Search NASA⌕ Search

SEARCH · Search NASA

Results for “Distributed training”

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 1,495 records · Page 83

Smart Methane Emission Detection System Development (Final Report)

Working with the Department of Energy's National Energy Technology Laboratory, Southwest Research Institute® (SwRI®) developed a system to identify methane leaks reliably, accurately, and autonomously at critical midstream sections of the natural gas distribution network in real-time for the purpose of mitigating methane emissions using Optical Gas Imaging (OGI) cameras. SwRI's Smart Leak Detection – Methane (SLED/M) adds a high degree of automation to the process of methane leak detection to minimize sources of human error, minimize response time to a leak event, and maximize midstream visibility. Furthermore, SwRI has been working towards integrating Quantitative OGI (QOGI) capabilities into this existing technology. By leveraging Deep Learning, SwRI now has the capability to estimate fugitive emission leak rates quickly and reliably, which allows operators to detect emissions, quantify leak rate, prioritize repairs, and validate the repairs in a single instrument. The next generation QOGI technology leverages the same cameras used in Leak Detection and Repair (LDAR) programs, with improvements in safety and speed for traditional quantification-based repairs, ultimately leading to less overhead cost for the operators. The goals for this research were to develop two types of models with the following goals: Run in real-time on the edge (≥ 12 Hz), Classification: Achieve less than 5% false positive detection, Classification: Achieve ≥ 95% methane plume detection rate, Regression: achieve ≤ 10 standard cubic feet per hour (scfh) prediction > 70% of the time. In order to achieve these results, multiple infrared (IR) and other sensors were investigated in tandem with the midwave IR (MWIR) OGI to provide additional information to train the underlying models. Information on atmospheric conditions including humidity, temperature, pressure, and solar radiation was provided by a weather station. Several machine learning and deep learning architectures and methods, including looking at quantized classification networks and regressions networks, were explored. As further data was collected, curated, and labeled, it allowed for more refined regressive networks to be adequately trained, leading to better insight into the true flow rates being observed. An important valuable deliverable of this research effort was the development of an advanced network which underwent multiple iterations capable of giving a continuous output. The current network has a predicted mean average percentage error (MAPE) of 12.3% just outside our target goal of 10.00%, but an accuracy of 97.78% at ±50 scfh, well within the overall goal for the Department of Energy (DOE) program. Upon closer inspection, it was observed that more than 10% of datapoints contributing to the MAPE predictions were the result of low flow rate predictions and are beyond the sensitivity of instrument measurement as a result of normal operational variation and noise.

03 NATURAL GAS↗

Predicting biomass comminution: Physical experiment, population balance model, and deep learning

An extended population balance model (PBM) and a deep learning-based enhanced deep neural operator (DNO+) model are introduced for predicting particle size distribution (PSD) of comminuted biomass through a large knife mill. Experimental tests using corn stalks with varied moisture contents, mill blade speeds, and discharge screen sizes are conducted to support model development. A novel mechanism in the extended PBM allows for including additional input parameters such as moisture content, which is not possible in the original PBM. The DNO+ model can include influencing factors of different data types such as moisture content and discharge screen size, which significantly extends the engineering applicability of the standard DNO model that only admits feed PSD and outcome PSD. Test results show that both models are remarkably accurate in the calibration or training parameter space and can be used as surrogate models to provide effective guidance for biomass preprocessing design.

09 BIOMASS FUELS↗

Finding simplicity: unsupervised discovery of features, patterns, and order parameters via shift-invariant variational autoencoders *

