Search NASA⌕ Search

SEARCH · Search NASA

Results for “Learning control”

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 793 records · Page 44
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Attitude Control Approach for Solar Cruiser, a Large, Deep Space Solar Sail

Solar Cruiser is a small satellite Technology Demonstration Mission (TDM) to mature solar sail propulsion technology using a solar sail larger than 1600 square meters, demonstrating performance as a propulsion system and a stable pointing platform for science observations in an artificial halo orbit sunward of the Sun-Earth Lagrange Point 1 (sub-L1). For the “sailcraft” to meet mission objectives, there are several unique attitude control challenges that the Attitude Determination and Control System (ADCS) must overcome. Large disturbance torques, primarily due to sail deformations coupled with off-sun pointing angles, make it more difficult to maintain adequate controls performance and manage accumulated momentum on the control actuators. The large-amplitude, low-frequency flexible body modes of the sail, in concert with noisy sensors and actuators, make it challenging to mitigate control-structure interactions and maintain fine pointing capabilities. Stability and control performance during sail deployment is complicated by rapidly and widely varying inertias. The Solar Cruiser ADCS effectively addresses these challenges using a simple, traditional control system – including momentum management actuators that use bang-bang control to constrain internal accumulated momentum, a low-pass controls filter and an attitude determination Kalman Filter (KF) that blends multiple star tracker solutions, and a reaction wheel assembly (RWA) with controller gains tuned to specific configurations or inertias. Solar Cruiser’s attitude control approach, and lessons learned from its development, establishes a state of the art of high value to future solar sail missions and other small spacecraft with large deployable structures operating in deep space.

solar sail↗
DOE OSTI2023

Integrated Operations for Nuclear: Work Reduction Opportunity Demonstration Strategy

EXECUTIVE SUMMARY The Light Water Reactor Sustainability Program Plant Modernization Pathway has been working with industry for a number of years to leverage digital technology to extend the life and improve the performance of the existing fleet through modernized technologies and improved processes for plant operation and power generation. This includes development of modernization solutions to improve reliability and economic performance while addressing US nuclear industry’s aging and obsolescence challenges. The objective of these efforts is to deliver a sustainable business model that enables US nuclear industry to remain competitive. Digital Infrastructure (DI) research has established a technical foundation for these efforts. This effort began with technical analysis and support for a safety-related instrumentation and control (I&C) pilot upgrade being performed at Constellation Energy Generation’s Limerick Nuclear Plant. The following publicly available reports were produced as part of this effort. • INL/EXT-20-61079, Vendor-Independent Design Requirements for a Boiling Water Reactor Safety System Upgrade [1] • INL/EXT-20-59371, Business Case Analysis for Digital Safety-Related Instrumentation & Control System Modernizations [2] • INL/EXT-20-59809, “Safety-Related Instrumentation and Control Pilot Upgrade: Initial Scoping Phase Implementation Report and Lessons Learned [3] • INL/RPT-23-72105, Safety-Related Instrumentation and Control Upgrade: Conceptual – Detailed Design Phase Report and Lessons Learned [4]

99 GENERAL AND MISCELLANEOUS↗
NASA NTRS2021

Spacecraft Line-Of-Sight Jitter Management and Mitigation Lessons Learned And Engineering Best Practices

Predicting, managing, controlling, and testing spacecraft line-of-sight (LoS) jitter caused by micro-vibrations due to on-board internal disturbance sources is a formidable multidisciplinary engineering task. It is especially challenging for those missions hosting high-performance (e.g., nano-radian/milli-arcsecond class), vibration-sensitive optical sensor payloads with stringent pointing stability requirements. The Nation Aeronautics and Space Administration (NASA) and the European Space Agency (ESA) are planning technically aggressive spaceflight missions that include ultra-high-performance optical payloads with delicate, highly vibration-sensitive scientific and observational instruments. The guidance, navigation, and control community of practice will need to leverage collective experiences and document their best practices and lessons learned to address future micro-vibration challenges. To identify lessons learned and best practices the NASA Engineering & Safety Center sponsored a 2-day Spacecraft LoS Jitter Workshop in late 2019. The workshop’s goal was to provide a multidisciplinary forum to elicit deeper understanding of the issues related to addressing the spacecraft LoS jitter/micro-vibration problem. The primary objective was to identify, document, and share lessons learned, best practices, and preferred options for jitter-related analysis and test activities. Representatives from NASA, ESA, along with NASA’s industrial partners, independent consultant subject matter experts, and members of academia participated in the workshop. This paper will describe the motivation for the workshop and summarize the identified findings and recommendations.

