that the tool is capable of predicting stalls 5–20 ms before they occur. With a high-speed controller capable of 5 ms time-steps, mitigating action could be taken to prevent compressor stall before it occurs.

42 ENGINEERING↗
DOE OSTI2023

Quantum Chemistry-Driven Machine Learning Approach for the Prediction of the Surface Tension and Speed of Sound in Ionic Liquids

Ionic liquids (ILs) have unique solvent properties and have thus garnered significant interest. However, exhaustive experimental determination of the physicochemical properties of ILs is unrealistic due to the large structural diversity of anions and cations, their high cost, the requirements of elevated temperature and pressure, and the time required. To circumvent these experimental costs, computational approaches to accurately calculate these properties have emerged. Here in the present study, we present a demonstration of two machine learning (ML) models for the prediction of two critical IL physical properties, the surface tension and the speed of sound, across a wide range of temperatures and pressures. The models make use of molecular descriptors derived from the COSMO-RS, a quantum chemical-based model. The ML models show excellent agreement with experimental observations, with an R2 value of 0.96–0.99 and RMSE of 1.71 mN/m and 16.12 m/s for the surface tension and speed of sound, respectively. This work paves the way for the development of COSMO-RS-informed ML models for the prediction of IL properties which can help to further optimize and accelerate technology development for ILs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Framework for Optimization, Quantification of Uncertainty and Surrogates – Updated Capabilities

This poster presents updated capabilities within the FOQUS Toolset, covering core features as well as project-based applications. Specifically, the poster highlights advancements in FOQUS integration with cloud-based clients, plugin support for machine learning tools and external simulations, sequential design of experiments, and advanced support for dynamic process models. The applications demonstrate the breadth of FOQUS predictive capabilities for chemical, energy and economic system analysis.

Paul, Brandon↗
DOE OSTI2025

Machine Learning-Driven Conservative-to-Primitive Conversion in Hybrid Piecewise Polytropic and Tabulated Equations of State

We present a novel machine learning (ML)-based method to accelerate conservative-to-primitive inversion, focusing on hybrid piecewise polytropic and tabulated equations of state. Traditional root-finding techniques are computationally expensive, particularly for large-scale relativistic hydrodynamics simulations. To address this, we employ feedforward neural networks (NNC2PS and NNC2PL), trained in PyTorch (2.0+) and optimized for GPU inference using NVIDIA TensorRT (8.4.1), achieving significant speedups with minimal accuracy loss. The NNC2PS model achieves 𝐿 1 and 𝐿 ∞ errors of 4.54 × 10 −7 and 3.44 × 10−6, respectively, while the NNC2PL model exhibits even lower error values. TensorRT optimization with mixed-precision deployment substantially accelerates performance compared to traditional root-finding methods. Specifically, the mixed-precision TensorRT engine for NNC2PS achieves inference speeds approximately 400 times faster than a traditional single-threaded CPU implementation for a dataset size of 1,000,000 points. Ideal parallelization across an entire compute node in the Delta supercomputer (dual AMD 64-core 2.45 GHz Milan processors and 8 NVIDIA A100 GPUs with 40 GB HBM2 RAM and NVLink) predicts a 25-fold speedup for TensorRT over an optimally parallelized numerical method when processing 8 million data points. Moreover, the ML method exhibits sub-linear scaling with increasing dataset sizes. We release the scientific software developed, enabling further validation and extension of our findings. By exploiting the underlying symmetries within the equation of state, these findings highlight the potential of ML, combined with GPU optimization and model quantization, to accelerate conservative-to-primitive inversion in relativistic hydrodynamics simulations.

conservative-to-primitive conversion↗
DOE OSTI2023

Data-driven, structure-preserving approximations to entropy-based moment closures for kinetic equations

In this study, we present a data-driven approach for approximating entropy-based closures of moment systems from kinetic equations. The proposed closure learns the entropy function by fitting the map between the moments and the entropy of the moment system, and thus does not depend on the spacetime discretization of the moment system or specific problem configurations such as initial and boundary conditions. With convex and C 2 approximations, this data-driven closure inherits several structural properties from entropy-based closures, such as entropy dissipation, hyperbolicity, and H-Theorem. We construct convex approximations to the Maxwell–Boltzmann entropy using convex splines and neural networks, test them on the plane source benchmark problem for linear transport in slab geometry, and compare the results to the standard, entropy-based systems which solve a convex optimization problem to find the closure. Numerical results indicate that these data-driven closures provide accurate solutions in much less computation time than that required by the optimization routine.

97 MATHEMATICS AND COMPUTING↗