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 1,495 records · Page 83

Efficient online quantum circuit learning with no upfront training

Optimization is a promising candidate for studying the utility of variational quantum algorithms (VQAs). However, evaluating cost functions using quantum hardware introduces runtime overheads that limit exploration. Surrogate-based methods can reduce calls to a quantum computer, yet existing approaches require hyperparameter pre-training and have been tested only on small problems. Here, we show that surrogate-based methods can enable successful optimization at scale, without pre-training, by using radial basis function interpolation (RBF) to construct an adaptive, hyperparameter-free surrogate. Using the surrogate as an acquisition function drives hardware queries to the vicinity of the true optima. For 16-qubit random 3-regular Max-Cut instances with the Quantum Approximate Optimization Algorithm (QAOA), our method outperforms state-of-the-art approaches, without considering their upfront training costs. Furthermore, we successfully optimize QAOA circuits for 127-qubit random Ising models on an IBM processor using 10 4 −10 5 measurements. Strong empirical performance demonstrates the promise of automated surrogate-based learning for large-scale VQA applications.

97 MATHEMATICS AND COMPUTING↗

Learning occupants’ indoor comfort temperature through a Bayesian inference approach for office buildings in United States

A carefully chosen indoor comfort temperature as the thermostat set-point is the key to optimizing building energy use and occupants’ comfort and well-being. ASHRAE Standard 55 or ISO Standard 7730 uses the PMV-PPD model or the adaptive comfort model that is based on small-sized or outdated sample data, which raises questions on whether and how ranges of occupant thermal comfort temperature should be revised using more recent larger-sized dataset. In this paper, a Bayesian inference approach has been used to derive new occupant comfort temperature ranges for U.S. office buildings using the ASHRAE Global Thermal Comfort Database. Bayesian inference can express uncertainty and incorporate prior knowledge. The comfort temperatures were found to be higher and less variable at cooling mode than at heating mode, and with significant overlapped variation ranges between the two modes. The comfort operative temperature of occupants varies between 21.9 and 25.4 °C for the cooling mode with a median of 23.7 °C, and between 20.5 and 24.9 °C for the heating mode with a median of 22.7 °C. These comfort temperature ranges are similar to the current ASHRAE standard 55 in the heating mode but 2–3 °C lower in the cooling mode. Finally, the results of this study could be adopted as more realistic thermostat set-points in building design, operation, control optimization, energy performance analysis, and policymaking.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Supporting shared resource usage for a diverse user community: The OSG experience and lessons learned

The Open Science Grid (OSG) supports a diverse community of new and existing users in adopting and making effective use of the Distributed High Throughput Computing (DHTC) model. The LHC user community has deep local support within the experiments. For other smaller communities and individual users the OSG provides consulting and technical services through the User Support area. We describe these sometimes successful and sometimes not so successful experiences and analyze lessons learned that are helping us improve our services. The services offered include forums to enable shared learning and mutual support, tutorials and documentation for new technology, and troubleshooting of problematic or systemic failure modes. For new communities and users, we bootstrap their use of the distributed high throughput computing technologies and resources available on the OSG by following a phased approach. We first adapt the application and run a small production campaign on a subset of friendly sites. Only then do we move the user to run full production campaigns across the many remote sites on the OSG, adding to the community resources up to hundreds of thousands of CPU hours per day. This scaling up generates new challenges - like no determinism in the time to job completion, and diverse errors due to the heterogeneity of the configurations and environments - so some attention is needed to get good results. We cover recent experiences with image simulation for the Large Synoptic Survey Telescope (LSST), small-file large volume data movement for the Dark Energy Survey (DES), civil engineering simulation with the Network for Earthquake Engineering Simulation (NEES), and accelerator modeling with the Electron Ion Collider group at BNL. We will categorize and analyze the use cases and describe how our processes are evolving based on lessons learned.

