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 37 records · Page 2

Adaptive Machine Learning for Robust Diagnostics and Control of Time-Varying Particle Accelerator Components and Beams

Machine learning (ML) is growing in popularity for various particle accelerator applications including anomaly detection such as faulty beam position monitor or RF fault identification, for non-invasive diagnostics, and for creating surrogate models. ML methods such as neural networks (NN) are useful because they can learn input-output relationships in large complex systems based on large data sets. Once they are trained, methods such as NNs give instant predictions of complex phenomenon, which makes their use as surrogate models especially appealing for speeding up large parameter space searches which otherwise require computationally expensive simulations. However, quickly time varying systems are challenging for ML-based approaches because the actual system dynamics quickly drifts away from the description provided by any fixed data set, degrading the predictive power of any ML method, and limits their applicability for real time feedback control of quickly time-varying accelerator components and beams. In contrast to ML methods, adaptive model-independent feedback algorithms are by design robust to un-modeled changes and disturbances in dynamic systems, but are usually local in nature and susceptible to local extrema. In this work, we propose that the combination of adaptive feedback and machine learning, adaptive machine learning (AML), is a way to combine the global feature learning power of ML methods such as deep neural networks with the robustness of model-independent control. We present an overview of several ML and adaptive control methods, their strengths and limitations, and an overview of AML approaches.

97 MATHEMATICS AND COMPUTING↗

Adaptive learning-driven high-throughput synthesis of oxygen reduction reaction Fe–N–C electrocatalysts

Reducing human reliance on inefficient energy systems and fossil fuels has become more urgent due to the consequences of global climate change. However, traditional trial-and-error approaches have hampered our ability to accelerate the discovery and implementation of functional materials for efficient energy conversion devices, such as polymer electrolyte fuel cells (PEFCs). To address this, we develop an adaptive learning framework that integrates machine learning and state-of-the-art capabilities in high-throughput synthesis to achieve expedited optimization of iron-nitrogen-carbon PEFC oxygen reduction reaction (ORR) electrocatalysts. We use statistical inference, uncertainty quantification, and global optimization to build a computational design-of-experiment tool that identifies the optimum compositions to be investigated next to reduce the demands placed on experimental materials discovery. We benchmark the ability of the proposed strategy to discover optimum catalyst synthesis conditions in a six-dimensional search space when starting with a thirty-six-sample database. By following the adaptive learning strategy, we synthesize fourteen new catalysts from approximately ten billion unique compositions and discover four catalysts that outperform all original samples. The best machine learning-optimized catalyst is 33% more active than the highest-performing one in the initial database, showing an ORR activity seven times larger than those typically reported for the same class of materials.

36 MATERIALS SCIENCE↗

MO-G-17A-05: PET Image Deblurring Using Adaptive Dictionary Learning

Purpose: The aim of this work is to deblur PET images while suppressing Poisson noise effects using adaptive dictionary learning (DL) techniques. Methods: The model that relates a blurred and noisy PET image to the desired image is described as a linear transform y=Hm+n where m is the desired image, H is a blur kernel, n is Poisson noise and y is the blurred image. The approach we follow to recover m involves the sparse representation of y over a learned dictionary, since the image has lots of repeated patterns, edges, textures and smooth regions. The recovery is based on an optimization of a cost function having four major terms: adaptive dictionary learning term, sparsity term, regularization term, and MLEM Poisson noise estimation term. The optimization is solved by a variable splitting method that introduces additional variables. We simulated a 128×128 Hoffman brain PET image (baseline) with varying kernel types and sizes (Gaussian 9×9, σ=5.4mm; Uniform 5×5, σ=2.9mm) with additive Poisson noise (Blurred). Image recovery was performed once when the kernel type was included in the model optimization and once with the model blinded to kernel type. The recovered image was compared to the baseline as well as another recovery algorithm PIDSPLIT+ (Setzer et. al.) by calculating PSNR (Peak SNR) and normalized average differences in pixel intensities (NADPI) of line profiles across the images. Results: For known kernel types, the PSNR of the Gaussian (Uniform) was 28.73 (25.1) and 25.18 (23.4) for DL and PIDSPLIT+ respectively. For blinded deblurring the PSNRs were 25.32 and 22.86 for DL and PIDSPLIT+ respectively. NADPI between baseline and DL, and baseline and blurred for the Gaussian kernel was 2.5 and 10.8 respectively. Conclusion: PET image deblurring using dictionary learning seems to be a good approach to restore image resolution in presence of Poisson noise. GE Health Care.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Distributed deep reinforcement learning for simulation control

