Search NASA⌕ Search

SEARCH · Search NASA

Results for “Learning 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 721 records · Page 40

Data-driven reinforcement learning–based real-time energy management system for plug-in hybrid electric vehicles

Plug-in hybrid electric vehicles (PHEVs) show great promise in reducing transportation-related fossil fuel consumption and greenhouse gas emissions. Designing an efficient energy management system (EMS) for PHEVs to achieve better fuel economy has been an active research topic for decades. Most of the advanced systems rely either on a priori knowledge of future driving conditions to achieve the optimal but not real-time solution (e.g., using a dynamic programming strategy) or on only current driving situations to achieve a real-time but nonoptimal solution (e.g., rule-based strategy). This paper proposes a reinforcement learning–based real-time EMS for PHEVs to address the trade-off between real-time performance and optimal energy savings. The proposed model can optimize the power-split control in real time while learning the optimal decisions from historical driving cycles. Here, a case study on a real-world commute trip shows that about a 12% fuel saving can be achieved without considering charging opportunities; further, an 8% fuel saving can be achieved when charging opportunities are considered, compared with the standard binary mode control strategy.

33 ADVANCED PROPULSION SYSTEMS↗

Machine Learning-Based PV Reserve Determination Strategy for Frequency Control on the WECC System: Preprint

Frequency control from Photovoltaic (PV) plants has great potential to address the frequency response challenge of the power system with high renewable penetration. However, using model-based approaches to determine the optimal PV headroom reserve requires significant online computation and is intractable for an interconnection level system. This paper proposes a machine learning based strategy, that is suitable for real-time operation, to determine the optimal PV reserve for frequency control. The proposed machine learning algorithm is trained and tested on 1,987 offline simulations of a 60% renewable penetration Western Electricity Coordinating Council (WECC) system. Furthermore, the proposed reserve determination strategy is applied on a realistic one-day operation profile of the WECC system and demonstrates over 40% PV headroom saving compared to a conservative approach. It is evident that the proposed strategy can efficiently and effectively determine the optimal PV frequency control reserve for realistic interconnection systems.

41 EE - Solar Energy Technologies Office (EE-4S)↗

Enhancing segmentation fairness through curriculum learning and progressive loss: a centralized and federated perspective on radiograph analysis

Bias in medical image segmentation can lead to unequal performance across demographic subgroups, raising concerns about fairness and reliability in clinical AI systems. While deep learning models have achieved high segmentation accuracy, ensuring equitable performance across race and gender remains a significant challenge, particularly in privacy-sensitive healthcare environments. This study investigates fairness-aware medical image segmentation for hip and knee radiographs using deep learning models evaluated in both centralized and Federated Learning (FL) settings. We introduce Curriculum Learning (CL) strategies and Progressive Loss (PL) functions to regulate sample difficulty during training. In addition, we propose two novel fairness-oriented federated learning algorithms, Federated Intersection over Union (FedIoU) and Federated Intersection over Union with Outlier Analysis (FedIoUoutlier). Experiments are conducted using multiple segmentation backbones and simulated multi-site data partitions derived from the Osteoarthritis Initiative dataset. Model performance is evaluated using Intersection over Union (IoU), IoU standard deviation, Skewed Error Ratio (SER), and Min-Max Disparity across race and gender subgroups. Statistical significance was verified using paired t-tests to compare per-sample IoU performance against baseline configurations. Across both hip and knee segmentation tasks, curriculum learning and progressive loss strategies consistently improved segmentation accuracy and reduced demographic performance disparities in centralized training. In federated settings, fairness-aware aggregation further enhanced performance. Notably, FedIoUoutlier combined with balanced curriculum learning and tiered progressive loss achieved the highest mean IoU while yielding the lowest SER and Min-Max Disparity, indicating improved fairness without sacrificing accuracy. In several configurations, federated models matched or exceeded the performance of optimized centralized models, with statistically significant improvements in per-sample IoU over baseline configurations. The results demonstrate that structured training strategies and fairness-aware federated aggregation can jointly improve accuracy, stability, and demographic fairness in medical image segmentation. By integrating curriculum learning, progressive loss, and novel FL algorithms, this work provides a practical pathway toward equitable and privacy-preserving AI systems for medical imaging.

