Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “adaptive learning”

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 1,135 records · Page 63

Electronic structure theory on modeling short-range noncovalent interactions between amino acids

While short-range noncovalent interactions (NCIs) are proving to be of importance in many chemical and biological systems, these atypical bindings happen within the so-called van der Waals envelope and pose an enormous challenge for current computational methods. We introduce SNCIAA, a database of 723 benchmark interaction energies of short-range noncovalent interactions between neutral/charged amino acids originated from protein x-ray crystal structures at the “gold standard” coupled-cluster with singles, doubles, and perturbative triples/complete basis set [CCSD(T)/CBS] level of theory with a mean absolute binding uncertainty less than 0.1 kcal/mol. Subsequently, a systematic assessment of commonly used computational methods, such as the second-order Møller−Plesset theory (MP2), density functional theory (DFT), symmetry-adapted perturbation theory (SAPT), composite electronic-structure methods, semiempirical approaches, and the physical-based potentials with machine learning (IPML) on SNCIAA is carried out. It is shown that the inclusion of dispersion corrections is essential even though these dimers are dominated by electrostatics, such as hydrogen bondings and salt bridges. Overall, MP2, ωB97M-V, and B3LYP+D4 turned out to be the most reliable methods for the description of short-range NCIs even in strongly attractive/repulsive complexes. SAPT is also recommended in describing short-range NCIs only if the δMP2 correction has been included. The good performance of IPML for dimers at close-equilibrium and long-range conditions is not transferable to the short-range. We expect that SNCIAA will assist the development/improvement/validation of computational methods, such as DFT, force-fields, and ML models, in describing NCIs across entire potential energy surfaces (short-, intermediate-, and long-range NCIs) on the same footing.

Chemistry↗

MO-FG-213-04: Managing the Relationship

Academic-industrial collaborations are playing an increasingly prominent role in medical physics research. Successful partnerships have enabled truly translational, benchtop-to-bedside research that has significantly and positively impacted our field. Given the current flat or depleting trends in federal funding, and the increasing emphasis in the medical physics community on demonstrating translatability, efficacy, and safety prior to widespread clinical use, it is anticipated that such partnerships will become ever more important in advancing our field. In the future, it is likely that neither academia alone, nor industry alone, will be able to bring significant innovation to successfully impact patient care in radiotherapy. Such innovation will require partnership and symbiosis through the lifetime of the project. This symposium is aimed at professionals from academia and the industry who are interested in either initiating new collaborations or maintaining and improving existing partnerships. We will hear from four thought leaders from industry and academia on critical aspects of building and maintaining successful academic-industrial research partnerships. The following topics will be addressed: 1. Motivations: The industry perspective [Kevin Brown, Elekta] What goes into the development of a company’s R&D roadmap What are the motivations and expectations of the industry from academic partners How can academics make a successful pitch by identifying projects aligned with a company’s roadmap Why are some relationships successful and some not Milestones and progress reports - Why do we need them 2. Motivations: The academic perspective [T. Rock Mackie, Univ. of Madison, WI; Tomotherapy] What are the motivations and expectations of academic partners from the industry What are the cultural differences in academia vs industry on sharing information, exchanging ideas Partnership or entrepreneurship? When does each option make sense 3. Industrial Grants and Research Contracts – a Primer [Norbert Pelc; Stanford University] What topics are suited to industrial grants vs. academic grants How do resources (personnel effort, tuition) get allocated and charged in academia What are reasonable resource and timeline requests for small and big projects; from small and big companies “Fair market value” considerations Sections of a research agreement (IP including software confidential information, publications, data, exclusivity, support) 4. Managing the relationship [John Wong, Johns Hopkins University] How to effectively collaborate with each other How to adapt to changes in the relationship How to renew funding with the industry Moderated Panel Discussion [All speakers + moderators] Learning Objectives: 1. Understand the motivations of each partner 2. Elements that contribute to a successful partnership 3. Avoiding pitfalls - Challenges on either side that may derail a project 4. How to frame a research grant and agreement 5. How to sustain the collaborative relationship Amit Sawant: Funding from: NIH. Varian, Elekta and VisionRT; Michelle Svatos: Employee and shareholder: Varian, Board Member, TopasMC; Kevin Brown: Employee, Elekta; T. Rock Mackie: Consultant, Accuray; Norbert Pelc: Research support: Philips Healthcare, GE Healthcare; John Wong: Funding from NIH, Xstrahl and Elekta. Co-founder and advisor for JPLC Associates, LLC.