Abstract Several applications in the scientific simulation of physical systems can be formulated as control/optimization problems. The computational models for such systems generally contain hyperparameters, which control solution fidelity and computational expense. The tuning of these parameters is non-trivial and the general approach is to manually ‘spot-check’ for good combinations. This is because optimal hyperparameter configuration search becomes intractable when the parameter space is large and when they may vary dynamically. To address this issue, we present a framework based on deep reinforcement learning (RL) to train a deep neural network agent that controls a model solve by varying parameters dynamically. First, we validate our RL framework for the problem of controlling chaos in chaotic systems by dynamically changing the parameters of the system. Subsequently, we illustrate the capabilities of our framework for accelerating the convergence of a steady-state computational fluid dynamics solver by automatically adjusting the relaxation factors of the discretized Navier–Stokes equations during run-time. The results indicate that the run-time control of the relaxation factors by the learned policy leads to a significant reduction in the number of iterations for convergence compared to the random selection of the relaxation factors. Our results point to potential benefits for learning adaptive hyperparameter learning strategies across different geometries and boundary conditions with implications for reduced computational campaign expenses 4 4 Data and codes available at https://github.com/Romit-Maulik/PAR-RL . .

42 ENGINEERING↗

MO-AB-BRA-10: Cancer Therapy Outcome Prediction Based On Dempster-Shafer Theory and PET Imaging

Purpose: In cancer therapy, utilizing FDG-18 PET image-based features for accurate outcome prediction is challenging because of 1) limited discriminative information within a small number of PET image sets, and 2) fluctuant feature characteristics caused by the inferior spatial resolution and system noise of PET imaging. In this study, we proposed a new Dempster-Shafer theory (DST) based approach, evidential low-dimensional transformation with feature selection (ELT-FS), to accurately predict cancer therapy outcome with both PET imaging features and clinical characteristics. Methods: First, a specific loss function with sparse penalty was developed to learn an adaptive low-rank distance metric for representing the dissimilarity between different patients’ feature vectors. By minimizing this loss function, a linear low-dimensional transformation of input features was achieved. Also, imprecise features were excluded simultaneously by applying a l2,1-norm regularization of the learnt dissimilarity metric in the loss function. Finally, the learnt dissimilarity metric was applied in an evidential K-nearest-neighbor (EK- NN) classifier to predict treatment outcome. Results: Twenty-five patients with stage II–III non-small-cell lung cancer and thirty-six patients with esophageal squamous cell carcinomas treated with chemo-radiotherapy were collected. For the two groups of patients, 52 and 29 features, respectively, were utilized. The leave-one-out cross-validation (LOOCV) protocol was used for evaluation. Compared to three existing linear transformation methods (PCA, LDA, NCA), the proposed ELT-FS leads to higher prediction accuracy for the training and testing sets both for lung-cancer patients (100+/−0.0, 88.0+/−33.17) and for esophageal-cancer patients (97.46+/−1.64, 83.33+/−37.8). The ELT-FS also provides superior class separation in both test data sets. Conclusion: A novel DST- based approach has been proposed to predict cancer treatment outcome using PET image features and clinical characteristics. A specific loss function has been designed for robust accommodation of feature set incertitude and imprecision, facilitating adaptive learning of the dissimilarity metric for the EK-NN classifier.

