Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “approximation algorithms”

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 1,549 records · Page 86

Use of the Lorentz-Boosted Frame Transformation to Simulate Free-Electron Laser Amplifier Physics

Recently [1]it has been pointed out that numerical simulation of some systems containing charged particles with highly relativistic directed motion can by speeded up by orders of magnitude by choice of the proper Lorentz boosted frame. A particularly good example is that of short wavelength free-electron lasers (FELs) in which a high energy (E0>_ 250 MeV) electron beam interacts with a static magnetic undulator. In the optimal boost frame with Lorentz factor gamma F, the red-shifted FEL radiation and blue shifted undulator have identical wavelengths and the number of required time-steps (presuming the Courant condition applies) decreases by a factor of g2 F for fullyelectromagnetic simulation. We have adapted the WARP code [2]to apply this method to several FEL problems including coherent spontaneous emission (CSE) from pre-bunched e-beams, and strong exponential gain in a single pass amplifier configuration. We discuss our results and compare with those from the"standard" FEL simulation approach which adopts the eikonal approximation for propagation ofthe radiation field.

43 PARTICLE ACCELERATORS↗

SU-E-T-27: A Tool for Routine Quality Assurance of Radiotherapy Dose Calculation Software

Purpose: Dose calculation software is thoroughly evaluated when it is commissioned; however, evaluation of periodic software updates is typically limited in scope due to staffing constraints and the need to quickly return the treatment planning system to clinical service. We developed a tool for quickly and comprehensively testing and documenting dose calculation software against measured data. Methods: A tool was developed using MatLab (The MathWorks, Natick, MA) for evaluation of dose calculation algorithms against measured data. Inputs to the tool are measured data, reference DICOM RT PLAN files describing the measurements, and dose calculations in DICOM format. The tool consists of a collection of extensible modules that can perform analysis of point dose, depth dose curves, and profiles using dose difference, distance-to-agreement, and the gamma-index. Each module generates a report subsection that is incorporated into a master template, which is converted to final form in portable document format (PDF). Results: After each change to the treatment planning system, a report can be generated in approximately 90 minutes. The tool has been in use for more than 5 years, spanning 5 versions of the eMC and 4 versions of the AAA. We have detected changes to the algorithms that affected clinical practice once during this period. Conclusion: Our tool provides an efficient method for quality assurance of dose calculation software, providing a complete set of tests for an update. Future work includes the addition of plan level tests, allowing incorporation of, for example, the TG-119 test suite for IMRT, and integration with the treatment planning system via an application programming interface. Integration with the planning system will permit fully-automated testing and reporting at scheduled intervals.

07 ISOTOPE AND RADIATION SOURCES↗

Increasing the solar photovoltaic energy capture on sunny and cloudy days

This report analyzes an extensive set of measurements of the solar irradiance made using four identical solar arrays and associated solar sensors (collectively referred to as solar collectors) with different tilt angles relative to the earth's surface, and thus the position of the sun, in order to determine an optimal tracking algorithm for capturing solar radiation. The study included a variety of ambient conditions including different seasons and both cloudy and cloud-free conditions. One set of solar collectors was always approximately pointed directly toward the sun (DTS) for a period around solar noon. These solar collectors thus captured the direct beam component of the solar radiation that predominates on sunny days. We found that on sunny days, solar collectors with a DTS configuration captured more solar energy in accordance with the well-known cosine dependence for the response of a flat-surfaced solar collector to the angle of incidence with direct beam radiation. In particular, a DTS orientation was found to capture up to twice as much solar energy as a horizontal (H) orientation in which the array is tilted toward the zenith. Another set of solar collectors always had an H orientation, and this best captured the diffuse component of the solar radiation that predominates on cloudy days. The dependence of the H/DTS ratio on the solar-collector tilt angle was in approximate agreement with the Isotropic Diffuse Model derived for heavily overcast conditions. During cloudy periods, we found that an H configuration increased the solar energy capture by nearly 40% compared to a DTS configuration during the same period, and we estimate the solar energy increase of an H configuration over a system that tracks the obscured solar disk could reach 50% over a whole heavily-overcast day. On an annual basis the increase is predicted to be much less, typically only about 1%, because the contribution of cloudy days to the total annual solar energy captured by a photovoltaic system is small. These results are consistent with the solar tracking algorithm optimized for cloudy conditions that we proposed in an earlier report and that was based on a much smaller data set. Improving the harvesting of solar energy on cloudy days deserves wider attention due to increasing efforts to utilize renewable solar energy. In particular, increasing the output of distributed solar power systems on cloudy days is important to developing solar-powered home fueling and charging systems for hydrogen-powered fuel-cell electric and battery-powered vehicles, respectively, because it reduces the system size and cost for solar power systems that are designed to have sufficient energy output on the worst (cloudy) days. (author)

