Engineering Papers⌕ Search

Engineering topics

Lawrence, Earl

Publications and source records attributed to Lawrence, Earl.

At least 19 records

Advanced Research Directions on AI for Science, Energy, and Security: Report on Summer 2022 Workshops

Over the past decade, fundamental changes in artificial intelligence (AI)—from foundational to applied—have delivered dramatic insights across a wide breadth of U.S. Department of Energy (DOE) mission space. AI is helping to augment and improve scientific and engineering workflows (e.g., for control, design, and dramatic performance gains through surrogate models) in national security, the Office of Science, and DOE’s applied energy programs. The progress and potential for AI in DOE science was captured in the 2020 “AI for Science” report from the DOE laboratory community in collaboration with academia and industry. Specific scientific areas ready to further leverage the power of AI ranged from the scale and performance of computational models to data analysis to creating new classes of observations using computer vision. Since that report, the scale and scope of scientific AI have accelerated, revealing new, emergent properties that yield insights that go beyond enabling opportunities to being potentially transformative in the way that scientific problems are posed and solved. Thus, under the guidance of both the Office of Science (SC) and the National Nuclear Security Administration (NNSA), the DOE national laboratories organized a series of workshops in 2022 to gather input on new and rapidly emerging opportunities and challenges of scientific AI. This 2023 report is a synthesis of those workshops. The scientific community believes AI can have a foundational impact on a broad range of DOE missions, including science, energy, and national security. Further, DOE has unique capabilities that enable the community to drive progress in scientific use of AI, building on long-standing DOE strengths and investments in computation, data, and communications infrastructure, spanning the Energy Sciences Network (ESnet), the Exascale Computing Project (ECP), and integrative programs such as the NNSA Office of Defense Programs Advanced Simulation and Computing (ASC) and the SC Scientific Discovery through Advanced Computing (SciDAC) programs.

97 MATHEMATICS AND COMPUTING↗

Temporal Characterization and Filtering of Sensor Data to Support Anomaly Detection

Here, we present an approach for characterizing complex temporal behavior in the sensor measurements of a system in order to support detection of anomalies in that system. We first characterize typical behavior by extending a hidden Markov model-based approach to time series alignment. We then use a trace of that learned behavior to develop a particle filter that enables efficient estimation of the filtering distribution on the state space. This produces filtered residuals that can then be used in an anomaly detection framework. Our motivating example is the daily behavior of a building’s heating, ventilation, and air conditioning (HVAC) system, using sensor measurements that arrive every minute and induce a state space with 15,120 states. We provide an end-to-end demonstration of our approach showing improved performance of anomaly detection after application of alignment and filtering compared to the unaligned data. The proposed model is implemented as a computationally efficient R package alignts (align time series) built with R and Fortran 95 with OpenMP support.

47 OTHER INSTRUMENTATION↗

Scaled Vecchia Approximation for Fast Computer-Model Emulation

Many scientific phenomena are studied using computer experiments consisting of multiple runs of a computer model while varying the input settings. Gaussian processes (GPs) are a popular tool for the analysis of computer experiments, enabling interpolation between input settings, but direct GP inference is computationally infeasible for large datasets. We adapt and extend a powerful class of GP methods from spatial statistics to enable the scalable analysis and emulation of large computer experiments. Specifically, we apply Vecchia’s ordered conditional approximation in a transformed input space, with each input scaled according to how strongly it relates to the computer-model response. The scaling is learned from the data by estimating parameters in the GP covariance function using Fisher scoring. Our methods are highly scalable, enabling estimation, joint prediction, and simulation in near-linear time in the number of model runs. In several numerical examples, our approach substantially outperformed existing methods.