60 APPLIED LIFE SCIENCES↗

Site Characterization and Adaptation - Lessons Learned from the Port Granby LLRW Project in Canada - 19255

The Port Hope Area Initiative (PHAI) is a community-based solution for the long-term management of historic low level radioactive waste (LLRW) resulting from 60 years of uranium and radium processing operations in the Town of Port Hope, Ontario. The Eldorado refinery, on the shores of Lake Ontario, began refining radium-226 from pitchblende ore, later transitioning to the refining of uranium ore. Process residues were deposited at the Welcome Waste Management Facility in the Town of Port Hope until the mid-1950's switching to the Port Granby Waste Management Facility (PGWMF), which continued to accept waste until the late 1980's. The Port Granby Project is part of the broader PHAI being completed by Canadian Nuclear Laboratories (CNL). The scope of remediation activities at Port Granby includes: 1) constructing a new, secure long-term waste management facility (LTWMF) and supporting infrastructure, and 2) excavating and relocating LLRW and marginally contaminated soil from the Port Granby Waste Management Facility (PGWMF) to the newly constructed LTWMF. Remediation design required extensive site investigations of waste buried in unmarked and unlined trenches across the PGWMF area. Investigations included reviews of historical records, interviews with past staff of Eldorado and Atomic Energy Canada Limited (the entity managing the PHAI cleanup prior to CNL), radiological surveys, and the advancement of boreholes across the site to facilitate the physical sampling of soil and groundwater contaminants indicative of the presence of LLRW. This paper describes the challenges associated with characterizing and quantifying waste prior to remediation and lessons-learned to adjust procedural adaptations during the characterization phase of a project to better anticipate the unanticipated nature and growth of the contamination profile that often occurs during remediation. Despite substantial pre-remediation investigative efforts, current PGWMF remediation activities have resulted in growth of both the areal and vertical extent of contamination. The contamination profile was shown to be dramatically influenced by leaching of contaminants deeper into native soils. In addition, remedial excavation discovered specific waste types such as unidentified compressed gas cylinders outside of the location indicated by historical records, as well as a quantity of unknown liquids and unmarked drums. These discoveries have resulted in higher waste quantity estimates and have required a partial redesign of the LTWMF mound (a higher final elevation and steeper slopes) to accommodate the increased waste volume. In addition, the means and methods necessary to execute the remedial construction had to be modified to account for the new conceptual model developed from the new contamination data. Despite extensive historical review and site characterization to define the nature and extent of LLRW buried at the PGWMF, actual conditions encountered during remediation efforts necessitated taking an adaptive approach to both the design of the LTWMF and the means and methods used for excavation and transport of waste. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Adaptive Reinforcement Learning Control for Power Distribution in Multi-Output Resonant Converters

This paper presents an adaptive reinforcement learning (ARL)-based control framework for efficient power distribution in a multi-output resonant converter for UAV applications. The proposed system is based on a high-frequency isolated resonant architecture, where a single energy source supplies multiple propulsion loads through independently controlled output rectifiers, addressing the need for coordinated multi-motor power management. The ARL framework dynamically allocates output power by learning optimal phase-shift control actions under varying load demands and operating conditions. The agent autonomously determines control parameters that maximize conversion efficiency while ensuring accurate power sharing among multiple outputs. In addition, the proposed approach enables adaptive operation without requiring detailed system modeling or manual tuning. Experimental results demonstrate stable and efficient performance over a wide range of operating conditions, confirming the effectiveness and robustness of the learning-based control strategy for multi-output resonant converter system.

Asa, Erdem [ORNL] (ORCID:0000000190884812)↗

Adaptive Reinforcement Learning (ARL) Control of a Multi-port Resonant Converter in UAV Systems

This study presents an adaptive reinforcement learning (ARL) control framework for a multi-port resonant converter used in hybrid unmanned aerial vehicle (UAV) power systems. The converter integrates high-frequency half-bridge input ports connected to a rectified engine–generator set and a battery energy storage system, along with a semi-bridgeless active rectifier supplying the propulsion load. A deep RL agent is trained to dynamically regulate inter-port phase-shift commands in real time based on flight conditions and load power demand. The ARL controller autonomously identifies phase-shift combinations that maximize conversion efficiency while maintaining stable and coordinated power flow, even under rapidly varying operating scenarios. This data-driven approach eliminates the need for explicit system modeling or extensive manual tuning and enables coordinated control among multiple power ports without inter-port communication. Experimental results validate that the ARL based strategy achieves reliable power sharing and consistently high-efficiency operation across diverse UAV operating conditions.