Abstract Recent advances in scanning tunneling and transmission electron microscopies (STM and STEM) have allowed routine generation of large volumes of imaging data containing information on the structure and functionality of materials. The experimental data sets contain signatures of long-range phenomena such as physical order parameter fields, polarization, and strain gradients in STEM, or standing electronic waves and carrier-mediated exchange interactions in STM, all superimposed onto scanning system distortions and gradual changes of contrast due to drift and/or mis-tilt effects. Correspondingly, while the human eye can readily identify certain patterns in the images such as lattice periodicities, repeating structural elements, or microstructures, their automatic extraction and classification are highly non-trivial and universal pathways to accomplish such analyses are absent. We pose that the most distinctive elements of the patterns observed in STM and (S)TEM images are similarity and (almost-) periodicity, behaviors stemming directly from the parsimony of elementary atomic structures, superimposed on the gradual changes reflective of order parameter distributions. However, the discovery of these elements via global Fourier methods is non-trivial due to variability and lack of ideal discrete translation symmetry. To address this problem, we explore the shift-invariant variational autoencoders (shift-VAEs) that allow disentangling characteristic repeating features in the images, their variations, and shifts that inevitably occur when randomly sampling the image space. Shift-VAEs balance the uncertainty in the position of the object of interest with the uncertainty in shape reconstruction. This approach is illustrated for model 1D data, and further extended to synthetic and experimental STM and STEM 2D data. We further introduce an approach for training shift-VAEs that allows finding the latent variables that comport to known physical behavior. In this specific case, the condition is that the latent variable maps should be smooth on the length scale of the atomic lattice (as expected for physical order parameters), but other conditions can be imposed. The opportunities and limitations of the shift VAE analysis for pattern discovery are elucidated.

97 MATHEMATICS AND COMPUTING↗

Robust Power System Stability Assessment Against Adversarial Machine Learning-Based Cyberattacks via Online Purification

The increasing complexity associated with renewable generation brings more challenges to power system stability assessment (SA). Data-driven approaches based on machine learning (ML) techniques for stability assessment have received significant research interest and shown their promising performance. However, ML-based models are recognized to be vulnerable to adversarial disturbances, where a slight perturbation to power system measurements could lead to unacceptable errors. To address this issue, this paper develops a novel lightweight mitigation strategy, i.e., robust online stability assessment (ROSA), to enhance the ML-based assessment model against both white-box and the black-box adversarial disturbances (i.e., purification) in the online implementation. The ROSA involves a supervised learning-based module for the primary stability assessment and a self-supervised learning-based module. Further, the two modules are trained jointly with different objective (loss) functions and implemented in sequence. A suitable purification objective and various time-series data augmentation methods are designed for SA applications to tackle adversarial disturbances adaptively. Case studies are performed, and the comparative results have clearly illustrated the competitive, robust accuracy against various adversarial scenarios and verified the effectiveness of the proposed online purification strategy.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Analog, Digital, or Enhanced Human-System Interfaces? Results of an Operator-in-the-Loop Study on Main Control Room Modernization for a Nuclear Power Plant

The second Operator Study of System Overviews (OSSO-2) was conducted with a three-person reactor operator crew in the Human Systems Simulation Laboratory (HSSL) at Idaho National Laboratory (INL) in August, 2017. The study supported control room modernization at a nuclear power plant and featured a benchmark comparison of three variants of a turbine control system (TCS): the existing analog TCS, a proposed standalone digital TCS with two displays, and the digital TCS with the addition of a third display consisting of a system overview screen. TCS prototypes were developed at INL to allow evaluation of operator performance and preferences during realistic turbine scenarios in the full-scope simulator. The study revealed that completion of turbine startup was several minutes faster with the digital TCS variants than with the conventional analog TCS. Eye tracking scan paths were more widely distributed in the overview vs. standalone TCS condition, suggesting the overview screen was cueing reactor operators to verify values across the boards. There was no significant difference in workload or situation awareness across the three interfaces. Reviewing key plant parameters showed smoother transitions during load following for the digital vs. analog TCS. Despite some performance advantages for the digital TCS variants, operators preferred the existing analog TCS. Open-ended responses suggested this finding may be more an artifact of familiarity than a reflection of dissatisfaction with the new TCS. OSSO-2 provides compelling evidence that the new digital TCS was used successfully by the operators without extensive training of rewriting of the operating procedures, suggesting high usability for the digital TCS design. Further advantages were realized through the addition of the system overview screen to provide crews with at-a-glance indicators of key turbine parameters.

