Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “algorithm selection”

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 955 records · Page 53

Online Adaptive Algorithm for Constraint Energy Minimizing Generalized Multiscale Discontinuous Galerkin Method

Here in this research, we propose an online basis enrichment strategy within the framework of a recently developed constraint energy minimizing generalized multiscale discontinuous Galerkin method. Combining the technique of oversampling, one makes use of the information of the current residuals to adaptively construct basis functions in the online stage to reduce the error of multiscale approximation. A complete analysis of the method is presented, which shows the proposed online enrichment leads to a fast convergence from multiscale approximation to the fine-scale solution. The error reduction can be made sufficiently large by suitably selecting oversampling regions and the number of oversampling layers. Further, the convergence rate of the enrichment algorithm depends on a factor of exponential decay regarding the number of oversampling layers and a user-defined parameter. Numerical results are provided to demonstrate the effectiveness and efficiency of the proposed online adaptive algorithm.

97 MATHEMATICS AND COMPUTING↗

Q4Q: Quantum Computation for Quantum Prediction of Materials and Molecular Properties

We aim at exploring the potential of quantum computation to solve practical problems that are currently targeted by “traditional” high performance computing (HPC). Along the way, we will develop theoretical frameworks and algorithmic strategies. We propose computational research activities that will advance the application of quantum computers to selected materials/molecular properties: (1) Band structure of solids; (2) Molecular electronic/vibronic properties; (3) Free energies and phase transitions.

36 MATERIALS SCIENCE↗

SU-E-T-793: Validation of COMPASS 3D Dosimetry as Pre Treatment Verification with Commercial TPS Algorithms

Purpose: The purpose of this study was to validate the advent of COMPASS 3D dosimetry as a routine pre treatment verification tool with commercially available CMS Monaco and Oncentra Masterplan planning system. Methods: Twenty esophagus patients were selected for this study. All these patients underwent radical VMAT treatment in Elekta Linac and plans were generated in Monaco v5.0 with MonteCarlo(MC) dose calculation algorithm. COMPASS 3D dosimetry comprises an advanced dose calculation algorithm of collapsed cone convolution(CCC). To validate CCC algorithm in COMPASS, The DICOM RT Plans generated using Monaco MC algorithm were transferred to Oncentra Masterplan v4.3 TPS. Only final dose calculations were performed using CCC algorithm with out optimization in Masterplan planning system. It is proven that MC algorithm is an accurate algorithm and obvious that there will be a difference with MC and CCC algorithms. Hence CCC in COMPASS should be validated with other commercially available CCC algorithm. To use the CCC as pretreatment verification tool with reference to MC generated treatment plans, CCC in OMP and CCC in COMPASS were validated using dose volume based indices such as D98, D95 for target volumes and OAR doses. Results: The point doses for open beams were observed <1% with reference to Monaco MC algorithms. Comparisons of CCC(OMP) Vs CCC(COMPASS) showed a mean difference of 1.82%±1.12SD and 1.65%±0.67SD for D98 and D95 respectively for Target coverage. Maximum point dose of −2.15%±0.60SD difference was observed in target volume. The mean lung dose of −2.68%±1.67SD was noticed between OMP and COMPASS. The maximum point doses for spinal cord were −1.82%±0.287SD. Conclusion: In this study, the accuracy of CCC algorithm in COMPASS 3D dosimetry was validated by compared with CCC algorithm in OMP TPS. Dose calculation in COMPASS is feasible within < 2% in comparison with commercially available TPS algorithms.

60 APPLIED LIFE SCIENCES↗

SU-F-J-115: Target Volume and Artifact Evaluation of a New Device-Less 4D CT Algorithm

Purpose: 4DCT is often used in radiation therapy treatment planning to define the extent of motion of the visible tumor (IGTV). Recent available software allows 4DCT images to be created without the use of an external motion surrogate. This study aims to compare this device-less algorithm to a standard device-driven technique (RPM) in regards to artifacts and the creation of treatment volumes. Methods: 34 lung cancer patients who had previously received a cine 4DCT scan on a GE scanner with an RPM determined respiratory signal were selected. Cine images were sorted into 10 phases based on both the RPM signal and the device-less algorithm. Contours were created on standard and device-less maximum intensity projection (MIP) images using a region growing algorithm and manual adjustment to remove other structures. Variations in measurements due to intra-observer differences in contouring were assessed by repeating a subset of 6 patients 2 additional times. Artifacts in each phase image were assessed using normalized cross correlation at each bed position transition. A score between +1 (artifacts “better” in all phases for device-less) and −1 (RPM similarly better) was assigned for each patient based on these results. Results: Device-less IGTV contours were 2.1 ± 1.0% smaller than standard IGTV contours (not significant, p = 0.15). The Dice similarity coefficient (DSC) was 0.950 ± 0.006 indicating good similarity between the contours. Intra-observer variation resulted in standard deviations of 1.2 percentage points in percent volume difference and 0.005 in DSC measurements. Only two patients had improved artifacts with RPM, and the average artifact score (0.40) was significantly greater than zero. Conclusion: Device-less 4DCT can be used in place of the standard method for target definition due to no observed difference between standard and device-less IGTVs. Phase image artifacts were significantly reduced with the device-less method.