07 ISOTOPE AND RADIATION SOURCES↗

A Robotics Enabled Eddy Current Testing System for Autonomous Inspection of Heat Exchanger Tubes

The objective of the project is to develop a robotics enabled eddy current testing system (REECTS) in automatic probe deployment, inspection, and data acquisition and analysis. The main functions of the REECTS are to: 1) identify geometry and locations of heat exchange tubes with assistance of an imaging recognition system; 2) precisely control the position and motion speed of ECT probes by an adaptive control system; 3) facilitate data analysis and real-time decision making for autonomous inspection assisted by machine learning algorithms.

20 FOSSIL-FUELED POWER PLANTS↗

Alternating prism exposure causes dual adaptation and generalization to a novel displacement

In two experiments, we examined the hypothesis that repeatedly adapting and readapting to two mutually conflicting sensory environments fosters the development of a separate adaptation to each situation (dual adaptation) as well as an increased ability to adapt to a novel displacement (adaptive generalization). In the preliminary study, subjects alternated between adapting their visuomotor coordination to 30-diopter prismatic displacement and readapting to normal vision. Dual adaptation was observed by the end of 10 alternation cycles. However, an unconfounded test of adaptive generalization was prevented by an unexpected prism-adaptive shift in preexposure baselines for the dual-adapted subjects. In the primary experiment, the subjects adapted and readapted to opposite 15-diopter displacements for a total of 12 cycles. Both dual adaptation and adaptive generalization to a 30-diopter displacement were obtained. These findings may be understood in terms of serial reversal learning and 'learning to learn'.

Welch, Robert B.↗

Earth System Grid Federation (ESGF) Compute Challenge

This Open Geospatial Consortium (OGC) Engineering Report (ER) describes the advancement of an Execution Management System (EMS) to support Web Processing Service (WPS) climate processes deployed on the Earth System Grid Federation (ESGF). The report introduces climate data, processes and applications into Common Workflow Language (CWL) workflows with the intent of advancing: application packaging, deployment and execution in clouds; interoperability of services in federated cyberinfrastructures; and geospatial workflows towards standardization. Work presented in this report is a direct continuation of the Earth Observation & Clouds (EOC) thread of Testbed-14. This report is expected to be of relevance to Testbed-15, both to the Earth Observation Process and Application Discovery (EOPAD) task and the Machine Learning task. This engineering report will describe: relevant work conducted in OGC Testbed-14; ESGF and its compute challenge; adaptations of existing climate processes into workflows; interoperability experiments with ESGF endpoints conforming to a common API.