Asa, Erdem [ORNL] (ORCID:0000000190884812)↗

Real-Time Adaptive Machine Learning Platform

As part of the Cyclotron Road program, Dauntless.io, Inc. extended its existing ultrafast (0.00001 – 0.001 second) real-time, adaptive machine learning platform for modeling & control of complex, high-value physical systems. Objectives included ability to scale ubiquitously & remain differentiated on a 5+ year horizon in the highly competitive, AI / machine learning market — a rapidly growing market that (overall) is estimated to be worth $191 B by 2024 at a CAGR of 37%.

97 MATHEMATICS AND COMPUTING↗

Adaptive, Active Learning, and Multifidelity Monte Carlo Methods in the MOOSE Stochastic Tools Module

MOOSE is an open-source computational platform for constructing multi-physics models and executing them in a massively parallel fashion. It has a stochastic tools module (STM) for forward/inverse uncertainty quantification (UQ) and surrogate modeling. This presentation details some recent developments to the STM with respect to the implementation of adaptive, active learning, and multifidelity Monte Carlo methods for forward UQ of computational models. Specifically, the adaptive Monte Carlo methods include Markov Chain Monte Carlo (MCMC)-driven algorithms like adaptive importance sampling and parallelized subset simulation for statistical QoI estimation, rare events analysis, and stochastic gradient-free optimization. The active learning methods include Gaussian Process (GP) surrogates and their training via Adam optimization, design of acquisition functions, and integration with samplers like Monte Carlo, adaptive importance, and parallelized subset simulation. These active learning methods are also designed to work in a batch mode, wherein, the required calls to the full computational model are executed in parallel whenever a user-specified batch size is met. The multifidelity methods in STM are broadly divided into two categories: hierarchical, where a defined hierarchy exists among the low-fidelity models, and peer, where all the low-fidelity models are treated equally. A GP surrogate is used to learn the differences between the low- and high-fidelity models in both multifidelity categories, and acquisition functions from the active learning classes are used to decide whether to rely on a low-fidelity model or call the expensive high-fidelity model. Alongside the software description and usage, applications are also presented to nuclear engineering computational models including a TRISO nuclear fuel particle, a reactor pressure vessel, and a heat-pipe microreactor.

97 MATHEMATICS AND COMPUTING↗

Unwrapping Radiograph Target Disassembly Using Adaptive Machine Learning [Slides]

Advanced analysis methods were developed to determine the electron temperature and density evolution of targets disassembly by intense electron beams at DARHT based on shearing interferometer and shadowgraph measurements. New methods have been developed to spatially and temporally resolve electron velocity and temperature, expanding on recently published first results of Aluminum target disassembly. A surrogate model has been created for use in developing Adaptive Machine Learning analysis techniques. The results will also be used to guide the development of the system and experimental approach to future measurements on DARHT Axis-I and a new system to be deployed on DARHT Axis-II.

43 PARTICLE ACCELERATORS↗

Feature‐shared adaptive‐boost deep learning for invasiveness classification of pulmonary subsolid nodules in CT images