97 MATHEMATICS AND COMPUTING↗

In Situ Inference for Earth System Predictability

Focal Area: Focal Area 3: Insight gleaned from complex simulated data using AI, big data analytics, and other advanced methods, including explainable AI and physics- or knowledge-guided AI. Science Challenge: An understanding of future evolution in precipitation extremes is critical to numerous DOE mission questions. Extreme events are by nature short time-scale events that are difficult to diagnose in available model data. Accurate modeling of extreme events necessarily requires high spatial resolution at the storm scale locally. However, the environment in which storms grow is dependent on global, remote, processes. These complex spatiotemporal relationships are impossible to diagnose at resolutions required to accurately model storms responsible for extreme precipitation. At exascale, climate simulations will produce results at fine enough resolution to investigate these relationships. However, the resulting data from these simulations will be far too large to save for post-simulation analysis. We advocate for fitting statistical models inside the simulations as they run, a context known as in situ, which will facilitate scientific investigations using the full fine-scale data stream. Figure 1 shows an example of the type of model we could consider, a Bayesian hierarchical spatial regression model. Precipitation extremes at each grid cell are modeled using extreme value distributions. Since extremes are rare, fitting models to individual grid cells can result in high variance and poor estimates. Instead, the model can be made more robust by smoothing the parameters of the extreme value model across space. Additionally, the parameters themselves can be functionally linked to other variables elsewhere in the simulation. Thus, we can use the fine-scale data to build more robust models for extremes that link extreme behavior to other climate patterns.

54 ENVIRONMENTAL SCIENCES↗

Penalized ensemble Kalman filters for high dimensional non-linear systems

The ensemble Kalman filter (EnKF) is a data assimilation technique that uses an ensemble of models, updated with data, to track the time evolution of a usually non-linear system. It does so by using an empirical approximation to the well-known Kalman filter. However, its performance can suffer when the ensemble size is smaller than the state space, as is often necessary for computationally burdensome models. This scenario means that the empirical estimate of the state covariance is not full rank and possibly quite noisy. To solve this problem in this high dimensional regime, we propose a computationally fast and easy to implement algorithm called the penalized ensemble Kalman filter (PEnKF). Under certain conditions, it can be theoretically proven that the PEnKF will be accurate (the estimation error will converge to zero) despite having fewer ensemble members than state dimensions. Further, as contrasted to localization methods, the proposed approach learns the covariance structure associated with the dynamical system. These theoretical results are supported with simulations of several non-linear and high dimensional systems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

ATOMIC Simulations and Experimental Data for CaCO3 Mixtures

This data consists of simulations and experimental measurements of laser-induced breakdown spectroscopy (LIBS). The simulations are produced by ATOMIC, a general purpose plasma modeling and kinetics code that has been designed to compute emission (or absorption) spectra from plasmas [1] and are used to develop a statistical characterization of matrix effects. Our overall suite of simulations includes contains several sets of simulations: training and validation sets of simulations for three and four element mixtures of calcium, carbon, oxygen, and nitrogen (included to account for atmosphere) along with simulations of the individual elements. The 4-element simulations include the mixture of all four elements mentioned and for each of the four individual elements. The 3-element simulations include output for the mixture of calcium, carbon, oxygen and for these three individual elements. The training data were produced using a 600-run design, shown in Figure 1, that varies input parameters temperature (T), electron density (Ne), and proportion of the elements calcium, carbon, oxygen, and nitrogen (Ca; C; O; N) for the 4 element output. The 3-element output includes all parameters except for the proportion of nitrogen. The element proportions (all the variables but T and Ne) sum to one and are unused in the single-element simulations. The validation data was produced with a 80-run design shown in Figure 2. The training and validation simulation outputs for the 4-element simulations for the mixture and for the single element calcium are shown as sample simulations in Figures 3 and 4 respectively. The simulations produce spectra over a range of 190nm - 950nm that roughly mimics the range collected by the SciAps Z-300 LIBS instrument that was used for the experimental data. The measured spectra for a CaCO3 (which may include contribution from Earth's atmosphere) in the experiment is shown in in Figure 5. All files are kept in directories whose names indicate the elemental composition (CaCO3, Ca, C, O, or N), number of elements (3 or 4), and purpose (training, which is not labeled in the file name, or validation) with file names numbered to indicate the line in the design files used to produce the simulation. The designs are provided as text files with names indicating their purpose. The experimental data is provided as a CSV file. [1] J Colgan, EJ Judge, DP Kilcrease, and JE Barefield II. Ab-initio modeling of an iron laser-induced plasma: Comparison between theoretical and experimental atomic emission spectra. Spectrochimica Acta Part B: Atomic Spectroscopy, 97:65{73}, 2014.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Fast strength model characterization using Bayesian statistics