54 ENVIRONMENTAL SCIENCES↗

Journal of Gravitational Physiology, Volume 12, Number 1

The following topics were covered: System Specificity in Responsiveness to Intermittent -Gx Gravitation during Simulated Microgravity in Rats; A Brief Overview of Animal Hypergravity Studies; Neurovestibular Adaptation to Short Radius Centrifugation; Effect of Artificial Gravity with Exercise Load by Using Short-Arm Centrifuge with Bicycle Ergometer as a Countermeasure Against Disused Osteoporosis; Perception of Body Vertical in Microgravity during Parabolic Flight; Virtual Environment a Behavioral and Countermeasure Tool for Assisted Gesture in Weightlessness: Experiments during Parabolic Flight; Artificial Gravity: Physiological Perspectives for Long-Term Space Exploration; Comparison of the Effects of DL-threo-Beta-Benzyloxyaspartate on the Glutamate Release from Synaptosomes before and after Exposure of Rats to Artificial Gravity; Do Perception and Postrotatory Vestibulo-Ocular Reflex Share the Same Gravity Reference?; Vestibular Adaptation to Changing Gravity Levels and the Orientation of Listing's Plane; Compound Mechanism Hypothesis on +Gz - Induced Brain Injury and Dysfunction of Learning and Memory; Environmental Challenge Impairs Prefrontal Brain Functions; Effect of 6-Days of Support Withdrawal on Characteristics of Balance Function; Hypergravity-Induced Changes of Neuronal Activities in CA1 Region of Rat Hippocampus; Audiological Findings in Antiorthostatic Position Modelling Microgravitation; Investigating Human Cognitive Performance during Spaceflight; The Relevance of the Minimization of Torque Exchange with the Environment in Weightlessness is Confirmed by Asimulation Study; Characteristics of the Eyes Pursuit Function during Readaptation to Terrestrial Gravity after Prolonged Flights Aboard the International Space Station; Comparison of Cognitive Performance Tests for Promethazine Pharmacodynamics in Human Subjects; Structural Reappraisal of Dendritic Tree of Cerebellar Purkinje Cell for Novel Functional Modeling of Elementary Sensorimotor Adaptive Processes; Orpheus 0 G or Ear in Microgravity to Establish Symptoms Concomittant of Inner and Middle Ear and Osteoporosis in Microgravity; Understanding Visual Perception in the Perspective of Gravity; Cortical Regions Associated with Orthostatic Stress in Conscious Humans; Restoration of Central Blood Volume: Application of a Simple Concept and Simple Device to Counteract Cardiovascular Instability in Syncope and Hemorrhage; WISE-2005: Integrative Cardiovascular Responses with LBNP during 60-Day Bed Rest in Women; Intracranial Pressure Increases during Weightlessness. A Parabolic Flights Study; Lower Limb & Portal Veins Echography for Predicting Risk of Thrombosis during a 90-D Bed Rest; Calf Tissue Liquid Stowage and Muscular and Deep Vein Distension in Orthostatic Tests after a 90-Day Head Down Bed Rest; Morphology of Brain Vessels in the Tail Suspended Rats Exposed to Intermittent 2 G; Alterations in Vasoreactivity of Femoral Artery Induced by Hindlimb Unweighting are Related to the Changes of Contractile Protein in Rats; and Respiratory Sinus Arrhythmia: A Marker of Decreased Parasympathetic Modulation after Short Duration.

Fuller, Charles A.↗

Enabling coordinated, distributed development of scientific software: A research agenda for adapting a team of teams approach

The “team of teams” concept is a strategy for coordinating activities within and across organizations, developed by General Stanley McChrystal and his colleagues based on their experiences leading the Joint Special Operations Task Force in Iraq from 2003-2008. These experiences, and the lessons learned from them, are recounted in the books Team of Teams (McChrystal 2015) and One Mission (Fussell 2017) (the authors and contributors to these books will hereafter be jointly referred to as McChrystal et al.). Essentially, their story is one of rapidly adapting a rigid, siloed organization to defeat the flexible, resilient, and fast-moving network of Al Qaeda in Iraq. As described in the books, when McChrystal took over the Task Force, he found a collection of highly competent, efficient teams of analysts and warfighters which nonetheless had difficulty operating in a unified manner due to close internal ties, distrust of outsiders, bureaucratic stove pipes, and cultural differences stemming from the diverse organizations Task Force teams were affiliated with, including Army Special Forces, CIA, and NSA.

