Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Sparsity”

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 523 records · Page 29

Weekly Mapping of Sea Ice Freeboard in the Ross Sea from ICESat-2

NASA’s ICESat-2 has been providing sea ice freeboard measurements across the polar regions since October 2018. In spite of the outstanding spatial resolution and precision of ICESat-2, the spatial sparsity of the data can be a critical issue for sea ice monitoring. This study employs a geostatistical approach (i.e., ordinary kriging) to characterize the spatial autocorrelation of the ICESat-2 freeboard measurements (ATL10) to estimate weekly freeboard variations in 2019 for the entire Ross Sea area, including where ICESat-2 tracks are not directly available. Three variogram models (exponential, Gaussian, and spherical) are compared in this study. According to the cross-validation results, the kriging-estimated freeboards show correlation coefficients of 0.56–0.57, root mean square error (RMSE) of ~0.12 m, and mean absolute error (MAE) of ~0.07 m with the actual ATL10 freeboard measurements. In addition, the estimated errors of the kriging interpolation are low in autumn and high in winter to spring, and low in southern regions and high in northern regions of the Ross Sea. The effective ranges of the variograms are 5–10 km and the results from the three variogram models do not show significant differences with each other. The southwest (SW) sector of the Ross Sea shows low and consistent freeboard over the entire year because of the frequent opening of wide polynya areas generating new ice in this sector. However, the southeast (SE) sector shows large variations in freeboard, which demonstrates the advection of thick multiyear ice from the Amundsen Sea into the Ross Sea. Thus, this kriging-based interpolation of ICESat-2 freeboard can be used in the future to estimate accurate sea ice production over the Ross Sea by incorporating other remote sensing data.

Satellite altimeter↗

Bayesian Model Selection for Reducing Bloat and Overfitting in Genetic Programming for Symbolic Regression

When performing symbolic regression using genetic programming, overfitting and bloat can negatively impact generalizability and interpretability of the resulting equations as well as increase computation times. A Bayesian fitness metric is introduced and its impact on bloat and overfitting during population evolution is studied and compared to common alternatives in the literature. The proposed approach was found to be more robust to noise and data sparsity in numerical experiments, guiding evolution to a level of complexity appropriate to the dataset. Further evolution of the population resulted not in overfitting or bloat, but rather in slight simplifications in model form. The ability to identify an equation of complexity appropriate to the scale of noise in the training data was also demonstrated. In general, the Bayesian model selection algorithm was shown to be an effective means of regularization which resulted in less bloat and overfitting when any amount of noise was present in the training data.

G F Bomarito↗

Bayesian Model Selection for Reducing Bloat and Overfitting in Genetic Programming for Symbolic Regression

When performing symbolic regression using genetic programming, overfitting and bloat can negatively impact generalizability and interpretability of the resulting equations as well as increase computation times. A Bayesian fitness metric is introduced and its impact on bloat and overfitting during population evolution is studied and compared to common alternatives in the literature. The proposed approach was found to be more robust to noise and data sparsity in numerical experiments, guiding evolution to a level of complexity appropriate to the dataset. Further evolution of the population resulted not in overfitting or bloat, but rather in slight simplifications in model form. The ability to identify an equation of complexity appropriate to the scale of noise in the training data was also demonstrated. In general, the Bayesian model selection algorithm was shown to be an effective means of regularization which resulted in less bloat and overfitting when any amount of noise was present in the training data.

Uncertainty quantification↗

Transonic Flutter Dips of the AGARD 445.6 Wing

The AGARD 445.6 configuration is the most popular validation test case for transonic flutter predictions, but the actual extent of truly nonlinear transonic flow for this case is unclear, due to the sparsity of the experimental data, and the thin profile of the wing. This work utilizes a combination of mesh adaptation and the linearized frequency-domain method to obtain high-quality viscous and inviscid flutter predictions; these solutions show a double flutter dip through the transonic Mach range driven by complex shock growth across the wing. A single experimental flutter point lies in this flutter dip area, which is not enough to assess the accuracy of these transonic flutter predictions. Modeling the boundary layer of the wind tunnel wall (as opposed to the commonly-assumed symmetry wall assumption) appears to have a large impact on the predicted flutter boundary, but true mesh convergence of this scenario is a challenge.