60 APPLIED LIFE SCIENCES↗

Exact and efficient hybrid Monte Carlo algorithm for accelerated Bayesian inference of gene expression models from snapshots of single-cell transcripts

Single cells exhibit a significant amount of variability in transcript levels, which arises from slow, stochastic transitions between gene expression states. Elucidating the nature of these states and understanding how transition rates are affected by different regulatory mechanisms require state-of-the-art methods to infer underlying models of gene expression from single cell data. A Bayesian approach to statistical inference is the most suitable method for model selection and uncertainty quantification of kinetic parameters using small data sets. However, this approach is impractical because current algorithms are too slow to handle typical models of gene expression. To solve this problem, we first show that time-dependent mRNA distributions of discrete-state models of gene expression are dynamic Poisson mixtures, whose mixing kernels are characterized by a piecewise deterministic Markov process. Here, we combined this analytical result with a kinetic Monte Carlo algorithm to create a hybrid numerical method that accelerates the calculation of time-dependent mRNA distributions by 1000-fold compared to current methods. We then integrated the hybrid algorithm into an existing Monte Carlo sampler to estimate the Bayesian posterior distribution of many different, competing models in a reasonable amount of time. We demonstrate that kinetic parameters can be reasonably constrained for modestly sampled data sets if the model is known a priori. If there are many competing models, Bayesian evidence can rigorously quantify the likelihood of a model relative to other models from the data. We demonstrate that Bayesian evidence selects the true model and outperforms approximate metrics typically used for model selection.

59 BASIC BIOLOGICAL SCIENCES↗

Comparative Approach of MRI-Based Brain Tumor Segmentation and Classification Using Genetic Algorithm

The detection of a brain tumor and its classification from modern imaging modalities is a primary concern, but a time-consuming and tedious work was performed by radiologists or clinical supervisors. The accuracy of detection and classification of tumor stages performed by radiologists is depended on their experience only, so the computer-aided technology is very important to aid with the diagnosis accuracy. In this study, to improve the performance of tumor detection, we investigated comparative approach of different segmentation techniques and selected the best one by comparing their segmentation score. Further, to improve the classification accuracy, the genetic algorithm is employed for the automatic classification of tumor stage. The decision of classification stage is supported by extracting relevant features and area calculation. The experimental results of proposed technique are evaluated and validated for performance and quality analysis on magnetic resonance brain images, based on segmentation score, accuracy, sensitivity, specificity, and dice similarity index coefficient. The experimental results achieved 92.03% accuracy, 91.42% specificity, 92.36% sensitivity, and an average segmentation score between 0.82 and 0.93 demonstrating the effectiveness of the proposed technique for identifying normal and abnormal tissues from brain MR images. The experimental results also obtained an average of 93.79% dice similarity index coefficient, which indicates better overlap between the automated extracted tumor regions with manually extracted tumor region by radiologists.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Studies of Front-End Electronics for High-Precision Timing Measurements with LGADs [Poster]

In Time-of-Arrival (ToA) measurements, simulations can be used to optimize precision detection and implementation methods of front-end electronics. The simulation tested in this project takes the response of a filter (in the s-domain) from a noisy input current signal from the detector, an LGAD (Low-Gain Avalanche Diode), and uses a threshold to determine the ToA value for the signal. Measuring the ToA values for a large sample of events and comparing them to a reference statistically produces the timing precision of the simulation. Here, the sub-algorithm to model LGAD events with piecewise linear (PWL) approximations (for more efficient Laplace transforms) was optimized. Using the Fourier Discrete Cosine Transform, 25 points could be selected to describe the primary behavior of the signal. Results showed that timing precision is on par with previous methods. We also found that the new algorithm is faster and equally consistent in producing accurate PWL approximations. Testi ng with larger samples is required to fully verify the new algorithm.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Enhancement of molecular sensitivity in positron emission tomography with quantum correlation of γ-ray photons