97 MATHEMATICS AND COMPUTING↗

RLScheduler: An Automated HPC Batch Job Scheduler Using Reinforcement Learning

Today’s high-performance computing (HPC) platforms are still dominated by batch jobs. Accordingly, effective batch job scheduling is crucial to obtain high system efficiency. Existing HPC batch job schedulers typically leverage heuristic priority functions to prioritize and schedule jobs. But, once configured and deployed by the experts, such priority functions can hardly adapt to the changes of job loads, optimization goals, or system settings, potentially leading to degraded system efficiency when changes occur. To address this fundamental issue, we present RLScheduler, an automated HPC batch job scheduler built on reinforcement learning. RLScheduler relies on minimal manual interventions or expert knowledge, but can learn high-quality scheduling policies via its own continuous ‘trial and error’. We introduce a new kernel-based neural network structure and trajectory filtering mechanism in RLScheduler to improve and stabilize the learning process. Through extensive evaluations, we confirm that RLScheduler can learn high-quality scheduling policies towards various workloads and various optimization goals with relatively low computation cost. Moreover, we show that the learned models perform stably even when applied to unseen workloads, making them practical for production use.

Zhang, Di↗

Response of a Memristive Biomembrane and Demonstration of Potential Use in Online Learning

The pervasive von Neumann architecture uses complex processor cores and sequential computation. In contrast, the brain is massively parallel and highly efficient, owing to the ability of the neurons and synapses to store and process information simultaneously and to adapt according to incoming information. These features have motivated researchers to develop a host of brain-inspired computers, devices, and models, collectively referred to as neuromorphic computing systems. The quest for synaptic materials capable of closely mimicking biological synapses has led to an alamethicin-doped, synthetic biomembrane with volatile memristive properties which can emulate key synaptic functions to facilitate learning and computation. In contrast to its solid-state counterparts, this two-terminal, biomolecular memristor features similar structure, switching mechanisms, and ionic transport modality as biological synapses while consuming considerably lower power. To use the device as a circuit element, it is important to understand its response to different kinds of input signals. Here we develop a simplified closed form analytical solution based on the underlying state equations for pulse and sine wave inputs. A Verilog-A model based on Runge-Kutta method was developed to incorporate the device in a circuit simulator. Finally, the paper demonstrates possible applications for short- and long-term learning using its unique volatile memristive properties.

Hasan, Md Sakib↗

Tools for unbinned unfolding

Machine learning has enabled differential cross section measurements that are not discretized. Going beyond the traditional histogram-based paradigm, these unbinned unfolding methods are rapidly being integrated into experimental workflows. Here, in order to enable widespread adaptation and standardization, we develop methods, benchmarks, and software for unbinned unfolding. For methodology, we demonstrate the utility of boosted decision trees for unfolding with a relatively small number of high-level features. This complements state-of-the-art deep learning models capable of unfolding the full phase space. To benchmark unbinned unfolding methods, we develop an extension of existing dataset to include acceptance effects, a necessary challenge for real measurements. Additionally, we directly compare binned and unbinned methods using discretized inputs for the latter in order to control for the binning itself. Lastly, we have assembled two software packages for the OmniFold unbinned unfolding method that should serve as the starting point for any future analyses using this technique. One package is based on the widely-used RooUnfold framework and the other is a standalone package available through the Python Package Index (PyPI).

47 OTHER INSTRUMENTATION↗

A Learning Framework for Control-Oriented Modeling of Buildings

Buildings consume a significant amount of energy worldwide. Several building optimization and control use cases require models of energy consumption which are control oriented, have high predictive capability, imposes minimal data pre-processing requirements, and have the ability to be adapted continuously to account for changing conditions as new data becomes available. Data driven modeling techniques, that have been investigated so far, while promising in the context of buildings, have been unable to simultaneously satisfy all the requirements mentioned above. In this context, deep learning techniques such as Recurrent Neural Networks (RNNs) hold promise, empowered by advanced computational capabilities and big data opportunities. In this paper, we propose a deep learning based methodology for the development of control oriented models for building energy management and test in on data from a real building. Results show that the proposed methodology outperforms other data driven modeling techniques significantly. We perform a detailed analysis of the proposed methodology along dimensions such as topology, sensitivity, and downsampling. Lastly, we conclude by envisioning a building analytics suite empowered by the proposed deep framework, that can drive several use cases related to building energy management.