Garzoglio, Gabriele↗

Integrated RMP-based ELM-crash-control process for plasma performance enhancement during ELM crash suppression in KSTAR

The integrated Resonant Magnetic Perturbation (RMP)-based Edge-Localized Mode (ELM)-crash-control process aims to enhance the plasma performance during the RMP-driven ELM crash suppression, where the RMP induces an unwanted confinement degradation. In this study, the normalized beta ($\beta_\textrm{N}$) is introduced as a metric for plasma performance. The integrated process incorporates the latest achievements in the RMP technique to enhance $\beta_\textrm{N}$ efficiently. The integrated process triggers the n = 1 Edge-localized RMP (ERMP) at the L–H transition timing using the real-time Machine Learning (ML) classifier. The pre-emptive RMP onset can reduce the required external heating power for achieving the same $\beta_\textrm{N}$ by over 10% compared to the conventional onset. During the RMP phase, the adaptive feedback RMP ELM controller, demonstrating its performance in previous experiments, plays a crucial role in maximizing $\beta_\textrm{N}$ during the suppression phase and sustaining the $\beta_\textrm{N}$-enhanced suppression state by optimizing the RMP strength. The integrated process achieves $\beta_\textrm{N}$ up to ~2.65 during the suppression phase, which is ~10% higher than the previous KSTAR record but ~6% lower than the target of the K-DEMO first phase ($\beta_\textrm{N}$ = 2.8), and maintains the suppression phase above the lower limit of target $\beta_\textrm{N}$ (= 2.4) for ~4 s (~60$\tau_\textrm{E}$). In addition to $\beta_\textrm{N}$ enhancement, the integrated process demonstrates quicker restoration of the suppression phase and recovery of $\beta_\textrm{N}$ compared to the adaptive control with the n = 1 Conventional RMP (CRMP). The post-analysis of the experiment shows the localized effect of the ERMP spectrum in radial and the close relationship between the evolution of $\beta_\textrm{N}$ and the electron temperature.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Adaptive sampling for accelerating neutron diffraction-based strain mapping *

Abstract Neutron diffraction is a useful technique for mapping residual strains in dense metal objects. The technique works by placing an object in the path of a neutron beam, measuring the diffracted signals and inferring the local lattice strain values from the measurement. In order to map the strains across the entire object, the object is stepped one position at a time in the path of the neutron beam, typically in raster order, and at each position a strain value is estimated. Typical dwell times at neutron diffraction instruments result in an overall measurement that can take several hours to map an object that is several tens of centimeters in each dimension at a resolution of a few millimeters, during which the end users do not have an estimate of the global strain features and are at risk of incomplete information in case of instruments outages. In this paper, we propose an object adaptive sampling strategy to measure the significant points first. We start with a small initial uniform set of measurement points across the object to be mapped, compute the strain in those positions and use a machine learning technique to predict the next position to measure in the object. Specifically, we use a Bayesian optimization based on a Gaussian process regression method to infer the underlying strain field from a sparse set of measurements and predict the next most informative positions to measure based on estimates of the mean and variance in the strain fields estimated from the previously measured points. We demonstrate our real-time measure-infer-predict workflow on additively manufactured steel parts—demonstrating that we can get an accurate strain estimate even with 30%–40% of the typical number of measurements—leading the path to faster strain mapping with useful real-time feedback. We emphasize that the proposed method is general and can be used for fast mapping of other material properties such as phase fractions from time-consuming point-wise neutron measurements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Exploring Capability of Multimodal Foundation Model for Image-based Fault Detection of Photovoltaic Modules

Multimodal Foundation Model (MFM), like ChatGPT and Gemini, have emerged as powerful tools for their exceptional natural language processing capabilities and their emerging potential in image analysis. This paper investigates the application of MFMs for photovoltaic (PV) fault detection through image analysis, focusing on ChatGPT 4.0 and Gemini 1.5 Pro. Three types of PV images and the corresponding common PV faults are detected: bird droppings using visible images, cell cracks via electroluminescence (EL) images, and hotspots using infrared (IR) images. Among the two models, Gemini 1.5 Pro demonstrated superior performance, achieving near-perfect results with an average F1 score of 0.97, consistently outperforming ChatGPT 4.0 in accuracy and reliability. Unlike traditional machine learning (ML) models, MFMs can operate in a zero shot manner that does not require additional training by the user, and the input images are not limited by size, angle, scope, or PV technology. The strong adaptability and user-friendliness make MFM a promising tool for analyzing PV images and advancing health monitoring for PV modules.