A variety of flow stress models exist with new models constantly being developed. These models aim to approximate the strength of materials in a variety of regimes from quasistatic loading through shock scenarios. All models contain an array of parameters which need to be tuned to the material under study. Some models perform well under limited conditions, requiring adjustment of the parameters when venturing outside of those predefined ranges. Other models perform well over a wide range of conditions with a set of parameters, but may be outperformed by other models optimized on a tighter range of conditions. Recent research by Los Alamos demonstrated the ability to optimize the Johnson Cook (JC) model using a set of 3 plate-impact experiments on Aluminum. They utilized Bayesian statistics and emulation to determine optimal parameters for the model with a quantification of parameter uncertainty. We present an extension of this capability to incorporate velocimetry from plate-impact tests, stress-strain data from split Hopkinson pressure bar and quasistatic compression tests, plus profiles from Taylor cylinders in a unified fashion. Statistically robust comparisons of the performance and uncertainty of different realizations of the JC flow stress model were carried out based on calibration to several possible combinations of these three different experiment types.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

ATOMIC Simulations and Experimental Data for Basalt-like Compounds

This data consists of simulations and experimental measurements of laser-induced breakdown spectroscopy (LIBS). The simulations are produced by ATOMIC, a general purpose plasma modeling and kinetics code that has been designed to compute emission (or absorption) spectra from plasmas [2]. The makeup of the plasma was considered to be divided into some proportion water, some proportion Martian atmosphere (CO2), and some proportion target (from the rock or object impacted by the laser), where these proportions add to 1. Based on expert knowledge, the proportion of water was kept in the range [0.0,0.5] and the proportion of atmosphere was kept in the range [0.02, 0.9]. Our overall suite of simulations contains six sets of simulations that differ in which elements were considered to make up the target. Within each set, we used uniformly drawn temperatures and log mass densities within pre-specified ranges. The temperature range was [0.5,1.5] eV and the log (base 10) mass density range was [-7,-4]. The proportion of water, atmosphere, and target were drawn from a symmetric Dirichlet distribution, but draws in which the propor- tion of water or atmosphere exceeded the pre-specified limits were rejected from the design. Up to eleven constituent elements (Si, Al, Fe, Mg, Ca, O, Ti, Mn, Na, K, P) were considered for the target, as they are the most common elements found in basalt compounds and were used in [1]. For each run, the proportions of the constituent elements making up the target were drawn from a symmetric Dirichlet distribution. We ran 1,350 simulations that included nonzero proportions of all eleven elements. We also ran simulations which excluded some of these elements. In particular, we ran 1,000 simulations that only included nonzero proportions for the six most common elements (Si, Al, Fe, Mg, Ca, O). We also ran five sets, each with 500 simulations, that only included nonzero proportions for five of the six most common elements (but where all sets included O). Thus, we generated a total of 4,850 spectra representing basalt-like compounds in which the target was comprised of oxygen and between four and ten other elements. The ATOMIC code produced spectra over a range of 240nm - 880nm that roughly mimics the range collected by the ChemCam instrument on the Mars rover Curiosity. Each spectra had 32,000 wavelengths split across three spectrometer ranges (to mimic ChemCam). The experimental data, described in [1], measures a prepared basalt sample. All files are kept in directories whose names indicate the set of elements considered for the target with file names numbered to indicate the line in the design files used to produce the simulation. The designs are provided as text files with names indicating their purpose. The experimental data is provided as a CSV file which contains a header with measurement information, followed by a collection of 50 shots across a collection of wavelengths, along with the computed median and mean across shots.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

