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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Machine learning in materials science: From explainable predictions to autonomous design

The advent of big data and algorithmic developments in the field of machine learning (and artificial intelligence, in general) have greatly impacted the entire spectrum of physical sciences, including materials science. Materials data, measured or computed, combined with various techniques of machine learning have been employed to address a myriad of challenging problems, such as, development of efficient and predictive surrogate models for a range of materials properties, screening and down-selection of novel candidate materials for targeted applications, new methodologies to improve and further expedite molecular and atomistic simulations, with likely many more important developments to come in the foreseeable future. While the applications thus far have provided a glimpse of the true potential data-enabled routes have to offer, it has also become clear that further progress in this direction hinges on our ability to understand, explain and rationalize findings of a machine learning model in light of the domain-knowledge. This focused review provides an overview of the main areas where machine learning has been widely and successfully used in materials science. Subsequently, a brief discussion of several techniques that have been helpful in extracting physically-meaningful insights, causal relationships and design-centric knowledge from materials data is provided. Finally, we identify some of the imminent opportunities and challenges that materials community faces in this exciting and rapidly growing field.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Towards physics-informed explainable machine learning and causal models for materials research

From emergent material descriptions to estimation of properties stemming from structures to optimization of process parameters for achieving best performance – all key facets of materials science and related fields have experienced tremendous growth with the introduction of data-driven models. This gradual progression goes at par with developments of machine learning workflows, from purely data-driven shallow models to those that are well-capable in encoding more complex graphs, symbolic representations, invariances, and positional embeddings. Furthermore, this perspective aims at summarizing strategic aspects of such transitions while providing insights into the requirements of bringing in explainable, interpretable predictive models, and causal learning to aid in materials design and discovery. Although the focus remains on a variety of functional materials by providing a handful of case studies, the applications of such integrated methodologies are universal to facilitate fundamental understandings of materials physics while enabling autonomous experiments.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Explaining drivers of housing prices with nonlinear hedonic regressions

Housing markets play a critical role in shaping the spatial and demographic evolution of urban areas. Simulating housing price dynamics can enhance projections of future urban development outcomes. However, traditional hedonic regressions for housing prices, which neglect nonlinear interactions among explanatory variables, often exhibit limited predictive performance. While machine learning (ML) methods can provide a more flexible representation of the relationships between predictors, they are often regarded as “black boxes” due to their complexity and lack of transparency. Interpretable ML techniques provide a promising route by combining the flexibility of ML methods with approaches to analyze the relationships between inputs and outputs. In this study, we employ interpretable ML to analyze the patterns driving the housing market in Baltimore, Maryland, USA. We train an Artificial Neural Network (ANN) to predict Baltimore housing prices based on structural characteristics (e.g., home size, number of stories) and locational attributes (e.g., distance to the city center). We then conduct sensitivity and Partial Dependence Plot (PDP) analyses to interpret the fitted ANN model. We find that the ML model achieves higher predictive accuracy and explains 16 % more of housing price variance than a traditional linear regression model. The interpretable ML model also reveals more nuanced and realistic nonlinear relationships between housing sales price and predictors as well as interactive effects underlying Baltimore home price dynamics. For instance, while the linear model indicates a steady housing price increase over time, our interpretable ML model detects a post-2008 decline, with smaller properties experiencing the sharpest drop.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Detecting thermodynamic phase transition via explainable machine learning of photoemission spectroscopy

Identifying thermodynamic signatures of electronic phases, such as superconductivity, is challenging in low-dimensional materials due to strong fluctuations and low probing volume. Spectroscopic methods are often used to identify new bulk phases, but their main measurable quantity—electronic energy gaps—is no longer an effective order parameter in low-dimensional and fluctuating systems. Combining angle-resolved photoemission with a domain-adversarial neural network, we report a data-driven method to identify thermodynamic phase transitions solely based on single-particle spectra. We demonstrate 97.6% accuracy in cuprate superconductor Bi 2 Sr 2 CaCu 2 O 8+δ with strong superconducting fluctuations. This model notably compensates for the scarcity of experimental data by leveraging virtually inexhaustible simulated data. Further, its explainability reveals the crucial role of in-gap spectral weight in detecting phase fluctuations and thermodynamic transitions. Our work pinpoints the spectroscopic signatures of fluctuating orders and enables using spectroscopy for machine-learning-assisted material discovery for low-dimensional and strong coupling systems.