Li, Baojie↗

AI-Based Upgrades to Observational Data Centers to Facilitate Data Interoperability

Focal Areas: (1) Data acquisition and assimilation enabled by machine learning, AI, and advanced methods including experimental/network design/optimization, unsupervised learning (including deep learning), and hardware-related efforts involving AI (e.g., edge computing). Focal areas 2 and 3 have critical dependencies to the modernization described. Key benefits to the focal areas: (1) Modernized observatory framework capable of agile adaptive observation, (2) Advanced instrument and data tagging supporting AI data acquisition for assimilation or validation, and (3) Widespread data interoperability bridging Earth system prediction scales

54 ENVIRONMENTAL SCIENCES↗

MULTI-LEADER: MULTI-source LEarning-Accelerated Design of high-Efficiency multi-stage compRessor (Final Technical Report)

The objective of MULTI-LEADER is to cut design costs by 80% while generating more energy-efficient designs of multi-stage compressors by developing and implementing novel machine learning (ML) techniques, which enable faster and fewer design iterations, improved solver performance, and concurrent multi-disciplinary design. Current industrial practices for the design of multi-stage compressors involve simulation-based design optimization with successive levels of model fidelity, iteratively evaluated between distinct disciplines, one stage at a time to tackle the high dimensional design variations. This project addresses these key design challenges: (1) concurrent optimization of multiple stages under many non-linear constraints; (2) multitude of evaluation of high-fidelity and expensive solvers and their gradients during optimization convergence in high-dimensional design; (3) multi-disciplinary design to maximize aerodynamic performance while guaranteeing structural integrity and additive manufacturability; (4) utilization of multiple fidelity of solvers with disparate parameterization and modeling assumptions. MULTI-LEADER achieved more than 5x speed up in detailed design of more energy-efficient compressors via these machine learning (ML) innovations: (i) rapid design surrogates by multi-source learning from diverse fidelities across multiple disciplines, (ii) physics-constrained data-augmented modeling for improved empiricism, (iii) generative manifold embedding for high dimensional concurrent design without gradient information; (iv) budget-constrained fidelity-adaptive sampling towards fewer design iterations.

33 ADVANCED PROPULSION SYSTEMS↗

Treatment of Activated Carbon, PVC, and Other Wastes using the THOR process - Pilot Plant Test Summary - 17399

This presentation will provide an overview of the first pilot plant study to be conducted for the Japanese Nuclear Market by Kobelco Studsvik Company (A new Joint Venture Company between Studsvik AB and Kobe Steel Ltd.). The test program was scheduled for October 2016, and it was designed to evaluate the treatment of activated carbon, PVC, and other waste types using Studsvik's THOR Process. This new study is being conducted to identify key design parameters required for application of the THOR process in the Japanese Waste Market and to confirm the effectiveness of the THOR technology in meeting the requirements for final waste acceptance. The findings of this pilot plant test will assist in the development of a revised THOR process in Japan that will treat not only spent ion exchange resins but other organic materials generated by Nuclear Power Plants (NPP) during normal operations and D and D activities. The information shared by this new pilot plant study will provide the audience with an understanding of the adaptability of the THOR process for treatment of various radioactive waste types and the results necessary to successful treat different types of carbon, plastic materials, and other waste types. In addition, it will share the lessons learned regarding how a pilot plant study should be planned and optimized to meet the specific requirements of the Japanese Nuclear waste market. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Quantum advantages for Pauli channel estimation

We show that entangled measurements provide an exponential advantage in sample complexity for Pauli channel estimation, which is both a fundamental problem and a practically important subroutine for benchmarking near-term quantum devices. The specific task we consider is to simultaneously learn all the eigenvalues of an n-qubit Pauli channel to ±ε precision. We give an estimation protocol with an n-qubit ancilla that succeeds with high probability using only O(n/ε 2 ) copies of the Pauli channel, while proving that any ancilla-free protocol (possibly with adaptive control and channel concatenation) would need at least Ω(2 n/3 ) rounds of measurement. We further study the advantages provided by a small number of ancillas. For the case that a k-qubit ancilla (k≤n) is available, we obtain a sample complexity lower bound of Ω(2 (n-k)/3 ) for any nonconcatenating protocol, and a stronger lower bound of Ω(n 2n-k ) for any nonadaptive, nonconcatenating protocol, which is shown to be tight. We also show how to apply the ancilla-assisted estimation protocol to a practical quantum benchmarking task in a noise-resilient and sample-efficient manner, given reasonable noise assumptions. Our results provide a practically interesting example for quantum advantages in learning and also bring insights for quantum benchmarking.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Accelerating computational fluid dynamics simulation of post-combustion carbon capture modeling with MeshGraphNets

