Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “learning based systems”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Machine Learning Based Parallel I/O Predictive Modeling: A Case Study on Lustre File Systems

Parallel I/O hardware and software infrastructure is a key contributor to performance variability for applications running on large-scale HPC systems. This variability confounds efforts to predict application performance for characterization, modeling, optimization, and job scheduling. We propose a modeling approach that improves predictive ability by explicitly treating the variability and by leveraging the sensitivity of application parameters on performance to group applications with similar characteristics. We develop a Gaussian process-based machine learning algorithm to model I/O performance and its variability as a function of application and file system characteristics. We demonstrate the effectiveness of the proposed approach using data collected from the Edison system at the National Energy Research Scientific Computing Center. The results show that the proposed sensitivity-based models are better at prediction when compared with application-partitioned or unpartitioned models. We highlight modeling techniques that are robust to the outliers that can occur in production parallel file systems. Using the developed metrics and modeling approach, we provide insights into the file system metrics that have a significant impact on I/O performance.

I/O performance variability↗
DOE OSTI2022

Post-Disaster Microgrid Formation for Enhanced Distribution System Resilience

This paper proposes a deep reinforcement learning (DRL) based approach for post-disaster critical load restoration in active distribution systems to form microgrids through network reconfiguration to minimize critical load curtailments. Distribution networks are represented as graph networks, and optimal network configurations with microgrids are obtained by searching for the optimal spanning forest. The constraints to the research question being explored are the radial topology and power balance. Unlike existing analytical and population-based approaches, which necessitate the repetition of entire analyses and computation for each outage scenario to find the optimal spanning forest, the proposed approach, once properly trained, can quickly determine the optimal, or near-optimal, spanning forest even when outage scenarios change. When multiple lines fail in the system, the proposed approach forms microgrids with distributed energy resources in active distribution systems to reduce critical load curtailment. The proposed DRL-based model learns the action-value function using the REINFORCE algorithm, which is a model-free reinforcement learning technique based on stochastic policy gradients. A case study was conducted on a 33-node distribution test system, demonstrating the effectiveness of the proposed approach for post-disaster critical load restoration.

active distribution systems↗
DOE OSTI2016

WE-H-BRC-09: Simulated Errors in Mock Radiotherapy Plans to Quantify the Effectiveness of the Physics Plan Review

Purpose: A standard tool for ensuring the quality of radiation therapy treatments is the initial physics plan review. However, little is known about its performance in practice. The goal of this study is to measure the effectiveness of physics plan review by introducing simulated errors into “mock” treatment plans and measuring the performance of plan review by physicists. Methods: We generated six mock treatment plans containing multiple errors. These errors were based on incident learning system data both within the department and internationally (SAFRON). These errors were scored for severity and frequency. Those with the highest scores were included in the simulations (13 errors total). Observer bias was minimized using a multiple co-correlated distractor approach. Eight physicists reviewed these plans for errors, with each physicist reviewing, on average, 3/6 plans. The confidence interval for the proportion of errors detected was computed using the Wilson score interval. Results: Simulated errors were detected in 65% of reviews [51–75%] (95% confidence interval [CI] in brackets). The following error scenarios had the highest detection rates: incorrect isocenter in DRRs/CBCT (91% [73–98%]) and a planned dose different from the prescribed dose (100% [61–100%]). Errors with low detection rates involved incorrect field parameters in record and verify system (38%, [18–61%]) and incorrect isocenter localization in planning system (29% [8–64%]). Though pre-treatment QA failure was reliably identified (100%), less than 20% of participants reported the error that caused the failure. Conclusion: This is one of the first quantitative studies of error detection. Although physics plan review is a key safety measure and can identify some errors with high fidelity, others errors are more challenging to detect. This data will guide future work on standardization and automation. Creating new checks or improving existing ones (i.e., via automation) will help in detecting those errors with low detection rates.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Verification of Spent Fuel Inside Dry Storage Casks by Cask Top Fast Neutron Mapping (FY2023 Mid-Year Report)