14 SOLAR ENERGY↗

Approximate photochemical dynamics of azobenzene with reactive force fields

We have fitted reactive force fields of the ReaxFF type to the ground and first excited electronic states of azobenzene, using global parameter optimization by genetic algorithms. Upon coupling with a simple energy-gap transition probability model, this setup allows for completely force-field-based simulations of photochemical cis→trans- and trans→cis-isomerizations of azobenzene, with qualitatively acceptable quantum yields. This paves the way towards large-scale dynamics simulations of molecular machines, including bond breaking and formation (via the reactive force field) as well as photochemical engines (presented in this work)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

THE EIGHTH DATA RELEASE OF THE SLOAN DIGITAL SKY SURVEY: FIRST DATA FROM SDSS-III

The Sloan Digital Sky Survey (SDSS) started a new phase in 2008 August, with new instrumentation and new surveys focused on Galactic structure and chemical evolution, measurements of the baryon oscillation feature in the clustering of galaxies and the quasar Ly{alpha} forest, and a radial velocity search for planets around {approx}8000 stars. This paper describes the first data release of SDSS-III (and the eighth counting from the beginning of the SDSS). The release includes five-band imaging of roughly 5200 deg{sup 2} in the southern Galactic cap, bringing the total footprint of the SDSS imaging to 14,555 deg{sup 2}, or over a third of the Celestial Sphere. All the imaging data have been reprocessed with an improved sky-subtraction algorithm and a final, self-consistent photometric recalibration and flat-field determination. This release also includes all data from the second phase of the Sloan Extension for Galactic Understanding and Exploration (SEGUE-2), consisting of spectroscopy of approximately 118,000 stars at both high and low Galactic latitudes. All the more than half a million stellar spectra obtained with the SDSS spectrograph have been reprocessed through an improved stellar parameter pipeline, which has better determination of metallicity for high-metallicity stars.

79 ASTRONOMY AND ASTROPHYSICS↗

Degradation of performance in ICF implosions due to Rayleigh–Taylor instabilities: A Hamiltonian perspective

The Rayleigh–Taylor instability (RTI) is an ubiquitous phenomenon that occurs in inertial-confinement-fusion (ICF) implosions and is recognized as an important limiting factor of ICF performance. To analytically understand the RTI dynamics and its impact on ICF capsule implosions, we develop a first-principle variational theory that describes an imploding spherical shell undergoing RTI. The model is based on a thin-shell approximation and includes the dynamical coupling between the imploding spherical shell and an adiabatically compressed fluid within its interior. Using a quasilinear analysis, we study the degradation trends of key ICF performance metrics (e.g., stagnation pressure, residual kinetic energy, and areal density) as functions of initial RTI parameters (e.g., the initial amplitude and Legendre mode), as well as the 1D implosion characteristics (e.g., the convergence ratio). We compare analytical results from the theory against nonlinear results obtained by numerically integrating the governing equations of this reduced model. Our findings emphasize the need to incorporate polar flows in the calculation of residual kinetic energy and demonstrate that higher convergence ratios in ICF implosions lead to significantly greater degradation of key performance metrics.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