Packed columns are commonly used in post-combustion processes to capture CO 2 emissions by providing enhanced contact area between a CO 2 -laden gas and CO 2 -absorbing solvent. To study and optimize solvent-based post-combustion carbon capture systems (CCSs), computational fluid dynamics (CFD) can be used to model the liquid–gas countercurrent flow hydrodynamics in these columns and derive key determinants of CO 2 -capture efficiency. However, the large design space of these systems hinders the application of CFD for design optimization due to its high computational cost. In contrast, data-driven modeling approaches can produce fast surrogates to study large-scale physics problems. We build our surrogates using MeshGraphNets (MGN), a graph neural network framework that efficiently learns and produces mesh-based simulations. We apply MGN to a random packed column modeled with over 160K graph nodes and a design space consisting of three key input parameters: solvent surface tension, inlet velocity, and contact angle. Our models can adapt to a wide range of these parameters and accurately predict the complex interactions within the system at rates over 1700 times faster than CFD, affirming its practicality in downstream design optimization tasks. This underscores the robustness and versatility of MGN in modeling complex fluid dynamics for large-scale CCS analyses.

97 MATHEMATICS AND COMPUTING↗

Trajectory design via unsupervised probabilistic learning on optimal manifolds

Abstract This article illustrates the use of unsupervised probabilistic learning techniques for the analysis of planetary reentry trajectories. A three-degree-of-freedom model was employed to generate optimal trajectories that comprise the training datasets. The algorithm first extracts the intrinsic structure in the data via a diffusion map approach. We find that data resides on manifolds of much lower dimensionality compared to the high-dimensional state space that describes each trajectory. Using the diffusion coordinates on the graph of training samples, the probabilistic framework subsequently augments the original data with samples that are statistically consistent with the original set. The augmented samples are then used to construct conditional statistics that are ultimately assembled in a path planning algorithm. In this framework, the controls are determined stage by stage during the flight to adapt to changing mission objectives in real-time.

42 ENGINEERING↗

Intelligent multi-zone residential HVAC control strategy based on deep reinforcement learning

Residential heating, ventilation, and air conditioning (HVAC) has been considered as an important demand response resource. However, the optimization of residential HVAC control is no trivial task due to the complexity of the thermal dynamic models of buildings and uncertainty associated with both occupant-driven heat loads and weather forecasts. In this paper, we apply a novel model-free deep reinforcement learning (RL) method, known as the deep deterministic policy gradient (DDPG), to generate an optimal control strategy for a multi-zone residential HVAC system with the goal of minimizing energy consumption cost while maintaining the users’ comfort. Here, the applied deep RL-based method learns through continuous interaction with a simulated building environment and without referring to any prior model knowledge. Simulation results show that compared with the state-of-art deep Q network (DQN), the DDPG-based HVAC control strategy can reduce the energy consumption cost by 15% and reduce the comfort violation by 79%; and when compared with a rule-based HVAC control strategy, the comfort violation can be reduced by 98%. In addition, experiments with different building models and retail price models demonstrate that the well-trained DDPG-based HVAC control strategy has high generalization and adaptability to unseen environments, which indicates its practicability for real-world implementation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Climate Information and Lessons Learned from the NASA Earth eXchange (NEX), Short-term Prediction Research and Transition Center (NASA SPoRT), and the NASA Regional Climate Model Evaluation System (RCMES)

Climate resilience can emerge from a combination of (1) climate projections that enable future-oriented planning and (2) contemporary observations that can be used for adaptation in the face of realized climate. NASA is uniquely positioned to provide global climate projections and satellite-based observations to users who must take a global perspective on climate risk (e.g., supply chain managers). Here, we will share the capabilities and experiences of three NASA projects. The first project, the NASA Earth eXchange (NEX) provides two downscaled climate projections, NEX-GDDP-CMIP6, a global 25 km resolution product, and NEX-DCP30-CMIP6, an 800 m resolution product covering the coterminous U.S. specifically designed for U.S. National Climate Assessment (NCA). NEX also offers land surface products that mirror the twice-daily products provided by the Moderate Resolution Imaging Spectroradiometer (MODIS) but are derived from operational geostationary weather satellite data collected at 5-10 minutes intervals. The second project, the Short-term Prediction Research and Transition Center (SPoRT), was established in 2002 to transition unique NASA satellite products and capabilities to stakeholders to improve decision-making that benefits society. SPoRT provides products and capabilities derived from NASA observations and models to end users in government, academia, and the private sector. SPoRT translates NASA Earth science into actionable solutions reaching over 400 stakeholders across 80 organizations. Finally, the Regional Climate Model Evaluation System (RCMES) is designed to facilitate regional-scale evaluations of climate models by providing standardized access to a vast and comprehensive set of observations, such as satellite, in-situ, and reanalyses, and modeling resources such as CMIP & CORDEX. These three projects form part of the core capabilities at the NASA Centers deliver climate information under the new NASA Earth Science to Action (ES2A) strategy.