97 MATHEMATICS AND COMPUTING↗

Fault Modeling of Extreme Scale Applications Using Machine Learning

Faults are commonplace in large scale systems. These systems experience a variety of faults such as transient, permanent and intermittent. Multi-bit faults are typically not corrected by the hardware resulting in an error. Here, this paper attempts to answer an important question: Given a multi-bit fault in main memory, will it result in an application error — and hence a recovery algorithm should be invoked — or can it be safely ignored? We propose an application fault modeling methodology to answer this question. Given a fault signature (a set of attributes comprising of system and application state), we use machine learning to create a model which predicts whether a multibit permanent/transient main memory fault will likely result in error. We present the design elements such as the fault injection methodology for covering important data structures, the application and system attributes which should be used for learning the model, the supervised learning algorithms (and potentially ensembles), and important metrics. Lastly, we use three applications — NWChem, LULESH and SVM — as examples for demonstrating the effectiveness of the proposed fault modeling methodology.

97 MATHEMATICS AND COMPUTING↗

Accelerating data acquisition with FPGA-based edge machine learning: a case study with LCLS-II

New scientific experiments and instruments generate vast amounts of data that need to be transferred for storage or further processing, often overwhelming traditional systems. Edge machine learning (EdgeML) addresses this challenge by integrating machine learning (ML) algorithms with edge computing, enabling real-time data processing directly at the point of data generation. EdgeML is particularly beneficial for environments where immediate decisions are required, or where bandwidth and storage are limited. In this paper, we demonstrate a high-speed configurable ML model in a fully customizable EdgeML system using a field programmable gate array (FPGA). Our demonstration focuses on an angular array of electron spectrometers, referred to as the ‘CookieBox,’ developed for the Linac Coherent Light Source II project. The EdgeML system captures 51.2 Gbps from a 6.4 GS s −1 analog to digital converter and is designed to integrate data pre-processing and ML inside an FPGA. Our implementation achieves an inference latency of 0.2 µs for the ML model, and a total latency of 0.4 µs for the complete EdgeML system, which includes pre-processing, data transmission, digitization, and ML inference. The modular design of the system allows it to be adapted for other instrumentation applications requiring low-latency data processing.

97 MATHEMATICS AND COMPUTING↗

Validation of Maui PSCAD Model: Motivation, Methodology, and Lessons Learned: Preprint

As power systems become increasingly dependent on inverter-based resources, the dominant physical and control interactions that determine system stability may no longer be accurately captured in positive sequence dynamic simulation tools; electromagnetic transient (EMT) simulations may be needed. This paper describes the process of creating and validating an EMT model of the entire Maui island transmission system in PSCAD software. The model is validated against the utility's PSSE model and against field data for an event that consisted of a single-phase fault followed by a generation trip. The methodology and lessons learned are summarized here.

41 EE - Solar Energy Technologies Office (EE-4S)↗

Machine learning for natural resource assessment: An application to the blind geothermal systems of Nevada

A study is underway to apply machine learning methods to evaluate natural resource potential. In particular, we are considering the search for blind geothermal systems in Nevada. Beginning with the data and experience from the previous Nevada play fairway analysis project, we are building models in TensorFlow/Keras and gaining experience toward predicting the geothermal resource potential as a probability map. During the first year of this project we have encountered several issues particular to using geological and geophysical data sets with these tools. Through an illustrative example we develop a promising workflow for future use as more data become available and are analyzed.

15 GEOTHERMAL ENERGY↗

Visibility-enhanced model-free deep reinforcement learning algorithm for voltage control in realistic distribution systems using smart inverters

Increasing integration of distributed solar photovoltaic (PV) into distribution networks could result in adverse effects on grid operation. Traditional model-based control algorithms require accurate model information that is difficult to acquire and thus are challenging to implement in practice. Here, this paper proposes a surrogate model-enabled grid visibility scheme to empower deep reinforcement learning (DRL) approach for distribution network voltage regulation using PV inverters with minimal system knowledge. In contrast to existing DRL methods, this paper presents and corroborates the adverse impact of missing load information on DRL performance and, based on this finding, proposes a surrogate model methodology to impute load information utilizing observable data. Additionally, a multi-fidelity neural network is utilized to construct the DRL training environment, chosen for its efficient data utilization and enhanced robustness to data uncertainty. The feasibility and effectiveness of the proposed algorithm are assessed by considering DRL testing across varying degrees of observable load information and diverse training environments on a realistic power system.

