SEARCH · Engineering Papers
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.
Public release of a version of the cross-reference capability of the UBID software that is feature-limited to cross-reference via Jaccard similarity coefficient (viz., intersection-over-union or "IoU") [2] over a quadtree-based spatial index only.
This scoping study synthesizes information that will help stakeholders understand the scope, scale, and drivers of recent increases in investor-owned utility (IOU) expenditures on local distribution power grids, while providing regulators and other decision-makers with potential strategies to keep electricity bills down. The study includes five distinct components. Drawing first on data from FERC Form 1, it summarizes key trends in past and recent IOU distribution costs. Next, through a review of a sample of distribution-system plans, it characterizes material drivers of planned distribution expenditures. Ultimately, regulators must approve cost recovery for IOU expenditures, including those for the distribution system. The study therefore also: examines trends in utility requests and regulatory approvals related to changes in retail rates and return on equity; identifies areas where utility shareholder and customer incentives may be misaligned; and develops a menu of options that state regulators might consider to optimize distribution system expenditures. Some of the key findings include: - IOU distribution spending at a national level has grown by 6%/yr since 2014 in real dollar terms, 4x faster than in the prior 20 years and consisting mostly of capital (not operating) expenditure. - On a per-kWh basis, increases in IOU distribution costs since 2014 represent over 30% of the overall national-average increase in retail electricity rates. - Regional spending growth has ranged from 2-8%/yr, with larger estimated rate impacts in CAISO, then NYISO & ISO-NE, and then the Southeast, MISO & PJM (see figure). - Some utilities are planning for significantly increased distribution system spending. Planned spending on managing the existing system (asset replacement, safety & reliability, and resilience are all important drivers) exceeds that for capacity expansion. - IOU rate increase requests ($18 billion in 2025) and public utility commission (PUC) approval levels (average of 64% of requested amounts from 2021-2025) have recently hit multi-decadal highs. - PUCs in New England and the Southeast have recently approved a greater fraction of rate requests (>75%, on average) than in ther regions, while PUCs in California and the Southeast have generally authorized higher equity returns than in other regions. - Regulators have many tools to tackle potential misalignments between utility and customer interests and, more specifically, to optimize and reduce distribution costs. Shorter-term options include those related to return on equity, capital structure, depreciation, trackers, construction work in progress, and securitization. Longer-term options include performance-based regulation and a wide variety of planning-related requirements. All options embed important tradeoffs.
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.
Zero-shot and prompt-based models have excelled at visual reasoning tasks by leveraging large-scale natural image corpora, but they often fail on sparse and domain-specific scientific image data. We introduce Zenesis, a no-code interactive computer vision platform designed to reduce data readiness bottlenecks in scientific imaging workflows. Zenesis integrates lightweight multimodal adaptation for zero-shot inference on raw scientific data, human-in-the-loop refinement, and heuristic-based temporal enhancement. We validate our approach on Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) datasets of catalyst-loaded membranes. Zenesis outperforms baselines, achieving an average accuracy of 0.947, Intersection over Union (IoU) of 0.858, and Dice score of 0.923 on amorphous catalyst samples; and 0.987 accuracy, 0.857 IoU, and 0.923 Dice on crystalline samples. These results represent a significant performance gain over conventional methods such as Otsu thresholding and standalone models like the Segment Anything Model (SAM). Zenesis enables effective image segmentation in domains where annotated datasets are limited, offering a scalable solution for scientific discovery.
Electricity prices reported in most electricity-system planning studies leave out many price components and do not translate into retail rates, making it difficult to interpret how projected electricity system changes will impact costs to consumers. Full transmission costs are left out of many studies; distribution and administration costs are similarly excluded or highly simplified. Here, we present a detailed bottom-up accounting method for projecting future retail electricity rates in the United States. Making the simplifying assumption that each state is served by an investor-owned utility (IOU), we translate projected generation and transmission capacity and costs from the Regional Energy Deployment System (ReEDS) capacity-expansion model into IOU balance sheet expenditures, accounting for depreciation, taxes, and the breakdown between operating and capitalized (rate-based) expenses. Distribution, administration, and intra-regional transmission costs are projected forward based on empirical trends over the past decade. Modeled bottom-up electricity rates are compared to historical rates from 2010-2019, and the sensitivity of modeled rates to a range of financing and modeling assumptions is explored. Distribution and administration rate components account for roughly 40% (4.4 ¢/kWh) of the projected national-average retail rate over 2020-2050 under central assumptions.
