SEARCH · Search NASA
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.
Using Deep Reinforcement Learning, we optimize the energy efficiency of traffic on a calibrated model of the I-210 in Los Angeles.
Plasma physics simulations are vital for a host of Sandia mission concerns, for fundamental science, and for clean energy in the form of fusion power. Sandia's most mature plasma physics simulation capabilities come in the form of particle-in-cell (PIC) models and magnetohydrodynamics (MHD) models. MHD models for a plasma work well in denser plasma regimes when there is enough material that the plasma approximates a fluid. PIC models, on the other hand, work well in lower-density regimes, in which there is not too much to simulate; error in PIC scales as the square root of the number of particles, making high-accuracy simulations expensive. Real-world applications, however, almost always involve a transition region between the high-density regimes where MHD is appropriate, and the low-density regimes for PIC. In such a transition region, a direct discretization of Vlasov is appropriate. Such discretizations come with their own computational costs, however; the phase-space mesh for Vlasov can involve up to six dimensions (seven if time is included), and to apply appropriate homogeneous boundary conditions in velocity space requires meshing a substantial padding region to ensure that the distribution remains sufficiently close to zero at the velocity boundaries. Moreover, for collisional plasmas, the right-hand side of the Vlasov equation is a collision operator, which is non-local in velocity space, and which may dominate the cost of the Vlasov solver. The present LDRD project endeavors to develop modern, foundational tools for the development of continuum-kinetic Vlasov solvers, using the discontinuous Petrov-Galerkin (DPG) methodology, for discretization of Vlasov, and machine-learning (ML) models to enable efficient evaluation of collision operators. DPG affords several key advantages. First, it has a built-in, robust error indicator, allowing us to adapt the mesh in a very natural way, enabling a coarse velocity-space mesh near the homogeneous boundaries, and a fine mesh where the solution has fine features. Second, it is an inherently high-order, high-intensity method, requiring extra local computations to determine so-called optimal test functions, which makes it particularly suited to modern hardware in which floating-point throughput is increasing at a faster rate than memory bandwidth. Finally, DPG is a residual-minimizing method, which enables high-accuracy computation: in typical cases, the method delivers something very close to the $L^2$ projection of the exact solution. Meanwhile, the ML-based collision model we adopt affords a cost structure that scales as the square root of a standard direct evaluation. Moreover, we design our model to conserve mass, momentum, and energy by construction, and our approach to training is highly flexible, in that it can incorporate not only synthetic data from direct-simulation Monte Carlo (DSMC) codes, but also experimental data. We have developed two DPG formulations for Vlasov-Poisson: a time-marching, backward-Euler discretization and a space-time discretization. We have conducted a number of numerical experiments to verify the approach in a 1D1V setting. In this report, we detail these formulations and experiments. We also summarize some new theoretical results developed as part of this project (published as papers previously): some new analysis of DPG for the convection-reaction problem (of which the Vlasov equation is an instance), a new exponential integrator for DPG, and some numerical exploration of various DPG-based time-marching approaches to the heat equation. As part of this work, we have contributed extensively to the Camellia open-source library; we also describe the new capabilities and their usage. We have also developed a well-documented methodology for single-species collision operators, which we applied to argon and demonstrated with numerical experiments. We summarize those results here, as well as describing at a high level a design extending the methodology to multi-species operators. We have released a new open-source library, MLC, under a BSD license; we include a summary of its capabilities as well.