Enhancement of molecular sensitivity in positron emission tomography (PET) has long been discussed with respect to imaging instrumentation and algorithms for data treatment. Here, the molecular sensitivity in PET is discussed on the basis of 2-dimensional coincident measurements of 511 keV γ ray photons resultant from two-photon annihilation. Introduction of an additional selection window based on the energy sum and difference of the coincidently measured γ ray photons, without any significant instrumental and algorithmic changes, showed an improvement in the signal-to-noise ratio (SNR) by an order of magnitude. Improvement of performance characteristics in the PET imaging system was demonstrated by an increase in the noise equivalent count rate (NECR) which takes both the SNR and the detection efficiency into consideration. A further improvement of both the SNR and the NECR is expected for the present system in real clinical and in-vivo environments, where much stronger positron sources are employed.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Deployment of inference as a service at the US CMS Tier-2 data centers

Coprocessors, especially GPUs, will be a vital ingredient of data production workflows at the HL-LHC. At CMS, the GPU-as-a-service approach for production workflows is implemented by the SONIC project (Services for Optimized Network Inference on Coprocessors). SONIC provides a mechanism for outsourcing computationally demanding algorithms, such as neural network inference, to remote servers, where requests from multiple clients are intelligently distributed across multiple GPUs by a load-balancing service. This talk highlights the recent progress in deploying SONIC at selected U.S. CMS Tier-2 data centers. Using realistic CMS Run3 data processing workflows, such as those containing transformer-based algorithms, we demonstrate how SONIC is integrated into the production-like environment to enable accelerated inference offloading. We will present developments from both the client and server sides, including production job and data center configurations for NVIDIA and AMD GPUs. We will also present performance scaling benchmarks and discuss the challenges of operating SONIC in CMS production, such as server discovery, GPU saturation, fallback server logic, etc.

Holzman, Burt↗

Performance of Heterogeneous Algorithm Scheduling in CMSSW

The CMS experiment started to utilize Graphics Processing Units (GPU) to accelerate the online reconstruction and event selection running on its High Level Trigger (HLT) farm in the 2022 data taking period. The projections of the HLT farm to the High-Luminosity LHC foresee a significant use of compute accelerators in the LHC Run 4 and onwards in order to keep the cost, size, and power budget of the farm under control. This direction of leveraging compute accelerators has synergies with the increasing use of HPC resources in HEP computing, as HPC machines are employing more and more compute accelerators that are predominantly GPUs today. In this work we review the features developed for the CMS data processing framework, CMSSW, to support the effective utilization of both compute accelerators and many-core CPUs within a highly concurrent task-based framework. We measure the impact of various design choices for the scheduling of heterogeneous algorithms on the event processing throughput, using the Run-3 HLT application as a realistic use case.

Bocci, Andrea↗

Seeking Echoes of Circumstellar Disks in Kepler Light Curves

Light echoes of flares on active stars offer the opportunity for direct detection of circumstellar dust. We revisit the problem of identifying faint echoes in postflare light curves, focusing on debris disks from ongoing planet formation. Starting with simulations, we develop an algorithm for estimating the radial extent and total mass from disk echo profiles. We apply this algorithm to light curves from over 2100 stars observed by NASA’s Kepler mission, selected for multiple short-lived flares in either the long-cadence or short-cadence data sets. While flux uncertainties in light curves from individual stars preclude useful mass limits on circumstellar disks, catalog-averaged light curves yield constraints on disk mass that are comparable to estimates from known debris disks. The average mass in micron- to millimeter-sized dust around the Kepler stars cannot exceed 10% of an Earth mass in exo-Kuiper belts or 10% of a lunar mass in the terrestrial zone. We group stars according to IR excess, based on WISE W1–W3 color, as an indicator for the presence of circumstellar dust. The mass limits are greater for stars with strong IR excess, a hint that echoes are lurking not far beneath the noise in postflare light curves. With increased sensitivity, echo detection will let time-domain astronomy complement spectroscopic and direct-imaging studies in mapping how, when, and where planets form.