Purpose In clinical practice, invasiveness is an important reference indicator for differentiating the malignant degree of subsolid pulmonary nodules. These nodules can be classified as atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IAC). The automatic determination of a nodule's invasiveness based on chest CT scans can guide treatment planning. However, it is challenging, owing to the insufficiency of training data and their interclass similarity and intraclass variation. To address these challenges, we propose a two‐stage deep learning strategy for this task: prior‐feature learning followed by adaptive‐boost deep learning. Methods The adaptive‐boost deep learning is proposed to train a strong classifier for invasiveness classification of subsolid nodules in chest CT images, using multiple 3D convolutional neural network (CNN)‐based weak classifiers. Because ensembles of multiple deep 3D CNN models have a huge number of parameters and require large computing resources along with more training and testing time, the prior‐feature learning is proposed to reduce the computations by sharing the CNN layers between all weak classifiers. Using this strategy, all weak classifiers can be integrated into a single network. Results Tenfold cross validation of binary classification was conducted on a total of 1357 nodules, including 765 noninvasive (AAH and AIS) and 592 invasive nodules (MIA and IAC). Ablation experimental results indicated that the proposed binary classifier achieved an accuracy of with an AUC of 81.3 . These results are superior compared to those achieved by three experienced chest imaging specialists who achieved an accuracy of , , and , respectively. About 200 additional nodules were also collected. These nodules covered 50 cases for each category (AAH, AIS, MIA, and IAC, respectively). Both binary and multiple classifications were performed on these data and the results demonstrated that the proposed method definitely achieves better performance than the performance achieved by nonensemble deep learning methods. Conclusions It can be concluded that the proposed adaptive‐boost deep learning can significantly improve the performance of invasiveness classification of pulmonary subsolid nodules in CT images, while the prior‐feature learning can significantly reduce the total size of deep models. The promising results on clinical data show that the trained models can be used as an effective lung cancer screening tool in hospitals. Moreover, the proposed strategy can be easily extended to other similar classification tasks in 3D medical images.

Wang, Jun↗

Accelerating the discovery of novel magnetic materials using machine learning–guided adaptive feedback

Magnetic materials are essential for energy generation and information devices, and they play an important role in advanced technologies and green energy economies. Currently, the most widely used magnets contain rare earth (RE) elements. An outstanding challenge of notable scientific interest is the discovery and synthesis of novel magnetic materials without RE elements that meet the performance and cost goals for advanced electromagnetic devices. Here, we report our discovery and synthesis of an RE-free magnetic compound, Fe 3 CoB 2 , through an efficient feedback framework by integrating machine learning (ML), an adaptive genetic algorithm, first-principles calculations, and experimental synthesis. Magnetic measurements show that Fe 3 CoB 2 exhibits a high magnetic anisotropy ( K 1 = 1.2 MJ/m 3 ) and saturation magnetic polarization ( J s = 1.39 T), which is suitable for RE-free permanent-magnet applications. Our ML-guided approach presents a promising paradigm for efficient materials design and discovery and can also be applied to the search for other functional materials.

36 MATERIALS SCIENCE↗

MACORD: Online Adaptive Machine Learning Framework for Silent Error Detection

Future HPC systems with ever-increasing resource capacity (such as compute cores, memory and storage) may significantly increase the risks on reliability. Silent data corruptions (SDCs) or silent errors are one of the major sources that corrupt HPC execution results. Unlike fail-stop errors, SDCs are rather harmful and dangerous in that they cannot be detected by hardware. We propose an online machine-learning based silent data corruption detection framework (abbreviated as MACORD) for detecting SDCs in HPC applications. In particular, we comprehensively investigate the prediction ability of a multitude of machine-learning algorithms in our study, and enable the detector to automatically select the best-fit algorithms at runtime to adapt to the data dynamics. Our learning framework exhibits low memory overhead (less than 1%), since it takes only spatial features (i.e., neighboring data values for each data point in the current time step) into the training data. Experiments based on real-world scientific applications/benchmarks show that our framework can get the detection sensitivity (i.e., recall) up to 99% while the false positive rate is limited to 0.1% in most cases, which is one order of magnitude improvement compared with the latest state-of-art spatial technique.

Subasi, Omer↗

TANTE: Time-adaptive operator learning via neural Taylor expansion

Operator learning for time-dependent partial differential equations (PDEs) has seen rapid progress in recent years, enabling efficient approximation of complex spatiotemporal dynamics. However, most existing methods rely on fixed time step sizes during rollout, which limits their ability to adapt to varying temporal complexity and often leads to error accumulation. In this work, we propose the Time-Adaptive Transformer with Neural Taylor Expansion (TANTE), a novel operator-learning framework that produces continuous-time predictions with adaptive step sizes. TANTE predicts future states by performing a Taylor expansion at the current state, where neural networks learn both the higher-order temporal derivatives and the local radius of convergence. This allows the model to dynamically adjust its rollout based on the local behavior of the solution, thereby reducing cumulative error and improving computational efficiency. We demonstrate the effectiveness of TANTE across a wide range of PDE benchmarks, achieving superior accuracy and adaptability compared to fixed-step baselines, delivering accuracy gains of 60-80 % and speed-ups of 30-40 % at inference time.

