Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “performance tuning”

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 613 records · Page 34

First characterisation of the MAGO cavity, a superconducting RF detector for kHz–MHz gravitational waves

Heterodyne detection using microwave cavities is a promising method for detecting high-frequency gravitational waves (GWs) or ultralight axion dark matter. In this work, we report on studies conducted on a spherical 2-cell cavity developed by the MAGO collaboration for high-frequency GWs detection. Although fabricated around 20 years ago, the cavity had not been used since. Due to deviations from the nominal geometry, we conducted a mechanical survey and performed room-temperature plastic tuning. Measurements and simulations of the mechanical resonances and electromagnetic properties were carried out, as these are critical for estimating the cavity’s GW coupling potential. Based on these results, we plan further studies in a cryogenic environment. The cavity characterisation does not only provide valuable experience for a planned physics run but also informs the future development of improved cavity designs.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Verification of induction zone models for wind farm annual energy production estimation

Fast engineering wake models are the backbone of wind farm annual energy production (AEP) estimators, whereas the addition of induction zone models in existing tools is a more recent response to rising concerns over wind farm blockage associated losses. Here, "blockage" describes the combined induction fields of all wind turbines inside a farm. Unlike the term might suggest, blockage not only reduces flow speeds, but also increases them; for instance along the outer edges of wind farms. Evaluating the overall impact on AEP of these gains and losses necessitates accurate wind farm induction models. Whilst engineering wake models are tuned to predict wind farm performance, existing induction zone models are all derived for accurately predicting the near field induction not the far field value—important for wind farm simulations. This paper presents the induction models implemented in the wind farm simulation tool PyWake, as well as results from novel analytical models and compares their far field predictions to RANS-AD simulations of different turbines. We demonstrate that when including blockage in AEP simulations, the downstream speed-ups need to be included to avoid an unrealistic bias toward AEP loss and that wake expansion significantly impacts induction at rated thrust levels.

17 WIND ENERGY↗

Vertex finding in neutrino-nucleus interaction: a model architecture comparison

We compare different neural network architectures for machine learning algorithms designed to identify the neutrino interaction vertex position in the MINERvA detector. The architectures developed and optimized by hand are compared with the architectures developed in an automated way using the package “Multi-node Evolutionary Neural Networks for Deep Learning” (MENNDL), developed at Oak Ridge National Laboratory. While the domain-expert hand-tuned network was the best performer, the differences were negligible and the auto-generated networks performed as well. There is always a trade-off between human, and computer resources for network optimization and this work suggests that automated optimization, assuming resources are available, provides a compelling way to save significant expert time.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Multiplicity dependent 𝐽/𝜓 and 𝜓⁡(2⁢𝑆) production at forward and backward rapidity in 𝑝 + 𝑝 collisions at $\sqrt{𝑠}$ = 200 GeV

Recent measurements of 𝐽/𝜓 production as a function of event charged-particle multiplicity at the collision energies of both the Large Hadron Collider (LHC) and the Relativistic Heavy Ion Collider (RHIC) show enhanced 𝐽/𝜓 production yields with increasing multiplicity. One potential explanation for this type of dependence is multiparton interactions (MPI). We present the first study of potential autocorrelations at RHIC energies and forward and backward rapidity of self-normalized 𝐽/𝜓 yields and 𝜓⁡(2⁢𝑆) to 𝐽/𝜓 ratio, as a function of self-normalized multiplicity in 𝑝 + 𝑝 collisions. In addition, detailed pythia studies tuned to RHIC energies were performed to investigate the MPI impacts. We find that the PHENIX data at RHIC are consistent with recent LHC measurements and can only be described by pythia calculations that include MPI effects. The forward and backward 𝜓⁡(2⁢𝑆) to 𝐽/𝜓 ratio is found to be less dependent on the charged-particle multiplicity.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

All-optical switching in a coupled cavity-atom system

We report an experimental observation of all-optical switching in a cavity QED system consisting of multiple three-level atoms confined in a cavity mode. A signal light is coupled into the normal mode of the cavity-atom system and its transmission through the cavity is switched on or off by a control laser coupled to the atoms from free space. Our experiment shows that the all-optical switching in the cavity QED system can be performed for the cavity mode tuned on resonance with or off resonance from the atomic transition frequency.

