Search NASA⌕ Search

SEARCH · Search NASA

Results for “reasoning about programs”

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 775 records · Page 43

Physical Security Model Development of an Electrochemical Facility

Nuclear facilities in the U.S. and around the world face increasing challenges in meeting evolving physical security requirements while keeping costs reasonable. The addition of security features after a facility has been designed and without attention to optimization (the approach of the past) can easily lead to cost overruns. Instead, security should be considered at the beginning of the design process in order to provide robust, yet efficient physical security designs. The purpose of this work is to demonstrate how modeling and simulation can be used to optimize the design of physical protection systems. A suite of tools, including Scribe3D and Blender, were used to model a generic electrochemical reprocessing facility. Physical protection elements such as sensors, portal monitors, barriers, and guard forces were added to the model based on best practices for physical security. Two theft scenarios (an outsider attack and insider diversion) as well as a sabotage scenario were examined in order to optimize the security design. Security metrics are presented. This work fits into a larger Virtual Facility Distributed Test Bed 2020 Milestone in the Material Protection, Accounting, and Control Technologies (MPACT) program through the Department of Energy (DOE). The purpose of the milestone is to demonstrate how a series of experimental and modeling capabilities across the DOE complex provide the capabilities to demonstrate complete Safeguards and Security by Design (SSBD) for nuclear facilities.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Comparison of Resonance Parameter Covariance Generation using CONRAD and SAMMY Computer Codes

Cross section evaluations in the resolved resonance region are based on formalisms derived from the R-matrix theory. As a result, the evaluations provide a set of resonance parameters that can be used to reproduce the experimental data reasonably well. The evaluated nuclear data are used in neutron transport calculations for the analysis and design of nuclear reactor systems, nuclear criticality safety analyses, etc. To achieve the desired accuracy on the nuclear system calculations, the questions frequently asked are how well the nuclear data are known and how the uncertainty in the nuclear data can be propagated into the final nuclear system results. There have been ongoing efforts at several research centers for generating data uncertainties in the resonance and high-energy regions. The biggest issue in relation to the covariance data is how good the calculated uncertainties are or whether the calculated uncertainties are in agreement with realistic uncertainties derived from an experimental nuclear system or nuclear benchmark. In this work an attempt is made to use two distinct and independently developed computer codes, CONRAD and SAMMY, to evaluate and generate covariance data in the resonance region. The verification study has been performed in support of the U.S. Nuclear Criticality Safety program (NCSP) as the NCSP is working to provide improved nuclear data files to support criticality safety analyses. The objective is to check the procedures and the methodologies used in the resonance region for covariance generation. The studies have been carried out using the 48Ti resolved resonance parameters.

97 MATHEMATICS AND COMPUTING↗

MLC quality assurance using EPID: A fitting technique with subpixel precision

Amorphous silicon based electronic portal imaging devices (EPIDs) have been shown to be a good alternative to radiographic film for routine quality assurance (QA) of multileaf collimator (MLC) positioning accuracy. In this work, we present a method of acquiring an EPID image of a traditional strip-test image using analytical fits of the interleaf and leaf abutment image signatures. After exposure, the EPID image pixel values are divided by an open field image to remove EPID response and radiation field variations. Profiles acquired in the direction orthogonal to the leaf motion exhibit small peaks caused by interleaf leakage. Gaussian profiles are fitted to the interleaf leakage peaks, the results of which are, using multiobjective optimization, used to calculate the image rotational angle with respect to the collimator axis of rotation. The relative angle is used to rotate the image to align the MLC leaf travel to the image pixel axes. The leaf abutments also present peaks that are fitted by heuristic functions, in this case modified Lorentzian functions. The parameters of the Lorentzian functions are used to parameterize the leaf gap width and positions. By imaging a set of MLC fields with varying gaps forming symmetric and asymmetric abutments, calibration curves with regard to relative peak height (RPH) versus nominal gap width are obtained. Based on this calibration data, the individual leaf positions are calculated to compare with the nominal programmed positions. The results demonstrate that the collimator rotation angle can be determined as accurate as 0.01 deg. . A change in MLC gap width of 0.2 mm leads to a change in RPH of about 10%. For asymmetrically produced gaps, a 0.2 mm MLC leaf gap width change causes 0.2 pixel peak position change. Subpixel resolution is obtained by using a parameterized fit of the relatively large abutment peaks. By contrast, for symmetrical gap changes, the peak position remains unchanged with a standard deviation of 0.05 pixels, or 0.026 mm. A trial run of 36 test images, each with gap widths varying from 0.4 to 1.4 mm, were used to analyze 8640 abutments. The leaf position variations were detected with a precision of 0.1 mm at a 95% confidence level, with a mean of 0.04 mm and a standard deviation of 0.03 mm. The proposed method is robust and minimizes the effect of image noise and pixel size and may help physicists to establish reliable and reasonable action levels in routine MLC QA.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Reflected solar radiance seen by satellite