NEX↗

SU-E-J-181: Magnetic Resonance Image-Guided Radiation Therapy Workflow: Initial Clinical Experience

Purpose: The aims of this work are to describe the workflow and initial clinical experience treating patients with an MRI-guided radiotherapy (MRIGRT) system. Methods: Patient treatments with a novel MR-IGRT system started at our institution in mid-January. The system consists of an on-board 0.35-T MRI, with IMRT-capable delivery via doubly-focused MLCs on three {sup 60} Co heads. In addition to volumetric MR-imaging, real-time planar imaging is performed during treatment. So far, eleven patients started treatment (six finished), ranging from bladder to lung SBRT. While the system is capable of online adaptive radiotherapy and gating, a conventional workflow was used to start, consisting of volumetric imaging for patient setup using visible tumor, evaluation of tumor motion outside of PTV on cine images, and real-time imaging. Workflow times were collected and evaluated to increase efficiency and evaluate feasibility of adding the adaptive and gating features while maintaining a reasonable patient throughput. Results: For the first month, physicians attended every fraction to provide guidance on identifying the tumor and an acceptable level of positioning and anatomical deviation. Average total treatment times (including setup) were reduced from 55 to 45 min after physician presence was no longer required and the therapists had learned to align patients based on soft-tissue imaging. Presently, the source strengths were at half maximum (7.7K Ci each), therefore beam-on times will be reduced after source replacement. Current patient load is 10 per day, with increase to 25 anticipated in the near future. Conclusion: On-board, real-time MRI-guided RT has been incorporated into clinical use. Treatment times were kept to reasonable lengths while including volumetric imaging, previews of tumor movement, and physician evaluation. Workflow and timing is being continuously evaluated to increase efficiency. In near future, adaptive and gating capabilities of the system will be implemented.

60 APPLIED LIFE SCIENCES↗

Neural networks for aircraft control

Current research in Artificial Neural Networks indicates that networks offer some potential advantages in adaptation and fault tolerance. This research is directed at determining the possible applicability of neural networks to aircraft control. The first application will be to aircraft trim. Neural network node characteristics, network topology and operation, neural network learning and example histories using neighboring optimal control with a neural net are discussed.

Linse, Dennis↗

Providing Effective Access to Shared Resources: A COIN Approach

Managers of systems of shared resources typically have many separate goals. Examples are efficient utilization of the resources among its users and ensuring no user's satisfaction in the system falls below a preset minimal level. Since such goals will usually conflict with one another, either implicitly or explicitly the manager must determine the relative importance of the goals, encapsulating that into an overall utility function rating the possible behaviors of the entire system. Here we demonstrate a distributed, robust, and adaptive way to optimize that overall function. Our approach is to interpose adaptive agents between each user and the system, where each such agent is working to maximize its own private utility function. In turn, each such agent's function should be both relatively easy for the agent to learn to optimize, and 'aligned' with the overall utility function of the system manager - an overall function that is based on but in general different from the satisfaction functions of the individual users. To ensure this we enhance the COllective INtelligence (COIN) framework to incorporate user satisfaction functions in the overall utility function of the system manager and accordingly in the associated private utility functions assigned to the users agents. We present experimental evaluations of different COIN-based private utility functions and demonstrate that those COIN-based functions outperform some natural alternatives.

Airiau, Stephane↗

Providing Effective Access to Shared Resources: A COIN Approach

Managers of systems of shared resources typically have many separate goals. Examples are efficient utilization of the resources among its users and ensuring no user s satisfaction in the system falls below a preset minimal level. Since such goals will usually conflict with one another, either implicitly or explicitly the manager must determine the relative importance of the goals, encapsulating that into an overall utility function rating the possible behaviors of the entire system. Here we demonstrate a distributed, robust, and adaptive way to optimize that overall function. Our approach is to interpose adaptive agents between each user and the system, where each such agent is working to maximize its own private utility function. In turn, each such agent's function should be both relatively easy for the agent to learn to optimize, and "aligned" with the overall utility function of the system manager - an overall function that is based on but in general different from the satisfaction functions of the individual users. To ensure this we enhance the Collective INtelligence (COIN) framework to incorporate user satisfaction functions in the overall utility function of the system manager and accordingly in the associated private utility functions assigned to the users agents. We present experimental evaluations of different COIN-based private utility functions and demonstrate that those COIN-based functions outperform some natural alternatives.

Airiau, Stephane↗