buildings↗

Data-driven materials research enabled by natural language processing and information extraction

Given the emergence of data science and machine learning throughout all aspects of society, but particularly in the scientific domain, there is increased importance placed on obtaining data. Data in materials science are particularly heterogeneous, based on the significant range in materials classes that are explored and the variety of materials properties that are of interest. This leads to data that range many orders of magnitude, and these data may manifest as numerical text or image-based information, which requires quantitative interpretation. The ability to automatically consume and codify the scientific literature across domains - enabled by techniques adapted from the field of natural language processing - therefore has immense potential to unlock and generate the rich datasets necessary for data science and machine learning. This review focuses on the progress and practices of natural language processing and text mining of materials science literature and highlights opportunities for extracting additional information beyond text contained in figures and tables in articles. Here, we discuss and provide examples for several reasons for the pursuit of natural language processing for materials, including data compilation, hypothesis development, and understanding the trends within and across fields. Current and emerging natural language processing methods along with their applications to materials science are detailed. We, then, discuss natural language processing and data challenges within the materials science domain where future directions may prove valuable.

36 MATERIALS SCIENCE↗

Adaptive hyperparameter updating for training restricted Boltzmann machines on quantum annealers

Restricted Boltzmann Machines (RBMs) have been proposed for developing neural networks for a variety of unsupervised machine learning applications such as image recognition, drug discovery, and materials design. The Boltzmann probability distribution is used as a model to identify network parameters by optimizing the likelihood of predicting an output given hidden states trained on available data. Training such networks often requires sampling over a large probability space that must be approximated during gradient based optimization. Quantum annealing has been proposed as a means to search this space more efficiently which has been experimentally investigated on D-Wave hardware. D-Wave implementation requires selection of an effective inverse temperature or hyperparameter (β) within the Boltzmann distribution which can strongly influence optimization. Here, we show how this parameter can be estimated as a hyperparameter applied to D-Wave hardware during neural network training by maximizing the likelihood or minimizing the Shannon entropy. We find both methods improve training RBMs based upon D-Wave hardware experimental validation on an image recognition problem. Neural network image reconstruction errors are evaluated using Bayesian uncertainty analysis which illustrate more than an order magnitude lower image reconstruction error using the maximum likelihood over manually optimizing the hyperparameter. The maximum likelihood method is also shown to out-perform minimizing the Shannon entropy for image reconstruction.

97 MATHEMATICS AND COMPUTING↗

Task 3: PNNL Visit by JAEA Researchers to Participate in TODAM Code Applications to Fukushima Rivers and to Evaluate the Feasibility of Adaptation of FLESCOT Code to Simulate Radionuclide Transport in the Pacific Ocean Coastal Water Around Fukushima

Four JAEA researchers visited PNNL for two weeks in February, 2013 to learn the PNNL-developed, unsteady, one-dimensional, river model, TODAM and the PNNL-developed, time-dependent, three dimensional, coastal water model, FLESCOT. These codes predict sediment and contaminant concentrations by accounting sediment-radionuclide interactions, e.g., adsorption/desorption and transport-deposition-resuspension of sediment-sorbed radionuclides.

Coastal Water Model↗

Ten questions concerning reinforcement learning for building energy management

As buildings account for approximately 40% of global energy consumption and associated greenhouse gas emissions, their role in decarbonizing the power grid is crucial. The increased integration of variable energy sources, such as renewables, introduces uncertainties and unprecedented flexibilities, necessitating buildings to adapt their energy demand to enhance grid resiliency. Consequently, buildings must transition from passive energy consumers to active grid assets, providing demand flexibility and energy elasticity while maintaining occupant comfort and health. This fundamental shift demands advanced optimal control methods to manage escalating energy demand and avert power outages. Reinforcement learning (RL) emerges as a promising method to address these challenges. Here, in this paper, we explore ten questions related to the application of RL in buildings, specifically targeting flexible energy management. We consider the growing availability of data, advancements in machine learning algorithms, open-source tools, and the practical deployment aspects associated with software and hardware requirements. Our objective is to deliver a comprehensive introduction to RL, present an overview of existing research and accomplishments, underscore the challenges and opportunities, and propose potential future research directions to expedite the adoption of RL for building energy management.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Machine Learning of Parameters for Accurate Semiempirical Quantum Chemical Calculations