14 SOLAR ENERGY↗

Planning for Resilient Power Distribution Systems using Risk-Based Quantification and Q-Learning

Grid hardening is one of the most effective approaches that reduce the component failures and restoration efforts thus increasing the resilience of the power systems against extreme events. However, hardening and upgrading the entire system is prohibitively expensive and hence the optimal design of a distribution network is challenging. This paper adopted a reinforcement learning algorithm to identify the optimal hardening strategy to enhance the resilience of power distribution systems. Adopting the Q-learning algorithm as the reinforcement learning technique, we found the sequential optimal action for hardening measures to enhance the grid's resilience for the given budget. To identify the optimal strategy through Q-learning, Conditional Value at Risk (CVaR) is used as a rewarding metric. A study on the IEEE 123-bus test feeder validate the effectiveness of the proposed model and show how to effectively allocate budget limited resources to plan a resilient power distribution network.

Paul, Shuva↗

A Machine Learning-Based Vulnerability Analysis for Cascading Failures of Integrated Power-Gas Systems

This article proposes a cascading failure simulation (CFS) method and a hybrid machine learning method for vulnerability analysis of integrated power-gas systems (IPGSs). The CFS method is designed to study the propagating process of cascading failures between the two systems, generating data for machine learning with initial states randomly sampled. The proposed method considers generator and gas well ramping, transmission line and gas pipeline tripping, island issue handling and load shedding strategies. Then, a hybrid machine learning model with a combined random forest (RF) classification and regression algorithms is proposed to investigate the impact of random initial states on the vulnerability metrics of IPGSs. Extensive case studies are carried out on three test IPGSs to verify the proposed models and algorithms. Simulation results show that the proposed models and algorithms can achieve high accuracy for the vulnerability analysis of IPGSs.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Sub-Area: 2.5 Demonstration of Promising Energy Storage Technologies - Project Type: Flywheel Energy Storage Demonstration (Revision V1.0)

In this program, Amber Kinetics designed, built, and tested a sub-­scale 5 kWh engineering prototype flywheel system. Applying lessons learned from the engineering prototype, Amber Kinetics then designed, built and tested full-­size, commercial-­scale 25 kWh flywheel systems. The systems underwent basic functional qualification testing before being installed, sequentially, at the company’s outdoor test site in Alameda, CA for full-­speed field-testing. The primary considerations in testing the prototype units were to demonstrate the functionality of the system, verify the frequencies of resonant modes, and quantify spinning losses and motor/generator efficiency.

42 ENGINEERING↗

DeepGraphONet: A Deep Graph Operator Network to Learn and Zero-Shot Transfer the Dynamic Response of Networked Systems

This article develops a deep graph operator network (DeepGraphONet) framework that learns to approximate the dynamics of a complex system (e.g., the power grid or traffic) with an underlying subgraph structure. Here, we build our DeepGraphONet by fusing the ability of graph neural networks to exploit spatially correlated graph information and deep operator networks to approximate the solution operator of dynamical systems. The resulting DeepGraphONet can then predict the dynamics within a given short/medium-term time horizon by observing a finite history of the graph state information. Furthermore, we design our DeepGraphONet to be resolution independent. That is, we do not require the finite history to be collected at the exact/same resolution. In addition, to disseminate the results from a trained DeepGraphONet, we design a zero-shot learning strategy that enables using it on a different subgraph. Finally, empirical results on the transient stability prediction problem of power grids and traffic flow forecasting problem of a vehicular system illustrate the effectiveness of the proposed DeepGraphONet.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Registration and fusion of large-scale melt pool temperature and morphology monitoring data demonstrated for surface topography prediction in LPBF