Swanson, Davin K.↗
NASA NTRS1991

Lessons learned for composite structures

Lessons learned for composite structures are presented in three technology areas: materials, manufacturing, and design. In addition, future challenges for composite structures are presented. Composite materials have long gestation periods from the developmental stage to fully matured production status. Many examples exist of unsuccessful attempts to accelerate this gestation period. Experience has shown that technology transition of a new material system to fully matured production status is time consuming, involves risk, is expensive and should not be undertaken lightly. The future challenges for composite materials require an intensification of the science based approach to material development, extension of the vendor/customer interaction process to include all engineering disciplines of the end user, reduced material costs because they are a significant factor in overall part cost, and improved batch-to-batch pre-preg physical property control. Historical manufacturing lessons learned are presented using current in-service production structure as examples. Most producibility problems for these structures can be traced to their sequential engineering design. This caused an excessive emphasis on design-to-weight and schedule at the expense of design-to-cost. This resulted in expensive performance originated designs, which required costly tooling and led to non-producible parts. Historically these problems have been allowed to persist throughout the production run. The current/future approach for the production of affordable composite structures mandates concurrent engineering design where equal emphasis is placed on product and process design. Design for simplified assembly is also emphasized, since assembly costs account for a major portion of total airframe costs. The future challenge for composite manufacturing is, therefore, to utilize concurrent engineering in conjunction with automated manufacturing techniques to build affordable composite structures. Composite design experience has shown that significant weight savings have been achieved, outstanding fatigue and corrosion resistance have been demonstrated, and in-service performance has been very successful. Currently no structural design show stoppers exist for composite structures. A major lesson learned is that the full scale static test is the key test for composites, since it is the primary structural 'hot spot' indicator. The major durability issue is supportability of thin skinned structure. Impact damage has been identified as the most significant issue for the damage tolerance control of composite structures. However, delaminations induced during assembly operations have demonstrated a significant nuisance value. The future challenges for composite structures are threefold. Firstly, composite airframe weight fraction should increase to 60 percent. At the same time, the cost of composite structures must be reduced by 50 percent to attain the goal of affordability. To support these challenges it is essential to develop lower cost materials and processes.

Whitehead, R. S.↗
NASA NTRS2020

Machine Learning For Planetary Mining Applications

Robotic mining could prove to be an efficient method of mining resources for extended missions on the Moon or Mars. One component of robotic mining is scouting an area for resources to be mined by other robotic systems. Writing controllers for scouting can be difficult due to the need for fault tolerance, inter-agent cooperation, and agent problem solving. Reinforcement learning could solve these problems by enabling the scouts to learn to improve their performance over time. This work is divided into two sections, with each section addressing the use of machine learning in this domain. The first contribution of this work focuses on the application of reinforcement learning to mining mission analysis. Various mission parameters were modified and control policies were learned. Then agent performance was used to assess the effect of the mission parameters on the performance of the mission. The second contribution of this work explores the potential use of reinforcement learning to learn a controller for the scouts. Through learning, these scouts would improve their ability to map their surroundings over time.

Cook, Joshua↗
DOE OSTI2020

Machine learning to predict biomass sorghum yields under future climate scenarios

Crop yield modeling is critical in the design of national strategies for agricultural production, particularly in the context of a changing climate. Forecasting yields of bioenergy crops at fine spatial resolutions can help to evaluate near-term and long-term pathways for scaling up bio-based fuel and chemical production, and for understanding the impacts of abiotic stressors such as severe droughts and temperature extremes on potential biomass supply. In this work we used a large dataset of 28,364 Sorghum bicolor yield samples (uniquely identified by county and year of observation), environmental variables, and multiple approaches to analyze historical trends in sorghum productivity across the USA. We selected the most accurate machine learning approach (a variation of the random forest approach) to predict future trends in sorghum yields under four greenhouse gas (GHG) emission scenarios and two irrigation regimes. We identified irrigation practices, vapor pressure deficit, and time (a proxy for technological improvement) as the most important predictors of sorghum productivity. Our results showed a decreasing trend of sorghum yields over future years (on average 2.7% from 2018 to 2099), with greater decline under a high GHG emissions scenario (3.8%) and in the absence of irrigation (4.6%). Geographically, we observed the steepest predicted declines in the Great Lakes (8.2%), Upper Midwest (7.5%), and Heartland (6.7%) regions. Our study demonstrates the use of machine learning to identify environmental controllers of sorghum biomass yield and predict yields with reasonable accuracy. These results can inform the development of more realistic biomass supply projections for bioenergy if sorghum production is scaled up. (c) 2020 Society of Chemical Industry and John Wiley & Sons, Ltd