Working together, Sandia National Laboratories, Southern California Edison (SCE) - an Investor-Owned Utility (IOU) - and the California Public Utilities Commission (CPUC) are studying how electric utilities can use equity and resilience metrics to help inform the prioritization and sequencing of resilience-driven infrastructure investments. To this end, this project evaluated “Social Burden,” an equitable resilience metric which measures the potential impact of disruptions in access to non-electric critical services on people and estimates community resilience to these disruptions. The Social Burden was expanded to incorporate SCE’s existing equity metric and applied to evaluate the potential impacts from a range of climate-informed hypothetical outage scenarios developed under SCE’s 2022 Climate Adaptation Vulnerability Assessment. One baseline (“blue-sky”) state and eight different outage scenarios were evaluated to measure the potential impacts of the outages on non-electric infrastructure, critical services, and people. Key findings include: 1) the Social Burden framework is flexible enough to adapt to and build upon existing utility equity and/or resilience metrics, 2) Social Burden results highlight the high degree of non-electric service redundancy within the SCE service area with most (6/8) hypothetical outage scenarios predicted to increase people’s Social Burden by less than 10%; however, 3) access to critical services and people’s ability to obtain them is unequal and spatially clustered, meaning that there are some hypothetical outage scenarios (2/8) that will exert a higher toll on communities directly experiencing the outage as well as some nearby communities with pre-existing vulnerabilities. The report concludes with recommendations for potential use cases of the expanded Social Burden metric and identifies priority follow-on work. Potential use cases may include incorporating equity into IOU’s prioritization of climate resilience investments. Additionally, Social Burden analysis may provide additional data and insights to augment grid planning, potentially by identifying additional needs and/or prioritizing previously identified needs.
Abstract. In the western United States, prolonged drought, a warming climate, and historical fuel buildup have contributed to larger and more intense wildfires as well as to longer fire seasons. As these costly wildfires become more common, new tools and methods are essential for improving our understanding of the evolution of fires and how extreme weather conditions, including heat waves, windstorms, droughts, and varying levels of active-fire suppression, influence fire spread. Here, we develop the Geostationary Operational Environmental Satellites (GOES)-Observed Fire Event Representation (GOFER) algorithm to derive the hourly fire progression of large wildfires and create a product of hourly fire perimeters, active-fire lines, and fire spread rates. Using GOES-East and GOES-West geostationary satellite detections of active fires, we test the GOFER algorithm on 28 large wildfires in California from 2019 to 2021. The GOFER algorithm includes parameter optimizations for defining the burned-to-unburned boundary and correcting for the parallax effect from elevated terrain. We evaluate GOFER perimeters using 12 h data from the Visible Infrared Imaging Radiometer Suite (VIIRS)-derived Fire Event Data Suite (FEDS) and final fire perimeters from the California's Fire and Resource Assessment Program (FRAP). Although the GOES imagery used to derive GOFER has a coarser resolution (2 km at the Equator), the final fire perimeters from GOFER correspond reasonably well to those obtained from FRAP, with a mean Intersection-over-Union (IoU) of 0.77, in comparison to 0.83 between FEDS and FRAP; the IoU indicates the area of overlap over the area of the union relative to the reference perimeters, in which 0 is no agreement and 1 is perfect agreement. GOFER fills a key temporal gap present in other fire tracking products that rely on low-Earth-orbit imagery, where perimeters are available at intervals of 12 h or longer or at ad hoc intervals from aircraft overflights. This is particularly relevant when a fire spreads rapidly, such as at maximum hourly spread rates of over 5 km h−1. Our GOFER algorithm for deriving the hourly fire progression using GOES can be applied to large wildfires across North and South America and reveals considerable variability in the rates of fire spread on diurnal timescales. The resulting GOFER product has a broad set of potential applications, including the development of predictive models for fire spread and the improvement of atmospheric transport models for surface smoke estimates. The resulting GOFER product has a broad set of potential applications, including the development of predictive models for fire spread and the improvement of atmospheric transport models for surface smoke estimates (https://doi.org/10.5281/zenodo.8327264, Liu et al., 2023).