A shadow molecular dynamics scheme for flexible charge models is presented, where the shadow Born-Oppenheimer potential is derived from a coarse-grained approximation of range-separated density functional theory. The interatomic potential, including the atomic electronegativities and the charge-independent short-range part of the potential and force terms, are modeled by the linear atomic cluster expansion (ACE), which provides a computationally efficient alternative to many machine learning methods. The shadow molecular dynamics scheme is based on extended Lagrangian (XL) Born-Oppenheimer molecular dynamics (BOMD) [Eur. Phys. J. B 94, 164 (2021)]. XL-BOMD provides a stable dynamics, while avoiding the costly computational overhead associated with solving an all-to-all system of equations, which normally is required to determine the relaxed electronic ground state prior to each force evaluation. To demonstrate the proposed shadow molecular dynamics scheme for flexible charge models using the atomic cluster expansion, we emulate the dynamics generated from self-consistent charge density functional tight-binding (SCC-DFTB) theory using a second-order charge equilibration (QEq) model. The charge-independent potentials and electronegativities of the QEq model are trained for a supercell of uranium oxide (UO2) and a molecular system of liquid water. The combined ACE + XL-QEq dynamics are stable over a wide range of temperatures both for the oxide and the molecular systems, and provide a precise sampling of the Born-Oppenheimer potential energy surfaces. Accurate ground Coulomb energies are produced by the ACE-based electronegativity model during an NVE simulation of UO 2 , predicted to be within 1 meV of those from SCC-DFTB on average during comparable simulations.
Abstract. Wildfire is an important ecosystem process, influencing land biogeophysical and biogeochemical dynamics and atmospheric composition. Fire-driven loss of vegetation cover, for example, directly modifies the surface energy budget as a consequence of changing albedo, surface roughness, and partitioning of sensible and latent heat fluxes. Carbon dioxide and methane emitted by fires contribute to a positive atmospheric forcing, whereas emissions of carbonaceous aerosols may contribute to surface cooling. Process-based modeling of wildfires in Earth system land models is challenging due to limited understanding of human, climate, and ecosystem controls on fire counts, fire size, and burned area. Integration of mechanistic wildfire models within Earth system models requires careful parameter calibration, which is computationally expensive and subject to equifinality. To explore alternative approaches, we present a deep neural network (DNN) scheme that surrogates the process-based wildfire model with the Energy Exascale Earth System Model (E3SM) interface. The DNN wildfire model accurately simulates observed burned area with over 90 % higher accuracy with a large reduction in parameterization time compared with the current process-based wildfire model. The surrogate wildfire model successfully captured the observed monthly regional burned area during validation period 2011 to 2015 (coefficient of determination, R2=0.93). Since the DNN wildfire model has the same input and output requirements as the E3SM process-based wildfire model, our results demonstrate the applicability of machine learning for high accuracy and efficient large-scale land model development and predictions.
Adjoint sensitivity equations are presented, which can be solved simultaneously (i.e., forward in time) with the dynamics of a nonlinear neural network. These equations provide the foundations for a new methodology which enables the implementation of temporal learning algorithms in a highly efficient manner.
Planning and scheduling is the area of artificial intelligence research that focuses on the determination of a series of operations to achieve some set of (possibly) interacting goals and the placement of those operations in a timeline that allows them to be accomplished given available resources. Work in this area at the NASA Ames Research Center ranging from basic research in constrain-based reasoning and machine learning, to the development of efficient scheduling tools, to the application of such tools to complex agency problems is described.
NASA's environment mirrors the changes taking place in the nation at large, i.e. workers are being asked to do more work with fewer resources. For software developers at NASA's Goddard Space Flight Center (GSFC), the effects of this change are that we must continue to produce quality code that is maintainable and reusable, but we must learn to produce it more efficiently and less expensively. To accomplish this goal, the Data Systems Technology Division (DSTD) at GSFC is trying a variety of both proven and state-of-the-art techniques for software development (e.g., object-oriented design, prototyping, designing for reuse, etc.). In order to evaluate the effectiveness of these techniques, the Software Process Assessment (SPA) program was initiated. SPA was begun under the assumption that the effects of different software development processes, techniques, and tools, on the resulting product must be evaluated in an objective manner in order to assess any benefits that may have accrued. SPA involves the collection and analysis of software product and process data. These data include metrics such as effort, code changes, size, complexity, and code readability. This paper describes the SPA data collection and analysis methodology and presents examples of benefits realized thus far by DSTD's software developers and managers.