We derive the integral expression for the reflected solar flux F (W/m 2 ) seen by a sensor deployed on a satellite circling the earth at elevation h t . The sensor’s responsivity is that of a Si photodiode, (0.3, 1.1) µm. We first ignore atmospheric absorption, but account for light emanating from all angles, zenith up to parallel to earth surface. The expression, which holds for all “Solar–Earth–Sensor” (elevation) angles φ e , corrects a long-used equation in the USNDS program, that models “background irradiance,” Ramsey [3], Eq.(2.2). Flux striking the sensor is proportional to the sea-level solar radiance I 0 and the average earth albedo A. Results are displayed for h t = 2.02 · 10 4 km. The integral varies with angle φ e , i.e., F = F(φ e ). When sensor and sun are aligned with earth center, F = F(0) = 0.620 I 0 AR 2 , where R = r e /(r e + h t ) and r e is the average earth radius. For the specified h t , |φ e | < 180 - θ 0 deg, where θ 0 = 13.90. For larger angles, F = 0 since the sensor cannot see any part of earth illuminated by the sun. The flux F(φ e ) decays as ϵ 3.5 when φ e = 180-θ 0 - ϵ and ϵ→ 0, i.e., as the sensor enters darkness. Our ϵ 3.5 result differs from Ramsey’s [3], Eq.(2.2), which has ϵ 4.3 . We conclude Eq.(2.2) is incorrect for two reasons: the erroneous decay and, more importantly, that it vanishes only at φ e = 180. Our result, Eq.(14), presents a simple approximation to F(φ e ), accurate to better than 0.5%. The approximation blends Ramsey’s expression for small φ e with an ϵ 3.5 decay for large angles. Assuming a solar radiance I 0 = 1120, W/m 2 Wikipedia [7], and albedo A = 0.3, NASA, yields F(0) = 3.8 W/m 2 , which is 80% less than the leading term in Ramsey [3], Eq.(2.2). If atmospheric absorption is not ignored our integrand is modified by the factor exp(-τ N atm ), where τ is the atmosphere’s optical depth and N atm accounts for slant path absorption. The result is spectrally dependent. Absorption significantly decreases the flux as φ e → 180 - θ 0 .

79 ASTRONOMY AND ASTROPHYSICS↗

Scaled Testing to Evaluate Pulse Jet Mixer Performance in Waste Treatment Plant Mixing Vessels