The Mira-Titan Universe. III. Emulation of the Halo Mass Function

We construct an emulator for the halo mass function over group and cluster mass scales for a range of cosmologies, including the effects of dynamical dark energy and massive neutrinos. The emulator is based on the recently completed Mira-Titan Universe suite of cosmological N-body simulations. The main set of simulations spans 111 cosmological models with 2.1 Gpc boxes. We extract halo catalogs in the redshift range z = [0.0, 2.0] and for masses M-200c >= 10(13)M(circle dot)/h. The emulator covers an eight-dimensional hypercube spanned by {Omega(m)h(2), Omega(b)h(2), Omega(nu)h(2), sigma(8), h, n(s), w(0), w(a)}; spatial flatness is assumed. We obtain smooth halo mass functions by fitting piecewise second-order polynomials to the halo catalogs and employ Gaussian process regression to construct the emulator while keeping track of the statistical noise in the input halo catalogs and uncertainties in the regression process. For redshifts z less than or similar to 1, the typical emulator precision is better than 2% for 10(13)-10(14)M(circle dot)/h and <10<^> M similar or equal to 101(circle dot)(54)/h. For comparison, fitting functions using the traditional universal form for the halo mass function can be biased at up to 30% at M similar or equal to 10(144)M(circle dot)/h for z = 0. Our emulator is publicly available at https://github.com/SebastianBocquet/MiraTitanHMFemulator.

cosmology: theory↗

ATOMIC Simulations and Experimental Data for Sodium and Copper Mixtures

This data consists of simulations and experimental measurements of laser-induced breakdown spectroscopy (LIBS). The simulations are produced by ATOMIC, a general purpose plasma modeling and kinetics code that has been designed to compute emission (or absorption) spectra from plasmas [1]. Our overall suite of simulations contains 12 sets of simulations: training and validation sets of simulations for each of two resolutions of each of sodium, copper, and mixtures of the two. The training data were produced using a 500-run design that varies input parameters temperature (T), electron density (Ne), proportion of sodium (pNa), and proportion of copper (pCu). The latter two variables sum to one and are unused in the single-element simulations. The validation data was produced with a 25-run design. The coarse simulations produce spectra over a range of 240nm - 880nm that roughly mimics the range collected by the ChemCam instrument on the Mars rover Curiosity. The fine resolution simulations produce spectra over a range of 550nm to 600nm and were produced to mimic the included experimental results. The experimental results cover four mixtures of sodium and copper. All data sets are kept in zip files whose names indicate the elemental composition (NaCu, Na, or Cu), resolution (coarse or fine), and purpose (training or validation) with file names numbered to indicate the line in the design files used to produce the simulation. The designs are provided as text files with names indicating their purpose. The experimental data is provided as a CSV file. See attached abstract for figures.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Advancing Fusion with Machine Learning Research Needs Workshop

The pursuit of fusion energy has required extensive experimental and theoretical science activities to develop the knowledge needed that will enable design of successful fusion power plants. Even today, following decades of research in many key areas including plasma physics and material science, much remains to be learned to enable optimization of the tokamak or other paths to fusion energy. Data science methods from the fields of machine learning and artificial intelligence (ML/AI) offer opportunities for enabling or accelerating progress toward the realization of fusion energy by maximizing the amount and usefulness of information extracted from experimental and simulation output data. Jointly supported by the Department of Energy Offices of Fusion Energy Science (FES) and Advanced Scientific Computing Research (ASCR), a workshop was organized to identify Priority Research Opportunities (PRO’s) for application of ML/AI methods to enable accelerated solution of fusion problems. The resulting “Advancing Fusion with Machine Learning Research Needs Workshop,” held in Gaithersburg, MD, April 30 – May 2, 2019, brought together ~ 60 experts in fields spanning fusion science, data science, statistical inference and mathematics, machine learning, and artificial intelligence, along with DOE program managers and technical experts, to identify key PRO’s.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Emulator Generation Gadget

This software automates the production of standalone emulators for complex computer models. The end result of running the software is a piece of code called an emulator, a statistical approximation which predicts the output of complex computer simulation, but at a much faster speed. The emulator is built using a training set of inputs and outputs from the simulations. The input-output pairs are used in a Bayesian scheme to estimate the parameters of a Gaussian process based regression model that learns the input-output relationship. The estimated parameters are combined with a generic C skeleton to produce a compile-able instantiation of the emulator.