The Space Technology 5(ST5) payload was successfully carried into orbit on an OSC Pegasus XL launch vehicle, which was carried aloft and dropped from the OSC Lockheed L-1011 from Vandenberg Air Force Base March 22,2006, at 9:03 am Eastern time, 6:03 am Pacific time. In order to reach the completion of the development and successful launch of ST 5, the systems integration and test(I&T) team determined that a different approach was required to meet the project requirements rather than the standard I&T approach used for single, room-sized satellites. The ST5 payload, part of NASA's New Millennium Program headquartered at JPL, consisted of three micro satellites (approximately 30 kg each) and the Pegasus Support Structure (PSS), the system that connected the spacecrafts to the launch vehicle and deployed the spacecrafts into orbit from the Pegasus XL launch vehicle. ST5 was a technology demonstration payload, intended to test six (6) new technologies for potential use for future space flights along with demonstrating the ability of small satellites to perform quality science. The main technology was a science grade magnetometer designed to take measurements of the earth's magnetic field. The three spacecraft were designed, integrated, and tested at NASA Goddard Space Flight Center with integration and environmental testing occurring in the Bldg. 7-1 0-15-29. The three spacecraft were integrated and tested by the same I&T team. The I&T Manager determined that there was insufficient time in the schedule to perform the three I&T spacecraft activities in series used standard approaches. The solution was for spacecraft #1 to undergo integration and test first, followed by spacecraft #2 and #3 simultaneously. This simultaneous integration was successful for several reasons. Each spacecraft had a Lead Test Conductor who planned and coordinated their spacecraft through its integration and test activities. One team of engineers and technicians executed the integration of all three spacecraft, learning and gaining knowledge and efficiency as spacecraft #1 integration and testing progressed. They became acutely familiar with the hardware, operation and processes for I&T, thus each team member had the experience and knowledge to safely execute I&T for spacecraft #2 and #3 together. The integration team was very versatile and each member could perform many different activities or work any spacecraft, when needed. Daily meetings between the three Lead TCs and technician team allowed the team to plan and implement activities efficiently. The three (3) spacecraft and PSS were successfully integrated and tested, shipped to the launch site, and ready for launch per the I&T schedule that was planned three years previously.
The ST5 technology demonstration mission led by GSFC of NASA's New Millennium Program managed by JPL consisted of three micro satellites (approximately 30 kg each) deployed into orbit from the Pegasus XL launch vehicle. In order to meet the launch date schedule of ST5, a different approach was required rather than the standard I&T approach used for single, room-sized satellites. The three spacecraft were designed, integrated, and tested at NASA Goddard Space Flight Center. It was determined that there was insufficient time in the schedule to perform three spacecraft I&T activities in series using standard approaches. The solution was for spacecraft #1 to undergo integration and test first, followed by spacecraft #2 and #3 simultaneously. This simultaneous integration was successful for several reasons. Each spacecraft had a Lead Test Conductor who planned and coordinated their spacecraft through its integration and test activities. One team of engineers and technicians executed the integration of all three spacecraft, learning and gaining knowledge and efficiency as spacecraft #1 integration and testing progressed. They became acutely familiar with the hardware, operation and processes for I&T, thus had the experience and knowledge to safely execute I&T for spacecraft #2 and #3. The integration team was extremely versatile; each member could perform many different activities or work any spacecraft, when needed. ST5 was successfully integrated, tested and shipped to the launch site per the I&T schedule that was planned three years previously. The I&T campaign was completed with ST5's successful launch on March 22, 2006.
