Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “explainable”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Efficient Reliability Analysis using Generalized Multifidelity Modeling and Explainable Active Learning

To assess the reliability of critical technologies like nuclear plants and infrastructure systems and improve the robustness of design, engineers have to quantify the uncertainties surrounding the system behavior accurately. However, the complexity of the problem can make standard reliability analysis algorithms prohibitively expensive, primarily due to the high computational cost of estimating the system response at each iteration. This cost can be greatly reduced by using multi-fidelity modeling and machine learning to build a surrogate model to replace the expensive response function. We propose a general and robust method for building surrogates from multiple Low Fidelity (LF) models coupled with machine learning to retain accuracy. Our framework first constructs “Corrected Low Fidelity models” (CLFs) by coupling a High Fidelity (HF) model inferred Gaussian Process correction term with each of the LF models. It then uses the correction terms to assign model probabilities to each of these CLFs in an explainable way before using them to assemble the final surrogate. No assumptions are made about the type of the LF models or their correlation with the HF model. The proposed surrogate modeling framework is used within the subset simulation algorithm (a variance-reduced MCMC-based reliability analysis algorithm) for enhanced efficiency. Additionally, an active learning step is added to the algorithm to adaptively decide when the surrogate is not sufficiently accurate, at which point the HF model is called and used to refine the surrogate. Through a frame buckling example, our method is shown to be highly efficient at reducing the expensive HF model calls while accurately estimating the failure probability.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation

Proximal remote sensing has the potential to provide critical information on vegetation biophysical factors that can predict land-atmosphere exchange of water and energy. Latent energy (LE) flux is traditionally estimated using process-based models which rely on vegetation parameters that change during the growing season. Data-driven models have the potential to address these issues by offering flexible predictor selection and more efficient utilization of the information in predictor sets. These models require careful choice of predictors to avoid redundancy and allow robust cross-validation. In this study we present a systematic and comprehensive evaluation of machine learning (ML) models to assess the capability of meteorological and proximal sensing data for predicting LE at a half-hourly temporal resolution across multiple growing seasons for an agricultural system. The results presented here demonstrate that a model using four environmental predictors in combination with two proximal sensing variables can capture 88 % of the variability in LE. ML models using only three predictors (one meteorological and two proximal remote sensing) captured 81 % of LE variability, offering the best trade-off between performance and complexity. An ML model utilizing only two predictors, one proximal remote sensing variable and downwelling radiation, captured 77 % of LE variability. These results demonstrate the power of proximal remote sensing and meteorological observations to estimate land-atmosphere water vapor exchange, providing a solution where more direct methods such as eddy covariance are not available and for evaluations of agronomic management and genotypic variations.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Investigating explainable transfer learning for battery lifetime prediction under state transitions

Battery lifetime prediction at early cycles is crucial for researchers and manufacturers to examine product quality and promote technology development. Machine learning has been widely utilized to construct data-driven solutions for high-accuracy predictions. However, the internal mechanisms of batteries are sensitive to many factors, such as charging/discharging protocols, manufacturing/storage conditions, and usage patterns. These factors will induce state transitions, thereby decreasing the prediction accuracy of data-driven approaches. Transfer learning is a promising technique that overcomes this difficulty and achieves accurate predictions by jointly utilizing information from various sources. Hence, we develop two transfer learning methods, Bayesian Model Fusion and Weighted Orthogonal Matching Pursuit, to strategically combine prior knowledge with limited information from the target dataset to achieve superior prediction performance. From our results, our transfer learning methods reduce root-mean-squared error by 41% through adapting to the target domain. Furthermore, the transfer learning strategies identify the variations of impactful features across different sets of batteries and therefore disentangle the battery degradation mechanisms and the root cause of state transitions from the perspective of data mining. These findings suggest that the transfer learning strategies proposed in our work are capable of acquiring knowledge across multiple data sources for solving specialized issues.