09 BIOMASS FUELS↗
NASA NTRS2003

Reinforcement Learning for Weakly-Coupled MDPs and an Application to Planetary Rover Control

Weakly-coupled Markov decision processes can be decomposed into subprocesses that interact only through a small set of bottleneck states. We study a hierarchical reinforcement learning algorithm designed to take advantage of this particular type of decomposability. To test our algorithm, we use a decision-making problem faced by autonomous planetary rovers. In this problem, a Mars rover must decide which activities to perform and when to traverse between science sites in order to make the best use of its limited resources. In our experiments, the hierarchical algorithm performs better than Q-learning in the early stages of learning, but unlike Q-learning it converges to a suboptimal policy. This suggests that it may be advantageous to use the hierarchical algorithm when training time is limited.

Bernstein, Daniel S.↗
DOE OSTI2022

A Hybrid Reinforcement Learning-MPC Approach for Distribution System Critical Load Restoration

This paper proposes a hybrid control approach for distribution system critical load restoration, combining deep reinforcement learning (RL) and model predictive control (MPC) aiming at maximizing total restored load following an extreme event. RL determines a policy for quantifying operating reserve requirements, thereby hedging against uncertainty, while MPC models grid operations incorporating RL policy actions (i.e., reserve requirements), renewable (wind and solar) power predictions, and load demand forecasts. We formulate the reserve requirement determination problem as a sequential decision-making problem based on the Markov Decision Process (MDP) and design an RL learning environment based on the OpenAI Gym framework and MPC simulation. The RL agent reward and MPC objective function aim to maximize and monotonically increase total restored load and minimize load shedding and renewable power curtailment. The RL algorithm is trained offline using a historical forecast of renewable generation and load demand. The method is tested using a modified IEEE 13-bus distribution test feeder containing wind turbine, photovoltaic, microturbine, and battery. Case studies demonstrated that the proposed method outperforms other policies with static operating reserves.

distribution system↗
NASA NTRS2014

Re-Engineering the ISS Payload Operations Control Center During Increased Utilization and Critical Onboard Events

With an increase in utilization and hours of payload operations being executed onboard the International Space Station (ISS), upgrading the NASA Marshall Space Flight Center (MSFC) Huntsville Operations Support Center (HOSC) ISS Payload Control Area (PCA) was essential to gaining efficiencies and assurance of current and future payload health and science return. PCA houses the Payload Operations Integration Center (POIC) responsible for the execution of all NASA payloads onboard the ISS. POIC Flight Controllers are responsible for the operation of voice, stowage, command, telemetry, video, power, thermal, and environmental control in support of ISS science experiments. The methodologies and execution of the PCA refurbishment were planned and performed within a four-month period in order to assure uninterrupted operation of ISS payloads and minimal impacts to payload operations teams. To vacate the PCA, three additional HOSC control rooms were reconfigured to handle ISS real-time operations, Backup Control Center (BCC) to Mission Control in Houston, simulations, and testing functions. This involved coordination and cooperation from teams of ISS operations controllers, multiple engineering and design disciplines, management, and construction companies performing an array of activities simultaneously and in sync delivering a final product with no issues that impacted the schedule. For each console operator discipline, studies of Information Technology (IT) tools and equipment layouts, ergonomics, and lines of sight were performed. Infusing some of the latest IT into the project was an essential goal in ensuring future growth and success of the ISS payload science returns. Engineering evaluations led to a state of the art Video Wall implementation and more efficient ethernet cabling distribution providing the latest products and the best solution for the POIC. These engineering innovations led to cost savings for the project. Constraints involved in the management of the project included executing over 450 crew-hours of ISS real-time payload operations including a major onboard communications upgrade, SpaceX un-berth, a Soyuz launch, roll-out of ISS live video and interviews from the POIC, annual BCC certification and hurricane season, and ISS simulations and testing. Continuous ISS payload operations were possible during the PCA facility modifications with the reconfiguration of four control rooms and standup of two temporary control areas. Another major restriction to the project was an ongoing facility upgrade that included a NASA Headquarters mandated replacement of all electrical and mechanical systems and replacement of an external generator. These upgrades required a facility power outage during the PCA upgrades. The project also encompassed console layout designs and ordering, amenities selections and ordering, excessing of old equipment, moves, disposal of old IT equipment, camera installations, facility tour re-schedules, and contract justifications. These were just some of the tasks needed for a successful project. This paper describes the logistics and lessons learned in upgrading a control center capability in the middle of complex real-time operations. Combining the efficiencies of controller interaction and new technology infusion were prime drivers for this upgrade to handle the increased utilization of science research on ISS. The success of this project could not jeopardize the current operations while these facility upgrades occurred.