2D materials↗
DOE OSTI2022

Explainable machine learning of the underlying physics of high-energy particle collisions

We present an implementation of an explainable and physics-aware machine learning model capable of inferring the underlying physics of high-energy particle collisions using the information encoded in the energy-momentum four-vectors of the final state particles. We demonstrate the proof-of-concept of our White Box AI approach using a Generative Adversarial Network (GAN) which learns from a DGLAP-based parton shower Monte Carlo event generator. The constrained generator network architecture mimics the structure of a parton shower exhibiting similarities with Recurrent Neural Networks (RNNs). We show, for the first time, that our approach leads to a network that is able to learn not only the final distribution of particles, but also the underlying parton branching mechanism, i.e. the Altarelli-Parisi splitting function, the ordering variable of the shower, and the scaling behavior. While the current work is focused on perturbative physics of the parton shower, we foresee a broad range of applications of our framework to areas that are currently difficult to address from first principles in QCD. Examples include nonperturbative and collective effects, factorization breaking and the modification of the parton shower in heavy-ion, and electron-nucleus collisions.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2023

Explaining the GeV excess with exploding black holes

Black holes may form in present-day collapse of microscopic structures of dark matter. We show that, if microstructure black holes (MSBH) with mass m ~ 10 13 g are produced, the spectrum of gamma rays from their evaporation agrees remarkably well with the GeV excess observed by Fermi Gamma-ray Space Telescope, while still avoiding all observational constraints. We also discuss the generic requirements for MSBHs to explain the GeV excess.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

An Application of ASP in Nuclear Engineering: Explaining the Three Mile Island Nuclear Accident Scenario

Abstract The paper describes an ongoing effort in developing a declarative system for supporting operators in the Nuclear Power Plant (NPP) control room. The focus is on two modules: diagnosis and explanation of events that happened in NPPs. We describe an Answer Set Programming (ASP) representation of an NPP, which consists of declarations of state variables, components, their connections, and rules encoding the plant behavior. We then show how the ASP program can be used to explain the series of events that occurred in the Three Mile Island, Unit 2 (TMI-2) NPP accident, the most severe accident in the USA nuclear power plant operating history. We also describe an explanation module aimed at addressing answers to questions such as “why an event occurs?” or “what should be done?” given the collected data.

Computer Science↗
DOE OSTI2025

Increasing the Scale of the Mass Spectrometry Query Language Compendium with Explainable AI

A significant bottleneck in metabolomics data interpretation is the effective use of domain knowledge to assign structural information based on fragmentation patterns. The mass spectrometry query language (MassQL) aims to make this process accessible and applicable across multiple analysis platforms. While advanced computational methods are capable of predicting compound structures from fragmentation data, AI/ML approaches often rely on complex, opaque criteria that are difficult to interpret or modify. As a result, their predictive patterns cannot be readily translated into human-readable rules, such as those used in MassQL. Here, in this study, we introduce ChemEcho, a machine learning embedding method that converts tandem mass spectrometry data into sparse feature vectors containing peak and neutral mass subformulae to enhance explainable AI/ML-based methods. An advantage of this approach is that decision trees trained using these feature vectors can be directly translated to MassQL. Using a battery of decision trees trained using ChemEcho embeddings to predict molecular attributes, we generated over 1500 MassQL queries for 765 molecular features and evaluated their precision and recall. From these queries, the 50 highest-performing queries were integrated into the MassQL compendium. This set of generated MassQL queries included environmentally and biologically relevant classes such as PFAS and molecules containing phosphate or sulfate substructures. To illustrate the impact these queries would have on a typical metabolomics experiment, these MassQL queries were applied to a public metabolomics data set─resulting in a marked increase in the structural information derived from tandem mass spectra. Access and reuse of these queries is expected to enhance structural annotation in untargeted experiments, leading to more specific claims and advancing many applications in metabolomics.