This project is developing a prototype scanner array verification system for detection of missing fuel assemblies in spent-fuel storage casks. The prototype consists of six fast-neutron scintillator detectors mounted to a linear actuator frame that is placed on the top of a spent fuel cask to scan across all fuel assembly positions. The scanner array was assembled and tested at LLNL in FY2022. A field test schedule has been requested at the Idaho National Laboratory (INL) Cask Farm site for FY2023. Note that the Cask Farm contractor determines this scheduling and not INL directly. Further system automation will be designed and implemented with the goal of obtaining a level of system operation that meets IAEA needs. This includes integration of the scanner array and data-acquisition control software into a single interface for operator use. In addition, commercial operators and the IAEA may have special requirements for portability, shipping, lifting, and installation. Prior to the Field Test at INL, the system will be operated at LLNL to exercise lifting procedure and linear actuators, monitor stability of detector energy and pulse-shape discrimination calibration, and test system software integration efforts. Following the Field Test, we will present results and discuss the technology with the IAEA. We will incorporate additional improvements to the system based on lessons learned from the field test and feedback from the IAEA. If successful, the technology can be transferred to the IAEA or other stakeholders for assessment.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

FY2024 Mid-Year Report: Verification of Spent Fuel Inside Dry Storage Casks by Cask Top Fast Neutron Mapping

This project is developing a prototype scanner array verification system for detection of missing fuel assemblies in spent-fuel storage casks. The prototype consists of six fast-neutron scintillator detectors mounted to a linear actuator frame that is placed on the top of a spent fuel cask to scan across all fuel assembly positions. The scanner array was assembled and tested at LLNL in prior years. A field test schedule has been requested at the Idaho National Laboratory (INL) Cask Farm site for FY2024. Following the Field Test, we will present results and discuss the technology with the IAEA. The IAEA may have special requirements for portability, shipping, lifting, and installation. We will incorporate additional improvements to the system based on lessons learned from the Field Test and feedback from the IAEA. If successful, the technology can be transferred to the IAEA or other stakeholders for assessment.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Fast Adaptive Neural Control of Resonant Extraction at Fermilab

We present the development of a machine learning (ML) based regulation system for third-order resonant beam extraction in the Mu2e experiment at Fermilab. Classical and ML-based controllers have been optimized using semi-analytic simulations and evaluated in terms of regulation performance and training efficiency. We compare several controller architectures and discuss the integration of neural control into an adaptive framework. We also present progress on surrogate models that predict the controller response given a spill intensity and controller action history. To enable real-time deployment, we report progress on implementing low-latency, edge-based inference suitable for hardware-constrained environments. Our results demonstrate the feasibility and advantages of ML-based control in managing complex, time-varying physical systems, with broader implications for accelerator operations and other domains requiring fast, adaptive regulation.

Berlioz, Jose Rene [Fermilab]↗
DOE OSTI2022

Dynamic energy system modeling using hybrid physics-based and machine learning encoder–decoder models

Three model configurations are presented for multi-step time series predictions of the heat absorbed by the water and steam in a thermal power plant. The models predict over horizons of 2, 4, and 6 steps into the future, where each step is a 5-minute increment. The evaluated models are a pure machine learning model, a novel hybrid machine learning and physics-based model, and the hybrid model with an incomplete dataset. The hybrid model deconstructs the machine learning into individual boiler heat absorption units: economizer, water wall, superheater, and reheater. Each configuration uses a gated recurrent unit (GRU) or a GRU-based encoder–decoder as the deep learning architecture. Mean squared error is used to evaluate the models compared to target values. The encoder–decoder architecture is over 11% more accurate than the GRU only models. The hybrid model with the incomplete dataset highlights the importance of the manipulated variables to the system. The hybrid model, compared to the pure machine learning model, is over 10% more accurate on average over 20 iterations of each model. Automatic differentiation is applied to the hybrid model to perform a local sensitivity analysis to identify the most impactful of the 72 manipulated variables on the heat absorbed in the boiler. The models and sensitivity analyses are used in a discussion about optimizing the thermal power plant.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2021

A Secure Learning Control Strategy via Dynamic Camouflaging for Unknown Dynamical Systems under Attacks

This paper presents a secure reinforcement learning (RL) based control method for unknown linear time-invariant cyber-physical systems (CPSs) that are subjected to compositional attacks such as eavesdropping and covert attack. We consider the attack scenario where the attacker learns about the dynamic model during the exploration phase of the learning conducted by the designer to learn a linear quadratic regulator (LQR), and thereafter, use such information to conduct a covert attack on the dynamic system, which we refer to as doubly learning-based control and attack (DLCA) framework. We propose a dynamic camouflaging based attack-resilient reinforcement learning (ARRL) algorithm which can learn the desired optimal controller for the dynamic system, and at the same time, can inject sufficient misinformation in the estimation of system dynamics by the attacker. The algorithm is accompanied by theoretical guarantees and extensive numerical experiments on a consensus multi-agent system and on a benchmark power grid model.