Dudley, Stephanie R. B.↗
NASA NTRS1994

POINTER: Portable Intelligent Trainer for External Robotics

Intelligent tutoring systems (ITS's) play an increasing role in training and education of people with different levels of skill and knowledge. As compared to conventional computer based training (CBT) an ITS provides more tailored instruction by trying to mimic the teaching behavior of a human instructor as much as possible and is therefore much more flexible. This paper starts with an introduction to ITS's, followed by the description of an ITS for training of an (astronaut) operator in monitoring and controlling robotic arm procedures. The robotic arm will be used for exchange of equipment between a space station and a space plane involving critical and accurate movements of the robotic arm. The ITS for this application, called Pointer, is developed by TNO Physics and Electronics Laboratory and is based upon an existing ITS that includes procedural training. Pointer has been developed on a workstation whereas the target platform was a portable computer. Therefore, a lot of attention had to be paid to scaling effects and keeping up with user friendliness of the much smaller user interface. Although the learning domain was the control of a robotic arm, it is clear that use of intelligent training technologies on a portable computer has many other applications (payload operations, operation control rooms, etc.). Training can occur at any time and place in an attractive and cost effective way.

Kuiper, Hilbert↗
DOE OSTI2021

A dimensionality reduction algorithm for mapping tokamak operational regimes using a variational autoencoder (VAE) neural network

Abstract A variational autoencoder (VAE) is a type of unsupervised neural network which is able to learn meaningful data representations in a reduced dimensional space. We present an application of VAE in identifying the operational stability boundary of tokamak plasma discharges. This model was implemented using a dataset of over 3000 discharges from the high beta tokamak-extended pulse (HBT-EP) device. We found the VAE model to be capable of forming a continuous low-dimensional operational space map and identifying the operational boundaries using a specified warning time window. By projecting the operational parameters onto the same reduced space, this provides an intuitive way for the machine operator or an automated control system to perform disruption avoidance using a relevant control actuator as a discharge approaches a boundary. Pre-programmed GPU control experiments were conducted to demonstrate this control technique using HBT-EP’s saddle control coils as a horizontal position actuator, showing the ability to avoid the oncoming disruptive event and extend the duration of the discharge.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS2006

Assessment of Postflight Locomotor Performance Utilizing a Test of Functional Mobility: Strategic and Adaptive Responses