Bret K Stanford↗

Computational Study of Low-Speed Boundary-Layer Laminar-Turbulent Transition over a 30º Swept Plate in a Channel

Reliable boundary-layer laminar-turbulent transition predictions are critical for the design of aircraft that use natural or controlled laminar flow to achieve low drag and better fuel efficiency. During cruise flight through low turbulence, the three-dimensional boundary layer over swept surfaces can transition prematurely due to stationary crossflow instabilities induced by surface roughness. Parabolized stability equations analyses have been successfully used to explain the different phases of the transition process under well controlled perturbations such as steps or discrete roughness elements. However, methodologies are still lacking to model the boundary layer receptivity to distributed random roughness, and the following non-linear instability stages leading to transition. For engineering applications, several semi-empirical methods based on linear amplification factors or local helicity have found growing adoption, but their reliability is limited by their level of simplification and the sparsity of data they are based on. To address this gap, a new study of the effect of distributed random surface roughness on crossflow transition was started, using a 30◦ swept plate as a test vehicle. The plate, which is mounted in the 2 Foot × 3 Foot Low-Speed Boundary-Layer Channel at the NASA Langley Research Center, has been tested with micro-scale roughness of varied geometry near the leading edge. In order to analyze the flow in the test section and prepare Direct Numerical Simulations with resolved roughness, Reynolds-Averaged Navier-Stokes (RANS) simulations of the flow over the plate installed in the wind tunnel test section were performed using the Langtry-Menter transition model with stationary crossflow extension. This paper presents the results of these simulations, with comparison to the experiment.

ARMD↗

A NASA GISTEMP Observational Uncertainty Ensemble: Regional and Monthly Uncertainty

The historical global temperature record is an essential data product for quantifying the variability and change of the Earth system. In recent years, better characterization of observational uncertainty in global and hemispheric trends has become available, but the methodologies are not necessarily applicable to analyses at smaller regional areas, or monthly means, where station sparsity and other systematic issues contribute to greater uncertainty. This work details a gridded uncertainty ensemble of historical temperature anomalies from the Goddard Institute for Space Studies (GISS) Surface Temperature product (GISTEMP) product. This ensemble characterizes the complex spatial and temporal correlation structure of uncertainty in gridded historical temperature, enabling proper uncertainty propagation for climate and social science at regional and monthly scales. This work details the methodology for generating the uncertainty ensemble, key statistics of the uncertainty evolution over space and time, and provides best practices for using the uncertainty ensemble in future studies. Summary statistics from the uncertainty ensemble are in good agreement with production GISTEMP. Two applications of the uncertainty ensemble are also presented. First, the warmest year on record is shown to most likely be 2016 with a 53.2% chance and 2020 as the second most likely with a 44.4% chance. Second, it is shown that the arctic is warming 2.5 - 5 times faster than the globe, significantly faster than the regularly quoted twice as fast.