99 GENERAL AND MISCELLANEOUS↗

CRADA with FedIMPACT, LLC (Project 1): Cooperative Research and Development CRADA Number CRD-17-00713 (Final Report)

The funds-in under the CRADA will fund a team of National Renewable Energy Laboratory (NREL) researchers to participate in Energy I-Corps (formerly known as Lab-Corps). Energy I-Corps pairs teams of researchers with industry mentors for an intensive two-month training where the researchers define technology value propositions, conduct customer discovery interviews, and develop viable market pathways for their technologies. FedIMPACT, LLC and its affiliate IP Group, Inc. will evaluate the work completed at Energy I-Corps to determine whether it would like to pursue further commercialization and development of related technologies and background intellectual property.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Control of a Three-Phase Grid-Connected Voltage-Sourced Converter Using Long Short-Term Memory Networks

With the rise of inverter-based resources (IBRs) within the power system, the control of grid-connected converters (GCCs) has become pertinent due to the fact they interface IBRs to the grid. The conventional method of control for a GCC such as the voltage-sourced converter (VSC) is through a decoupled control loop in the synchronous reference frame. However, this model-based control method is sensitive to parameter changes causing deterioration in controller performance. Data-driven approaches such as machine learning can be utilized to design controllers that are capable of operating GCCs in various system conditions. This work explores a deep learning-based control method for a three-phase grid-connected VSC, specifically utilizing a long short-term memory (LSTM) network for robust control. Simulations of a conventional controlled VSC are conducted using Simulink to collect data for training the LSTM-based controller. The LSTM model is built and trained using the Keras and TensorFlow libraries in Python and tested in Simulink. The performance of the LSTM-based controller is evaluated under different case studies and compared to the conventional method of control. Simulation results demonstrate the effectiveness of this approach by outperforming the conventional controller and maintaining stability under different system parameter changes.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Quantifying the structural uncertainty of the aerosol mixing state representation in a modal model

Abstract. Aerosol mixing state is an important emergent property that affects aerosol radiative forcing and aerosol–cloud interactions, but it has not been easy to constrain this property globally. This study aims to verify the global distribution of aerosol mixing state represented by modal models. To quantify the aerosol mixing state, we used the aerosol mixing state indices for submicron aerosol based on the mixing of optically absorbing and non-absorbing species (χo), the mixing of primary carbonaceous and non-primary carbonaceous species (χc), and the mixing of hygroscopic and non-hygroscopic species (χh). To achieve a spatiotemporal comparison, we calculated the mixing state indices using output from the Community Earth System Model with the four-mode version of the Modal Aerosol Module (MAM4) and compared the results with the mixing state indices from a benchmark machine-learned model trained on high-detail particle-resolved simulations from the particle-resolved stochastic aerosol model PartMC-MOSAIC. The two methods yielded very different spatial patterns of the mixing state indices. In some regions, the yearly averaged χ value computed by the MAM4 model differed by up to 70 percentage points from the benchmark values. These errors tended to be zonally structured, with the MAM4 model predicting a more internally mixed aerosol at low latitudes and a more externally mixed aerosol at high latitudes compared to the benchmark. Our study quantifies potential model bias in simulating mixing state in different regions and provides insights into potential improvements to model process representation for a more realistic simulation of aerosols towards better quantification of radiative forcing and aerosol–cloud interactions.

54 ENVIRONMENTAL SCIENCES↗

Developing Smart Building Technology Modules to Enhance Workforce Preparedness: A Case for AI-Driven Academic and Professional Education