25 ENERGY STORAGE↗
DOE OSTI2024

An Explainable Classification Framework for Determining and Understanding the Suitability of Solvent Extraction for Bioproduct Recovery

Lignocellulosic biomass is an abundant feedstock for producing sustainable fuels and chemicals. However, a key challenge in most biomass utilization strategies is the recovery of products from a dilute, typically aqueous, phase. In this respect, liquid–liquid extraction, which relies on a solvent to transfer a product of interest from one liquid phase to another (solvent-rich) phase, is a technology that can reduce the energy requirements for product recovery. To reduce solvent consumption, liquid–liquid extraction needs to be combined with another separation method (e.g., distillation) to recycle the solvent. Despite the research on solvent extraction, there are limited system-wide methods that allow us to determine when extraction is well suited to carry out a specific separation. Accordingly, we present a classification framework to predict whether extraction, coupled with distillation, is feasible and more economical than distillation. Our framework is based on features such as feed composition, liquid–liquid equilibrium constants, relative volatilities, and solvent price, and leads to trained classifiers that show good prediction accuracy. We further study how specific features influence the suitability of extraction. Furthermore, to showcase the applicability of the framework, we use it to analyze the separation of acetic acid from water.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Explainable AI for a No-Teardown Vehicle Component Cost Estimation: A Top-Down Approach

The broader ambition of this article is to popularize an approach for the fair distribution of the quantity of a system's output to its subsystems while allowing for underlying complex subsystem level interactions. Particularly, we present the use of this framework on a very specific (but generalizable) application, interesting for a more general AI audience. Here, we detail a data-driven approach to vehicle price modeling and its component price estimation by leveraging a combination of concepts from machine learning and game theory. We show an alternative to common teardown methodologies and surveying approaches for component and vehicle price estimation at the manufacturer's suggested retail price (MSRP) level that has the advantage of bypassing uncertainties involved in gathering teardown data, the need to perform expensive and biased surveying, and the need to perform retail price equivalent or indirect cost multiplier adjustments to mark up direct manufacturing costs to MSRP. This novel exercise not only provides accurate pricing of the technologies at the customer level, but also shows the, a priori known, large gaps in pricing strategies between manufacturers, vehicle classes, market segments, etc. There is also clear interaction between the price of technologies and other specifications present in vehicles. Those results are indication that old methods of manufacturer-level component costing, aggregation, and application of flat and rigid adjustment factors should be carefully examined. The findings are based on a database developed by Argonne, which includes over 64,000 vehicles covering MY1990 to MY2020 with hundreds of vehicle specs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Reward Driven Workflows for Unsupervised Explainable Analysis of Phases and Ferroic Variants From Atomically Resolved Imaging Data

Rapid progress in aberration corrected electron microscopy necessitates development of robust methods for the identification of phases, ferroic variants, and other pertinent aspects of materials structure from imaging data. While unsupervised methods for clustering and classification are widely used for these tasks, their performance can be sensitive to hyperparameter selection in the analysis workflow. In this study, the effects of descriptors and hyperparameters are explored on the capability of unsupervised ML methods to distill local structural information, exemplified by the discovery of polarization and lattice distortion in Sm − dopped BiFeO 3 (BFO) thin films. It is demonstrated that a reward-driven approach can be used to optimize these key hyperparameters across the full workflow, where rewards are designed to reflect domain wall continuity and straightness, ensuring that the analysis aligns with the material's physical behavior. This approach allows the discovery of local descriptors that are best aligned with the specific physical behavior, providing insight into the fundamental physics of materials. The reward driven workflow is further extended to disentangle structural factors of variation via an optimized variational autoencoder (VAE). Lastly, the importance of well-defined rewards is explored as a quantifiable measure of the success of the workflow.