Mukherjee, Sayak↗
DOE OSTI2024

Jamming Detection for Low-Resolution SC-FDE Systems: A Machine Learning Approach

Jammers interfere with communication between base stations (BSs) and legitimate users, leading to degradation of wireless system performance. Our study focuses on jamming detection for wideband single-carrier frequency domain equalization (SC-FDE) systems with low-resolution analog-to digital converters (ADCs). In such systems, jamming detection is challenging because traditional analytical approaches cannot be directly applied due to the delay dispersion in wideband channels and the non-linearity induced by low-resolution ADCs. We propose a machine learning (ML)-based jamming detection method that directly uses the quantized receive signals. Significantly, our ML-based detector can be integrated into existing standard frameworks, such as unique word (UW)-based SC-FDE systems, as it uses existing pilots without requiring additional pilots for jamming detection. Through numerical simulations, we show that two or more bits provide satisfactory performance compared to unquantized scenarios. Additionally, we demonstrate that using more and well-separated pilot symbols improves performance.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2025

FL‐ADS: Federated learning anomaly detection system for distributed energy resource networks

Abstract With the ongoing development of Distributed Energy Resources (DER) communication networks, the imperative for strong cybersecurity and data privacy safeguards is increasingly evident. DER networks, which rely on protocols such as Distributed Network Protocol 3 and Modbus, are susceptible to cyberattacks such as data integrity breaches and denial of service due to their inherent security vulnerabilities. This paper introduces an innovative Federated Learning (FL)‐based anomaly detection system designed to enhance the security of DER networks while preserving data privacy. Our models leverage Vertical and Horizontal Federated Learning to enable collaborative learning while preserving data privacy, exchanging only non‐sensitive information, such as model parameters, and maintaining the privacy of DER clients' raw data. The effectiveness of the models is demonstrated through its evaluation on datasets representative of real‐world DER scenarios, showcasing significant improvements in accuracy and F1‐score across all clients compared to the traditional baseline model. Additionally, this work demonstrates a consistent reduction in loss function over multiple FL rounds, further validating its efficacy and offering a robust solution that balances effective anomaly detection with stringent data privacy needs.

Purohit, Shaurya [Iowa State University Ames Iowa ↗
DOE OSTI2026

Lyapunov-Based Iterative Learning of Regions of Attraction for Autonomous Systems

This paper proposes a novel algorithm for estimating the region of attraction of equilibrium points for nonlinear discrete-time autonomous systems. The method iteratively expands an initial estimate of the region of attraction by constructing unions of sublevel sets of learned functions parametrized as neural networks. Unlike conventional techniques that rely on a single global Lyapunov function, the proposed approach provides a collection of local Lyapunov-like functions, enabling richer representations and potentially larger region of attraction estimates. These functions are trained using sampled state-space data, and their Lipschitz continuity ensures that desirable properties extend beyond the training samples. The devised strategy is tested via numerical simulations, demonstrating the effectiveness of the proposed approach.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Automatic Determination of the Need for Intravenous Contrast in Musculoskeletal MRI Examinations Using IBM Watson’s Natural Language Processing Algorithm

Magnetic resonance imaging (MRI) protocoling can be time- and resource-intensive, and protocols can often be suboptimal dependent upon the expertise or preferences of the protocoling radiologist. Providing a best-practice recommendation for an MRI protocol has the potential to improve efficiency and decrease the likelihood of a suboptimal or erroneous study. The goal of this study was to develop and validate a machine learning-based natural language classifier that can automatically assign the use of intravenous contrast for musculoskeletal MRI protocols based upon the free-text clinical indication of the study, thereby improving efficiency of the protocoling radiologist and potentially decreasing errors. We utilized a deep learning-based natural language classification system from IBM Watson, a question-answering supercomputer that gained fame after challenging the best human players on Jeopardy! in 2011. We compared this solution to a series of traditional machine learning-based natural language processing techniques that utilize a term-document frequency matrix. Each classifier was trained with 1240 MRI protocols plus their respective clinical indications and validated with a test set of 280. Ground truth of contrast assignment was obtained from the clinical record. For evaluation of inter-reader agreement, a blinded second reader radiologist analyzed all cases and determined contrast assignment based on only the free-text clinical indication. In the test set, Watson demonstrated overall accuracy of 83.2% when compared to the original protocol. This was similar to the overall accuracy of 80.2% achieved by an ensemble of eight traditional machine learning algorithms based on a term-document matrix. When compared to the second reader’s contrast assignment, Watson achieved 88.6% agreement. When evaluating only the subset of cases where the original protocol and second reader were concordant (n = 251), agreement climbed further to 90.0%. The classifier was relatively robust to spelling and grammatical errors, which were frequent. Implementation of this automated MR contrast determination system as a clinical decision support tool may save considerable time and effort of the radiologist while potentially decreasing error rates, and require no change in order entry or workflow.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2022