79 ASTRONOMY AND ASTROPHYSICS↗

Fast analysis of molecular dynamics trajectories with graphics processing units-Radial distribution function histogramming

The calculation of radial distribution functions (RDFs) from molecular dynamics trajectory data is a common and computationally expensive analysis task. The rate limiting step in the calculation of the RDF is building a histogram of the distance between atom pairs in each trajectory frame. Here we present an implementation of this histogramming scheme for multiple graphics processing units (GPUs). The algorithm features a tiling scheme to maximize the reuse of data at the fastest levels of the GPU's memory hierarchy and dynamic load balancing to allow high performance on heterogeneous configurations of GPUs. Several versions of the RDF algorithm are presented, utilizing the specific hardware features found on different generations of GPUs. We take advantage of larger shared memory and atomic memory operations available on state-of-the-art GPUs to accelerate the code significantly. The use of atomic memory operations allows the fast, limited-capacity on-chip memory to be used much more efficiently, resulting in a fivefold increase in performance compared to the version of the algorithm without atomic operations. The ultimate version of the algorithm running in parallel on four NVIDIA GeForce GTX 480 (Fermi) GPUs was found to be 92 times faster than a multithreaded implementation running on an Intel Xeon 5550 CPU. On this multi-GPU hardware, the RDF between two selections of 1,000,000 atoms each can be calculated in 26.9 s per frame. The multi-GPU RDF algorithms described here are implemented in VMD, a widely used and freely available software package for molecular dynamics visualization and analysis.

97 MATHEMATICS AND COMPUTING↗

Application of deep learning methods for beam size control during user operation at the Advanced Light Source

Past research at the Advanced Light Source (ALS) provided a proof-of-principle demonstration that deep learning methods could be effectively employed to compensate for the significant perturbations to the transverse electron beam size induced by user-controlled adjustments of the insertion devices. However, incorporating these methods into the ALS’ daily operations has faced notable challenges. The complexity of the system’s operational requirements and the significant upkeep demands has restricted their sustained application during user operation. Here, we introduce the development of a more robust neural network (NN)-based algorithm that utilizes a novel online fine-tuning approach and its systematic integration into the day-to-day machine operations. Our analysis emphasizes the process of NN model selection, demonstrates the superior performance of the NN-based method over traditional feedback methods, and examines the effectiveness and resilience of the new algorithm during user-operation scenarios. Published by the American Physical Society 2024

43 PARTICLE ACCELERATORS↗

Efficient Joins with Compressed Bitmap Indexes

We present a new class of adaptive algorithms that use compressed bitmap indexes to speed up evaluation of the range join query in relational databases. We determine the best strategy to process a join query based on a fast sub-linear time computation of the join selectivity (the ratio of the number of tuples in the result to the total number of possible tuples). In addition, we use compressed bitmaps to represent the join output compactly: the space requirement for storing the tuples representing the join of two relations is asymptotically bounded by min(h; n . cb), where h is the number of tuple pairs in the result relation, n is the number of tuples in the smaller of the two relations, and cb is the cardinality of the larger column being joined. We present a theoretical analysis of our algorithms, as well as experimental results on large-scale synthetic and real data sets. Our implementations are efficient, and consistently outperform well-known approaches for a range of join selectivity factors. For instance, our count-only algorithm is up to three orders of magnitude faster than the sort-merge approach, and our best bitmap index-based algorithm is 1.2x-80x faster than the sort-merge algorithm, for various query instances. We achieve these speedups by exploiting several inherent performance advantages of compressed bitmap indexes for join processing: an implicit partitioning of the attributes, space-efficiency, and tolerance of high-cardinality relations.

97 MATHEMATICS AND COMPUTING↗

Stochastic simulation of electron avalanches on supercomputers

In the paper, we present a three-dimensional parallel Monte Carlo algorithm named ELSHOW which is developed for simulation of electron avalanches in gases. Parallel implementation of the ELSHOW was made on supercomputers with different architectures (massive parallel and hybrid ones). Using the ELSHOW, calculations of such integral characteristics as the number of particles in an avalanche, the coefficient of impact ionization, the drift velocity, and the others were made. Also, special precise computations were made to select an appropriate size of the time step using the technique of dependent statistical tests. Particularly, the algorithm consists of special methods of distribution modeling, a lexicographic implementation scheme for “branching” of trajectories, justified estimation of functionals. A comparison of the obtained results for nitrogen with previously published theoretical and experimental data was made.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Optical CT imaging of solid radiochromic dosimeters in mismatched refractive index solutions using a scanning laser and large area detector