COWALKER:EFFECTIVE TRANSPORT PROPERTIES OF COMPOSITE MATERIALS

SF-23-026 This software computes effective transport properties of composite materials involving fibers and nanoparticles using a random-walk algorithm that efficiently scales to an arbitrary number of processes and cores. Effective transport properties (thermal, electrical) are key to bridge the microstructure of complex materials with its macroscopic behavior. Traditional approaches either use effective medium approximations (closed mathematical expressions that are approximation for certain conditions) or continuum simulation models such as finite element or finite volume, which require the generation of a mesh for each configuration explored. cowalker leverages the equivalence between laplacian or heat equation-based models and random walks to compute the asymptotic transport properties from an ensemble of first sojourn times of a random walker moving through the composite material. This allows us to directly define a composite material as a collection of particles and use algorithms developed for molecular dynamics to quickly compute the intersection of the walker with the different interfaces in the material. cowalker is developed in C++, and it relies on the GNU Scientific Library for random generation. cowalker is currently delivered as source code, so the GSL library is not included in cowalker's distribution. A more userfriendly version, cowalker.jl is currently in development and will be released as part of cowalker.

YANGUAS-GIL, ANGEL↗

thornado-hydro: A Discontinuous Galerkin Method for Supernova Hydrodynamics with Nuclear Equations of State

This paper describes algorithms for non-relativistic hydrodynamics in the toolkit for high-order neutrino radiation hydrodynamics (thornado), which is being developed for multiphysics simulations of core-collapse supernovae (CCSNe) and related problems with Runge–Kutta discontinuous Galerkin (RKDG) methods. More specifically, thornado employs a spectral type nodal collocation approximation, and we have extended limiters — a slope limiter to prevent non-physical oscillations and a bound-enforcing limiter to prevent non-physical states — from the standard RKDG framework to be able to accommodate a tabulated nuclear equation of state (EoS). To demonstrate the efficacy of the algorithms with a nuclear EoS, we first present numerical results from basic test problems in idealized settings in one and two spatial dimensions, employing Cartesian, spherical-polar, and cylindrical coordinates. Then, we apply the RKDG method to the problem of adiabatic collapse, shock formation, and shock propagation in spherical symmetry, initiated with a 15 M ⊙ progenitor. Herein, we find that the extended limiters improve the fidelity and robustness of the RKDG method in idealized settings. The bound-enforcing limiter improves robustness of the RKDG method in the adiabatic collapse application, while we find that slope limiting in characteristic fields is vulnerable to structures in the EoS — more specifically, in the phase transition from nuclei and nucleons to bulk nuclear matter. The success of these applications marks an important step toward applying RKDG methods to more realistic CCSN simulations with thornado in the future.

Nuclear astrophysics (1129)↗

Spreader-Bar Radiation Detection System Enhancements: A Modeling and Simulation Study