74 ATOMIC AND MOLECULAR PHYSICS↗

Designing Monte Carlo Simulation and an Optimal Machine Learning to Optimize and Model Space Missions

This paper investigates applying artificial intelligence (AI) algorithms to attitude control system of satellites to optimally tune the controller using high performance computing. This methodology is applied to the Virtual Telescope for X-ray Observation mission, which is a precise formation of two separate spacecraft observing multiple objects in the space in the X-ray domain. The mission is divided into phases based on the instrumentation and the mission goal. To reach an stable precise formation robust to stochastic slew and slew rate (i.e., Euler angles and angular velocities) in a minimal constrained time T , consumed energy of the attitude control system, denoted as E, and root-mean-square state error of attitude control system, denoted as e, are minimized. Monte-Carlo simulation is used for the sensitivity analysis of optimization and designing a controller. Deep neural networks (DNN), Gaussian processes (GP), and support vector regression (SVR) learn this optimization as a surrogate model, while their hyperparameters are optimized in a novel approach. THETA supercomputer at Argonne Leadership Computing Facility (ALCF) is used for optimizing the hyperparameters of DNN. The surrogate model meets the requirements of the mission, and it shows a better performance over the optimization and Monte-Carlo. The optimal DNN can satisfy the mission requirements e and T while reducing E for 90% compared to the other given methods.

42 ENGINEERING↗

A Step towards Energy Efficient Computing: Redesigning a Hydrodynamic Application on CPU-GPU, In: 2014 IEEE 28th International Parallel and Distributed Processing Symposium

Power and energy consumption are becoming an increasing concern in high performance computing. Compared to multi-core CPUs, GPUs have a much better performance per watt. In this paper we discuss efforts to redesign the most computation intensive parts of BLAST, an application that solves the equations for compressible hydrodynamics with high order finite elements, using GPUs BLAST, Dobrev. In order to exploit the hardware parallelism of GPUs and achieve high performance, we implemented custom linear algebra kernels. We intensively optimized our CUDA kernels by exploiting the memory hierarchy, which exceed the vendor's library routines substantially in performance. We proposed an auto tuning technique to adapt our CUDA kernels to the orders of the finite element method. Compared to a previous base implementation, our redesign and optimization lowered the energy consumption of the GPU in two aspects: 60% less time to solution and 10% less power required. Compared to the CPU-only solution, our GPU accelerated BLAST obtained a 2.5× overall speedup and 1.42× energy efficiency (green up) using 4th order (Q_4) finite elements, and a 1.9× speedup and 1.27× green up using 2nd order (Q2) finite elements.

Computer Science↗

Phasing Magnet for CSX-2 Beamline at NSLS-II

A phasing magnet has been developed at National Synchrotron Light Source II (NSLS-II) for the Coherent Soft X-ray Scattering (CSX) beamline. The phasing magnet will be located at the center of the straight section in between two identical and independent variably polarized APPLE-II devices. Based on Permanent Magnet technology, the phasing magnet has been designed to achieve the required electron beam delay to properly adjust the phase matching of these two consecutive EPUs (Elliptically Polarizing Undulators) and ensure a positive interference between the photon beam emitted in each device. Also this paper will describe the mechanical and magnetic design together with the final field measurements and magnetic tuning results. Also, the spectral performance of the two EPUs and the method used to properly set the field strength of the phasing magnet for any given radiation wavelength and polarization mode will be presented as well.

36 MATERIALS SCIENCE↗

Implementation of a Model Predictive Control Strategy to Regulate Temperature Inside Plug-Flow Solar Reactor With Countercurrent Flow

Abstract Solar-driven thermochemical energy storage systems are proven to be promising energy carriers (solar fuels) to utilize solar energy by using reactive solid-state pellets. However, the production of solar fuel requires a quasi-steady-state process temperature, which represents the main challenge due to the transient nature of solar power. In this work, an adaptive model predictive controller (MPC) is presented to regulate the temperature inside a tubular solar reactor to produce solid-state solar fuel for long-term thermal storage systems. The solar reactor system consists of a vertical tube heated circumferentially over a segment of its length by concentrated solar power, and the reactive pellets (MgMn2O4) are fed from the top end and flow downwards through the heated tube. A countercurrent flowing gas supplied from the lower end interacts with flowing pellets to reduce it thermochemically at a temperature range of 1000—1500 °C. A low-order physical model was developed to simulate the dynamics of the solar reactor including the reaction kinetics, and the proposed model was validated numerically by using a 7-kW electric furnace. The numerical model then was utilized to design the MPC controller, where the control system consists of an MPC code linked to an adaptive system identification code that updates system parameters online to ensure system robustness against external disturbances (sudden change in the flow inside the reactor), model mismatches, and uncertainty. The MPC controller parameters are tuned to enhance the system performance with minimum steady-state error and overshoot. The controller is tested to track different temperature ranges between 500 °C and 1400 °C with different particles/gas mass flowrates and ramping temperature profiles. Results show that the MPC controller successfully regulated the reactor temperature within ± 1 °C of its setpoint and maintained robust performance with minimum input effort when subjected to sudden changes in the amount of flowing media and the presence of chemical reaction.

