Search NASA⌕ Search

SEARCH · Search NASA

Results for “Vectorized algorithm”

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 1,027 records · Page 57

Overview of Algorithms for Using Particle Morphology in Pre-Detonation Nuclear Forensics

A major goal in pre-detonation nuclear forensics is to infer the processing conditions and/or facility type that produced radiological material. This review paper focuses on analyses of particle size, shape, texture (“morphology”) signatures that could provide information on the provenance of interdicted materials. For example, uranium ore concentrates (UOC or yellowcake) include ammonium diuranate (ADU), ammonium uranyl carbonate (AUC), sodium diuranate (SDU), magnesium diuranate (MDU), and others, each prepared using different salts to precipitate U from solution. Once precipitated, UOCs are often dried and calcined to remove adsorbed water. The products can be allowed to react further, forming uranium oxides UO3, U3O8, or UO2 powders, whose surface morphology can be indicative of precipitation and/or calcination conditions used in their production. This review paper describes statistical issues and approaches in using quantitative analyses of measurements such as particle size and shape to infer production conditions. Statistical topics include multivariate t tests (Hotelling’s T 2 ), design of experiments, and several machine learning (ML) options including decision trees, learning vector quantization neural networks, mixture discriminant analysis, and approximate Bayesian computation (ABC). ABC is emphasized as an attractive option to include the effects of model uncertainty in the selected and fitted forward model used for inferring processing conditions.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

STAR-GALAXY CLASSIFICATION IN MULTI-BAND OPTICAL IMAGING

Ground-based optical surveys such as PanSTARRS, DES, and LSST will produce large catalogs to limiting magnitudes of r {approx}> 24. Star-galaxy separation poses a major challenge to such surveys because galaxies-even very compact galaxies-outnumber halo stars at these depths. We investigate photometric classification techniques on stars and galaxies with intrinsic FWHM <0.2 arcsec. We consider unsupervised spectral energy distribution template fitting and supervised, data-driven support vector machines (SVMs). For template fitting, we use a maximum likelihood (ML) method and a new hierarchical Bayesian (HB) method, which learns the prior distribution of template probabilities from the data. SVM requires training data to classify unknown sources; ML and HB do not. We consider (1) a best-case scenario (SVM{sub best}) where the training data are (unrealistically) a random sampling of the data in both signal-to-noise and demographics and (2) a more realistic scenario where training is done on higher signal-to-noise data (SVM{sub real}) at brighter apparent magnitudes. Testing with COSMOS ugriz data, we find that HB outperforms ML, delivering {approx}80% completeness, with purity of {approx}60%-90% for both stars and galaxies. We find that no algorithm delivers perfect performance and that studies of metal-poor main-sequence turnoff stars may be challenged by poor star-galaxy separation. Using the Receiver Operating Characteristic curve, we find a best-to-worst ranking of SVM{sub best}, HB, ML, and SVM{sub real}. We conclude, therefore, that a well-trained SVM will outperform template-fitting methods. However, a normally trained SVM performs worse. Thus, HB template fitting may prove to be the optimal classification method in future surveys.

79 ASTRONOMY AND ASTROPHYSICS↗

Turbulent flow and vortex characteristics in a blocked subchannel of a helically wrapped rod bundle