An investigation of radiation detectors on crane-mounted platforms was performed to understand the limitations and optimal capability of these types of systems. Crane-mounted systems are primarily focused on crane-mounted radiation detector systems, where the detectors are mounted on a spreader-bar platform. Spreader bars are attached with cables to a crane at seaports to load and unload containers from cargo ships. The bars can spread from 20 feet to 53 feet using hydraulics, allowing a single bar to move a variety of container lengths. Spreader-bars are also used on straddle carriers, which move containers around the laydown yard in the port. Spreader bar radiation detection (SBRD) systems are commercially available from at least two companies. These systems employ a set of spectroscopic sensors placed on the underside of a gantry crane and held near the top of a cargo container for detection. These systems have been previously tested and some issues and limitations observed. Radiation detectors mounted on spreader-bars are single sided systems (observing cargo only from the top) and it is therefore challenging to obtain the same sensitivity as double-sided systems (for example, portal monitors) especially when considering the possible amounts of intervening cargo that could be present. The approximately exponential attenuation of gamma rays in intervening dense cargo presents an inherent limitation to detection. However, SBRD systems hold detectors closer to the container, have a longer dwell time, and may have lower background (over the ship); all of these aspects help mitigate the single-sided issue. In addition, there are operational considerations that make spreader-bar radiation systems attractive. These include scanning for radiation during other operations (loading, unloading or container movement) which minimizes additional time for scanning, incorporating the radiation detection into existing equipment to minimize initial cost, and in the case of ship unloading, scanning the containers before reaching the laydown yard. This study is focused on researching optimized configurations of SBRD systems, including number of detectors, locations, and best use of detector data, to determine a reasonable limit on the capability of these types of systems. The results will help support decisions on where and how these systems might be effectively deployed, and could be used for future specification development for SBRD acquisition. The next phase of this research and investigation into SBRD systems and quantification of the possible improved capability with an enhanced SBRD is the focus of this report. Supporting this investigation, detailed models of an enhanced SBRD system were created using a 3D radiation transport code, and models were run to simulate the data from sources and background for a realistic cargo-loading port environment. The enhanced SBRD had increased detector material (numbers of detectors) compared to the commercial systems, with a distribution providing more uniform coverage. Measurements using small industrial sources were made with an enhanced SBRD system developed by Sandia National Laboratory to support model validation to establish credibility for the simulations of these threat and background scenarios. A significant simulation effort was undertaken to simulate sources of interest within a 40-foot cargo container. In order to estimate the performance of a generalized SBRD system in scenarios including background and cargo, a highly enriched uranium (HEU) test source (with 3x10 -8 232 U by weight) within a wood cargo model was simulated for a 40-foot cargo container. This source was placed at a variety of locations, and the flux from the source was multiplied as necessary to create estimated minimal detectable quantities (MDQ) figures for various quantities of HEU. Simulated data sets were produced corresponding to a complete container offloading scenario for a large number of statistical realizations. Significant work was performed on both the algorithm development and the container cargo distributions entering the US, and the result of this modeling, validation, and simulation effort was a set of estimates of the performance of a variety of radiation detection algorithms in this cargo-screening environment. In the case of algorithms, a variety of contemporary radiation detection and identification approaches were employed, and the optimal approaches to selecting time and location subsets of detector data were explored. In all cases, algorithm performance was compared, and the best approaches were suggested along with the best expected performance from the enhanced SBRD. In general, template-matching schemes employing a limited set of regression parameters tend to provide best results. The improvements of an enhanced SBRD with increased detector material and improved algorithms result in an effective MDQ of 10 kg of HEU in ~50 percent of the imported cargo in 40-foot containers, assuming a random source location. This is approximately two times the capability of the estimated performance of the commercial systems. For an MDQ of the 10 kg HEU source located in the worst location (bottom of the 40-foot container), the volume of cargo scanned drops to 10% for the enhanced SBRD, again a factor of two or more larger than the commercial systems. Enhancing the SBRD systems by adding more detectors, providing a more uniform coverage, and incorporating more advanced algorithms will result in a significant increase in capability. Although this investigation provided some indication of the performance capabilities of an enhanced SBRD system, additional work is needed to estimate the capability for other container configurations, such as the 20 and 45 foot containers. Since the 20-foot container weight limit is similar to the 40-foot, more dense cargo is typically shipped in 20-foot containers. This will result in a smaller percentage of cargo scanned for the 20-foot containers (for a particular source quantity) compared to the 40-foot containers. In addition to other configuration, other source material needs to be investigated to understand the capabilities for the entire threat space, where only HEU is considered in this work.

crane-mounted↗

Identifying Important Ions and Positions in Mass Spectrometry Imaging Data Using CUR Matrix Decompositions