Barakati, Kamyar [University of Tennessee, Knoxvil↗
DOE OSTI2024

Suppressed Lone Pair Electrons Explain Unconventional Rise of Lattice Thermal Conductivity in Defective Crystalline Solids

Manipulating thermal properties of materials can be interpreted as the control of how vibrations of atoms (known as phonons) scatter in a crystal lattice. Compared to a perfect crystal, crystalline solids with defects are expected to have shorter phonon mean free paths caused by point defect scattering, leading to lower lattice thermal conductivities than those without defects. While this is true in many cases, alloying can increase the phonon mean free path in the Cd-doped AgSnSbSe 3 system to increase the lattice thermal conductivity from 0.65 to 1.05 W m –1 K –1 by replacing 18% of the Sb sites with Cd. It is found that the presence of lone pair electrons leads to the off-centering of cations from the centrosymmetric position of a cubic lattice. X-ray pair distribution function analysis reveals that this structural distortion is relieved when the electronic configuration of the dopant element cannot produce lone pair electrons. Furthermore, a decrease in the Grüneisen parameter with doping is experimentally confirmed, establishing a relationship between the stereochemical activity of lone pair electrons and the lattice anharmonicity. The observed “harmonic” behavior with doping suggests that lone pair electrons must be preserved to effectively suppress phonon transport in these systems.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Thermodynamic Water Activity Explains the Unusual Electrochemical Stability of Aqueous Deep Eutectic Solvents

The presence of water in nonaqueous deep eutectic solvent (DES) electrolytes has been debated in recent years, with efforts ranging from its complete removal to willful addition. It was shown that controlled amounts of water can be beneficial, as it not only enhances the physicochemical properties of these electrolytes but also has no significant detrimental effects on their electrochemical stability. Despite these advantages, there is still limited understanding of how water interacts with DES systems at the molecular level. This study examines the water activity in ethylene glycol and glycerol, as well as their binary mixtures with choline chloride to form the DESs ethaline and glyceline, respectively. In this work, we show that the high electrochemical stability of glyceline is related to its lower water activity compared to ethaline and can be attributed to the robust H-bonding network formed by the three hydroxyl groups of glycerol. Its 3D H-bond network effectively integrates water molecules within its solvent structure, reducing degradation and maintaining stability at higher water contents. The deviations from the ideal Raoult's law behavior are reflected in the water activity and activity coefficients, which highlight the intricate H-bond interactions within DES-water mixtures. Water acts like a lubricant within the more viscous DES mixtures without being detrimental to their electrochemical performance. The presented results emphasize the necessity of customizing DES-water compositions to enhance their performance as electrolytes, especially in flow battery applications where electrochemical stability, ionic conductivity, and fluidity are of utmost importance.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Leaf traits and canopy structure together explain canopy functional diversity: an airborne remote sensing approach

Plant functional diversity is strongly connected to photosynthetic carbon assimilation in terrestrial ecosystems. However, many of the plant functional traits that regulate photosynthetic capacity, including foliar nitrogen concentration and leaf mass per area, vary significantly between and within plant functional types and vertically through forest canopies, resulting in considerable landscape-scale heterogeneity in three dimensions. Hyperspectral imagery has been used extensively to quantify functional traits across a range of ecosystems but is generally limited to providing information for top-of-canopy leaves only. On the other hand, lidar data can be used to retrieve the vertical structure of forest canopies. Because these data are rarely collected at the same time, there are unanswered questions about the effect of forest structure on the 3-D spatial patterns of functional traits across ecosystems. In the United States, the National Ecological Observatory Network's Airborne Observation Platform (NEON AOP) provides an opportunity to address this structure-function relationship by collecting lidar and hyperspectral data together across a variety of ecoregions. With a fusion of hyperspectral and lidar data from the NEON AOP and field-collected foliar trait data, we assessed the impacts of forest structure on spatial patterns of N. In addition, we examine the influence of abiotic gradients and management regimes on top of canopy %N and total canopy N (i.e. the total amount of N (g/m 2 ) within a forest canopy) at a NEON site consisting of a mosaic of open longleaf pine and dense broadleaf deciduous forests. Here, our resulting maps suggest that in contrast with top-of-canopy values, total canopy N variation is dampened across this landscape resulting in relatively homogeneous spatial patterns. At the same time, we found that leaf functional diversity and canopy structural diversity showed distinct dendritic patterns related to the spatial distribution of plant functional types.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Explainability and extrapolation of machine learning models for predicting the glass transition temperature of polymers