Space flight induces adaptive modification in sensorimotor function, allowing crewmembers to operate in the unique microgravity environment. This adaptive state, however, is inappropriate for a terrestrial environment. During a re-adaptation period upon their return to Earth, crewmembers experience alterations in sensorimotor function, causing various disturbances in perception, spatial orientation, posture, gait, and eye-head coordination. Following long duration space flight, sensorimotor dysfunction would prevent or extend the time required to make an emergency egress from the vehicle; compromising crew safety and mission objectives. We are investigating two types of motor learning that may interact with each other and influence a crewmember's ability to re-adapt to Earth's gravity environment. In strategic learning, crewmembers make rapid modifications in their motor control strategy emphasizing error reduction. This type of learning may be critical during the first minutes and hours after landing. In adaptive learning, long-term plastic transformations occur, involving morphological changes and synaptic modification. In recent literature these two behavioral components have been associated with separate brain structures that control the execution of motor strategies: the strategic component was linked to the posterior parietal cortex and the adaptive component was linked to the cerebellum (Pisella, et al. 2004). The goal of this paper was to demonstrate the relative contributions of the strategic and adaptive components to the re-adaptation process in locomotor control after long duration space flight missions on the International Space Station (ISS). The Functional Mobility Test (FMT) was developed to assess crewmember s ability to ambulate postflight from an operational and functional perspective. Sixteen crewmembers were tested preflight (3 sessions) and postflight (days 1, 2, 4, 7, 25) following a long duration space flight (approx 6 months) on the ISS. We have further analyzed the FMT data to characterize strategic and adaptive components during the postflight readaptation period. Crewmembers walked at a preferred pace through an obstacle course set up on a base of 10 cm thick medium density foam (Sunmate Foam, Dynamic Systems, Inc., Leicester, NC). The 6.0m X 4.0m course consisted of several pylons made of foam; a Styrofoam barrier 46.0cm high that crewmembers stepped over; and a portal constructed of two Styrofoam blocks, each 31cm high, with a horizontal bar covered by foam and suspended from the ceiling which was adjusted to the height of the crewmember s shoulder. The portal required crewmembers to bend at the waist and step over a barrier simultaneously. All obstacles were lightweight, soft and easily knocked over. Crewmembers were instructed to walk through the course as quickly and as safely as possible without touching any of the objects on the course. This task was performed three times in the clockwise direction and three times in the counterclockwise direction that was randomly chosen. The dependent measures for each trial were: time to complete the course (seconds) and the number of obstacles touched or knocked down. For each crewmember, the time to complete each FMT trial from postflight days 1, 2, 4, 7 and 25 were further analyzed. A single logarithmic curve using a least squares calculation was fit through these data to produce a single comprehensive curve (macro). This macro curve composed of data spanning 25 days, illustrates the re-adaptive learning function over the longer time scale term. Additionally, logarithmic curves were fit to the 6 data trials within each individual post flight test day to produce 5 separate daily curves. These micro curves, produced from data obtained over the course of minutes, illustrates the strategic learning function exhibited over a relative shorter time scale. The macro curve for all subjects exhibited adaptive motor learning patterns over the 25 day period. Howev, 9/16 crewmembers exhibited significant strategic motor learning patterns in their micro curves, as defined by m > 1 in the equation of the line y=m*LN(x) +b. These data indicate that postflight recovery in locomotor function involves both strategic and adaptive mechanisms. Future countermeasures will be designed to enhance both recovery processes.

Warren, L. E.↗
NASA NTRS1990

Genetic algorithm based fuzzy control of spacecraft autonomous rendezvous

The U.S. Bureau of Mines is currently investigating ways to combine the control capabilities of fuzzy logic with the learning capabilities of genetic algorithms. Fuzzy logic allows for the uncertainty inherent in most control problems to be incorporated into conventional expert systems. Although fuzzy logic based expert systems have been used successfully for controlling a number of physical systems, the selection of acceptable fuzzy membership functions has generally been a subjective decision. High performance fuzzy membership functions for a fuzzy logic controller that manipulates a mathematical model simulating the autonomous rendezvous of spacecraft are learned using a genetic algorithm, a search technique based on the mechanics of natural genetics. The membership functions learned by the genetic algorithm provide for a more efficient fuzzy logic controller than membership functions selected by the authors for the rendezvous problem. Thus, genetic algorithms are potentially an effective and structured approach for learning fuzzy membership functions.

Karr, C. L.↗
DOE OSTI2023

Certifiably Robust Neural ODE With Learning-Based Barrier Function

Neural Ordinary Differential Equations (ODEs) have gained traction in many applications. While recent studies have focused on empirically increasing the robustness of neural ODEs against natural or adversarial attacks, certified robustness is still lacking. In this work, we propose a framework for training a neural ODE using barrier functions and demonstrate improved robustness for classification problems. Finally, we further provide the first generalization guarantee of robustness against adversarial attacks using a wait-and-judge scenario approach.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Optimal Control of SOEC-Based Hydrogen Production Systems for Demand Response Using Deep Reinforcement Learning in Smart Grids