Mass spectrometry imaging enables label-free, high-resolution spatial mapping of the chemical composition of complex, biological samples. Typical experiments require selecting ions and/or positions from the images: ions for fragmentation studies to identify keystone compounds and positions for follow up validation measurements using microdissection or other orthogonal techniques. Unfortunately, with modern imaging machines, these must be selected from an overwhelming amount of raw data. Existing techniques to reduce the volume of data, the most popular of which are principle component analysis and non-negative matrix factorization, have the disadvantage that they return difficult-to-interpret linear combinations of actual data elements. In this work, we show that CX and CUR matrix decompositions can be used directly to address this selection need. CX and CUR matrix decompositions use empirical statistical leverage scores of the input data to provide provably good low-rank approximations of the measured data that are expressed in terms of actual ions and actual positions, as opposed to difficult-to-interpret eigenions and eigenpositions. We show that this leads to effective prioritization of information for both ions and positions. In particular, important ions can be found either by using the leverage scores as a ranking function and using a deterministic greedy selection algorithm or by using the leverage scores as an importance sampling distribution and using a random sampling algorithm; however, selection of important positions from the original matrix performed significantly better when they were chosen with the random sampling algorithm. Also, we show that 20 ions or 40 locations can be used to reconstruct the original matrix to a tolerance of 17% error for a widely studied image of brain lipids; and we provide a scalable implementation of this method that is applicable for analysis of the raw data where there are often more than a million rows and/or columns, which is larger than SVD-based low-rank approximation methods can handle. Lastly, these results introduce the concept of CX/CUR matrix factorizations to mass spectrometry imaging, describing their utility and illustrating principled algorithmic approaches to deal with the overwhelming amount of data generated by modern mass spectrometry imaging.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Cutting planes for the multistage stochastic unit commitment problem

As renewable energy penetration rates continue to increase in power systems worldwide, new challenges arise for system operators in both regulated and deregulated electricity markets to solve the security-constrained coal-fired unit commitment problem with intermittent generation (due to renewables) and uncertain load, in order to ensure system reliability and maintain cost effectiveness. In this paper, we study a security-constrained coal-fired stochastic unit commitment model, which we use to enhance the reliability unit commitment process for day-ahead power system operations. In our approach, we first develop a deterministic equivalent formulation for the problem, which leads to a large-scale mixed-integer linear program. Then, we verify that the turn on/off inequalities provide a convex hull representation of the minimum-up/down time polytope under the stochastic setting. Next, we develop several families of strong valid inequalities mainly through lifting schemes. In particular, by exploring sequence independent lifting and subadditive approximation lifting properties for the lifting schemes, we obtain strong valid inequalities for the ramping and general load balance polytopes. Lastly, branch-and-cut algorithms are developed to employ these valid inequalities as cutting planes to solve the problem. Our computational results verify the effectiveness of the proposed approach.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Matrix-free subcell residual distribution for Bernstein finite element discretizations of linear advection equations

In this work, we introduce a new residual distribution (RD) framework for the design of bound-preserving high-resolution finite element schemes. The continuous and discontinuous Galerkin discretizations of the linear advection equation are modified to construct local extremum diminishing (LED) approximations. To that end, we perform mass lumping and redistribute the element residuals in a manner which guarantees the LED property. The hierarchical correction procedure for high-order Bernstein finite element discretizations involves localization to subcells and definition of bound-preserving weights for subcell contributions. Using strong stability preserving (SSP) Runge–Kutta methods for time integration, we prove the validity of discrete maximum principles under CFL-like time step restrictions. The low-order version of our method has roughly the same accuracy as the one derived from a piecewise (multi)-linear approximation on a submesh with the same nodal points. In high-order extensions, we use an element-based flux-corrected transport (FCT) algorithm which can be interpreted as a nonlinear RD scheme. The proposed LED corrections are tailor-made for matrix-free implementations which avoid the rapidly growing cost of matrix assembly for high-order Bernstein elements. The results for 1D, 2D, and 3D test problems compare favorably to those obtained with the best matrix-based approaches.