97 MATHEMATICS AND COMPUTING↗

Adaptive Graph Learning with Transformer for Multi-Reservoir Inflow Prediction

Reservoir inflow prediction is crucial for water resource management, yet existing approaches mainly focus on single-reservoir models that ignore spatial dependencies among interconnected reservoirs. We introduce AdaTrip as an adaptive, time-varying graph learning framework for multi-reservoir inflow forecasting. AdaTrip constructs dynamic graphs where reservoirs are nodes with directed edges reflecting hydrological connections, employing attention mechanisms to automatically identify crucial spatial and temporal dependencies. Evaluation on thirty reservoirs in the Upper Colorado River Basin demonstrates superiority over existing baselines, with improved performance for reservoirs with limited records through parameter sharing. Additionally, AdaTrip provides interpretable attention maps at edge and time-step levels, offering insights into hydrological controls to support operational decision-making. Our code is available at https://github.com/humphreyhuu/AdaTrip.

Hu, Pengfei [ORNL] (ORCID:0009000367130950)↗

A resolution independent neural operator

The Deep operator network (DeepONet) is a powerful yet simple neural operator architecture that utilizes two deep neural networks to learn mappings between infinite-dimensional function spaces. This architecture is highly flexible, allowing the evaluation of the solution field at any location within the desired domain. However, it imposes a strict constraint on the input space, requiring all input functions to be discretized at the same locations; this limits its practical applications. Here, in this work, we introduce a general framework for operator learning from input–output data with arbitrary number and locations of sensors. This begins by introducing a resolution-independent DeepONet (RI-DeepONet), enabling it to handle input functions that are arbitrarily, but sufficiently finely, discretized. To this end, we propose two dictionary learning algorithms to adaptively learn a set of appropriate continuous basis functions, parameterized as implicit neural representations (INRs), from correlated signals defined on arbitrary point cloud data. These basis functions are then used to project arbitrary input function data as a point cloud onto an embedding space (i.e., a vector space of finite dimensions) with dimensionality equal to the dictionary size, which can be directly used by DeepONet without any architectural changes. In particular, we utilize sinusoidal representation networks (SIRENs) as trainable INR basis functions. The introduced dictionary learning algorithms are then used in a similar way to learn an appropriate dictionary of basis functions for the output function data, which defines a new neural operator architecture referred to as the R esolution I ndependent N eural O perator (RINO). In the RINO, the operator learning task simplifies to learning a mapping from the coefficients of input basis functions to the coefficients of output basis functions. We demonstrate the robustness and applicability of RINO in handling arbitrarily (but sufficiently richly) sampled input and output functions during both training and inference through several numerical examples.

Deep operator network (DeepONet)↗

Reinforcement Learning‐Based Adaptation of Grid Following Inverter's Internal Controller to Networked Microgrids' Strengths

The varying topological configurations, generator commitments and dispatches, and dynamic load demand lead to changing system's strengths during the operations of networked microgrids. When the system's strengths significantly change, the fixed control gains at large devices may result in unsatisfactory system performance; this necessitates the tuning of the control gains at large devices to adapt to the changing system's strengths. In this paper, observer-based reinforcement learning (RL) is utilised to automatically tune the proportional-integral (PI) gains of phase lock loop (PLL) controller of grid-following (GFL) inverters to adapt to the changing strengths of microgrids and networked microgrids. The RL agent in this framework augments an observer predicting system's strengths, from which the RL control policy will adjust accordingly to tune the PLL controller's gains towards the system's strengths. Also, to enhance the control performance, the recently introduced Barrier function-based RL framework is leveraged for the design of reward function to prevent the high frequency nadir. An operational 26 kV electric distribution system, which is modelled as networked microgrids, is used to illustrate the need and effectiveness of the proposed RL-tuned control.

frequency response↗