Harwood, Thomas V. [USDOE Joint Genome Institute (↗
DOE OSTI2024

Microbial Proxies for Anoxic Microsites Vary with Management and Partially Explain Soil Carbon Concentration

Anoxic microsites are potentially important but unresolved contributors to soil organic carbon (C) storage. How anoxic microsites vary with soil management and the degree to which anoxic microsites contribute to soil C stabilization remain unknown. Sampling from four long-term agricultural experiments in the central United States, we examined how anoxic microsites varied with management (e.g., cultivation, tillage, and manure amendments) and whether anoxic microsites determine soil C concentration in surface (0–15 cm) soils. We used a novel approach to track anaerobe habitat space and, hence, anoxic microsites using DNA copies of anaerobic functional genes over a confined volume of soil. No-till practices inconsistently increased anoxic microsite extent compared to conventionally tilled soils, and within one site organic matter amendments increased anaerobe abundance in no-till soils. Across all long-term tillage trials, uncultivated soils had ~2–4 times more copies of anaerobic functional genes than their cropland counterparts. Finally, anaerobe abundance was positively correlated to soil C concentration. Even when accounting for other soil C protection mechanisms, anaerobe abundance, our proxy for anoxic microsites, explained 41% of the variance and 5% of the unique variance in soil C concentration in cropland soils, making anoxic microsites the strongest management-responsive predictor of soil C concentration. Our results suggest that careful management of anoxic microsites may be a promising strategy to increase soil C storage within agricultural soils.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Distinct Chemistries Explain Decoupling of Slip and Wettability in Atomically Smooth Aqueous Interfaces

Despite essentially identical crystallography and equilibrium structuring of water, nanoscopic channels composed of hexagonal boron nitride and graphite exhibit an order-of-magnitude difference in fluid slip. We investigate this difference using molecular dynamics simulations, demonstrating that its origin is in the distinct chemistries of the two materials. In particular, the presence of polar bonds in hexagonal boron nitride, absent in graphite, leads to Coulombic interactions between the polar water molecules and the wall. We demonstrate that this interaction is manifested in a large typical lateral force experienced by a layer of oriented hydrogen atoms in the vicinity of the wall, leading to the enhanced friction in hexagonal boron nitride. The fluid adhesion to the wall is dominated by dispersive forces in both materials, leading to similar wettabilities. Finally, our results rationalize recent observations that the difference in frictional characteristics of graphite and hexagonal boron nitride cannot be explained on the basis of the minor differences in their wettabilities.

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

Single Nanoflake Photoelectrochemistry Reveals Intrananoflake Doping Heterogeneity That Explains Ensemble-Level Photoelectrochemical Behavior