Space vehicle staging and separation events require pyrotechnic devices. They are single-use mechanisms that cannot be tested, nor can failure-tolerant performance be demonstrated in actual flight articles prior to flight use. This necessitates the implementation of a robust design and test approach coupled with a fully redundant, failure-tolerant explosive mechanism to ensure that the system functions even in the event of a single failure. Historically, NASA has followed the single failure-tolerant (SFT) design philosophy for all human-rated spacecraft, including the Space Shuttle Program. Following the end of this program, aerospace companies proposed building the next generation human-rated vehicles with off-the-shelf, non-redundant, zero-failure-tolerant (ZFT) separation systems. Currently, spacecraft and launch vehicle providers for both the Orion and Commercial Crew Programs (CCPs) plan to deviate from the heritage safety approach and NASA's SFT human rating requirements. Both programs' partners have base-lined ZFT frangible joints for vehicle staging and fairing separation. These joints are commercially available from pyrotechnic vendors. Non-human-rated missions have flown them numerous times. The joints are relatively easy to integrate structurally within the spacecraft. In addition, the separation event is debris free, and the resultant pyro shock is lower than that of other design solutions. It is, however, a serious deficiency to lack failure tolerance. When used for critical applications on human-rated vehicles, a single failure could potentially lead to loss of crew (LOC) or loss of mission (LOM)). The Engineering and Safety & Mission Assurance directorates within the NASA Johnson Space Center took action to address this safety issue by initiating a project to develop a fully redundant, SFT frangible joint design, known as the Flat H. Critical to the ability to retrofit on launch vehicles being developed, the SFT mechanisms must fit within the same three-dimensional envelope as current designs as well as meet structural loads requirements. There is increased mass associated with the redundant design, and the goal is to minimize the weight impact as much as possible. These requirements presented significant challenges, both technically and financially; these challenges will be explored in this paper. Perhaps greater than the technical issues confronted during this design process, were the financial considerations. These were a significant part of the story of this design and development plan. Insufficient financial and labor resources were formidable barriers to completing this project. Nevertheless, JSC personnel successfully conducted several test series at JSC with very useful results. The many lessons learned drove design improvements, performance efficiency, and increased functional reliability. This paper examines the significant technical and financial challenges that these requirements posed to the project team. It discusses the evolution of the SFT frangible joint design, including optimization, testing, and successful partnering of the Johnson Space Center (JSC) engineering and JSC safety organizations, to enhance the flight safety margin for America's next generation of human-rated space vehicles.
Cardiac magnetic resonance imaging provides high spatial resolution, enabling improved extraction of important functional and morphological features for cardiovascular disease staging. Segmentation of ventricular cavities and myocardium in cardiac cine sequencing provides a basis to quantify cardiac measures such as ejection fraction. A method is presented that curtails the expense and observer bias of manual cardiac evaluation by combining semantic segmentation and disease classification into a fully automatic processing pipeline. The initial processing element consists of a robust dilated convolutional neural network architecture for voxel-wise segmentation of the myocardium and ventricular cavities. The resulting comprehensive volumetric feature matrix captures diagnostic clinical procedure data and is utilized by the final processing element to model a cardiac pathology classifier. Our approach evaluated anonymized cardiac images from a training data set of 100 patients (4 pathology groups, 1 healthy group, 20 patients per group) examined at the University Hospital of Dijon. The top average Dice index scores achieved were 0.940, 0.886, 0.849 for structure segmentation of the left ventricle (LV), myocardium and right ventricle (RV) respectively. A 5-ary pathology classification accuracy of 90% was recorded on an independent test set using the trained model. Performance results demonstrate potential for advanced machine learning methods to deliver accurate, efficient and reproducible cardiac pathological assessment.
The Mars Entry, Descent, and Landing Instrumentation (MEDLI2) sensor suite collected data during entry of the Mars 2020 Perseverance rover into Mars’ atmosphere. An inverse estimation of the backshell and heatshield surface aeroheating was performed, using the data from the MEDLI2 Instrumented Sensor Plugs, a network of thermocouples embedded within the thermal protection system across the aeroshell. Monte Carlo analysis was conducted to assess the sensitivity of the surface heat rate, temperature, and heat load to uncertainties in thermocouple depth and material properties. In this paper, a variance decomposition method using Sobol indices was employed to understand the relative contributions of each uncertainty parameter. Performing this analysis using results from the inverse analysis tool FIAT_Opt was found to require incredibly high computation time, and thus machine learning models were trained and evaluated as a surrogate model for FIAT_Opt. This paper demonstrates that machine learning models can be an efficient, accurate alternative to state-of-the-art inverse analysis tools like FIAT_Opt, especially for computationally-expensive processes. Using these models, the sensitivity analysis showed that uncertainties in heat capacity and thermal conductivity were the main drivers for the overall uncertainty in peak reconstructed heating and heat load.