Smart building technologies are resources that improve building energy efficiency and resilience, reduce carbon emissions, and provide load flexibility to the grid. However, in both academic curricula and building professionals’ continuing education, there is a lack of systematic instruction on methods to integrate multiple energy systems including distributed energy resources (DER), smart building technologies, AI (Artificial Intelligence) tools and key concepts, components, and controls, including “Internet of Things” (IoT) devices. In today’s dynamic workforce, this major gap in smart building technology education prevents stakeholders from being able to attract talent with an understanding and preparation to adopt smart building technologies in building design and operations. A federally funded project included a partnership between Slipstream and Texas A&M University (TAMU) to develop a semester-long smart building curriculum for engineering college students with the ability to adapt the contents for workforce development of professionals in building services. The final product consists of 16 training videos adapted for building professionals and the public. The educational content and training materials cover the benefits of building energy systems, the latest sensor technologies and IoT devices, all with a focus on smart building technologies. The key drivers are on topics related to smart building controls (i.e., energy management information systems), smart building control platforms, cybersecurity, grid-interactive-efficient buildings (GEBs), smart building control methods, and occupant-centric control. Although not explicitly included the technologies nod to the need for AI driven technologies to prepare engineers and industry professionals to be future ready. This paper describes the project approach, provides outlines of the training materials, and identifies lessons learned in creating the content for this course. The authors suggest ways to scale the instruction of smart building concepts to empower the workforce to accelerate the adoption of smart building technologies and AI-based teaching and learning in higher education and building sector.

99 GENERAL AND MISCELLANEOUS↗

Variable Valve Actuation

Many approaches exist to enable advanced mode, low temperature combustion systems for diesel engines - such as premixed charge compression ignition (PCCI), Homogeneous Charge Compression Ignition (HCCI) or other HCCI-like combustion modes. The fuel properties and the quantity, distribution and temperature profile of air, fuel and residual fraction in the cylinder can have a marked effect on the heat release rate and combustion phasing. Figure 1 shows that a systems approach is required for HCCI-like combustion. While the exact requirements remain unclear (and will vary depending on fuel, engine size and application), some form of substantially variable valve actuation is a likely element in such a system. Variable valve actuation, for both intake and exhaust valve events, is a potent tool for controlling the parameters that are critical to HCCI-like combustion and expanding its operational range. Additionally, VVA can be used to optimize the combustion process as well as exhaust temperatures and impact the after treatment system requirements and its associated cost. Delphi Corporation has major manufacturing and product development and applied R&D expertise in the valve train area. Historical R&D experience includes the development of fully variable electro-hydraulic valve train on research engines as well as several generations of mechanical VVA for gasoline systems. This experience has enabled us to evaluate various implementations and determine the strengths and weaknesses of each. While a fully variable electro-hydraulic valve train system might be the 'ideal' solution technically for maximum flexibility in the timing and control of the valve events, its complexity, associated costs, and high power consumption make its implementation on low cost high volume applications unlikely. Conversely, a simple mechanical system might be a low cost solution but not deliver the flexibility required for HCCI operation. After modeling more than 200 variations of the mechanism it was determined that the single cam design did not have enough flexibility to satisfy three critical OEM requirements simultaneously, (maximum valve lift variation, intake valve opening timing and valve closing duration), and a new approach would be necessary. After numerous internal design reviews including several with the OEM a dual cam design was developed that had the flexibility to meet all motion requirements. The second cam added complexity to the mechanism however the cost was offset by the deletion of the electric motor required in the previous design. New patent applications including detailed drawings and potential valve motion profiles were generated and alternate two cam designs were proposed and evaluated for function, cost, reliability and durability. Hardware was designed and built and testing of sample hardware was successfully completed on an engine test stand. The mechanism developed during the course of this investigation can be applied by Original Equipment Manufacturers, (OEM), to their advanced diesel engines with the ultimate goal of reducing emissions and improving fuel economy. The objectives are: (1) Develop an optimal, cost effective, variable valve actuation (VVA) system for advanced low temperature diesel combustion processes. (2) Design and model alternative mechanical approaches and down-select for optimum design. (3) Build and demonstrate a mechanism capable of application on running engines.