Solid oxide electrolysis cell (SOEC) hydrogen production technology can range in size from small, appliance-size equipment to large-scale, central production facilities that can be tied directly to renewable or non-greenhouse-gas-emitting forms of electricity production, making it an ideal resource for demand response (DR). The SOEC hydrogen production system is a complex integrated system that encompasses fluid dynamics, electrical dynamics, and electrochemical and thermal dynamics, all of which involve non-linearity and non-convexity. Proper control of the SOEC hydrogen production system is crucial to enable its participation in the DR program. Here, to overcome the difficulty of designing an explicit control law for such nonlinear systems with nonconvex optimization features in DR applications, deep reinforcement learning (DRL) is explored to achieve the optimal control of the SOEC system for DR participation. Specifically, a twin delayed deterministic policy gradient (TD3) control framework is applied to achieve optimal response performance during DR events by considering power tracking error and hydrogen production efficiency with a suitable reward function. Two case studies with grid connections for tracking different DR commands were investigated. The first case study involved operating conditions reaching the boundaries, while the second involved operating conditions within the boundaries. The results showed that the proposed DRL-based control for SOEC can track the DR signal in a timely manner while maintaining high energy efficiency.

08 HYDROGEN↗
DOE OSTI2025

Active Learning of Microgrid Frequency Dynamics Using Neural Ordinary Differential Equations

Accurate frequency modelling of inverter‐based resource (IBR)‐dominated power systems is crucial for ensuring stable, reliable and resilient operations, particularly given their inherent low‐inertia characteristics and fast dynamics that traditional swing equation‐based models inadequately capture. This paper explores neural ordinary differential equations (Neural ODEs) as a computationally efficient, data‐driven framework for modelling power system frequency dynamics, specifically within microgrids integrating high penetrations of distributed energy resources (DERs). The developed neural ODEs framework incorporates a neural network architecture designed to capture input dynamics. By actively perturbing the system with a known signal, the Python‐based neural ODEs framework was trained using measured system states and inputs, without the need for detailed system information. The framework, tested on a model of the Cordova, AK, microgrid, achieved a goodness of fit ranging from 60% to 99% across different state variables and maintained a mean square error in the 10 -6 p.u. range under square and step excitation signals. The proposed approach demonstrated robustness to measurement noise and initial condition variations while maintaining low computational complexity suitable for real‐time power system control applications. Furthermore, transfer learning enabled the neural ODEs model to adapt to the following changes in system topology or generator dispatch, highlighting its effectiveness for dynamic microgrids with frequently evolving configurations and diverse DERs.

Aryal, Tara [South Dakota State Univ., Brookings, ↗
DOE OSTI2021

Learning Optimal Power Flow Solutions using Linearized Models in Power Distribution Systems

Solving nonlinear optimal power flow (OPF) problem is computationally expensive, and poses scalability challenges for power distribution networks. An alternative to solving the original nonlinear OPF is the linear approximated OPF models. Although, these linear approximated OPF models are fast, the resulting solutions may result in significant optimality gap. Lately, the application of machine learning (ML) methods in successfully solving the nonlinear OPF has been reported. These methods learn and estimate the nonlinear control policies using a purely data-driven approach. In this paper, we propose an approach to complements the ML based approach to solving OPF using solutions from known linearized OPF model. Specifically, we use supervised learning to map the solutions of linear OPF to nonlinear control variables. Unlike, the traditional ML based methods for OPF that approximate the full distribution feeder model using function approximation, our approach uses a two-node approximation of radial networks. The proposed approach is validated using IEEE 123 bus test system for OPF solutions obtained using the nonlinear OPF models.

optimal power flow, power distribution systems, su↗
DOE OSTI2023

Data-driven Whitney forms for structure-preserving control volume analysis

Control volume analysis models physics via the exchange of generalized fluxes between subdomains. Here, we introduce a scientific machine learning framework adopting a partition of unity architecture to identify physically-relevant control volumes, with generalized fluxes between subdomains encoded via Whitney forms. The approach provides a differentiable parameterization of geometry which may be trained in an end-to-end fashion to extract reduced models from full field data while exactly preserving physics. The architecture admits a data-driven finite element exterior calculus allowing discovery of mixed finite element spaces with closed form quadrature rules. An equivalence between Whitney forms and graph networks reveals that the geometric problem of control volume learning is equivalent to an unsupervised graph discovery problem. The framework is developed for manifolds in arbitrary dimension, with examples provided for H(div) problems in $\mathbb{R}$ establishing convergence and structure preservation properties. Finally, we consider a lithium-ion battery problem where we discover a reduced finite element space encoding transport pathways from high-fidelity microstructure resolved simulations. The approach reduces the 5.89M finite element simulation to 136 elements while reproducing pressure to under 0.1% error and preserving conservation.

97 MATHEMATICS AND COMPUTING↗