GISTEMP↗

A Machine Learning Approach to Improve Air Traffic Management Initiatives

Collaborating closely with commercial air carriers and related organizations, the Federal Aviation Administration(FAA) regulates air traffic and ensures the safety and efficiency of air operations. Air traffic controllers make strategic decisions, such as delaying, rerouting, or canceling flights, partly based on guidance provided by the FAA’s Air TrafficControl System Command Center (ATCSCC). The guidance includes, among other things, control measures known asTraffic Management Initiatives (TMIs) designed to enhance safety and improve operational efficiency. TMIs play a crucial role in managing the demand and capacity within the U.S. National Airspace System (NAS). Two major TMIs that are routinely used (primarily to mitigate the adverse effects of bad weather) are Ground Delay Programs (GDPs) andGround Stops (GSs). In a GDP, flights destined for airports facing thunderstorm activity experience delays at their origin airports. This proactive approach minimizes the risk of routing aircraft through hazardous weather conditions and also replaces (fuel burning) airborne delays with ground delays. In a GS, a temporary restriction is imposed on the departure or arrival of aircraft at a specific airport or within a designated airspace. Although other TMIs (e.g., miles-in-trail) are also implemented as part of (air) traffic flow management in the NAS, the focus of this work is on GDPs and GSs. Since TMIs, by design, lead to flight delays or cancellations, it is crucial to put in place the right set of parameters(e.g., scope and duration of the GDP). For example, when the end time of a GDP extends beyond what is necessary, it imposes unnecessary delays on departing flights. This situation could occur as a result of inaccurate prediction of the(required) duration of the GDP based on the weather forecast. On the other hand, if a GDP ends prematurely before the underlying capacity constraints are resolved at the destination airport, it may result in airborne holding. The delicate balance lies in matching the termination of the GDP precisely with the resolution of capacity constraints, avoiding both the imposition of unnecessary ground delays and the need for airborne holding due to premature program termination.Failing to specify the right parameters for TMIs also leads to flight delays, creating a significant obstacle in managing the increasing traffic volumes causing increased work load for the controllers. To address this issue, we propose the integration of Machine Learning (ML) models in the traffic flow management(TFM) pipeline. In current operations, decisions are made by human experts based on extensive training, historical patterns, available traffic and weather data. Since we have an abundance of data from past events that tell us the likely impact of various TMIs, by ingesting historical data, properly trained ML models can offer valuable insights and aid human decision-making. With the FAA increasingly exploring advanced analytics, ML emerges as a focal point for enhancing TFM within the National Airspace System (NAS). As a first step, this study aims to provide traffic controllers with decision-making support for the issuance and adjustment of TMIs. Data analytics and machine learning have been previously employed to address some of the challenges associated with TMIs. Numerous studies have concentrated on various facets of TMI issuance, exploring factors influencing TMI parameters, including arrival rate, airport capacity, and delay prediction. For example, using weather forecasts, several statistical methods were used to produce probabilistic capacity profiles which in conjunction with deterministic models provided insights into the GDP planning process [1–4]. The downside of using deterministic models is that they rely on fixed inputs and predetermined rules, which lack the ability to account for the inherent uncertainty and variability present in real-world scenarios. In a separate series of studies, researchers aimed to predict the occurrences of GDPs and GSs. The majority of these studies utilized various supervised learning methods, including Decision Trees, Naive Bayes, Support VectorMachines, and Random Forests to analyze the influence of weather conditions and arrival demand on TMI incidents[5–8]. However, these studies primarily focused on predicting the incidence of TMIs without explicitly addressing the scope of TMIs, including their duration and their geographical coverage. Furthermore, the emphasis of these studies was largely on GDPs, given their higher frequency and longer duration when compared to GSs. A limited number of studies focused on predicting the parameters of TMIs, specifically addressing their duration and extent. In one such study focusing on optimizing the TMI parameters at San Francisco International Airport (SFO),the authors utilized a probabilistic forecast of fog [9]. They simulated various capacity scenarios based on the (fog)burn-off forecasts, selecting GDP parameters that minimized airborne and overall ground delays. However, this approach exclusively emphasizes stratus (fog) burn-off as the primary determinant of GDP and GS, neglecting other influential factors like severe weather events, runway closures, lower capacity than traffic demand, and other important variables. Given the complexity of predicting the TMI and determining its scope, we seek a more holistic approach. We aim to consider all significant factors that could impact TMIs and their parameters. What sets this research apart is the fusion of all data sources relevant