02 PETROLEUM↗

Dynamics of Confined Flexible and Unentangled Polymer Melts in Highly Adsorbing Cylindrical Pores

Inspired by the recent neutron spin echo experiments (NSE) designed to address the dynamic phenomena in polymer melts that are induced by interactions with a confining surface, we performedcoarse-grained molecular dynamics simulations to replicate the experimental results in order to provide new molecular insight for the observations. The results show excellent agreement in the values obtained for the normalized coherent single chain dynamic structure factor, S(Q;Dt)/S(Q;0) , between experiments and simulations. The simulations indicate that using different chain molecular weights that are used to achieve scattering contrast results in the uneven distribution of scatterers in the radial direction because lower molecular weight chains prefer to adsorb in the confining surface. In the bulk configuration, both simulations and experiments confirm that the polymer chains follow Rouse dynamics. However, under confinement, the Rouse modes are suppressed. The mean-squared radius of gyration, R2g, and the average relative shape anisotropy k2 of the conformation of the polymer chains indicate a pancake-like conformation near the surface and a bulk-like conformation near the center of the confining cylinder. Direct visualization of the polymers in the simulation confirm the pancake-like conformation of the adsorbed chains and the presence of trains, loops and tails in the region between the adsorbed chains and the chains not in contact with the surface. Despite the presence of these different conformations, the average form factor of the confined chains still follows the Debye function which describes linear ideal chains, which is in agreement with small angle neutron scattering (SANS) experiments. The experimentally inaccessible mean squared displacement of the confined monomers, calculated as a function of radial distance from the pore surface, was obtained in the simulations. The simulation shows a gradual increase of the MSD from the adsorbed, but mobile layer, to that similar to the bulk far away from the surface.

Carrillo, Jan-Michael Y.↗

Temperature dependence of nuclear quantum effects on liquid water via artificial neural network model based on SCAN meta-GGA functional

We investigate the temperature dependence of nuclear quantum effects (NQEs) on structural and dynamic properties of liquid water by training a neural network force field using first-principles molecular dynamics (FPMD) based on the strongly constrained and appropriately normed meta-generalized gradient approximation exchange-correlation approximation. The FPMD simulation based on density functional theory has become a powerful computational approach for studying a wide range of condensed phase systems. However, its large computational cost makes it difficult to incorporate NQEs in the simulation and investigate temperature dependence of various properties. To circumvent this difficulty, we use an artificial neural network model and employ the thermostatted ring polymer MD approach for studying the temperature dependence of NQEs on various properties. The NQEs generally bring the radial distribution functions closer to the experimental measurements. Translational diffusivity and rotational dynamics of water molecules are both slowed down by the NQEs. The competing inter-molecular and intra-molecular quantum effects on hydrogen bonds, as discussed by Habershon, Markland, and Manolopoulos [J. Chem. Phys. 131(2), 024501 (2019)], can explain the observed temperature dependence of the NQEs on the dynamical properties in our simulation.

Yao, Yi (ORCID:0000000299576979)↗

Multiscale Based Characterization and Classification of Urban Land-Use

Machine learning and deep learning provide a means for generating urban land-use maps with relatively little human effort compared to manually digitizing images. This is especially important for supporting global and regional initiatives focused on sustainability, planning, health, pro-poor policy, infrastructure, and population distribution estimates. Many of these initiatives work in areas where geospatial data is scarce, such as the global south, and often use land-use maps to help achieve their goals. In this study, we develop a typology for automated labeling of urban land-use data that captures the variation in structural patterns within cities. A comparison of classification accuracy between convolutional neural networks (CNNs) and support vector machines (SVMs) coupled with handcrafted features is conducted. Through experimental validation on two highly dense cities in Africa, we report on new insights and the potential benefits offered by both multiscale handcrafted features and multiscale-CNNs even with limited training data.