Engineering↗

HPC Programming on Intel Many-Integrated-Core Hardware with MAGMA Port to Xeon Phi

This paper presents the design and implementation of several fundamental dense linear algebra (DLA) algorithms for multicore with Intel Xeon Phi coprocessors. In particular, we consider algorithms for solving linear systems. Further, we give an overview of the MAGMA MIC library, an open source, high performance library, that incorporates the developments presented here and, more broadly, provides the DLA functionality equivalent to that of the popular LAPACK library while targeting heterogeneous architectures that feature a mix of multicore CPUs and coprocessors. The LAPACK-compliance simplifies the use of the MAGMA MIC library in applications, while providing them with portably performant DLA. High performance is obtained through the use of the high-performance BLAS, hardware-specific tuning, and a hybridization methodology whereby we split the algorithm into computational tasks of various granularities. Execution of those tasks is properly scheduled over the heterogeneous hardware by minimizing data movements and mapping algorithmic requirements to the architectural strengths of the various heterogeneous hardware components. Our methodology and programming techniques are incorporated into the MAGMA MIC API, which abstracts the application developer from the specifics of the Xeon Phi architecture and is therefore applicable to algorithms beyond the scope of DLA.

97 MATHEMATICS AND COMPUTING↗

Simulation-Based Inference for Neutrino Parameter Tuning

This code trains a simulation-based inference (SBI) model for neutrino interaction parameter tuning and subsequently evaluates its performance on MicroBooNE, T2K, and NuWro datasets.

Tame-Narvaez, Karla [Fermi National Accelerator La↗

Quantifying the Impact of Vertical Resolution on the Representation of Marine Boundary Layer Physics for Global-Scale Models

Abstract While GCM horizontal resolution has received the majority of scale improvements in recent years, ample evidence suggests that a model’s vertical resolution exerts a strong control on its ability to accurately simulate the physics of the marine boundary layer. Here we show that, regardless of parameter tuning, the ability of a single-column model (SCM) to simulate the subtropical marine boundary layer improves when its vertical resolution is improved. We introduce a novel objective tuning technique to optimize the parameters of an SCM against profiles of temperature and moisture and their turbulent fluxes, horizontal winds, cloud water, and rainwater from large-eddy simulations (LES). We use this method to identify optimal parameters for simulating marine stratocumulus and shallow cumulus. The novel tuning method utilizes an objective performance metric that accounts for the uncertainty in the LES output, including the covariability between model variables. Optimization is performed independently for different vertical grid spacings and value of time step, ranging from coarse scales often used in current global models (120 m, 180 s) to fine scales often used in parameterization development and large-eddy simulations (10 m, 15 s). Uncertainty-weighted disagreement between the SCM and LES decreases by a factor of ∼5 when vertical grid spacing is improved from 120 to 10 m, with time step reductions being of secondary importance. Model performance is shown to converge at a vertical grid spacing of 20 m, with further refinements to 10 m leading to little further improvement. Significance Statement In successive generations of computer models that simulate Earth’s atmosphere, improvements have been mainly accomplished by reducing the horizontal sizes of discretized grid boxes, while the vertical grid spacing has seen comparatively lesser refinements. Here we advocate for additional attention to be paid to the number of vertical layers in these models, especially in the model layers closest to Earth’s surface where climatologically important marine stratocumulus and shallow cumulus clouds reside. Our experiments show that the ability of a one-dimensional model to represent physical processes important to these clouds is strongly dependent on the model’s vertical grid spacing.

Meteorology & Atmospheric Sciences↗

femto-PIXAR: a self-supervised neural network method for reconstructing femtosecond X-ray free electron laser pulses

X-ray Free Electron Lasers (X-FELs) operate in a wide range of lasing configurations for a broad variety of scientific applications at ultrafast time-scales such as structural biology, materials science, and atomic and molecular physics. Shot-by-shot characterization of the X-FEL pulses is crucial for analysis of many experiments as well as tuning the X-FEL performance. However, for the weak pulses found in advanced configurations, e.g. those needed for coherent, two-pulse studies of quantum materials, there is no current method for reliably resolving pulse profiles. Here we show that a physics-based U-net model can reconstruct the individual pulse power profiles for sub-picosecond pulse separation without the need for simulations. Using experimental data from weak X-FEL pulse pairs, we demonstrate we can learn the pulse characteristics on a shot-by-shot basis when conventional methods fail.

43 PARTICLE ACCELERATORS↗

Dynamic Model of a supercritical CO2 10MW Recompression Closed Brayton Cycle

This model of the 10MW recompression closed Brayton cycle (RCBC) was developed in conjunction with the DOE’s Supercritical Transformational Electric Power (STEP) project. A high-fidelity dynamic model was used extensively to study the dynamic characteristics of the cycle and develop the process control architecture and strategies for start-up and shutdown procedures. This version of the model has been simplified from the original version to be more accessible for a variety of applications and research. The controllers developed for the original model have been maintained for this version and tuned to give a similar performance to the original model. This model can be used to perform similar studies as those performed in Liese et al (2020). The overall cycle performance will not be identical but will perform similarly.