The Waste Treatment and Immobilization Plant (WTP) at Hanford is being designed and built to pre-treat and vitrify the waste in Hanford’s 177 underground waste storage tanks. Numerous process vessels will hold waste at various stages in the WTP. These vessels have pulse jet mixer (PJM) systems. A test program was developed to evaluate the adequacy of mixing system designs in the solids-containing vessels in the WTP. The program focused mainly on non-cohesive solids behavior. Specifically, the program addressed the effectiveness of the mixing systems to suspend settled solids off the vessel bottom, and distribute the solids vertically. Experiments were conducted at three scales using various particulate simulants. A range of solids loadings and operational parameters were evaluated, including jet velocity, pulse volume, and duty cycle. In place of actual PJMs, the tests used direct injection from tubes with suction at the top of the tank fluid. This gave better control over the discharge duration and duty cycle and simplified the facility requirements. The mixing system configurations represented in testing varied from 4 to 12 PJMs with various jet nozzle sizes. In this way the results collected could be applied to the broad range of WTP vessels with varying geometrical configurations and planned operating conditions. Data for “just-suspended velocity”, solids cloud height, and solids concentration vertical profile were collected, analyzed, and correlated. The correlations were successfully benchmarked against previous large-scale test results, then applied to the WTP vessels using reasonable assumptions of anticipated waste properties to evaluate adequacy of the existing mixing system designs.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

ROSE::FTTransform - A Source-to-Source Translation Framework for Exascale Fault-Tolerance Research

Exascale computing systems will require sufficient resilience to tolerate numerous types of hardware faults while still assuring correct program execution. Such extreme-scale machines are expected to be dominated by processors driven at lower voltages (near the minimum 0.5 volts for current transistors). At these voltage levels, the rate of transient errors increases dramatically due to the sensitivity to transient and geographically localized voltage drops on parts of the processor chip. To achieve power efficiency, these processors are likely to be streamlined and minimal, and thus they cannot be expected to handle transient errors entirely in hardware. Here we present an open, compiler-based framework to automate the armoring of High Performance Computing (HPC) software to protect it from these types of transient processor errors. We develop an open infrastructure to support research work in this area, and we define tools that, in the future, may provide more complete automated and/or semi-automated solutions to support software resiliency on future exascale architectures. Results demonstrate that our approach is feasible, pragmatic in how it can be separated from the software development process, and reasonably efficient (0% to 30% overhead for the Jacobi iteration on common hardware; and 20%, 40%, 26%, and 2% overhead for a randomly selected subset of benchmarks from the Livermore Loops [1]).

97 MATHEMATICS AND COMPUTING↗

The diurnal cycle of clouds and precipitation at the ARM SGP site: Cloud radar observations and simulations from the multiscale modeling framework

Millimeter Wavelength Cloud Radar (MMCR) data from December 1996 to December 2010, collected at the U.S. Department of Energy Atmospheric Radiation Measurement (ARM) program Southern Great Plains (SGP) site, are used to examine the diurnal cycle of hydrometeor occurrence. These data are categorized into clouds (-40 dBZ e ≤ reflectivity < -10 dBZ e ), drizzle and light precipitation (-10 dBZ e ≤ reflectivity < 10 dBZ e ), and heavy precipitation (reflectivity ≥ 10 dBZ e ). The same criteria are implemented for the observation-equivalent reflectivity calculated by feeding outputs from a Multiscale Modeling Framework (MMF) climate model into a radar simulator. The MMF model consists of the National Center for Atmospheric Research Community Atmosphere Model with conventional cloud parameterizations replaced by a cloud-resolving model. We find that a radar simulator combined with the simple reflectivity categories can be an effective approach for evaluating diurnal variations in model hydrometeor occurrence. It is shown that the MMF only marginally captures observed increases in the occurrence of boundary layer clouds after sunrise in spring and autumn and does not capture diurnal changes in boundary layer clouds during the summer. Above the boundary layer, the MMF captures reasonably well diurnal variations in the vertical structure of clouds and light and heavy precipitation in the summer but not in the spring.

54 ENVIRONMENTAL SCIENCES↗

Assessment of the MAMMOTH and PRONGHORN Pebble Bed Reactor Modeling Capabilities for the HTR-PM Design