97 MATHEMATICS AND COMPUTING↗

Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations

Hejre, we introduce physics-informed neural networks – neural networks that are trained to solve supervised learning tasks while respecting any given laws of physics described by general nonlinear partial differential equations. In this work, we present our developments in the context of solving two main classes of problems: data-driven solution and data-driven discovery of partial differential equations. Depending on the nature and arrangement of the available data, we devise two distinct types of algorithms, namely continuous time and discrete time models. The first type of models forms a new family of data-efficient spatio-temporal function approximators, while the latter type allows the use of arbitrarily accurate implicit Runge–Kutta time stepping schemes with unlimited number of stages. Moreover, the effectiveness of the proposed framework is demonstrated through a collection of classical problems in fluids, quantum mechanics, reaction–diffusion systems, and the propagation of nonlinear shallow-water waves.

97 MATHEMATICS AND COMPUTING↗

Description of the code ANVIL (ANisotropic Vhf Impulse Location)

In this paper, we have developed a general, three-dimensional method to locate sources of earth-directed radiation which takes into account the flight paths of reflected signals. While time-of-arrival algorithms exist for locating radio sources using line-of-sight propagation paths, radio sources originating above the surface of the earth will not necessarily emit strong power along direct paths to satellites. A combination of direct and ground-reflected pulses or only ground-reflected signals from such sources could be received by satellite-borne sensors. The work presented here applies to satellite detection of sub-ionospheric radio sources in a vacuum environment with ideal reflection off the earth's surface. Because satellites are not static receivers, their configuration with respect to a radio source is not always optimal. Therefore, a statistical study is performed using 1000 randomized configurations of 24 satellites in middle earth orbit. For each configuration, the radio source latitude and longitude is fixed, and its altitude is varied from 1 to 97 km. An analytic direct-path algorithm using the five satellites nearest the radio source provides an initial guess for the radio source latitude and longitude. We find that, using this approximation, the mean error in the calculated nadir location has a maximum value of 134 m for the 97-km source altitude. The coordinates of the initial guess are used to define the origin of a grid within which an all-points search refinement is performed. Finally, using this procedure, the overall maximum mean error in radio source position is found to be on the order of the computational grid size.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Prediction of vacancy defect diffusion paths in high entropy alloys via machine learning on molecular dynamics data

Identifying the diffusion path of point defects is a critical step in understanding their evolution and the mechanisms of related phenomena. Defect diffusion occurs at small length and time scales, with impacts on material properties that may continue to evolve over ns to μs, ms, and the continuum scale (s, min, etc., and cm, m, etc.). The time scale accessible to molecular dynamics (MD) simulations is limited by small step sizes, typically in the fs range. Thus, surrogate models of MD simulations through machine learning (ML)-based algorithms are of great interest, especially for complex systems such as high entropy alloys (HEAs). In this work, dynamics governing vacancy migration in HEA were approximated with graph convolutional network (GCN) models as ansatzes for kinetic Monte Carlo (KMC) rate catalogs. Network design considered that diffusion in crystalline solids generally depends on interactions between defects and their immediate neighbor atoms. Graphs represented the vacancy surroundings, MD-generated trajectories provided training and comparison datasets, and unsupervised GCN models approximated interatomic dynamics governing vacancy migration in HEAs as ansatzes for KMC. A proof-of-concept model trained on MD data for the Fe, Ni, Cr, Co, and Cu HEA environment was used with two different neighbor interactions to assess the feasibility of training a GCN to predict vacancy defect transition rates in the HEA environment. The resulting setup rapidly generated MD-formatted synthetic trajectories based on dynamics learned from the MD training set, with a time acceleration of roughly two orders of magnitude and a similar diffusion coefficient to MD observations. Additionally, Nudged Elastic Band (NEB) calculations were performed on randomly generated FeNiCrCoCu HEA structures to determine vacancy migration barriers across nearest-neighbor sites. Transition probabilities for each jump, categorized by atomic type, were extracted from these calculations. NEB-based and GCN-based approaches led to similar outcomes.