to the issuance and adjustment of TMIs and it represents the first comprehensive attempt to optimize TMIs in this manner. Since this comprehensive solution involves various aspects, we break down the problem into smaller components and input all parameters into a unified model called the “TMI Adjuster”. Figure 1 shows the overall framework and the list of datasets used in each model. The objective of the TMI Adjuster module is to deliver reliable, consistent and expedited recommendations for the progression, adjustment, and termination of TMIs. The ML solution entails developing a pipeline capable of predicting the necessity of a TMI (e.g., GS or GDP) along with its various parameters. For example, in the case of a GS, this includes the scope of the GS either in terms of distance from the destination airport or based on pre-defined airspace sectors. Here, scope refers to those regions and departing airports that are subject to the GS. In this paper, we concentrate on the issuance of GSs in the three major airports in the New York area — LaGuardia(LGA), John F. Kennedy International (JFK), and Newark Liberty International (EWR). We fuse traffic, weather and other relevant aviation data from years 2017 to 2019 to train and validate the ML models. In particular, we use the following datasets: •Terminal Aerodrome Forecast (TAF): meteorological forecasts specific to each airport, issued four times a day, covering predefined time periods. •TMI data: includes all GSs and GDPs along with their respective parameters. •Aviation System Performance Metrics (ASPM): includes traffic related data such as aircraft delays, arrival, and departure rates. •Notices to Airmen (NOTAMs): utilized to extract runway closure data and manage interdependencies between terminals in close proximity. •Flight cancellation data •Airspace Flow Programs (AFP): includes information on flight airborne holdings caused by TMIs. The data preprocessing entails transforming ASPM, TMI, AFP, NOTAMs, and weather data into an hourly format and consolidating all datasets by merging them based on date and time as the primary key. The TMI Adjuster framework comprises two parallel models: one dedicated to GS and a second model focused on GDP. As previously mentioned, our specific focus is on the GS model as a multi-classification problem. In this framework, each data point of the GS model input summarizes ten hours of data. Specifically, the data loader for the GS model generates the input and output of the model as follows: at a given time step, the input includes the actual traffic, weather, and TMI data from the two-hour window before the time step, alongside the weather forecast and scheduled traffic for the next 8 hours starting from the time step. Based on this information, the output of the GS model for each time interval consists of three dimensions. The first dimension represents a binary decision on whether there should be a GS in place for the next hour or not. The second dimension is related to the scope of the GS in the United States, and the third dimension is related to the scope of the GS in Canada (i.e., to determine if the GS impacts airports in Canada).One of the challenges with TMI modeling is the sparsity of TMI events, particularly regarding its scope. To address this challenge in the scope of the GS model output, we implement grouping. The GS scope for the US region is defined based on a list of centers that should be included when the GS is in place. With 20 centers in the US, we utilized historical data to group them into 4 categories. In particular, we summarized our historical data in a graph format where nodes represent centers, and link weights are defined based on the co-occurrence of centers in the scope parameter ofTMIs. By identified strongly connected components in this graph, we were able to partition the centers into four groups. We consider two model structures for the GS Model. Firstly, a hierarchical classification model [10], where the human decision-making for a GS is of hierarchical nature. The decision-maker first decides whether there is a need fora GS, and if the answer is yes, determines the scope. A hierarchical classification model organizes the problem into a class hierarchy, typically a tree or a Directed Acyclic Graph (DAG) structure, and considers the dependency of the decision in the previous step to the next component [10]. Here, we employ the local classifier per level approach, which involves training one multi-class classifier for each level of the class hierarchy. The second structure is the independent structure. In this setting, as the name suggests, we do not consider the dependency of the decisions in the different dimensions of the output of the model. Instead, for each dimension, we train a multi-class classifier independently. Table 1 summarizes GS model statistics for training, validation and testing. The table documents the effect of limiting data to the time steps when there was actually a TMI in place or when a TMI had just terminated. This resulted in a more balanced distribution of the GS class(GS positive class)versus “No GS”(GS negative class), which might help the training process. While JFK and LGA follow very similar distributions, with 40% and 42% GS positive class respectively, EWR has proportionally fewer GS incidents at 28%. Our subsequent phase involves evaluating the performance of both hierarchical structure and independent structure using different state-of-the-art multi-class classifier models such as Random Forest, Decision Trees, K-nearest Neighbors, and Logistic Regression and forecast the duration and scope of the GSs.