This assessment explores the current analysis methodologies of various external and MOOSE-based tools available at the Idaho National Laboratory to model pebble bed reactor multiphysics phenomena. The study involves three MOOSE-based, coupled codes: 1) MAMMOTH for reactor physics, 2) Rattlesnake for neutron transport and 3) Pronghorn for thermal fluids and strongly coupled conjugate heat transfer. These advanced tools are being developed under DOE-NE’s Nuclear Energy Advanced Modeling and Simulation program (NEAMS). They currently contain a collection of advanced modeling capabilities that allow high-fidelity simulations of advanced reactor concepts including gas and fluoride-salt cooled pebble bed reactors. Software development of these tools follows NQA-1 standards ensuring software quality that many of the currently deployed legacy codes lack. In this work, the MAMMOTH Pronghorn tools are compared to the legacy codes PEBBED-THERMIX-KONVEK. The external codes in use are LAMMPS for pebble mechanics and Serpent for cross section preparation. The coupling under MOOSE enables significant flexibility for modern engineering tools since MOOSE applications can be seamlessly coupled with various levels of resolution. This report focuses on low-resolution (r-z) and midresolution (3-D) homogenized geometries. The high-resolution Pebble Tracking Transport solution is not pursued herein and is postponed to future work. The results from this assessment show that the new MOOSE-based tools compare well with the results from legacy tools (r-z) and produce reasonable accuracy limited by the methods in the legacy tools. In the 3-D studies, the PJFNK-SPH methodology available in MAMMOTH allows the reproduction of Monte Carlo results with a transport-corrected diffusion solution. This has implications for the coupled solutions since without the SPH correction the temperature coefficient of reactivity is under-predicted and leads to lower temperature feedback and a significantly higher eigenvalue showing a difference of 2000 pcm.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

THE MASS-LOSS RETURN FROM EVOLVED STARS TO THE LARGE MAGELLANIC CLOUD. II. DUST PROPERTIES FOR OXYGEN-RICH ASYMPTOTIC GIANT BRANCH STARS

We model multi-wavelength broadband UBVIJHK{sub s} and Spitzer IRAC and MIPS photometry and Infrared Spectrograph spectra from the SAGE and SAGE-Spectroscopy observing programs of two oxygen-rich asymptotic giant branch (O-rich AGB) stars in the Large Magellanic Cloud (LMC) using radiative transfer (RT) models of dust shells around stars. We chose a star from each of the bright and faint O-rich AGB populations found by earlier studies of the SAGE sample in order to derive a baseline set of dust properties to be used in the construction of an extensive grid of RT models of the O-rich AGB stars found in the SAGE surveys. From the bright O-rich AGB population, we chose HV 5715, and from the faint O-rich AGB population we chose SSTISAGE1C J052206.92-715017.6 (SSTSAGE052206). We found the complex indices of refraction of oxygen-deficient silicates from Ossenkopf et al. and a power law with exponential decay grain size distribution like what Kim et al. used but with {gamma} of -3.5, a {sub min} of 0.01 {mu}m, and a {sub 0} of 0.1 {mu}m to be reasonable dust properties for these models. There is a slight indication that the dust around the faint O-rich AGB may be more silica-rich than that around the bright O-rich AGB. Simple models of gas emission suggest a relatively extended gas envelope for the faint O-rich AGB star modeled, consistent with the relatively large dust shell inner radius for the same model. Our models of the data require the luminosity of SSTSAGE052206 and HV 5715 to be {approx}5100 L {sub sun} and {approx}36,000 L {sub sun}, respectively. This, combined with the stellar effective temperatures of 3700 K and 3500 K, respectively, that we find best fit the optical and near-infrared data, suggests stellar masses of {approx}3 M {sub sun} and {approx}7 M {sub sun}. This, in turn, suggests that HV 5715 is undergoing hot-bottom burning and that SSTSAGE052206 is not. Our models of SSTSAGE052206 and HV 5715 require dust shells of inner radius {approx}17 and {approx}52 times the stellar radius, respectively, with dust temperatures there of 900 K and 430 K, respectively, and with optical depths at 10 {mu}m through the shells of 0.095 and 0.012, respectively. The models compute the dust mass-loss rates for the two stars to be 2.0 x 10{sup -9} M{sub sun} yr{sup -1} and 2.3 x 10{sup -9} M{sub sun} yr{sup -1}, respectively. When a dust-to-gas mass ratio of 0.002 is assumed for SSTSAGE052206 and HV 5715, the dust mass-loss rates imply total mass-loss rates of 1.0 x 10{sup -6} M{sub sun} yr{sup -1} and 1.2 x 10{sup -6} M{sub sun} yr{sup -1}, respectively. These properties of the dust shells and stars, as inferred from our models of the two stars, are found to be consistent with properties observed or assumed by detailed studies of other O-rich AGB stars in the LMC and elsewhere.