Sweeney, Christine↗

The Mira-Titan Universe. II. Matter Power Spectrum Emulation

We introduce a new cosmic emulator for the matter power spectrum covering eight cosmological parameters. Targeted at optical surveys, the emulator provides accurate predictions out to a wavenumber k ~ 5Mpc -1 and redshift z ≤ 2. Besides covering the standard set of CDM parameters, massive neutrinos and a dynamical dark energy of state are included. The emulator is built on a sample set of 36 cosmological models, carefully chosen to provide accurate predictions over the wide and large parameter space. For each model, we have performed a high-resolution simulation, augmented with sixteen medium-resolution simulations and TimeRG perturbation theory results to provide accurate coverage of a wide k-range; the dataset generated as part of this project is more than 1.2Pbyte. With the current set of simulated models, we achieve an accuracy of approximately 4%. Because the sampling approach used here has established convergence and error-control properties, follow-on results with more than a hundred cosmological models will soon achieve ~1% accuracy. We compare our approach with other prediction schemes that are based on halo model ideas and remapping approaches. The new emulator code is publicly available.

97 MATHEMATICS AND COMPUTING↗

The Nonhomogeneous Poisson Process for Fast Radio Burst Rates

This paper presents the non-homogeneous Poisson process (NHPP) for modeling the rate of fast radio bursts (FRBs) and other infrequently observed astronomical events. The NHPP, well-known in statistics, can model dependence of the rate on both astronomical features and the details of an observing campaign. This is particularly helpful for rare events like FRBs because the NHPP can combine information across surveys, making the most of all available information. The goal of the paper is two-fold. First, it is intended to be a tutorial on the use of the NHPP. Second, we build an NHPP model that incorporates beam patterns and a power law flux distribution for the rate of FRBs. Using information from 12 surveys including 15 detections, we find an all-sky FRB rate of 587 events per sky per day above a flux of 1 Jy (95% CI: 272, 924) and a flux power-law index of 0:91 (95% CI: 0.57, 1.25).

79 ASTRONOMY AND ASTROPHYSICS↗

CosmicEmu, Version 2.0

The CosmicEmu implements a Gaussian process emulation scheme to predict the dark matter and dark matter + neutrino power spectra over a range of physical parameters that describe the composition and evolution of the Universe. The procedure can produce thousands of spectra per second, much faster than the complex n-body simulations used to produce the training set, which can take weeks per simulation. The training simulations, their design, and the general scheme is described in Lawrence et al. (2010), Heitmann et al. (2014), Heitmann et al. (2016), and an upcoming paper.

Lawrence, Earl↗

CosmicEmu, Version 2.0

The CosmicEmu implements a Gaussian process emulation scheme to predict the dark matter and dark matter + neutrino power spectra over a range of physical parameters that describe the composition and evolution of the Universe. The procedure can produce thousands of spectra per second, much faster than the complex n-body simulations used to produce the training set, which can take weeks per simulation. The training simulations, their design, and the general scheme is described in Lawrence et al. (2010), Heitmann et al. (2014), Heitmann et al. (2016), and an upcoming paper.

Lawrence, Earl↗

NHPP for FRBs, Version 1.0

This code implements the non-homogeneous poisson process model for estimating the rate of fast radio bursts. It includes modeling terms for the distribution of events in the Universe and the detection sensitivity of the radio telescopes and arrays used in observation. The model is described in LA-UR-16-26261.

Lawrence, Earl↗