Reimer, C↗

On the use of HDR {sup 60}Co source with the MammoSite Radiation Therapy System

This work summarizes Monte Carlo results in order to evaluate the potential of using HDR {sup 60}Co sources in accelerated partial breast irradiation (APBI) with the MammoSite applicator. Simulations have been performed using the MCNP5 Monte Carlo Code, in simple geometries comprised of two concentric spheres; the internal consisting of selected concentrations, C, of a radiographic contrast solution in water (Omnipaque 300) to simulate the MammoSite balloon and the external consisting of water to simulate surrounding tissue. The magnitude of the perturbation of delivered dose due to the radiographic contrast medium used in the MammoSite applicator is calculated. At the very close vicinity of the balloon surface, a dose build-up region is observed, which leads to a dose overestimation by the treatment planning system (TPS) which depends on Omnipaque 300 solution concentration (and is in order of 2.3%, 3.0%, and 4.5%, respectively, at 1 mm away from the balloon - water interface, for C=10%, 15%, and 20%). However, dose overestimation by the TPS is minimal for points lying at the prescription distance (d=1 cm) or beyond, for all simulated concentrations and radii of MammoSite balloon. An analytical estimation of the integral dose outside the CTV in the simple geometries simulated shows that dose to the breast for MammoSite applications is expected to be comparable using HDR {sup 60}Co and {sup 192}Ir sources, and higher than that for {sup 169}Yb. The higher enegies of {sup 60}Co sources result to approximately twice radiation protection requirements as compared to {sup 169}Ir sources. However, they allow for more accurate dosimetry calculation with currently used treatment planning algorithms for {sup 60}Co sources, compared to {sup 169}Ir.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Clustering at Massive Scale

ClaMS provides hierarchical clustering technology for use on massive, high-dimensional datasets that require distributed memory for processing. The algorithm employed is inspired by the popular HDBSCAN algorithm but makes use of computational kernels better suited for distributed computing. ClaMS is built on scalable nearest neighbor graph construction, metric forest completion, and approximate minimum spanning tree techniques.

Stanley, ThomasA [Lawrence Livermore National Labo↗

Insights into cosmological structure formation with machine learning.

Our modern understanding of cosmological structure formation posits that small matter density fluctuations present in the early Universe, as traced by the cosmic microwave background, grow via gravitational instability to form extended haloes of dark matter. A theoretical understanding of the structure, evolution and formation of dark matter haloes is an essential step towards unravelling the intricate connection between halo and galaxy formation, needed to test our cosmological model against data from upcoming galaxy surveys. Physical understanding of the process of dark matter halo formation is made difficult by the highly non-linear nature of the haloes' evolution. I describe a new approach to gain physical insight into cosmological structure formation based on machine learning. This approach combines the ability of machine learning algorithms to learn non-linear relationships, with techniques that enable us to physically interpret the learnt mapping. I describe applicatio ns of th e method, with the aim of investigating which aspects of the early universe density field impact the later formation of dark matter haloes. First I present a case where the process of halo formation is turned into a binary classification problem; the algorithm predicts whether or not dark matter `particles' in the initial conditions of a simulation will collapse into haloes of a given mass range. Second, I present its generalization to regression, where the algorithm infers the final mass of the halo to which each particle will later belong. I show that the initial tidal shear does not play a significant role compared to the initial density field in establishing final halo masses. Finally, I demonstrate that extending the framework to deep learning algorithms such as convolutional neural networks allows us to explore connections between the early universe and late time haloes beyond those studied by existing analytic approximations of halo collapse.

79 ASTRONOMY AND ASTROPHYSICS↗