Transition metal dichalcogenide (TMD) nanoflake thin films are attractive electrode materials for photoelectrochemical (PEC) solar energy conversion and sensing applications, but their photocurrent quantum yields are generally lower than those of bulk TMD electrodes. The poor PEC performance has been primarily attributed to enhanced charge carrier recombination at exposed defect and edge sites introduced by the exfoliation process. Here, a single nanoflake PEC approach reveals how an alternative effect, doping heterogeneity, limits ensemble-level PEC performance. Photocurrent mapping and local photocurrent–potential (i–E) measurements of MoS 2 nanoflakes exfoliated from naturally occurring bulk crystals revealed the presence of n- and p-type domains within the same nanoflake. Interestingly, the n- and p-type domains in the natural MoS 2 nanoflakes were equally efficient for iodide oxidation and tri-iodide reduction (IQE values exceed 80%). At the single domain-level, the natural MoS 2 nanoflakes were nearly as efficient as nanoflakes exfoliated from synthetic n-type MoS 2 crystals. Single domain-level i–E measurements explain why natural MoS 2 nanoflakes exhibit an n-type to p-type photocurrent switching effect in ensemble-level measurements: the n- and p-type diode currents from individual domains oppose each other upon illuminating the entire nanoflake, resulting in zero photocurrent at the switching potential. Furthermore, the doping heterogeneity effect is likely due to nonideal stoichiometry, where p-type domains are S-rich according to XPS measurements. Although this doping heterogeneity effect limits photoanode or photocathode performance, these findings open the possibility to synthesize efficient TMD nanoflake photocatalysts with well-defined lateral p- and n-type domains for enhanced charge separation.

14 SOLAR ENERGY↗
DOE OSTI2023

Biphasic Janus Particles Explain Self-Healing in Pt–Pd Diesel Oxidation Catalysts

The addition of Pd to Pt-based diesel oxidation catalysts is known to enhance performance and restrict the anomalous growth of Pt nanoparticles when subjected to aging at high temperatures in oxidative environments. To gain a mechanistic understanding, we studied the transport of the mobile Pt and Pd species to the vapor phase, since vapor phase transport is the primary route for sintering in these catalysts. The results are surprising: there is a 30-fold drop in the effective vapor pressure of Pt in the Pt-Pd catalysts compared to monometallic Pt. At the same time, there is a significant enhancement in the vapor pressure of Pd, compared to PdO, which otherwise has a negligible vapor pressure at the aging temperature. Such behavior cannot be explained simply by alloying Pt and Pd in the metallic phase, or a core-shell morphology where a PdO shell covers a Pt core. Transmission electron microscopic examination of catalysts aged up to 50 h in air at 800 °C shows that the particles exhibit a biphasic “Janus”-like structure. The metal and oxide phases are conjoined, exposing a metal and an oxide face to the gas phase. The high mobility of the Pt and Pd allows them to be partitioned into the metal and oxide phases, in apparent thermodynamic equilibrium. The PdO helps to trap mobile PtO2 and as a result contains high concentrations of Pt oxide, consistent with its role in mitigating the transport of Pt to the vapor phase and preventing the growth of anomalously large particles. In turn, Pt allows Pd to remain metallic, allowing the catalyst to retain both metal and oxide functionality for catalysis. The regeneration of deactivated catalysts typically requires an external input, such as a change in the working environment from reducing to oxidizing or vice-versa. Here, we show that the mobile species, which are primary contributors to catalyst sintering are effectively returned to the active site, hence our use of the term “selfhealing”. The detailed insights into the inner workings of the Pt-Pd diesel oxidation catalysts can help provide clues to the design of robust and durable heterogeneous catalysts.

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

Silver Adsorption on Calcium Niobate(001) Nanosheets: Calorimetric Energies That Explain Sinter-Resistant Support