Farzan Masrour Shalmani↗

Spatiotemporal Analysis of High-Resolution Pressure-Sensitive Paint Measurements on a Launch Vehicle Wind Tunnel Model

An accurate understanding of unsteady aerodynamic loads is critical for launch vehicle structural design and for development of the vibro-acoustic environment for payload components. Analysts commonly use point-wise surface pressure measurements from wind tunnel testing to build statistical models of the in-flight turbulent boundary layer flow structures, often extrapolating over large swaths of the surface due to measurement sparsity. We present novel results from a transonic wind tunnel test using fast-response pressure-sensitive paint (PSP) and show how it can be used to visualize and quantify localized turbulent boundary layer wall pressure statistics with unprecedented spatial resolution. The wind tunnel model is a scaled model of the Space Launch System (SLS) rocket Block 1B Crew configuration and analysis focused on the compression corner between the Service Module fairing and the Universal Stage Adapter.

Corcos Model↗

Evaluating the Efficacy of Conditional Variational Autoencoders in Generating Synthetic Single Nuclei RNA-Seq Data for Space Biology Research

Astronauts are subject to unique stressors during spaceflight, leading to changes in their cellular function. However, neither astronauts nor model organisms respond the same to spaceflight, and research implicates a contribution of omics components in differential responses. Understanding how gene expression affects astronaut health is critical for the success of long-term space missions, prompting interest in developing personalized predictive models leveraging artificial intelligence (AI) and machine learning (ML) techniques. Developing such models requires extensive data, which is challenging to obtain and share. This study explores the use of conditional variational autoencoders (CVAEs) to synthetically generate single-nuclei RNA-seq (snRNA-seq) data. CVAEs build on standard variational autoencoders (VAEs) by conditioning data generation on covariates like sample identity and mission parameters, enhancing the relevance of generated data for specific contexts. For our work, we built two CVAEs with varying degrees of sparsity to optimize both interpretability and generative power. We train and validate models on existing snRNA-seq data collected from the brain tissue of mice subjected to spaceflight conditions and their ground control counterparts. We evaluate model performance using statistical tests and visualizations to compare synthetic data to real data. We aim to demonstrate that these prototype CVAE architectures could be used in future space biology work and that this is a method worth further exploring.

Sarah Golts↗

Contextualizing Air Traffic Management Conversations using Natural Language Understanding

Efficient management of air traffic and mitigation of delays depend on extracting actionable information from unstructured data, such as dialogues from the Federal Aviation Administration’s (FAA’s) Air Traffic Control System Command Center (ATCSCC) telecons. This study presents a pipeline utilizing Natural Language Processing (NLP) methods for Intent Classification (IC) and Slot Filling (SF) to identify and extract Traffic Management Initiatives (TMIs) from aviation-specific dialogues. We leveraged DeBERTa, a pre-trained transformer model, and fine-tuned it to the nuances of the aviation domain. Despite challenges posed by annotation complexities, the IC model achieved promising results with a weighted average F1-score of 0.81. Our results are close to those of human annotators, which demonstrates the model’s strong alignment with human-level performance. The SF model also showed strong performance, achieving a weighted F1-score of 0.97, which demonstrates its effectiveness in accurately predicting key slots. Our analysis revealed limitations in handling less frequent intents and slot labels due to data sparsity, motivating future efforts to adopt joint IC-SF modeling and data augmentation strategies. This research highlights the potential of domain-specific NLP to streamline decision-making in the aviation industry and improve the management of TMIs.