The study of flow and heat transfer through a bundle of rods is important, as this type of configuration is widely used in many engineering applications, for example, in heat exchangers, clusters of support structures, steam generators, and nuclear power reactors. This work experimentally investigated the flow-field characteristics of rod bundles with helically wrapped wires (wire-wrapped rod assemblies) proposed for advanced burner reactors. Subchannel blockage is one of the most interesting accident scenarios for liquid metal fast reactors as different mechanisms can lead to a partial or total subchannel blockage. It is of paramount important to understand the thermal-hydraulic behavior of the flow in the vicinity of such blockages. The flow mixing characteristics in two interior subchannels located near the core of the bundle, one without and one with the presence of a blockage, were experimentally investigated. The velocity fields within the spatial gaps of two subchannels were obtained by applying the matched-index-of-refraction and time-resolved particle image velocimetry (TR-PIV) technique for Reynolds numbers of 4000 and 17,000. The first- and second-order statistics, including the mean velocity, root-mean-square fluctuating velocity, and Reynolds stress, were computed from the obtained TR-PIV velocity vector fields. In the blocked sub-channel, the TR-PIV results revealed a large recirculation flow region in the vicinity of the blockage that was created by the interaction between inflows from neighboring subchannels. Spectral analysis was performed and the vortex shedding frequency indicated by the Strouhal number was found to be St = 0.16 for the two Reynolds numbers investigated. The flow characteristics of the interior subchannels were investigated via the two-point cross-correlation of fluctuating velocities. In addition, proper orthogonal decomposition (POD) analysis was applied to the instantaneous vorticity fields to extract the statistically dominant flow structures. Lastly, the statistical characteristics of the vortex, such as the vortex populations, spatial distributions, and vortex strengths, were acquired by combining POD analysis and the vortex identification algorithm.

42 ENGINEERING↗

Simultaneously improving accuracy and computational cost under parametric constraints in materials property prediction tasks

Abstract Modern data mining techniques using machine learning (ML) and deep learning (DL) algorithms have been shown to excel in the regression-based task of materials property prediction using various materials representations. In an attempt to improve the predictive performance of the deep neural network model, researchers have tried to add more layers as well as develop new architectural components to create sophisticated and deep neural network models that can aid in the training process and improve the predictive ability of the final model. However, usually, these modifications require a lot of computational resources, thereby further increasing the already large model training time, which is often not feasible, thereby limiting usage for most researchers. In this paper, we study and propose a deep neural network framework for regression-based problems comprising of fully connected layers that can work with any numerical vector-based materials representations as model input. We present a novel deep regression neural network, iBRNet, with branched skip connections and multiple schedulers, which can reduce the number of parameters used to construct the model, improve the accuracy, and decrease the training time of the predictive model. We perform the model training using composition-based numerical vectors representing the elemental fractions of the respective materials and compare their performance against other traditional ML and several known DL architectures. Using multiple datasets with varying data sizes for training and testing, We show that the proposed iBRNet models outperform the state-of-the-art ML and DL models for all data sizes. We also show that the branched structure and usage of multiple schedulers lead to fewer parameters and faster model training time with better convergence than other neural networks. Scientific contribution: The combination of multiple callback functions in deep neural networks minimizes training time and maximizes accuracy in a controlled computational environment with parametric constraints for the task of materials property prediction.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A Low-Rank QTT-based Finite Element Method for Elasticity Problems

We present an efficient and robust numerical algorithm for solving the linear elasticity problem that combines the Quantized Tensor Train format and a domain partitioning strategy. This approach makes it possible to solve the linear elasticity problem on a computational domain that is more general than a square. By integrating Z-ordering and subdomain concatenation, our method substantially decreases memory usage and achieves a notable reduction in rank compared to established Finite Element implementations like the FEniCS platform. This efficiency is maintained while still guaranteeing exponential convergence with respect to the number of degrees of freedom. This performance gain, however, requires a fundamental rethinking of how core finite element operations are implemented. This includes changes to mesh discretization, node and degree of freedom ordering, stiffness matrix and internal nodal force assembly, and the execution of algebraic matrix-vector operations. In this work, we discuss all these aspects in detail and assess the method’s performance in the numerical approximation of three representative test cases.