Nucleons in short-range correlated (SRC) pairs, due to their close proximity and high relative momentum, can provide insight into the short-range part of the strong nuclear interaction. In particular, the prevalence of np pairs is due to the dominance of a tensor term for correlated nucleons with momenta of approximately 400?600 MeV/c. This dissertation comprises two studies advancing the community?s understanding of the isospin composition of SRC pairs. First, I performed a study of proton and neutron knockout from initially low-momentum and high-momentum states in 3He. Previous work has shown that protons are disproportionately represented in high-momentum states in neutron-rich nuclei. I demonstrate that spectral functions for the proton-rich nucleus 3He predict, in agreement with data, that neutrons are disproportionately represented in high-momentum states, but that 3He does not display the same strong prevalence of np pairs that is observed in larger nuclei. Second, Generalized Contact Formalism (GCF), a well-supported theory for predicting SRC behavior, predicts the transition from an isospin-dependent, tensor-dominant interaction at intermediate distances to a scalar-dominant, isospin-independent interaction at very short distances. This dissertation uses data from the CLAS12 Nuclear Targets Experiments in Hall B at Jefferson Lab to measure the relative abundances of pp and pn pairs for increasing relative momentum and decreasing separation. I provide an independent confirmation of the previously-observed increase in pp pairs at increasing momentum of the struck nucleon. I also contribute to the application of the new CLAS12 Central Neutron Detector by precisely measuring the neutron detection efficiency and developing a machine learning model for rejecting charged particle background.
Liquid cooling is critical for thermal management in high-density data centers with the rising AI workloads. However, machine learning-based controllers are essential to unlock greater energy efficiency and reliability, promoting sustainability. We present LC-Opt, a Sustainable Liquid Cooling (LC) benchmark environment, for reinforcement learning (RL) control strategies in energy-efficient liquid cooling of high-performance computing (HPC) systems. Built on the baseline of a high-fidelity digital twin of Oak Ridge National Lab's Frontier Supercomputer cooling system, LC-Opt provides detailed Modelica-based end-to-end models spanning site-level cooling towers to data center cabinets and server blade groups. RL agents optimize critical thermal controls like liquid supply temperature, flow rate, and granular valve actuation at the IT cabinet level, as well as cooling tower (CT) setpoints through a Gymnasium interface, with dynamic changes in workloads. This environment creates a multi-objective real-time optimization challenge balancing local thermal regulation and global energy efficiency, and also supports additional components like a heat recovery unit (HRU). We benchmark centralized and decentralized multi-agent RL approaches, demonstrate policy distillation into decision and regression trees for interpretable control, and explore LLM-based methods that explain control actions in natural language through an agentic mesh architecture designed to foster user trust and simplify system management. LC-Opt democratizes access to detailed, customizable liquid cooling models, enabling the ML community, operators, and vendors to develop sustainable data center liquid cooling control solutions.
During the last few years, deep learning in neural networks has demonstrated impressive successes in the areas of computer vision, speech and image recognition, text generation, and many others. However, sensitive engineering areas such as nuclear engineering benefited less from these efficient techniques. In this work, deep learning expert systems are utilized to model and predict time series progression of a design-basis nuclear accident, featuring a loss of coolant accident. Two major findings are accomplished in this work. First, the ability to train expert systems with high accuracy, which could help nuclear power plant operators to figure out plant responses during the accident. Second, building fast, efficient, and accurate deep models to simulate nuclear phenomena, which could be valuable to nuclear computational science. In this work, large amount of time series data is obtained from simulation tools by simulating different conditions of the base-case/nominal accident scenario. Four critical outputs/responses are monitored during the accident (e.g. temperature, pressure, break flow rate, water level). Two approaches are adopted in this work. The first approach is to use feedforward deep neural networks (DNN) to fit all time steps and outputs in a single model. The second approach is to use long short-term memory (LSTM) to fit all time steps together for each reactor response separately. Both DNN and LSTM demonstrate very good performance in predicting the test and base-case scenarios, with accuracy as low as 92% and as high as 99%, where these test scenarios are unknown to the expert systems and are not included in the model training. In addition, both approaches demonstrate a significant reduction in computational costs, as the deep expert system is able to accurately predict the accident 100,000 times faster than the original simulation tool. Given sufficient data, the methodology adopted in this study demonstrates that DNN/LSTM expert systems can be used as a decision support system to model advanced time series phenomena within nuclear power plants with high accuracy and negligible computational costs.