Air Traffic Control Management↗

Analog system for computing sparse codes

A parallel dynamical system for computing sparse representations of data, i.e., where the data can be fully represented in terms of a small number of non-zero code elements, and for reconstructing compressively sensed images. The system is based on the principles of thresholding and local competition that solves a family of sparse approximation problems corresponding to various sparsity metrics. The system utilizes Locally Competitive Algorithms (LCAs), nodes in a population continually compete with neighboring units using (usually one-way) lateral inhibition to calculate coefficients representing an input in an over complete dictionary.

Rozell, Christopher John↗

Scalable methods for representing, characterizing, and generating large graphs.

Goal - design methods to characterize and identify a low dimensional representation of graphs. Impact - enabling predictive simulation; monitoring dynamics on graphs; and sampling and recovering network structure from limited observations. Areas to explore are: (1) Enabling technologies - develop novel algorithms and tailor existing ones for complex networks; (2) Modeling and generation - Identify the right parameters for graph representation and develop algorithms to compute these parameters and generate graphs from these parameters; and (3) Comparison - Given two graphs how do we tell they are similar? Some conclusions are: (1) A bad metric can make anything look good; (2) A metric that is based an edge-by edge prediction will suffer from the skewed distribution of present and absent edges; (3) The dominant signal is the sparsity, edges only add a noise on top of the signal, the real signal, structure of the graph is often lost behind the dominant signal; and (4) Proposed alternative: comparison based on carefully chosen set of features, it is more efficient, sensitive to selection of features, finding independent set of features is an important area, and keep an eye on us for some important results.

99 GENERAL AND MISCELLANEOUS↗

Uncertainty quantification given discontinuous climate model response and a limited number of model runs.

Uncertainty quantification in complex climate models is challenged by the sparsity of available climate model predictions due to the high computational cost of model runs. Another feature that prevents classical uncertainty analysis from being readily applicable is bifurcative behavior in climate model response with respect to certain input parameters. A typical example is the Atlantic Meridional Overturning Circulation. The predicted maximum overturning stream function exhibits discontinuity across a curve in the space of two uncertain parameters, namely climate sensitivity and CO2 forcing. We outline a methodology for uncertainty quantification given discontinuous model response and a limited number of model runs. Our approach is two-fold. First we detect the discontinuity with Bayesian inference, thus obtaining a probabilistic representation of the discontinuity curve shape and location for arbitrarily distributed input parameter values. Then, we construct spectral representations of uncertainty, using Polynomial Chaos (PC) expansions on either side of the discontinuity curve, leading to an averaged-PC representation of the forward model that allows efficient uncertainty quantification. The approach is enabled by a Rosenblatt transformation that maps each side of the discontinuity to regular domains where desirable orthogonality properties for the spectral bases hold. We obtain PC modes by either orthogonal projection or Bayesian inference, and argue for a hybrid approach that targets a balance between the accuracy provided by the orthogonal projection and the flexibility provided by the Bayesian inference - where the latter allows obtaining reasonable expansions without extra forward model runs. The model output, and its associated uncertainty at specific design points, are then computed by taking an ensemble average over PC expansions corresponding to possible realizations of the discontinuity curve. The methodology is tested on synthetic examples of discontinuous model data with adjustable sharpness and structure.

54 ENVIRONMENTAL SCIENCES↗

Computationally Efficient Use of Derivatives in Emulation of Complex Computational Models

We will investigate the use of derivative information in complex computer model emulation when the correlation function is of the compactly supported Bohman class. To this end, a Gaussian process model similar to that used by Kaufman et al. (2011) is extended to a situation where first partial derivatives in each dimension are calculated at each input site (i.e. using gradients). A simulation study in the ten-dimensional case is conducted to assess the utility of the Bohman correlation function against strictly positive correlation functions when a high degree of sparsity is induced.