97 MATHEMATICS AND COMPUTING↗

Machine learning for design principles for single atom catalysts towards electrochemical reactions

Machine learning (ML) integrated density functional theory (DFT) calculations have recently been used to accelerate the design and discovery of heterogeneous catalysts such as single atom catalysts (SACs) through the establishment of deep structure–activity relationships. Here, this review provides recent progress in the ML-aided rational design of heterogeneous catalysts with the focus on SACs in terms of structure–activity relationships, feature importance analysis, high-throughput screening, stability, and metal–support interactions for electrochemistry. Support vector machine (SVM), random forest regression (RFR), and deep neural networks (DNN) along with atomic properties are mainly used for the design of SACs. The ML results have shown that the number of electrons in the d orbital, oxide formation enthalpy, ionization energy, Bader charge, d-band center, and enthalpy of vaporization are mainly the most important parameters for the defining of the structure–activity relationships for electrochemistry. However, the black-box nature of ML techniques occasionally makes a physical interpretation of descriptors, such as the Bader charge, d-band center, and enthalpy of vaporization, non-trivial. At the current stage, ML application is limited by the lack of a large and high-quality database. Future prospects for the development of a large database and a generalized ML algorithm for SAC design are discussed to give insights for further studies in this field.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Benchmarking Current and Emerging Approaches to Infrasound Signal Classification

Low frequency sound ≤ 20 Hz, known as infrasound, is generated by a variety of natural and anthropogenic sources. Following an event, infrasonic waves travel through a dynamic atmosphere that can change on the order of minutes. This makes infrasound event classification a difficult problem as waveforms from the same source type can look drastically different. Event classification usually requires ground truth information from seismic or other methods. This is time consuming, inefficient, and does not allow for a classification if the event locates somewhere other than a known source, the location accuracy is poor, or ground truth from seismic data is lacking. Here we compare the performance of the state of the art for infrasound event classification, support vector machine (SVM), to the performance of a convolutional neural network (CNN), a method that has been proven in tangential fields such as seismology. For a 1-class catalog consisting of only volcanic activity and earthquake events, the 4-fold average SVM classification accuracy is 75%, while it is 74% when using a CNN. Classification accuracies from the 4-class catalog consisting of the most common infrasound events detected at the global scale are 55% and 56% for the SVM and CNN architectures, respectively. These results demonstrate that using a CNN does not increase performance for infrasound event classification. This suggests that SVM should be the preferred classification method as it is a simpler and more trustworthy architecture and can be tied to the physical properties of the waveforms. The SVM and CNN algorithms described in this paper are not yet generalizable to other infrasound event catalogs. We anticipate this study to be a starting point for the development of large and comprehensive, systematically labeled, infrasound event catalogs as such catalogs will be necessary to provide an increase in the value of deep learning on event classification.

47 OTHER INSTRUMENTATION↗

Synchronized multiartifact reduction with tomographic reconstruction (SMART-RECON): A statistical model based iterative image reconstruction method to eliminate limited-view artifacts and to mitigate the temporal-average artifacts in time-resolved CT