Abstract Machine learning (ML) offers promising tools to develop surrogate models for polymers' structure–property relations. Surrogate models can be built upon existing polymer data and are useful for rapidly predicting the properties of unknown polymers. The accuracy of such ML models appears to depend on the feature space representation of polymers, the range of training data, and learning algorithms. Here, we establish connections between these factors for predicting the glass transition temperature (T g ) of polymers. Our analysis suggests linear models with fewer fitting parameters are as accurate as nonlinear models with many hidden and unexplainable parameters. Also, the performance of a monomer topology‐based ML model is found to be qualitatively identical to that of a physicochemical descriptor‐based ML model. We find that the ML models's performance in the extrapolative region is enhanced as the property range of the training data increases. Moreover, we establish newT g – polymer chemistry correlations via ML. Our work illustrates how ML can advance the fundamental understanding of polymer structure–property correlations and its efficacy for extrapolation problems.

Polymer Science↗
DOE OSTI2021

Explaining and predicting human behavior and social dynamics in simulated virtual worlds: reproducibility, generalizability, and robustness of causal discovery methods

Ground Truth program was designed to evaluate social science modeling approaches using simulation test beds with ground truth intentionally and systematically embedded to understand and model complex Human Domain systems and their dynamics Lazer et al. (Science 369:1060–1062, 2020). Our multidisciplinary team of data scientists, statisticians, experts in Artificial Intelligence (AI) and visual analytics had a unique role on the program to investigate accuracy, reproducibility, generalizability, and robustness of the state-of-the-art (SOTA) causal structure learning approaches applied to fully observed and sampled simulated data across virtual worlds. In addition, we analyzed the feasibility of using machine learning models to predict future social behavior with and without causal knowledge explicitly embedded. In this paper, we first present our causal modeling approach to discover the causal structure of four virtual worlds produced by the simulation teams—Urban Life, Financial Governance, Disaster and Geopolitical Conflict. Our approach adapts the state-of-the-art causal discovery (including ensemble models), machine learning, data analytics, and visualization techniques to allow a human-machine team to reverse-engineer the true causal relations from sampled and fully observed data. We next present our reproducibility analysis of two research methods team’s performance using a range of causal discovery models applied to both sampled and fully observed data, and analyze their effectiveness and limitations. We further investigate the generalizability and robustness to sampling of the SOTA causal discovery approaches on additional simulated datasets with known ground truth. Our results reveal the limitations of existing causal modeling approaches when applied to large-scale, noisy, high-dimensional data with unobserved variables and unknown relationships between them. We show that the SOTA causal models explored in our experiments are not designed to take advantage from vasts amounts of data and have difficulty recovering ground truth when latent confounders are present; they do not generalize well across simulation scenarios and are not robust to sampling; they are vulnerable to data and modeling assumptions, and therefore, the results are hard to reproduce. Finally, when we outline lessons learned and provide recommendations to improve models for causal discovery and prediction of human social behavior from observational data, we highlight the importance of learning data to knowledge representations or transformations to improve causal discovery and describe the benefit of causal feature selection for predictive and prescriptive modeling.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Explainable discrepancy checker and diagnosis for digital Twin-based supervisory control system

By virtually representing a physical object and process, a digital twin (DT) enables optimal autonomous operations by combining classical and novel frameworks in sensors, state predictions, and multi-input/multi-output systems. A DT’s values depend on how well models estimate quantities of interest and on how uncertainty is handled. Moreover, DTs often combine physics-based and data-driven models with mixed fidelities, where classical uncertainty quantification (UQ) struggles with many sources of uncertainty and real-time constraints. Here, this work presents a UQ-based discrepancy checking and diagnosis tool for a DT-based supervisory control system. The tool is developed using metadata from an automated DT development process to learn correlations between sources of uncertainties and outcomes. During operation, it compares predictions with measurements, attributes discrepancies to dominant sources, and recommends parameter and configuration updates. We verify the workflow on a synthetic temperature-control problem and deploy it on a virtual Thermal Energy Delivery System, reducing mismatch and improving control robustness.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