97 MATHEMATICS AND COMPUTING↗

Incomplete Data in Smart Grid: Treatment of Values in Electric Vehicle Charging Data

In this paper, five imputation methods namely Constant (zero), Mean, Median, Maximum Likelihood, and Multiple Imputation methods have been applied to compensate for missing values in Electric Vehicle (EV) charging data. The outcome of each of these methods have been used as the input to a prediction algorithm to forecast the EV load in the next 24 hours at each individual outlet. The data is real world data at the outlet level from the UCLA campus parking lots. Given the sparsity of the data, both Median and Constant (=zero) imputations improved the prediction results. Since in most missing value cases in our database, all values of that instance are missing, the multivariate imputation methods did not improve the results significantly compared to univariate approaches.

Majipour, Mostafa↗

Tensor-Dictionary Learning with Deep Kruskal-Factor Analysis

We introduce new dictionary learning methods for tensor-variate data of any order. We represent each data item as a sum of Kruskal decomposed dictionary atoms within the framework of beta-process factor analysis (BPFA). Our model is nonparametric and can infer the tensor-rank of each dictionary atom. This Kruskal-Factor Analysis (KFA) is a natural generalization of BPFA. We also extend KFA to a deep convolutional setting and develop online learning methods. We test our approach on image processing and classification tasks achieving state of the art results for 2D & 3D inpainting and Caltech 101. The experiments also show that atom-rank impacts both overcompleteness and sparsity.

Stevens, Andrew J.↗

An Embedded 3D Fracture Modeling Approach for Simulating Fracture-Dominated Fluid Flow and Heat Transfer in Geothermal Reservoirs

An efficient modeling approach is described for incorporating arbitrary 3D, discrete fractures, such as hydraulic fractures or faults, into modeling fracture-dominated fluid flow and heat transfer in fractured geothermal reservoirs. This technique allows 3D discrete fractures to be discretized independently from surrounding rock volume and inserted explicitly into a primary fracture/matrix grid, generated without including 3D discrete fractures in prior. An effective computational algorithm is developed to discretize these 3D discrete fractures and construct local connections between 3D fractures and fracture/matrix grid blocks of representing the surrounding rock volume. The constructed gridding information on 3D fractures is then added to the primary grid. This embedded fracture modeling approach can be directly implemented into a developed geothermal reservoir simulator via the integral finite difference (IFD) method or with TOUGH2 technology This embedded fracture modeling approach is very promising and computationally efficient to handle realistic 3D discrete fractures with complicated geometries, connections, and spatial distributions. Compared with other fracture modeling approaches, it avoids cumbersome 3D unstructured, local refining procedures, and increases computational efficiency by simplifying Jacobian matrix size and sparsity, while keeps sufficient accuracy. Several numeral simulations are present to demonstrate the utility and robustness of the proposed technique. Our numerical experiments show that this approach captures all the key patterns about fluid flow and heat transfer dominated by fractures in these cases. Thus, this approach is readily available to simulation of fractured geothermal reservoirs with both artificial and natural fractures.

15 GEOTHERMAL ENERGY↗

Novel deep learning methods for track reconstruction

For the past year, the HEP.TrkX project has been investigating machine learning solutions to LHC particle track reconstruction problems. A variety of models were studied that drew inspiration from computer vision applications and operated on an image-like representation of tracking detector data. While these approaches have shown some promise, image-based methods face challenges in scaling up to realistic HL-LHC data due to high dimensionality and sparsity. In contrast, models that can operate on the spacepoint representation of track measurements ("hits") can exploit the structure of the data to solve tasks efficiently. In this paper we will show two sets of new deep learning models for reconstructing tracks using space-point data arranged as sequences or connected graphs. In the first set of models, Recurrent Neural Networks (RNNs) are used to extrapolate, build, and evaluate track candidates akin to Kalman Filter algorithms. Such models can express their own uncertainty when trained with an appropriate likelihood loss function. The second set of models use Graph Neural Networks (GNNs) for the tasks of hit classification and segment classification. These models read a graph of connected hits and compute features on the nodes and edges. They adaptively learn which hit connections are important and which are spurious. The models are scaleable with simple architecture and relatively few parameters. Results for all models will be presented on ACTS generic detector simulated data.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