79 ASTRONOMY AND ASTROPHYSICS↗

Oxy-Combustion System Process Optimization

The overall objective of this work is to develop a new chemical absorbent-based, high pressure, CO 2 purification system to remove the residual oxygen that currently contaminates the recovered CO 2 , and to optimize the Pressurized Oxy-Combustion (POxC) process to minimize the Cost of Electricity (COE) generated in this advanced combustion process. TDA developed and validated the performance of the oxygen removal system for CO 2 purification. In collaboration with the Advanced Power & Energy Program (APEP) of University of California, Irvine (UCI), we optimized the POxC process, including thermal management, heat integration, and power cycle optimization using process design and modeling supported with Aspen Plus® process simulations. The techno-economic analysis results indicate that the pressurized oxycombustion coal power plant with Ion-transport membrane (ITM) air separation unit (ASU) (Case 2 – 30.55%) does not show an advantage over a cryogenic ASU (Case 1 – 31.24%) while TDA’s sorption-based ASU (Case 4 - 32.61%) shows a significant advantage over the cryogenic ASU (Case 1 – 331.24%). The specific plant costs show a wide range with a low of $2544/kW for Case 11C (TDA ASU, co-sequestering the SO x , and ultra-supercritical steam cycle) to a high of $2975/kW for Case 12A (cryogenic ASU and sCO 2 cycle). In general, the ITM ASU based cases have lower specific plant costs than the corresponding cryogenic ASU based cases while the TDA ASU based cases show the lowest specific plant costs. The main reason for lowering these costs is the higher overall plant thermal efficiency which decreases the plant cost on a $ per kW basis. Next comparing the cases with different power cycle working fluid conditions in terms of temperature and pressure while all utilizing steam, similar trends as the plant costs may be observed. However, with the supercritical CO 2 (sCO 2 ) cycle, the increase in thermal efficiency of the sCO 2 cycle was not able to offset its increase in plant cost making the plant costs higher than those of the corresponding steam cycle cases. The Cost of Electricity (COE) again shows similar trends as the specific plant costs. The COE for Case 11C at $110.1/MWh is also the lowest, but among all cases that do not co-sequester the SO x , Case 12C (TDA ASU and sCO 2 cycle) has the lowest COE at $\$$117.5/MWh while the highest is for Case 8A (cryogenic ASU and supercritical steam cycle) at $130.4/MWh.

20 FOSSIL-FUELED POWER PLANTS↗

SU-E-T-613: Dosimetric Consequences of Systematic MLC Leaf Positioning Errors

Purpose: The purpose of this study is to determine the dosimetric consequences of systematic MLC leaf positioning errors for clinical IMRT patient plans so as to establish detection tolerances for quality assurance programs. Materials and Methods: Dosimetric consequences were simulated by extracting mlc delivery instructions from the TPS, altering the file by the specified error, reloading the delivery instructions into the TPS, recomputing dose, and extracting dose-volume metrics for one head-andneck and one prostate patient. Machine error was simulated by offsetting MLC leaves in Pinnacle in a systematic way. Three different algorithms were followed for these systematic offsets, and are as follows: a systematic sequential one-leaf offset (one leaf offset in one segment per beam), a systematic uniform one-leaf offset (same one leaf offset per segment per beam) and a systematic offset of a given number of leaves picked uniformly at random from a given number of segments (5 out of 10 total). Dose to the PTV and normal tissue was simulated. Results: A systematic 5 mm offset of 1 leaf for all delivery segments of all beams resulted in a maximum PTV D98 deviation of 1%. Results showed very low dose error in all reasonably possible machine configurations, rare or otherwise, which could be simulated. Very low error in dose to PTV and OARs was shown in all possible cases of one leaf per beam per segment being offset (<1%), or that of only one leaf per beam being offset (<.2%). The errors resulting from a high number of adjacent leaves (maximum of 5 out of 60 total leaf-pairs) being simultaneously offset in many (5) of the control points (total 10–18 in all beams) per beam, in both the PTV and the OARs analyzed, were similarly low (<2–3%). Conclusions: The above results show that patient shifts and anatomical changes are the main source of errors in dose delivered, not machine delivery. These two sources of error are “visually complementary” and uncorrelated (albeit not additive in the final error) and one can easily incorporate error resulting from machine delivery in an error model based purely on tumor motion.