Purpose: In x-ray computed tomography (CT), a violation of the Tuy data sufficiency condition leads to limited-view artifacts. In some applications, it is desirable to use data corresponding to a narrow temporal window to reconstruct images with reduced temporal-average artifacts. However, the need to reduce temporal-average artifacts in practice may result in a violation of the Tuy condition and thus undesirable limited-view artifacts. In this paper, the authors present a new iterative reconstruction method, synchronized multiartifact reduction with tomographic reconstruction (SMART-RECON), to eliminate limited-view artifacts using data acquired within an ultranarrow temporal window that severely violates the Tuy condition. Methods: In time-resolved contrast enhanced CT acquisitions, image contrast dynamically changes during data acquisition. Each image reconstructed from data acquired in a given temporal window represents one time frame and can be denoted as an image vector. Conventionally, each individual time frame is reconstructed independently. In this paper, all image frames are grouped into a spatial–temporal image matrix and are reconstructed together. Rather than the spatial and/or temporal smoothing regularizers commonly used in iterative image reconstruction, the nuclear norm of the spatial–temporal image matrix is used in SMART-RECON to regularize the reconstruction of all image time frames. This regularizer exploits the low-dimensional structure of the spatial–temporal image matrix to mitigate limited-view artifacts when an ultranarrow temporal window is desired in some applications to reduce temporal-average artifacts. Both numerical simulations in two dimensional image slices with known ground truth and in vivo human subject data acquired in a contrast enhanced cone beam CT exam have been used to validate the proposed SMART-RECON algorithm and to demonstrate the initial performance of the algorithm. Reconstruction errors and temporal fidelity of the reconstructed images were quantified using the relative root mean square error (rRMSE) and the universal quality index (UQI) in numerical simulations. The performance of the SMART-RECON algorithm was compared with that of the prior image constrained compressed sensing (PICCS) reconstruction quantitatively in simulations and qualitatively in human subject exam. Results: In numerical simulations, the 240{sup ∘} short scan angular span was divided into four consecutive 60{sup ∘} angular subsectors. SMART-RECON enables four high temporal fidelity images without limited-view artifacts. The average rRMSE is 16% and UQIs are 0.96 and 0.95 for the two local regions of interest, respectively. In contrast, the corresponding average rRMSE and UQIs are 25%, 0.78, and 0.81, respectively, for the PICCS reconstruction. Note that only one filtered backprojection image can be reconstructed from the same data set with an average rRMSE and UQIs are 45%, 0.71, and 0.79, respectively, to benchmark reconstruction accuracies. For in vivo contrast enhanced cone beam CT data acquired from a short scan angular span of 200{sup ∘}, three 66{sup ∘} angular subsectors were used in SMART-RECON. The results demonstrated clear contrast difference in three SMART-RECON reconstructed image volumes without limited-view artifacts. In contrast, for the same angular sectors, PICCS cannot reconstruct images without limited-view artifacts and with clear contrast difference in three reconstructed image volumes. Conclusions: In time-resolved CT, the proposed SMART-RECON method provides a new method to eliminate limited-view artifacts using data acquired in an ultranarrow temporal window, which corresponds to approximately 60{sup ∘} angular subsectors.

60 APPLIED LIFE SCIENCES↗

Nonlinear models for coupling the effects of stimulated Raman scattering to inertial confinement fusion codes

Laser plasma instabilities (LPI) reduce driver-target coupling, alter implosion symmetry, and therefore can fundamentally limit fusion performance in inertial confinement fusion (ICF). Developing a predictive modeling capability for LPI effects can critically advance the success of the field. We perform vector particle-in-cell simulations of multi-speckled laser beams undergoing stimulated Raman scattering (SRS) at various densities and intensities relevant to mainly indirectly driven and a subset of parameter space for directly driven ICF systems, focusing on the regimes with intensities above the onset of electron trapping. Based on the wavenumber of the SRS daughter electron plasma wave, we identify several regions with underpinning SRS saturation physics: the electron-trapping dominated region with intermediate kλD values, the strong Landau damping region at larger kλD values, and the region where the Langmuir decay instability arises at lower kλD values. We develop a nonlinear SRS reflectivity model that features the base trapping-dominated scaling of (kλD)−4 and its modifications. Electron trapping deforms the initialized electron distribution functions, and we have developed a new δf-Gaussian-mixture algorithm for an accurate characterization of the trapped hot electron population. With this SRS hot electron description, we construct a nonlinear energy deposition model and a hot electron source model—based on a modified Manley–Rowe relation—suitable for including SRS effects as a sub-grid module in a high-fidelity ICF design code.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

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↗

Search for vector-like T quarks decaying to top quarks and Higgs bosons in the all-hadronic channel using jet substructure

We performed a search for a vector-like heavy T quark that is produced in pairs and that decays to a top quark and a Higgs boson. The data analysed correspond to an integrated luminosity of 19.7 fb -1 collected with the CMS detector in proton-proton collisions at √s=8 TeV. For T quarks with large mass values the top quarks and Higgs bosons can have significant Lorentz boosts, so that their individual decay products often overlap and merge. Methods are applied to resolve the substructure of such merged jets. We also derived upper limits on the production cross section of a T quark with mass between 500 and 1000 GeV/c 2 . If the T quark decays exclusively to tH, the observed (expected) lower limit on the mass of the T quark is 745 (773) GeV/c 2 at 95% confidence level. For the first time an algorithm is used for tagging boosted Higgs bosons that is based on a combination of jet substructure information and b tagging.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Stage-parallel fully implicit Runge–Kutta solvers for discontinuous Galerkin fluid simulations