In-situ monitoring technologies for laser powder bed fusion (LPBF) additive manufacturing often face one key challenge, extracting the ultrafast melt pool (MP) signatures for understanding the localized part properties. Further, the spatial information of each monitored MP signature is essential for correlating the MP – part property. This spatial information is often unavailable especially from commercial LPBF printers. Many MP monitoring methods have been reported and utilized. However, very few of these have the MP’s spatial information. To overcome this challenge, in this work we report a method for spatially registering the key MP signatures (MP intensity, temperature, and area) to the monitored print parts. The MP signatures are obtained from our coaxial high-speed single-camera based two-wavelength imaging pyrometry (STWIP) system and the MP spatial information is obtained from an off-axis camera system. A machine learning aided image analysis method is employed to retrieve the spatial distribution of MPs within the corresponding part’s coordinates system. Then, the MP signature maps (MPSMs) are reconstructed by mapping the STWIP measured MP signatures to the registered MP coordinates. Further, a long short-term memory (LSTM) neural network is developed for estimating the layer surface topography from the registered MPSMs. The obtained results indicate that the layer surface topography can be more accurately estimated by using MP temperature signature rather than MP intensity and/or area signatures as in common practice. Finally, our developed methods for MP monitoring, registration, and MP-surface topography prediction offer advanced capabilities for the online detection of process anomalies and part defects.

36 MATERIALS SCIENCE↗

Short Term Plasticity for Artificial Neural Networks

Achieving efficient learning for AI systems was identified as a major challenge in the DOE's recently released, AI for Science , report. The human brain is capable of efficient and low-powered learning. It is likely that implementing brain-like principles will lead to more efficient AI systems. In this LDRD, I aim to contribute to this goal by creating a foundation for implementing and studying a brain phenomenon termed short term plasticity (STP) in spiking artificial neural networks within Sandia. First, data collected by the Allen Institute for Brain Science (AIBS) was analyzed to see if STP could be classified into types using the data collected. Although the data was inadequate at the time, AIBS has updated their database and created models that could be utilized in the future. Second, I began creating a software package to assess the ability of a Boltzmann machine utilizing STP to sample from national security data.

59 BASIC BIOLOGICAL SCIENCES↗

Interactive Machine Learning at Scale with CHISSL

We demonstrate CHISSL, a scalable client-server system for real-time interactive machine learning. Our system is capa- ble of incorporating user feedback incrementally and imme- diately without a structured or pre-defined prediction task. Computation is partitioned between a lightweight web-client and a heavyweight server. The server relies on representation learning and agglomerative clustering to learn a dendrogram, a hierarchical approximation of a representation space. The client uses only this dendrogram to incorporate user feedback into the model via transduction. Distances and predictions for each unlabeled instance are updated incrementally and deter- ministically, with O(n) space and time complexity. Our al- gorithm is implemented in a functional prototype, designed to be easy to use by non-experts. The prototype organizes the large amounts of data into recommendations. This allows the user to interact with actual instances by dragging and drop- ping to provide feedback in an intuitive manner. We applied CHISSL to several domains including cyber, social media, and geo-temporal analysis.

Arendt, Dustin L.↗

Learning emergent partial differential equations in a learned emergent space

We propose an approach to learn effective evolution equations for large systems of interacting agents. This is demonstrated on two examples, a well-studied system of coupled normal form oscillators and a biologically motivated example of coupled Hodgkin-Huxley-like neurons. For such types of systems there is no obvious space coordinate in which to learn effective evolution laws in the form of partial differential equations. In our approach, we accomplish this by learning embedding coordinates from the time series data of the system using manifold learning as a first step. In these emergent coordinates, we then show how one can learn effective partial differential equations, using neural networks, that do not only reproduce the dynamics of the oscillator ensemble, but also capture the collective bifurcations when system parameters vary. The proposed approach thus integrates the automatic, data-driven extraction of emergent space coordinates parametrizing the agent dynamics, with machine-learning assisted identification of an emergent PDE description of the dynamics in this parametrization.

97 MATHEMATICS AND COMPUTING↗

Learning physics-based reduced-order models from data using nonlinear manifolds

Here we present a novel method for learning reduced-order models of dynamical systems using nonlinear manifolds. First, we learn the manifold by identifying nonlinear structure in the data through a general representation learning problem. The proposed approach is driven by embeddings of low-order polynomial form. A projection onto the nonlinear manifold reveals the algebraic structure of the reduced-space system that governs the problem of interest. The matrix operators of the reduced-order model are then inferred from the data using operator inference. Numerical experiments on a number of nonlinear problems demonstrate the generalizability of the methodology and the increase in accuracy that can be obtained over reduced-order modeling methods that employ a linear subspace approximation.

97 MATHEMATICS AND COMPUTING↗