Effective monitoring of global water resources is increasingly critical due to climate change and population growth. Advancements in remote sensing technology, specifically in spatial, spectral, and temporal resolutions, are revolutionizing water resource monitoring, leading to more frequent and high-quality surface water extent maps using various techniques such as traditional image processing and machine learning algorithms. However, satellite imagery datasets contain trade-offs that result in inconsistencies in performance, such as disparities in measurement principles between optical (e.g., Sentinel-2) and radar (e.g., Sentinel-1) sensors and differences in spatial and spectral resolutions among optical sensors. Therefore, developing accurate and robust surface water mapping solutions requires independent validations from multiple datasets to identify potential biases within the imagery and algorithms. However, high-quality validation datasets are expensive to build, and few contain information on water resources. For this purpose, we introduce a globally sampled, high-spatial-resolution dataset labeled using 3 m PlanetScope imagery. Our surface water extent dataset comprises 100 images, each with a size of 1024×1024 pixels, which were sampled using a stratified random sampling strategy covering all 14 biomes. We highlighted urban and rural regions, lakes, and rivers, including braided rivers and coastal regions. We evaluated two surface water extent mapping methods using our dataset – Dynamic World, based on Sentinel-2, and the NASA IMPACT model, based on Sentinel-1. Dynamic World achieved a mean intersection over union (IoU) of 72.16 % and F1 score of 79.70 %, while the NASA IMPACT model had a mean IoU of 57.61 % and F1 score of 65.79 %. Performance varied substantially across biomes, highlighting the importance of evaluating models on diverse landscapes to assess their generalizability and robustness. Our dataset can be used to analyze satellite products and methods, providing insights into their advantages and drawbacks. Our dataset offers a unique tool for analyzing satellite products, aiding the development of more accurate and robust surface water monitoring solutions. The dataset can be accessed via https://doi.org/10.25739/03nt-4f29.
Accurate building footprint extraction is critical for applications ranging from population estimation to disaster management. Although optical imagery provides detailed spectral information, it often struggles with shadows, occlusions, and background clutter in dense urban environments. Lidar data, by contrast, offer precise elevation and structural attributes but face challenges such as variable point density and noise. This study integrates multispectral imagery from the U.S. Department of Agriculture (USDA) National Agriculture Imagery Program (NAIP) with lidar-derived feature height and intensity from the U.S. Geological Survey (USGS) 3D Elevation Program (3DEP) to improve footprint extraction using a U-Net–based deep learning model. A six-band input stack (RGB, near-infrared, height, intensity) was developed, normalized, and tiled for training and evaluation against Microsoft Global Building Footprints (GBF). Results from the Houston, TX test site show that the six-band model achieved a precision of 0.86, recall of 0.88, F1 score of 0.87, and Intersection-over-Union (IoU) of 0.76, consistently outperforming four-band baselines by reducing false positives while maintaining sensitivity. Predictions on withheld Houston tiles confirmed strong within-region generalization, yielded a precision of 0.78, recall of 0.81, F1 score of 0.79, and IoU of 0.66. Qualitative analysis further revealed limitations stemming from both training label quality and vegetation–building confusion. These findings demonstrate the complementary value of integrating spectral and structural information for robust building footprint extraction and how domain adaptation strategies can be used to enhance cross-regional transferability.
There are two main algorithmic approaches to autonomous driving systems: (1) An end-to-end system in which a single deep neural network learns to map sensory input directly into appropriate warning and driving responses. (2) A mediated hybrid recognition system in which a system is created by combining independent modules that detect each semantic feature. While some researchers believe that deep learning can solve any problem, others believe that a more engineered and symbolic approach is needed to cope with complex environments with less data. Deep learning alone has achieved state-of-the-art results in many areas, from complex gameplay to predicting protein structures. In particular, in image classification and recognition, deep learning models have achieved accuracies as high as humans. But sometimes it can be very difficult to debug if the deep learning model doesn't work. Deep learning models can be vulnerable and are very sensitive to changes in data distribution. Generalization can be problematic. It's usually hard to prove why it works or doesn't. Deep learning models can also be vulnerable to adversarial attacks. Here, we combine deep learning-based object recognition and tracking with an adaptive neurosymbolic network agent, called the Non-Axiomatic Reasoning System (NARS), that can adapt to its environment by building concepts based on perceptual sequences. We achieved an improved intersection-over-union (IOU) object recognition performance of 0.65 in the adaptive retraining model compared to IOU 0.31 in the COCO data pre-trained model. We improved the object detection limits using RADAR sensors in a simulated environment, and demonstrated the weaving car detection capability by combining deep learning-based object detection and tracking with a neurosymbolic model.