In this paper, we develop new techniques for solving the large, coupled linear systems that arise from fully implicit Runge–Kutta methods. This method makes use of the iterative preconditioned GMRES algorithm for solving the linear systems, which has seen success for fluid flow problems and discontinuous Galerkin discretizations. By transforming the resulting linear system of equations, one can obtain a method which is much less computationally expensive than the untransformed formulation, and which compares competitively with other time-integration schemes, such as diagonally implicit Runge–Kutta (DIRK) methods. We develop and test several ILU-based preconditioners effective for these large systems. We additionally employ a parallel-in-time strategy to compute the Runge–Kutta stages simultaneously. Numerical experiments are performed on the Navier–Stokes equations using Euler vortex and 2D and 3D NACA airfoil test cases in serial and in parallel settings. In conclusion, the fully implicit Radau IIA Runge–Kutta methods compare favorably with equal-order DIRK methods in terms of accuracy, number of GMRES iterations, number of matrix–vector multiplications, and wall-clock time, for a wide range of time steps.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Large-scale 3D geoelectromagnetic modeling using parallel adaptive high-order finite element method

Here, we have investigated the use of the adaptive high-order finite-element method (FEM) for geoelectromagnetic modeling. Because high-order FEM is challenging from the numerical and computational points of view, most published finite-element studies in geoelectromagnetics use the lowest order formulation. Solution of the resulting large system of linear equations poses the main practical challenge. We have developed a fully parallel and distributed robust and scalable linear solver based on the optimal block-diagonal and auxiliary space preconditioners. The solver was found to be efficient for high finite element orders, unstructured and nonconforming locally refined meshes, a wide range of frequencies, large conductivity contrasts, and number of degrees of freedom (DoFs). Furthermore, the presented linear solver is in essence algebraic; i.e., it acts on the matrix-vector level and thus requires no information about the discretization, boundary conditions, or physical source used, making it readily efficient for a wide range of electromagnetic modeling problems. To get accurate solutions at reduced computational cost, we have also implemented goal-oriented adaptive mesh refinement. The numerical tests indicated that if highly accurate modeling results were required, the high-order FEM in combination with the goal-oriented local mesh refinement required less computational time and DoFs than the lowest order adaptive FEM.

58 GEOSCIENCES↗

Online adaptive basis refinement and compression for reduced-order models via vector-space sieving

In many applications, projection-based reduced-order models (ROMs) have demonstrated the ability to provide rapid approximate solutions to high-fidelity full-order models (FOMs). However, there is no a priori assurance that these approximate solutions are accurate; their accuracy depends on the ability of the low-dimensional trial basis to represent the FOM solution. As a result, ROMs can generate inaccurate approximate solutions, e.g., when the FOM solution at the online prediction point is not well represented by training data used to construct the trial basis. To address this fundamental deficiency of standard model-reduction approaches, this work proposes a novel online-adaptive mechanism for efficiently enriching the trial basis in a manner that ensures convergence of the ROM to the FOM, yet does not incur any FOM solves. The mechanism is based on the previously proposed adaptive $h$-refinement method for ROMs, but improves upon this work in two crucial ways. First, the proposed method enables basis refinement with respect to any orthogonal basis (not just the Kronecker basis), thereby generalizing the refinement mechanism and enabling it to be tailored to the physics characterizing the problem at hand. Second, the proposed method provides a fast online algorithm for periodically compressing the enriched basis via an efficient proper orthogonal decomposition (POD) method, which does not incur any operations that scale with the FOM dimension. We report these two features allow the proposed method to serve as (1) a failsafe mechanism for ROMs, as the method enables the ROM to satisfy any prescribed error tolerance online (even in the case of inadequate training), and (2) an efficient online basis-adaptation mechanism, as the combination of basis enrichment and compression enables the basis to adapt online while controlling its dimension.