With the recent demand for decarbonization and energy efficiency, advanced HVAC control using Deep Reinforcement Learning (DRL) becomes a promising solution. Due to its flexible structures, DRL has been successful in energy reduction for many HVAC systems. However, only a few researches applied DRL agents to manage the entire central HVAC system and control multiple components in both the water loop and the air loop, owing to its complex system structures. Moreover, those researches have not extended their applications by incorporating the indoor air quality, especially both CO2 and PM2.5concentrations, on top of energy saving and thermal comfort, as achieving those objectives simultaneously can cause multiple control conflicts. What's more, DRL agents are usually trained on the simulation environment before deployment, so another challenge is to develop an accurate but relatively simple simulator. Therefore, we propose a DRL algorithm for a central HVAC system to co-optimize energy consumption, thermal comfort, indoor CO2 level, and indoor PM2.5 level in an office building. To train the controller, we also developed a hybrid simulator that decoupled the complex system into multiple simulation models, which are calibrated separately using laboratory test data. The hybrid simulator combined the dynamics of the HVAC system, the building envelope, as well as moisture, CO2, and particulate matter transfer. Three control algorithms (rule-based, MPC, and DRL) are developed, and their performances are evaluated on the hybrid simulator environment with a realistic scenario (i.e., with stochastic noises). The test results showed that, the DRL controller can save 21.4 % of energy compared to a rule-based controller, and has improved thermal comfort, reduced indoor CO2 concentration. The MPC controller showed an 18.6 % energy saving compared to the DRL controller, mainly due to savings from comfort and indoor air quality boundary violations caused by unmeasured disturbances, and it also highlights computational challenges in real-time control due to non-linear optimization. Finally, we provide the practical considerations for designing and implementing the DRL and MPC controllers based on their respective pros and cons.
Standard structural health monitoring techniques face well-known difficulties for comprehensive defect diagnosis in real-world structures that have structural, material, or geometric complexity. This motivates the exploration of machine-learning-based structural health monitoring methods in complex structures. However, creating sufficient training data sets with various defects is an ongoing challenge for data-driven machine (deep) learning algorithms. The ability to transfer the knowledge of a trained neural network from one component to another or to other sections of the same component would drastically reduce the required training data set. Also, it would facilitate computationally inexpensive machine learning based inspection systems. In this work, a machine-learning-based multi-level damage characterization is demonstrated with the ability to transfer trained knowledge within the sparse sensor network. A novel network spatial assistance and an adaptive convolution technique are proposed for efficient knowledge transfer within the deep learning algorithm. Proposed structural health monitoring method is experimentally evaluated on an aluminum plate with artificially induced defects. It was observed that the method improves the performance of knowledge transferred damage characterization by 50% during localization and 24% during severity assessment. Further, experiments using time windows with and without multiple edge reflections are studied. Results reveal that multiply scattered waves contain rich and deterministic defect signatures that can be mined using deep learning neural networks, improving the accuracy of both identification and quantification. In the case of a fixed sensor network, using multiply scattered waves shows 100% prediction accuracy at all levels of damage characterization.
Plug loads are responsible for a significant portion of the energy consumed in commercial buildings, yet their distributed and ever-changing nature makes them one of the most challenging building end uses to manage. Plug load management systems exist today that utilize smart plugs to meter and control devices at the outlet level, however, their uptake has been relatively slow in part due to the significant labor required for installation and maintenance. Learning behavior algorithms and automatic and dynamic load detection have been identified as two technology areas that could accelerate the adoption of plug load management systems by reducing these labor demands and providing additional energy efficiency and non-energy benefits. Learning behavior algorithms learn occupant behavior and adjust plug load management systems accordingly, allowing for the automatic creation of optimized control schedules. Automatic and dynamic load detection allows a plug load management system to identify devices as they are plugged in to a building and keeps the system up to date as devices are moved throughout a building. In this paper, we present our findings with respect to the current state of these two technologies based on a review of existing research and patents, as well as a series of interviews with companies working in the plug load space. We have found that, as of now, no commercialized solutions exist for these plug load technologies and that more work is needed to bring them to market. In addition, we summarize our findings related to the technology challenges, market barriers, drivers, and opportunities for these technologies moving forward.