60 APPLIED LIFE SCIENCES↗

NCERC-CL TREAT Free Field Characterization Foil Reaction Rate Results

This report summarizes the analysis and preliminary MCNP reaction rate modeling of free field characterization measurements performed in the Transient Reaction Test Facility (TREAT) at Idaho National Laboratory (INL) between 12/18/2023 and 2/22/2024. A series of seven irradiations were performed using the Big-BUSTER (Broad Use Specimen Transient Experiment Rig) core configuration. Foil sets including 19.5% LEU-Zr alloy wire, S, Au, Fe, Ni, Co, Ti, and Zr were deployed in each irradiation. Identical foil sets were supplied to Los Alamos National Laboratory (LANL), Lawrence Livermore National Laboratory (LLNL), and INL. All foils were supplied by INL. Additional details on the experiment can be found in [2]. The foil sets were shipped to the National Criticality Experiments Research Center Counting Laboratory (NCERC-CL) branch at LANL (NCERC-CL NISC) and were received on 3/25/2024. Additional details on receipt and counting are reported in [3]. All reactor dosimetry measurements were performed in adherence to the ASTM standards applicable to reactor dosimetry. Measured reaction rates from Au, Co, Fe, Ni, and Ti are reported. Select reaction rate ratios and simulated reaction rates and ratios are discussed. Reaction rates for all measured reaction products were modeled with MCNP 6.3.1 and a TREAT input deck supplied by Edward Lum. Neutron emission estimates calculated using the measured and simulated reaction rates show excellent agreement between the three fast-threshold n-p reactions with the percent differences being < 9% between the three reactions. The capture reactions had poorer, but still reasonable, agreement with < 20% percent differences between the neutron emission estimates of the three capture reactions. The percent difference between the fast-threshold and capture reactions is ∼ 100% for each reaction. This supports the TREAT model is not accurately modeling the neutron spectrum and additional measurements are required to characterize TREAT and match measurements to simulation.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

GlusterFS One Storage Server to Rule Them All

GlusterFS is a Linux based distributed file system, designed to be highly scalable and serve many clients. Some reasons to use GlusterFS are: No centralized metadata server, Scalability, Open Source, Dynamic and live service modifications, Can be used over Infiniband or Ethernet, Can be tuned for speed and/or resilience and Flexible administration. It's useful for enterprise environments - virtualization; high performance computing (HPC) and it works with Mac, Linux and Windows clients. Conclusions are: (1) GlusterFS proved to have widespread capabilities as a virtual file system; (2) Scalability is very dependent upon the underlying hardware; (3) Lack of built-in encryption and security paradigm; and (4) Best suited in a general purpose computing environment.