42 ENGINEERING↗

In vivo 3D modeling of the femoropopliteal artery in human subjects based on x-ray angiography: Methodology and validation

Endovascular revascularization of the femoropopliteal (FP) artery has been limited by high rates of restenosis and stent fracture. The unique physical forces that are applied to the FP artery during leg movement have been implicated in these phenomena. The foundation for measuring the effects of physical forces on the FP artery in a clinically relevant environment is based on the ability to develop 3D models of this vessel in different leg positions in vivo in patients with peripheral arterial disease (PAD). By acquiring paired angiographic images of the FP artery, and using angiography-based 3D modeling algorithms previously validated in the coronary arteries, the authors generated 3D models of ten FP arteries in nine patients with PAD with the lower extremity in straight leg (SL) and crossed leg (CL) positions. Due to the length of the FP artery, overlapping paired angiographic images of the entire FP artery were required to image the entire vessel, which necessitated the development of a novel fusion process in order to generate a 3D model of the entire FP artery. The methodology of angiographic acquisition and 3D model generation of the FP artery is described. In a subset of patients, a third angiographic view (i.e., validation view) was acquired in addition to the standard paired views for the purpose of validating the 3D modeling process. The mean root-mean-square (rms) error of the point-to-point distances between the centerline of the main FP artery from the 2D validation view and the centerline from the 3D model placed in the validation view for the SL and CL positions were 0.93{+-}0.19 mm and 1.12{+-}0.25 mm, respectively. Similarly, the mean rms error of the same comparison for the main FP artery and sidebranches for the SL and CL positions were 1.09{+-}0.38 mm and 1.21{+-}0.25 mm, respectively. A separate validation of the novel fusion process was performed by comparing the 3D model of the FP artery derived from fusion of 3D models of adjacent FP segments with the 2D validation view incorporating the region of fusion. The mean rms error of vessel centerline points of the main FP artery, the main FP artery plus directly connected sidebranches, and the mean rms error of upstream, downstream, and sidebranch directional vectors at bifurcation points in the overlap region were 1.41{+-}0.79 mm, 2.13{+-}1.12 mm, 3.16{+-}3.72 deg., 3.60{+-}5.39 deg., and 8.68{+-}8.42 deg. in the SL position, respectively, and 1.29{+-}0.35 mm, 1.61{+-}0.78 mm, 4.68{+-}4.08 deg., 3.41{+-}2.23 deg., and 5.52{+-}4.41 deg. in the CL position, respectively. Inter- and intraobserver variability in the generation of 3D models of individual FP segments and the fusion of overlapping FP segments were assessed. The mean rms errors between the centerlines of nine 3D models of individual FP segments generated by two independent observers, and repeated measurement by the same observer were 2.78{+-}1.26 mm and 3.50{+-}1.15 mm, respectively. The mean rms errors between the centerline of four 3D models of fused overlapping FP segments generated by two independent observers, and repeated measurement by the same observer were 4.99{+-}0.99 mm and 5.98{+-}1.22 mm, respectively. This study documents the ability to generate 3D models of the entire FP artery in vivo in patients with PAD in both SL and CL positions using routine angiography, and validates the methodologies used.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

SU-G-BRC-15: The Potential Clinical Significance of Dose Mapping Error for Intra- Fraction Dose Mapping for Lung Cancer Patients