Metal nanoparticles deposited on oxide supports are essential to many technologies, including catalysts, fuel cells and electronics. Therefore, understanding the chemical bonding strength between metal nanoparticles and oxide surfaces is of great interest. The adsorption energetics, adhesion energy and adsorbate structure of Ag on dehydrated HCa 2 Nb 3 O 10 (001) nanosheets at 300 K have been studied using metal adsorption calorimetry and surface spectroscopies. These dehydrated (“dh”) calcium niobate nanosheets “(dh-HCa 2 Nb 3 O 10 (001))” have stoichiometry Ca 4 Nb 6 O 19 . They impart unusual stability to metal nanoparticles when used as catalyst supports and are easy-to-prepare by Langmuir-Blodgett (LB) techniques, highly ordered, and essentially single-crystal surfaces of mixed oxides with a huge ratio of terrace to edge sites. Below monolayer coverage, Ag grows on dh-HCa 2 Nb 3 O 10 (001) as 2D islands of thickness ~2 layers. The differential heat of Ag adsorption is initially ~303 kJ/mol, increasing slowly to ~338 kJ/mol by 0.8 ML. At higher coverages, Ag atoms mainly add on top of these 2D islands, growing 3D nanoparticles of increasing thickness, as the heat decreases asymptotically towards silver’s heat of sublimation (285 kJ/mol). The adhesion energy of Ag(s) to this Ca niobate surface is estimated to be 4.33 J/m 2 , larger than on any oxide surface previously measured. This explains the sinter resistance reported for metal nanoparticles on this support. Electron transfer from Ag into the calcium niobate is also measured. Furthermore, these results demonstrate an easy way to do singlecrystal-type surface science studies - and especially thermochemical measurements - on the complex surfaces of mixed oxides: using LB-deposited perovskite nanosheets and ultrahigh vacuum annealing in O 2 .

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

Sub-Cloud Turbulence Explains Cloud-Base Updrafts for Shallow Cumulus Ensembles: First Observational Evidence

Sub-cloud turbulent kinetic energy has been used to parameterize the cloud-base updraft velocity (w b ) in cumulus parameterizations. The validity of this idea has never been proved in observations. Instead, it was challenged by recent Doppler Lidar observations showing a poor correlation between the two. We argue that the low correlation is likely caused by the difficulty of a fixed-point Lidar to measure ensemble properties of cumulus fields. Taking advantage of the stationarity and ergodicity of early afternoon convection, we developed a Lidar sampling methodology to measure w b of a shallow cumulus (ShCu) ensemble (not a single ShCu). By analyzing 128 ShCu ensembles over the Southern Great Plains, we show that the ensemble properties of sub-cloud turbulence explain nearly half of the variability in ensemble-mean wb, demonstrating the ability of sub-cloud turbulence to dictate w b . Lastly, the derived empirical formulas will be useful for developing cumulus parameterizations and satellite inference of w b .

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Mineral Protection and Resource Limitations Combine to Explain Profile-Scale Soil Carbon Persistence

The fate of soil carbon (C) is largely controlled by microbial oxidation of organic matter (OM), which is constrained by a variety of mechanisms. OM association with soil minerals provides pronounced protection against microbial decomposition. However, factors such as climate, occlusion, and resource limitations also contribute to OM preservation. Here, we explore the factors explaining C distribution and age within an upland rainforest soil in Hawai'i, a site with abundant preferential flow paths (PFPs) and high short-range order (SRO) mineral content. We characterized lateral and vertical changes in Δ 14 C, SRO mineral content, C-functional group chemistry, and microbial community composition to elucidate the contributions of multiple protection mechanisms to OM preservation. Consistent with our expectation, SRO mineral content and Δ 14 C were strongly correlated (R 2 = 0.95), indicating strong mineral protection of OM throughout the profile. Surprisingly, distance from PFP was also a significant predictor of Δ 14 C and improved model fit, particularly in the shallow horizons (R 2 = 0.97). Elevated C/N ratios, decreased microbial abundance, and greater SRO mineral content suggest nitrogen limitations and enhanced mineral protection constrain OM turnover with distance from PFPs in deep, subsurface mineral horizons. Steady microbial abundance, increasing putative anaerobe abundance, and changes in C-functional group chemistry indicate oxygen limitations constrain OM turnover in the matrix of shallow mineral horizons. Given that oxygen and nutrient limitations contribute to OM preservation in this high SRO system—an exemplar of mineral protection—resource limitations may play an even more important role in OM preservation in other well-structured soils.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