99 GENERAL AND MISCELLANEOUS↗

PETSc Users Manual (Rev. 3.3)

This manual describes the use of PETSc for the numerical solution of partial differential equations and related problems on high-performance computers. The Portable, Extensible Toolkit for Scientific Computation (PETSc) is a suite of data structures and routines that provide the building blocks for the implementation of large-scale application codes on parallel (and serial) computers. PETSc uses the MPI standard for all message-passing communication. PETSc includes an expanding suite of parallel linear, nonlinear equation solvers and time integrators that may be used in application codes written in Fortran, C, C++, Python, and MATLAB (sequential). PETSc provides many of the mechanisms needed within parallel application codes, such as parallel matrix and vector assembly routines. The library is organized hierarchically, enabling users to employ the level of abstraction that is most appropriate for a particular problem. By using techniques of object-oriented programming, PETSc provides enormous flexibility for users. PETSc is a sophisticated set of software tools; as such, for some users it initially has a much steeper learning curve than a simple subroutine library. In particular, for individuals without some computer science background, experience programming in C, C++ or Fortran and experience using a debugger such as gdb or dbx, it may require a significant amount of time to take full advantage of the features that enable efficient software use. However, the power of the PETSc design and the algorithms it incorporates may make the efficient implementation of many application codes simpler than “rolling them” yourself; For many tasks a package such as MATLAB is often the best tool; PETSc is not intended for the classes of problems for which effective MATLAB code can be written. PETSc also has a MATLAB interface, so portions of your code can be written in MATLAB to “try out” the PETSc solvers. The resulting code will not be scalable however because currently MATLAB is inherently not scalable; and PETSc should not be used to attempt to provide a “parallel linear solver” in an otherwise sequential code. Certainly all parts of a previously sequential code need not be parallelized but the matrix generation portion must be parallelized to expect any kind of reasonable performance. Do not expect to generate your matrix sequentially and then “use PETSc” to solve the linear system in parallel. Since PETSc is under continued development, small changes in usage and calling sequences of routines will occur. PETSc is supported; see the web site http://www.mcs.anl.gov/petsc for information on contacting support. A http://www.mcs.anl.gov/petsc/publications may be found a list of publications and web sites that feature work involving PETSc. We welcome any reports of corrections for this document.

97 MATHEMATICS AND COMPUTING↗

PETSc Users Manual (Rev. 3.4)

This manual describes the use of PETSc for the numerical solution of partial differential equations and related problems on high-performance computers. The Portable, Extensible Toolkit for Scientific Computation (PETSc) is a suite of data structures and routines that provide the building blocks for the implementation of large-scale application codes on parallel (and serial) computers. PETSc uses the MPI standard for all message-passing communication. PETSc includes an expanding suite of parallel linear, nonlinear equation solvers and time integrators that may be used in application codes written in Fortran, C, C++, Python, and MATLAB (sequential). PETSc provides many of the mechanisms needed within parallel application codes, such as parallel matrix and vector assembly routines. The library is organized hierarchically, enabling users to employ the level of abstraction that is most appropriate for a particular problem. By using techniques of object-oriented programming, PETSc provides enormous flexibility for users. PETSc is a sophisticated set of software tools; as such, for some users it initially has a much steeper learning curve than a simple subroutine library. In particular, for individuals without some computer science background, experience programming in C, C++ or Fortran and experience using a debugger such as gdb or dbx, it may require a significant amount of time to take full advantage of the features that enable efficient software use. However, the power of the PETSc design and the algorithms it incorporates may make the efficient implementation of many application codes simpler than “rolling them” yourself; For many tasks a package such as MATLAB is often the best tool; PETSc is not intended for the classes of problems for which effective MATLAB code can be written. PETSc also has a MATLAB interface, so portions of your code can be written in MATLAB to “try out” the PETSc solvers. The resulting code will not be scalable however because currently MATLAB is inherently not scalable; and PETSc should not be used to attempt to provide a “parallel linear solver” in an otherwise sequential code. Certainly all parts of a previously sequential code need not be parallelized but the matrix generation portion must be parallelized to expect any kind of reasonable performance. Do not expect to generate your matrix sequentially and then “use PETSc” to solve the linear system in parallel. Since PETSc is under continued development, small changes in usage and calling sequences of routines will occur. PETSc is supported; see the web site http://www.mcs.anl.gov/petsc for information on contacting support. A http://www.mcs.anl.gov/petsc/publications may be found a list of publications and web sites that feature work involving PETSc. We welcome any reports of corrections for this document.