A novel image moment-based model for shape estimation and tracking of an extended target moving with a complex trajectory is presented. The proposed extended object tracking algorithm is based on multiple noisy measurement points sampled from the target at each time step. The shape of the object, approximated by an ellipse, is estimated using a combination of image moments. Dynamic models of image moments for constant velocity and coordinated turn motions are mathematically derived. An unscented Kalman filter - interacting multiple model (UKF-IMM) method is used to track the object and estimate its shape. A likelihood function based on average log-likelihood is derived for the IMM filter. Simulation results of the proposed UKF-IMM algorithm with the image momentbased models are presented that show the estimation of the shape of the object moving in a complex trajectory. The intersection over union (IoU), and the root mean square errors (RMSEs) of the position and velocity of the centroid of the ellipse are used as metrics. The comparison results of the proposed algorithm with a benchmark algorithm from literature based on the IoU and RMSE metrics are presented.
Four state-of-the-art Deep Learning-based Convolutional Neural Networks (DCNN) were applied to automate the semantic segmentation of a 3D Transmission x-ray Microscopy (TXM) nanotomography image data. The standard U-Net architecture as baseline along with UNet++, PSPNet, and DeepLab v3+ networks were trained to segment the microstructural features of an AA7075 micropillar. A workflow was established to evaluate and compare the DCNN prediction dataset with the manually segmented features using the Intersection of Union (IoU) scores, time of training, confusion matrix, and visual assessment. Comparing all model segmentation accuracy metrics, it was found that using pre-trained models as a backbone along with appropriate training encoder-decoder architecture of the Unet++ can robustly handle large volumes of x-ray radiographic images in a reasonable amount of time. This opens a new window for handling accurate and efficient image segmentation of in situ time-dependent 4D x-ray microscopy experimental datasets.
Part qualification is often a critical and labor-intensive process in additive manufacturing, particularly in the detection of defects such as porosity, which stands to benefit significantly from advancements in machine learning. We present a deep learning approach for quantifying and localizing ex-situ porosity within Laser Powder Bed Fusion fabricated samples utilizing in-situ thermal image monitoring data. Our goal is to build the real time porosity map of parts based on thermal images acquired during the build. The quantification task builds upon the established Convolutional Neural Network model architecture to predict pore count and the localization task leverages the spatial and temporal attention mechanisms of the novel Video Vision Transformer model to indicate areas of expected porosity. Our model for porosity quantification achieved a R 2 score of 0.57 and our model for porosity localization produced an average Intersection over Union (IoU) score of 0.32 and a maximum of 1.0. This work is setting the foundations of part porosity “Digital Twins” based on additive manufacturing monitoring data and can be applied downstream to reduce time-intensive post-inspection and testing activities during part qualification and certification. In addition, we seek to accelerate the acquisition of crucial insights normally only available through ex-situ part evaluation by means of machine learning analysis of in-situ process monitoring data.
Segmentation of scanning electron microscopy (SEM) images is critical yet time-consuming for geological analyses, as it needs to differentiate the boundaries for different mineral objects to facilitate subsequent analyses, such as porosity calculation. Recently, various machine learning methods, especially convolutional neural networks (CNNs), have been explored to segment SEM images of fine-grained shale samples. However, we found that general CNNs do not yield optimal performance due to insufficient training data and imbalanced objects in SEM images. This work has revised the U-Net architecture, a popular approach for biomedical image analyses, by incorporating a loss function that addresses the imbalance issue. Furthermore, we used the ensemble learning method to train multiple models and combined the results to improve the overall performance of segmentation. We prepared 2162 sub-images from raw SEM images in our experiments and divided them into training, validation, and testing datasets. The overall results show that our method improves the average Intersection over Union (IOU) of mineral objects from 0.49 to 0.58, compared to the original U-Net model. Our method can clearly distinguish each object from others with boundaries, even in highly imbalanced images. Training our models takes less than three minutes using a single GPU, while manual labeling can take up to three hours for each image. Furthermore, the method helps geoscientists gain insights quickly and effectively by building neural network models from a small dataset of SEM images.