Deep reinforcement learning based optimization for a tightly coupled nuclear renewable integrated energy system

New ways to integrate energy systems to maximize efficiency are being sought to meet carbon emissions goals. Nuclear-renewable integrated energy system (NR-IES) concepts are a leading solution that couples a nuclear power plant with renewable energy, hydrogen generation plants, and energy storage systems, such that thermal and electrical power are dispatchable to fulfill grid-flexibility requirements while also producing hydrogen and maximizing revenue. Here, this paper introduces a deep reinforcement learning (DRL)-based framework to address the complex decision-making tasks for NR-IES. The objective is to maximize revenue by generating and selling hydrogen and electricity simultaneously according to their time-varying prices while keeping the energy flow in the subsystems in balance. A Python-based simulator for a NR-IES concept has been developed to integrate with OpenAI Gym and Ray/RLlib to enable an efficient and flexible computational framework for DRL research and development. Three state-of-the-art DRL algorithms have been investigated, including two-delayed deep deterministic policy gradient (TD3), soft-actor critic (SAC), proximal policy optimization (PPO), to illustrate DRL’s superiority for controlling NR-IES by comparing it with a conventional control approach, particle swarm optimization (PSO). In this effort, PPO has shown more-stable performance and also better generalization capability than SAC and TD3. Comparisons with PSO have demonstrated that, on average, PPO can achieve 13.9% more mean episode returns from the training process and 29.4% more mean episode returns from the testing process when different hydrogen-production targets are applied.

08 HYDROGEN↗
DOE OSTI2022

Projecting the Thermal Response in a HTGR-Type System during Conduction Cooldown Using Graph-Laplacian Based Machine Learning

Accurate prediction of an off-normal event in a nuclear reactor is dependent upon the availability of sensory data, reactor core physical condition, and understanding of the underlying phenomenon. This work presents a method to project the data from some discrete sensory locations to the overall reactor domain during conduction cooldown scenarios similar to High Temperature Gas-cooled Reactors (HTGRs). The existing models for conductive cooldown in a heterogeneous multi-body system, such as an assembly of prismatic blocks or pebble beds relies on knowledge of the thermal contact conductance, requiring significant knowledge of local thermal contacts and heat transport possibilities across those contacts. With a priori knowledge of bulk geometry features and some discrete sensors, a machine learning approach was devised. The presented work uses an experimental facility to mimic conduction cooldown with an assembly of 68 cylindrical rods initially heated to 1200 K. High-fidelity temperature data were collected using an infrared (IR) camera to provide training data to the model and validate the predicted temperature data. The machine learning approach used here first converts the macroscopic bulk geometry information into Graph-Laplacian, and then uses the eigenvectors of the Graph-Laplacian to develop Kernel functions. Support vector regression (SVR) was implemented on the obtained Kernels and used to predict the thermal response in a packed rod assembly during a conduction cooldown experiment. The usage of SVR modeling differs from most models today because of its representation of thermal coupling between rods in the core. When trained with thermographic data, the average normalized error is less than 2% over 400 s, during which temperatures of the assembly have dropped by more than 500 K. The rod temperature prediction performance was significantly better for rods in the interior of the assembly compared to those near the exterior, likely due to the model simplification of the surroundings.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2022

Data-driven cyber-attack detection for photovoltaic systems: A transfer learning approach

With increasing exposure to software-based sensing and control, power systems are facing higher risks of cyber/physical attacks. Here, to ensure system stability and minimize the potential economic losses, it is imperative to monitor the operating states and detect those attacks at the early stage. In this paper, a transfer learning method is proposed to detect cyber-attacks in photovoltaic (PV) systems with much less training data. First of all, two PV systems with a different number of PV inverters and power ratings are analyzed and their attack models are studied. Next, an attack detection Convolutional Neural Network (CNN) model was trained with rich amount of data from PV #1. Then, transfer learning was proposed to transfer the well-trained features from PV #1 to PV #2. Lastly, the attack detection model on PV #2 was trained based on the transferred CNN model. The experiment results show that the proposed transfer learning method achieves better accuracy and a faster convergence rate with a much less training dataset than conventional deep learning.

14 SOLAR ENERGY↗