Purpose: To evaluate the dose-mapping error (DME) inherent to conventional dose-mapping algorithms as a function of dose-matrix resolution. Methods: As DME has been reported to be greatest where dose-gradients overlap tissue-density gradients, non-clinical 66 Gy IMRT plans were generated for 11 lung patients with the target edge defined as the maximum 3D density gradient on the 0% (end of inhale) breathing phase. Post-optimization, Beams were copied to 9 breathing phases. Monte Carlo dose computed (with 2*2*2 mm{sup 3} resolution) on all 10 breathing phases was deformably mapped to phase 0% using the Monte Carlo energy-transfer method with congruent mass-mapping (EMCM); an externally implemented tri-linear interpolation method with voxel sub-division; Pinnacle’s internal (tri-linear) method; and a post-processing energy-mass voxel-warping method (dTransform). All methods used the same base displacement-vector-field (or it’s pseudo-inverse as appropriate) for the dose mapping. Mapping was also performed at 4*4*4 mm{sup 3} by merging adjacent dose voxels. Results: Using EMCM as the reference standard, no clinically significant (>1 Gy) DMEs were found for the mean lung dose (MLD), lung V20Gy, or esophagus dose-volume indices, although MLD and V20Gy were statistically different (2*2*2 mm{sup 3}). Pinnacle-to-EMCM target D98% DMEs of 4.4 and 1.2 Gy were observed ( 2*2*2 mm{sup 3}). However dTransform, which like EMCM conserves integral dose, had DME >1 Gy for one case. The root mean square RMS of the DME for the tri-linear-to- EMCM methods was lower for the smaller voxel volume for the tumor 4D-D98%, lung V20Gy, and cord D1%. Conclusion: When tissue gradients overlap with dose gradients, organs-at-risk DME was statistically significant but not clinically significant. Target-D98%-DME was deemed clinically significant for 2/11 patients (2*2*2 mm{sup 3}). Since tri-linear RMS-DME between EMCM and tri-linear was reduced at 2*2*2 mm{sup 3}, use of this resolution is recommended for dose mapping. Interpolative dose methods are sufficiently accurate for the majority of cases. J.V. Siebers receives funding support from Varian Medical Systems.

60 APPLIED LIFE SCIENCES↗

Forecasting district-scale energy dynamics through integrating building network and long short-term memory learning algorithm

With the development of data-driven techniques, building energy prediction on a district level has attracted increasing attention in recent years for revealing energy use patterns and reduction potentials. However, data acquisition in large building groups is difficult and adjacent buildings also interact with each other. To reduce data cost and incorporate the inter-building impact with the data-driven building energy model, this study proposes a deep learning predictive approach that fuses the building network model with a long short-term memory learning model for district-scale building energy modeling. The building network was constructed based on correlations between the energy use intensity of buildings, which can significantly reduce the computational complexity of the deep learning models for energy dynamic prediction. Five typical building groups with energy use data from 2015 to 2018 on two institutional campuses were selected to perform the validation experiment with TensorFlow. Based on the prediction error assessments, the results suggest that for total building energy use intensity prediction, the proposed model can achieve a mean absolute percentage error of 6.66% and a root mean square error of 0.36 kWh/m 2 , compared to 12.05% and 0.63 kWh/m 2 of the conventional artificial neural network model and to 11.06% and 0.89 kWh/m 2 for the support vector regression model.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Verification and validation of developed short-term forecasting models

Recent advancements in machine learning (ML) and artificial intelligence (AI) technologies provide an opportunity for leveraging data-driven algorithms to predict future nuclear power plant (NPP) operating conditions by using recorded plant process data. Successfully implementing these models can lead to cost-reducing, conditioned-based predictive maintenance through optimized maintenance schedules and a reduction of unnecessary maintenance activities. This report discusses the verification and validation of short-term forecasting processes (i.e., data cleaning, feature selection, model optimization, and forecasting) developed in previous reports. The verification and validation (V&V) process demonstrates the expected precision and accuracy when the ML model encounters new datasets from different systems. Shapley additive explanations were used as the primary means of feature selection across these different data set. Individual models were trained for each data set, then validated through a cross-validation procedure. In this report, two different ML models were tasked to predict variables from three different plant process data sets with varying prediction horizons. The results indicate that support vector regression (SVR) outperformed long short-term memory (LSTM) neural networks in regard to each data set and each prediction horizon in this study, but further tuning and optimization could improve long short-term memory results. However, each forecasting model showed reduced performance as the prediction horizon was extended from 1 hour to 1 day ahead. Research is ongoing to evaluate the optimal input variable space, which is based on a given set of process parameters, to further improve forecasting accuracy.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