97 MATHEMATICS AND COMPUTING↗

PETSc Users Manual (Rev. 3.5)

This manual describes the use of PETSc for the numerical solution of partial differential equations and related problems on high-performance computers. The Portable, Extensible Toolkit for Scientific Computation (PETSc) is a suite of data structures and routines that provide the building blocks for the implementation of large-scale application codes on parallel (and serial) computers. PETSc uses the MPI standard for all message-passing communication. PETSc includes an expanding suite of parallel linear, nonlinear equation solvers and time integrators that may be used in application codes written in Fortran, C, C++, Python, and MATLAB (sequential). PETSc provides many of the mechanisms needed within parallel application codes, such as parallel matrix and vector assembly routines. The library is organized hierarchically, enabling users to employ the level of abstraction that is most appropriate for a particular problem. By using techniques of object-oriented programming, PETSc provides enormous flexibility for users. PETSc is a sophisticated set of software tools; as such, for some users it initially has a much steeper learning curve than a simple subroutine library. In particular, for individuals without some computer science background, experience programming in C, C++ or Fortran and experience using a debugger such as gdb or dbx, it may require a significant amount of time to take full advantage of the features that enable efficient software use. However, the power of the PETSc design and the algorithms it incorporates may make the efficient implementation of many application codes simpler than “rolling them” yourself. ;For many tasks a package such as MATLAB is often the best tool; PETSc is not intended for the classes of problems for which effective MATLAB code can be written. PETSc also has a MATLAB interface, so portions of your code can be written in MATLAB to “try out” the PETSc solvers. The resulting code will not be scalable however because currently MATLAB is inherently not scalable; and PETSc should not be used to attempt to provide a “parallel linear solver” in an otherwise sequential code. Certainly all parts of a previously sequential code need not be parallelized but the matrix generation portion must be parallelized to expect any kind of reasonable performance. Do not expect to generate your matrix sequentially and then “use PETSc” to solve the linear system in parallel. Since PETSc is under continued development, small changes in usage and calling sequences of routines will occur. PETSc is supported; see the web site http://www.mcs.anl.gov/petsc for information on contacting support. A http://www.mcs.anl.gov/petsc/publications may be found a list of publications and web sites that feature work involving PETSc. We welcome any reports of corrections for this document.

97 MATHEMATICS AND COMPUTING↗

Deep-Learning-Based Multi-Timescale Load Forecasting in Buildings: Opportunities and Challenges from Research to Deployment

Electricity load forecasting for buildings and campuses is becoming increasingly important as the penetration of distributed energy resources (DERs) grows. Efficient operation and dispatch of DERs require reasonably accurate predictions of future energy consumption in order to conduct near-real-time optimized dispatch of on-site generation and storage assets. Electric utilities have traditionally performed load forecasting for load pockets spanning large geographic areas, and therefore, forecasting has not been a common practice by buildings and campus operators. Given the growing trends of research and prototyping in the grid-interactive efficient buildings domain, characteristics beyond simple algorithm forecast accuracy are important in determining the algorithm's true utility for smart buildings. Other characteristics include the overall design of the deployed architecture and the operational efficiency of the forecasting system. In this work, we present a deep-learning-based load forecasting system that predicts the building load at 1-hour intervals for 18 hours in the future. We also discuss challenges associated with the real-time deployment of such systems as well as the research opportunities presented by a fully functional forecasting system that has been developed within the National Renewable Energy Laboratory's Intelligent Campus program.

building load forecasting↗

Specifying structural constraints of architectural patterns in the ARCHERY language

ARCHERY is an architectural description language for modelling and reasoning about distributed, heterogeneous and dynamically reconfigurable systems in terms of architectural patterns. The language supports the specification of architectures and their reconfiguration. This paper introduces a language extension for precisely describing the structural design decisions that pattern instances must respect in their (re)configurations. The extension is a propositional modal logic with recursion and nominals referencing components, i.e., a hybrid µ-calculus. Its expressiveness allows specifying safety and liveness constraints, as well as paths and cycles over structures. Refinements of classic architectural patterns are specified.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