Arndt, Jacob↗

Random Forests as a Viable Method to Select and Discover High-redshift Quasars

We present a method of selecting quasars up to redshift ≈6 with random forests, a supervised machine-learning method, applied to Pan-STARRS1 and WISE data. We find that, thanks to the increasing set of known quasars, we can assemble a training set that enables supervised machine-learning algorithms to become a competitive alternative to other methods up to this redshift. We present a candidate set for the redshift range 4.8–6.3, which includes the region around z = 5.5 where selecting quasars is difficult due to their photometric similarity to red and brown dwarfs. We demonstrate that, under our survey restrictions, we can reach a high completeness (66% ± 7% below redshift 5.6/83{sub -9}{sup +6}% above redshift 5.6) while maintaining a high selection efficiency (78{sub -8}{sup +10}%/94{sub -8}{sup +5}%). Our selection efficiency is estimated via a novel method based on the different distributions of quasars and contaminants on the sky. The final catalog of 515 candidates includes 225 known quasars. We predict the candidate catalog to contain additional 148{sub -33}{sup +41} new quasars below redshift 5.6 and 45{sub -8}{sup +5} above, and we make the catalog publicly available. Spectroscopic follow-up observations of 37 candidates led us to discover 20 new high redshift quasars (18 at 4.6 ≤ z ≤ 5.5, 2 z ~ 5.7). These observations are consistent with our predictions on efficiency. We argue that random forests can lead to higher completeness because our candidate set contains a number of objects that would be rejected by common color cuts, including one of the newly discovered redshift 5.7 quasars.

79 ASTRONOMY AND ASTROPHYSICS↗

Space-Time Quantum Information from the Entangled States of Magnetic Molecule (STI Product)

This collaborative project combines synthesis, measurement, and theory by three faculty members at the Eddleman Quantum Institute of UC Irvine to effectively investigate the quantum properties of molecules in the space, time, and frequency domains. Through synthetic chemistry, molecules are tailored for their magnetic and coherent properties. By combining femtosecond (fs) terahertz (THz) light and a continuous wave (cw) THz laser with a low temperature scanning tunneling microscope (STM), quantum phenomena are probed with simultaneous femtosecond temporal and atomic-scale spatial resolution. In particular, the invention of the quantum superposition microscope (QSM) advances quantum sensing for enhanced spectroscopy and imaging capabilities. Coupling theory to the experimental efforts offers a deeper understanding and predictive power for the molecular systems. The phenomena of superposition, entanglement, and coherence is central to quantum information science and can be realized in qubit states. Many systems can be modeled by a double-well potential in which two levels are formed in the two lowest energy states interacting with the environment and external radiation. In focusing on molecules as two-level systems, the underlying expectation is that their tunable composition and structure allows an effective parameter space to optimize their use as qubits for quantum sensing and computing. The THz radiation induces the superposition between the two states, appearing as temporal oscillations that damp in amplitude. Enhanced spectroscopy and imaging in the time and frequency domains is achieved through the extreme sensitivity of the frequency and damping of coherence of two-level systems to its environment. A single hydrogen molecule trapped in the STM tunneling gap experiences a double-well potential and absorption of THz femtosecond pulses of light creates the superposition of its two levels, appearing as damped oscillations in the light induced direct current (DC). The oscillation frequency depends sensitively on the electric field distribution of the copper nitride (Cu 2 N) surface, through the Stark effect, and associated with the different charge distributions at the copper and nitrogen sites and in between. This QSM can resolve variation in the surface electric field with 0.02 nanometer resolution. In addition, the single hydrogen molecule entaes with nearby hydrogen molecules as seen in the avoided level crossings of energy (oscillation frequency) versus the voltage across the tunneling gap. Thus, the first application of the QSM senses and images the surface electric field at the atomic scale. Results from this project advance fundamental understanding of quantum phenomena, develop novel synthesis, measurement, and theory, provide the knowledge foundation for molecule-based qubits and sensing that enable the development of the QSM and emergent technologies. This project trained researchers in quantum information science, extended knowledge in classrooms, and outreached to the community.