Purpose: The practical use of the PRESAGE® solid plastic dosimeter is limited by the inconvenience of immersing it in high-viscosity oils to achieve refractive index matching for optical computed tomography (CT) scanning. The oils are slow to mix and difficult to clean from surfaces, and the dosimeter rotation can generate dynamic Schlieren inhomogeneity patterns in the reference liquid, limiting the rotational and overall scan speed. Therefore, it would be beneficial if lower-viscosity, water-based solutions with slightly unmatched refractive index could be used instead. The purpose of this work is to demonstrate the feasibility of allowing mismatched conditions when using a scanning laser system with a large acceptance angle detector. A fiducial-based ray path measurement technique is combined with an iterative CT reconstruction algorithm to reconstruct images. Methods: A water based surrounding liquid with a low viscosity was selected for imaging PRESAGE® solid dosimeters. Liquid selection was optimized to achieve as high a refractive index as possible while avoiding rotation-induced Schlieren effects. This led to a refractive index mismatch of 6% between liquid and dosimeters. Optical CT scans were performed with a fan-beam scanning-laser optical CT system with a large area detector to capture most of the refracted rays. A fiducial marker placed on the wall of a cylindrical sample occludes a given light ray twice. With knowledge of the rotation angle and the radius of the cylindrical object, the actual internal path of each ray through the dosimeter can be calculated. Scans were performed with 1024 projections of 512 data samples each, and rays were rebinned to form 512 parallel-beam projections. Reconstructions were performed on a 512 × 512 grid using 100 iterations of the SIRT iterative CT algorithm. Proof of concept was demonstrated with a uniformly attenuating solution phantom. PRESAGE® dosimeters (11 cm diameter) were irradiated with Cobalt-60 irradiator to achieve either a uniform dose or a 2-level “step-dose” pattern. Results: With 6% refractive index mismatching, a circular field of view of 85% of the diameter of a cylindrical sample can be reconstructed accurately. Reconstructed images of the test solution phantom were uniform (within 3%) inside this radius. However, the dose responses of the PRESAGE® samples were not spatially uniform, with variations of at least 5% in sensitivity. The variation appears as a “cupping” artifact with less sensitivity in the middle than at the periphery of the PRESAGE® cylinder. Polarization effects were also detected for these samples. Conclusions: The fiducial-based ray path measurement scheme, coupled with an iterative reconstruction algorithm, enabled optical CT scanning of PRESAGE® dosimeters immersed in mismatched refractive index solutions. However, improvements to PRESAGE® dose response uniformity are required.

60 APPLIED LIFE SCIENCES↗

Entropic description of gas hydrate ice/liquid equilibrium via enhanced sampling of coexisting phases

Metastable β ice holds small guest molecules in stable gas hydrates, so its solid/liquid equilibrium is of interest. However, aqueous crystal/liquid transitions are very difficult to simulate. A new MD algorithm generates trajectories in a generalized NPT ensemble and equilibrates states of coexisting phases with a selectable enthalpy. Furthermore, with replicas spanning the range between β ice and liquid water we find the statistical temperature from the enthalpy histograms and characterize the transition by the entropy, introducing a general computational procedure for first-order transitions.

36 MATERIALS SCIENCE↗

Probing other solar systems with current and future adaptive optics

Over the past decade, the study of extrasolar planets through indirect techniques--primarily Doppler measurements--has revolutionized our understanding of other solar systems. The next major step in this field will be the direct detection and characterization, via imaging and spectroscopy, of the planets themselves. To achieve this, we must separate the light from the faint planet from the extensive glare of its parent star. We pursued this goal using the current generation of adaptive optics (AO) systems on large ground-based telescopes, using infrared imaging to search for the thermal emission from young planets and developing image processing techniques to distinguish planets from telescope-induced artifacts. Our new Angular Differential Imaging (ADI) technique, which uses the sidereal rotation of the Earth and telescope, is now standard for ground-based high-contrast imaging. Although no young planets were found in our surveys, we placed the strongest limits yet on giant planets in wide orbits (>30 AU) around young stars and characterized planetary companion candidates. The imaging of planetary companions on solar-system-like scales (5-30 AU) will require a new generation of advanced AO systems that are an order of magnitude more powerful than the LLNL-built Keck AO system. We worked to develop and test the key technologies needed for these systems, including a spatially-filtered wavefront sensor, efficient and accurate wavefront reconstruction algorithms, and precision AO wavefront control at the sub-nm level. LLNL has now been selected by the Gemini Observatory to lead the construction of the Gemini Planet Imager, a $24M instrument that will be the most advanced AO system in the world.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