We investigate possible improvements in the accuracy of semiempirical quantum chemistry (SQC) methods through the use of machine learning (ML) models for the parameters. For a given class of compounds, ML techniques require sufficiently large training sets to develop ML models that can be used for adapting SQC parameters to reflect changes in molecular composition and geometry. The ML-SQC approach allows the automatic tuning of SQC parameters for individual molecules, thereby improving the accuracy without deteriorating transferability to molecules with molecular descriptors very different from those in the training set. The performance of this approach is demonstrated for the semiempirical OM2 method using a set of 6095 constitutional isomers C7H10O2, for which accurate ab initio atomization enthalpies are available. The ML-OM2 results show improved average accuracy and a much reduced error range compared with those of standard OM2 results, with mean absolute errors in atomization enthalpies dropping from 6.3 to 1.7 kcal/mol. They are also found to be superior to the results from specific OM2 reparameterizations (rOM2) for the same set of isomers. The ML-SQC approach thus holds promise for fast and reasonably accurate high-throughput screening of materials and molecules.

Dral, Pavlo O.↗

Implementation and (Inverse Modified) Error Analysis for Implicitly Templated ODE-Nets

We focus on learning unknown dynamics from data using ODE-nets templated on implicit numerical initial value problem solvers. First, we perform inverse modified error analysis of the ODE-nets using unrolled implicit schemes for ease of interpretation. It is shown that training an ODE-net using an unrolled implicit scheme returns a close approximation of an inverse modified differential equation (IMDE). In addition, we establish a theoretical basis for hyperparameter selection when training such ODE-nets, whereas current strategies usually treat numerical integration of ODE-nets as a black box. We thus formulate an adaptive algorithm which monitors the level of error and adapts the number of (unrolled) implicit solution iterations during the training process, so that the error of the unrolled approximation is less than the current learning loss. This helps accelerate training while maintaining accuracy. Several numerical experiments are performed to demonstrate the advantages of the proposed algorithm compared to nonadaptive unrollings and validate the theoretical analysis. Here, we also note that this approach naturally allows for incorporating partially known physical terms in the equations, giving rise to what is termed “gray box” identification.

ODE-nets↗

Adaptive Fault Detection Based on Neural Networks and Multiple Sampling Points for Distribution Networks and Microgrids

Smart networks such as microgrid (MG) and active distribution networks (ADN) have been recently playing an important role in power system operation. The design and implementation of appropriate protection systems for such networks must be addressed, which imposes new technical challenges. This paper presents the implementation and validation aspects of an adaptive fault detection strategy based on neural networks (NNs) and multiple sampling points for ADN and MG. The solution is implemented on an edge device. Artificial NNs are used to derive a data-driven model that uses only local measurements to detect fault states of the network without the need for communication infrastructure. Multiple sampling points are used to derive a data-driven model, which allows the generalization considering the implementation in physical systems. The adaptive fault detector model is implemented on a Jetson Nano system, which is a single-board computer (SBC) with a small Graphic Processing Unit (GPU) intended to run machine learning loads at the edge. The proposed method is tested in a physical, real-life, low-voltage network located at Universidad del Norte, Colombia. This testing network is based on the IEEE-13 Node Test Feeder scaled down to 220 V. The validation in a simulation environment shows the accuracy and dependability above 99.6%, while the real-time tests show the accuracy and dependability of 95.5% and 100%, respectively. Without hard-to-derive parameters, the easy-to-implement embedded model highlights the potential for real-life applications.

42 ENGINEERING↗