Controls,Power Cycles,Process Systems Engineering,↗

Investigation of the LiCaAlF 6 :Eu Scintillator Family for Next-Generation Neutron Detection (Annual Report)

LiCaAlF 6 :Eu (LiCAF:Eu) is a promising scintillator that possesses several fundamental advantages over CsLiYCl 6 :Ce (CLYC). This scintillator, however, is just one member of a much broader family of materials known as colquiriites, designated as LiMe II Me III F 6 , where Me II and Me III can be any number of divalent and trivalent cations, respectively. The LiCAF system has incredible potential for a wide variety of substitutions in the Me II and Me III sites, and its members are recognized as excellent hosts for luminescing activator dopants (e.g., Ce 3+ , Eu 2+ , Cr 3+ , Pb 2+ and Yb 2+ ) in order to tune the nuclear and luminescence performance in several ways. We intend to fully exploit this versatility by investigating the effects of substituting other alkali, alkali-earth, rare-earth, Group III, and Group XIII elements. The objective of this work is to explore the colquiriite system of scintillating materials, most notably LiCaAlF 6 :Eu, to identify and characterize new scintillators with improved scintillation performance, simplified crystal growth, and tunable neutron and gamma sensitivities.

Harrison, Mark J.↗

Convolutional Neural Networks for Signal Detection

Currently, traditional methods such as short-term average/long-term average (STA/LTA) are used to detect arrivals in three-component seismic waveform data. Accurately establishing the identity and arrival of these waves is helpful in detecting and locating seismic events. Convolutional Neural Networks (CNNs) have been shown to significantly improve performance at local distances. This work will expand the use of CNNs to more remote distances and lower magnitudes. Sandia National Labs (SNL) will explore the advantages and limits of a particular approach and investigate requirements for expanding this technique to different types, distances, and magnitudes of events in the future. The team will describe detailed performance results of this method tuned on a curated dataset from Utah with its expert-defined arrival picks.

58 GEOSCIENCES↗

Development of an Advanced Hydrogen Energy Storage System using Aerogel in a Cryogenic Flux Capacitor

The Cryogenic Flux Capacitor (CFC) is a cold, dense fluid storage core with integrated design features that afford the designer flexibility and provide new possibilities for the storage and discharge of energy. The stored energy, in this case, is represented by hydrogen physically bonded within the nanoscale pores within the aerogel composite blanket material, and the process of bonding or debonding is governed by the principles of physical adsorption (physisorption) and thermodynamics. The large surface area afforded by the nanoporous aerogel (~1,000 m 2 /g) allows for storage densities close to, or in some cases exceeding, that of normal boiling point liquids. Its performance easily exceeds what can be achieved via ambient temperature and high-pressure gas storage for an equivalent volume. CFC storage is predicted to be easily scalable, constructed from readily available commercial materials, and lends itself to a range of pressure applications. The project team designed and manufactured the CFC, integrating it into a set of temperature-controlled flow loops. The CFC test setup was validated against existing data from previous NASA work. Tests used bottled hydrogen gas for test validation. A cryogenic fluid, nitrogen, flowed through a metered control system. Hydrogen temperature, pressure, and flow rates, as well as the temperature control targets, provided all necessary control targets. The team analyzed the recorded data and updated the Technology Maturation Plan based on the results of the test, making recommendations for a follow-up test, scale-up, and maturation pathway. For the engineering development and maturation of CFC hydrogen storage technology, a Techno-Economic Assessment (TEA) and Commercialization Plan for the integration of this technology with various power generation assets were produced. A companion analytical model was developed, experimentally validated, compared against performance metric, and used to tune the model and scale up the technology.

08 HYDROGEN↗

G2PDeep-v2: A Web-Based Deep-Learning Framework for Phenotype Prediction and Biomarker Discovery for All Organisms Using Multi-Omics Data

Multi-omics data offers rich insights into complex traits across organisms, yet integrating and analyzing these datasets for phenotype prediction and marker discovery remains challenging. Researchers need accessible tools that combine deep learning, hyperparameter optimization, visualization, and downstream analysis in a unified web platform. To address this, we developed G2PDeep-v2, a web-based platform powered by deep learning for phenotype prediction and marker discovery from multi-omics data across a wide range of organisms, including humans and plants. The server provides multiple services for researchers to create deep-learning models through an interactive interface and train these models using an automated hyperparameter tuning algorithm on high-performance computing resources. Users can visualize the results of phenotype and markers predictions and perform Gene Set Enrichment Analysis for the significant markers to provide insights into the molecular mechanisms underlying complex diseases, conditions and other biological phenotypes being studied.

59 BASIC BIOLOGICAL SCIENCES↗