PM 2.5 Is Insufficient to Explain Personal PAH Exposure

To understand how chemical exposure can impact health, researchers need tools that capture the complexities of personal chemical exposure. In practice, fine particulate matter (PM 2.5 ) air quality index (AQI) data from outdoor stationary monitors and Hazard Mapping System (HMS) smoke density data from satellites are often used as proxies for personal chemical exposure, but do not capture total chemical exposure. Silicone wristbands can quantify more individualized exposure data than stationary air monitors or smoke satellites. However, it is not understood how these proxy measurements compare to chemical data measured from wristbands. In this study, participants wore daily wristbands, carried a phone that recorded locations, and answered daily questionnaires for a 7-day period in multiple seasons. We gathered publicly available daily PM 2.5 AQI data and HMS data. We analyzed wristbands for 94 organic chemicals, including 53 polycyclic aromatic hydrocarbons. Wristband chemical detections and concentrations, behavioral variables (e.g., time spent indoors), and environmental conditions (e.g., PM 2.5 AQI) significantly differed between seasons. Machine learning models were fit to predict personal chemical exposure using PM 2.5 AQI only, HMS only, and a multivariate feature set including PM 2.5 AQI, HMS, and other environmental and behavioral information. On average, the multivariate models increased predictive accuracy by approximately 70% compared to either the AQI model or the HMS model for all chemicals modeled. This study provides evidence that PM 2.5 AQI data alone or HMS data alone is insufficient to explain personal chemical exposures. Our results identify additional key predictors of personal chemical exposure.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗
DOE OSTI2024

Fast Seismic Anomalies Under Continents Explained by the Delaminated Lower Continental Crust—Implications From High Pressure‐Temperature Elasticity of Jadeite

Seismic tomography has shown that the shear wave velocities (Vs) under continents, especially under cratons, are extremely fast at 100–200 km depth, which is difficult to explain by low temperatures or high Mg#. Alternatively, delaminated eclogitic lower continental crust has been proposed to account for these fast seismic anomalies. However, the thermoelastic properties of jadeite which constitutes up to 60–80 mol% of clinopyroxene in the potentially delaminated lower continental crust are not well constrained. In this study, we measured the single-crystal elasticity of jadeite by Brillouin spectroscopy under simultaneous high pressure and temperature conditions for the first time. We found that the temperature dependence of Vs of jadeite is extremely small if not negligible. The seismic velocities of the potentially delaminated lower continental crusts were subsequently modeled and found to match the widely observed fast seismic anomalies under cratons between 100 and 200 km depth.

Geology↗
DOE OSTI2023

Projection of Future Fire Emissions Over the Contiguous US Using Explainable Artificial Intelligence and CMIP6 Models

Abstract Increasing temperature and water cycle changes due to warming climate may increase the frequency and intensity of wildfires. Fire emission projections are useful for informing strategies for adaptation and mitigation of fire impacts on societies and ecosystems. Here, we construct a neural network (NN) model explained by the Shapley Additive explanation to predict fire PM 2.5 emissions change and understand their drivers over the contiguous US (CONUS) in the mid‐21st century under a high greenhouse gas emissions scenario (SSP5‐8.5). Using future meteorology and leaf area index (LAI) simulated by eight global climate models from the Coupled Model Intercomparison Project Phase 6, future population density, and present‐day land use and land cover (LULC) as input to the NN model, the total fire PM 2.5 emissions over CONUS are projected to increase by 4%–75% (model spread). Among different regions, fire emissions in the western US are projected to increase more significantly in June‐July‐August than in other seasons and regions, with the median ratios of future to present‐day fire emissions ranging from 1.67 to 2.86. The increases in fire emissions are mainly driven by increasing normalized temperature (23%–29%) and decreasing soil moisture (2%–10%) in the future. When future LULC change is considered, the projected fire emissions further increase by 58%–83% over the western US compared to projections without LULC change because of future increases in vegetation fraction. The results highlight the important role of warmer temperature, decreasing soil moisture, and LULC change in increasing fire emissions in the future.

54 ENVIRONMENTAL SCIENCES↗