47 OTHER INSTRUMENTATION↗

Effect of reaction media on hydrogenolysis of polyethylene plastic waste: Polymer-surface interactions in small alkane/polymer blends

The polymer reaction media and its properties can be altered by recycling a fraction of liquid products or adding alkane solvents. Less clear is whether this strategy affects hydrogenolysis. Herein, we investigated the effect of short-chain alkanes C n consisting of n carbons (n=8, 16, and 32) on the upcycling of high-density polyethylene (HDPE) plastic waste to lubricant-range products over Ru/TiO 2 catalysts by multiscale simulations and experiments. First, we trained a force field for polymer/surface interactions on a Ru 22 nanoparticle (NP) supported on TiO 2 . Using replica exchange molecular dynamics simulations, we studied the effect of small hydrocarbons on the adsorption of a surrogate polymer, C 142 , on the catalyst. We found segregation of long chains (C 142 ) at the catalyst surface due to the enthalpy gained by adsorbing more C-C bonds of the long chains, compensating for entropic losses upon adsorption. Short-chain molecules decrease the adsorbed carbons of long chains on the Ru NP due to blocking Ru active sites. Compared to the bulk chains, competitive adsorption results in a broader, heavy-tailed distribution of end-to-end distance of adsorbed chains. Our experiments demonstrated that catalyst activity declines significantly beyond simple dilution due to changes in polymer adsorption, and tuning the reaction media by creating suitable blends impacts hydrogenolysis. Density distributions for a 50:50%wt mixture of PP and PE show that PE chains are segregated at the surface, so they are prone to C-C bond breaking much faster than PP chains. H/D exchange experiments show preferential deuteration of PE, while CH 3 groups of PP remain undeuterated. Furthermore, this may be explained by the preferential sorption of PE over PP, leading to specific distribution in the polymer blend.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

End-to-end online performance data capture and analysis for scientific workflows

With the increased prevalence of employing workflows for scientific computing and a push towards exascale computing, it has become paramount that we are able to analyze characteristics of scientific applications to better understand their impact on the underlying infrastructure and vice-versa. Such analysis can help drive the design, development, and optimization of these next generation systems and solutions. Here, we present the architecture, integrated with existing well-established and newly developed tools, to collect online performance statistics of workflow executions from various, heterogeneous sources and publish them in a distributed database (Elasticsearch). Using this architecture, we are able to correlate online workflow performance data, with data from the underlying infrastructure, and present them in a useful and intuitive way via an online dashboard. We have validated our approach by executing two classes of real-world workflows, both under normal and anomalous conditions. The first is an I/O-intensive genome analysis workflow; the second, a CPU- and memory-intensive material science workflow. Based on the data collected in Elasticsearch, we are able to demonstrate that we can correctly identify anomalies that we injected. The resulting end-to-end data collection of workflow performance data is an important resource of training data for automated machine learning analysis.

97 MATHEMATICS AND COMPUTING↗

Identifying schools at high-risk for elevated lead in drinking water using only publicly available data

Estimating the risk of lead contamination of schools' drinking water at the State level is a complex, important, and unexplored challenge. Variable water quality among water systems and changes in water chemistry during distribution affect lead dissolution rates from pipes and fittings. In addition, the locations of lead-bearing plumbing materials are uncertain. We tested the capability of six machine learning models to predict the likelihood of lead contamination of drinking water at the schools' taps using only publicly available datasets. The predictive features used in the models correspond to those with a proven correlation to the dominant, but commonly unavailable, factors that govern lead leaching: the presence of lead-bearing plumbing materials and water quality conducive to lead corrosion. By combining water chemistry data from public reports, socioeconomic information from the US census, and spatial features using Geographic Information Systems, we trained and tested models to estimate the likelihood of lead contaminated tap water in over 8,000 schools across California and Massachusetts. Our best-performing model was a Random Forest, with a 10-fold cross validation score of 0.88 for Massachusetts and 0.78 for California using the average Area Under the Receiver Operating Characteristic Curve (ROC AUC) metric. The model was then used to assign a lead leaching risk category to half of the schools across California (the other half was used for training). There was good agreement between the modeled risk categories and the actual lead leaching outcomes for every school; however, the model overestimated the lead leaching risk in up to 17% of the schools. This model is the first of its kind to offer a tool to predict the risk of lead leaching in schools at the State level. Further use of this model can help deploy limited resources more effectively to prevent childhood lead exposure from school drinking water.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