Improvements to synchrotron-based micro-computed tomography scanning capabilities have gifted researchers the ability to characterize 4D material thermomechanical responses more thoroughly than ever before. These advancements, however, have brought about new challenges in analyzing the resulting deluge of data. We report on a nickel-based superalloy specimen imaged 26 times in-situ during cyclic loading at Argonne National Laboratory Advanced Photon Source beamline 1ID, in order to monitor crack growth within the microstructure. Therefore, several deep learning approaches which utilize convolutional neural networks are implemented to segment crack features from reconstructed tomography scans. U-Net architecture implementations are found to be especially effective, achieving IoU = 0.995 +/- 0.004 and Matthews correlation coefficient scores of Φ = 0.826 +/- 0.085. These advancements broaden possibilities for scientists seeking to automate segmentation analyses of similar large datasets.
Image segmentation is a critical enabler for tasks ranging from medical diagnostics to autonomous driving. However, the correct segmentation semantics — where are boundaries located? what segments are logically similar? — change depending on the domain, such that state-of-the-art foundation models can generate meaningless and incorrect results. Moreover, in certain domains, fine-tuning and retraining techniques are infeasible: obtaining labels is costly and time-consuming; domain images (micrographs) can be exponentially diverse; and data sharing (for third-party retraining) is restricted. To enable rapid adaptation of the best segmentation technology, we propose the concept of semantic boosting: given a zero-shot foundation model, guide its segmentation and adjust results to match domain expectations. Here, we apply semantic boosting to the Segment Anything Model (SAM) to obtain microstructure segmentation for transmission electron microscopy. Our booster, SAM-I-Am, serves as a post-processing engine that extracts geometric and textural features of various intermediate masks to perform mask removal and mask merging operations. We demonstrate a zero-shot performance increase of (absolute) +21.35%, +12.6%, +5.27% in mean IoU, and a -9.91%, -18.42%, -4.06% drop in mean false positive masks across images of three difficulty classes over vanilla SAM (ViT-L).
The fast and reliable layer identification of two-dimensional transition metal dichalcogenide (TMD), such as WSe 2 , is essential to investigating their thickness-dependent electronic and optical properties. This article presents efficient optical image thresholding methodology designed to segment the mono, bi, and tri-layer regions of WSe 2 flakes mechanically exfoliated onto a SiO 2 /Si substrate. The optical images were first preprocessed to exclude the background effect and analyzed using the pixel medians and interquartile ranges for fundamental color channels—red, green, and blue (RGB). The analysis of red channel pixel intensities yielded three distinct ranges, serving as thresholds for layer segmentation: monolayer (111.0–118.0), bilayer (103.0–110.0), and tri-layer (93.0–103.0). Similarly, thresholds were established for each color channel, facilitating a comparative study of the segmentation performances. Further, the intersection-over-union ($IoU$) calculations revealed that the red and green channels demonstrated greater than 99 % and 90 % accuracy in differentiating each layer, respectively. This approach yields remarkable results without substantial data calibration that utilizes time-intensive heuristic techniques. Moreover, the proposed methodology offers the flexibility to compare performances across different color channels, expanding the applicability for other 2D material systems.
Imaging nanoscale radiation-induced defects using the transmission electron microscope (TEM) is a key factor in the successful implementation of materials for nuclear energy structural applications. Analyzing each defect in a TEM micrograph is currently a manual task. To identify the defects in a single image can take anywhere from 15 min to an hour and a project can require the analysis of anywhere from tens to ≥ 100 images. Here, we use artificial intelligence (AI) models to automate this task. For simplification, we evaluated images with only a single type of defect; helium bubbles. Additionally, we performed semantic segmentation of these helium bubble defects in electron microscopy images of irradiated FeCrAl alloys using a deep learning DefectSegNet model. This model, which was previously used to classify crystal defects, is inspired by the classic DenseNet and U-Net image segmentation models. It claims high spatial resolution, but has poor performance at object boundaries. Our paper improves the DefectSegNet model’s application by adding two new features. First, the DefectSegNet model is applied not only to perform calculation pixel-wise but also object (or feature) wise. Because object-wise metrics are directly relevant to our final goal of detecting bubbles, whereas pixel-wise classification is only an intermediate step, it is an important part of our study. Second, a distance map loss (DML) function has been added to increase its performance at object boundaries. It is crucial to accurately represent defects boundaries, especially bubbles, in order to track the bubble-induced swelling caused by irradiation. The boundary-focused DML function is also compared to other loss functions like Cross-entropy, Weighted Binary Cross Entropy (WBCE), Dice and Intersection over Union (IOU). Finally, by incorporating new features, we found a marked improvement on segmentation quality and better shape preservation at the boundaries and areas of the bubbles.