Search NASA⌕ Search

SEARCH · Search NASA

Results for “predictive model”

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 1,459 records · Page 81
All sourcesNASA NTRSDOE OSTI
NASA NTRS2008

Model Effects on GLAS-Based Regional Estimates of Forest Biomass and Carbon

ICESat/GLAS waveform data are used to estimate biomass and carbon on a 1.27 million sq km study area. the Province of Quebec, Canada, below treeline. The same input data sets and sampling design are used in conjunction with four different predictive models to estimate total aboveground dry forest biomass and forest carbon. The four models include nonstratified and stratified versions of a multiple linear model where either biomass or (square root of) biomass serves as the dependent variable. The use of different models in Quebec introduces differences in Provincial biomass estimates of up to 0.35 Gt (range 4.942+/-0.28 Gt to 5.29+/-0.36 Gt). The results suggest that if different predictive models are used to estimate regional carbon stocks in different epochs, e.g., y2005, y2015, one might mistakenly infer an apparent aboveground carbon "change" of, in this case, 0.18 Gt, or approximately 7% of the aboveground carbon in Quebec, due solely to the use of different predictive models. These findings argue for model consistency in future, LiDAR-based carbon monitoring programs. Regional biomass estimates from the four GLAS models are compared to ground estimates derived from an extensive network of 16,814 ground plots located in southern Quebec. Stratified models proved to be more accurate and precise than either of the two nonstratified models tested.

Nelson, Ross↗
DOE OSTI2022

Language models for the prediction of SARS-CoV-2 inhibitors

The COVID-19 pandemic highlights the need for computational tools to automate and accelerate drug design for novel protein targets. We leverage deep learning language models to generate and score drug candidates based on predicted protein binding affinity. We pre-trained a deep learning language model (BERT) on ∼9.6 billion molecules and achieved peak performance of 603 petaflops in mixed precision. Our work reduces pre-training time from days to hours, compared to previous efforts with this architecture, while also increasing the dataset size by nearly an order of magnitude. For scoring, we fine-tuned the language model using an assembled set of thousands of protein targets with binding affinity data and searched for inhibitors of specific protein targets, SARS-CoV-2 Mpro and PLpro. We utilized a genetic algorithm approach for finding optimal candidates using the generation and scoring capabilities of the language model. Our generalizable models accelerate the identification of inhibitors for emerging therapeutic targets.

Blanchard, Andrew E.↗
NASA NTRS2018

Nasa GEOS Model for Sub-Seasonal to Seasonal Predictions: The Major Teleconnections, Tropical Cyclone Activity, and ENSO

Ability of the NASA GEOS seasonal prediction model system (the latest version) in predicting the large-scale teleconnections, tropical cyclone (TC) activity, and ENSO are validated. The major large-scale teleconnections include the North Atlantic Oscillation (NAO), the Arctic Oscillation (AO), and the Pacific North American (PNA) that span vast geographical area across the North Pacific/Atlantic and North America. Predictive skill of TC activity is assessed by Genesis potential index (GPI). Anomaly correlations are greater than 0.5 for winter teleconnections at up to 2 month lead and for the first four month (June- July-August-September) GPIs over the North Atlantic and the Western Pacific. Correlation for the long-range prediction of the ENSO (Nino3.4 SST) maturity reaches 0.9 at 6 month lead and 0.8 at 9 month lead.

Lim, Young-Kwon↗
NASA NTRS1995

The use of a block diagram simulation language for rapid model prototyping

The research performed this summer focussed on the development of a predictive model for the loading of liquid oxygen (LO2) into the external tank (ET) of the shuttle prior to launch. A predictive model can greatly aid the operational personnel since instrumentation aboard the orbiter and ET is limited due to weight constraints. The model, which focuses primarily on the orbiter section of the system was developed using a block diagram based simulation language known as VisSim. Simulations were run on LO2 loading data for shuttle flights STS50 and STS55 and the model was demonstrated to accurately predict the sensor data recorded for these flights. As a consequence of the simulation results, it can be concluded that the software tool can be very useful for rapid prototyping of complex models.

Whitlow, Jonathan E.↗
DOE OSTI2023

Dimensionally reduced machine learning model for predicting single component octanol–water partition coefficients

Abstract MF-LOGP, a new method for determining a single component octanol–water partition coefficients ( $$LogP$$ LogP ) is presented which uses molecular formula as the only input. Octanol–water partition coefficients are useful in many applications, ranging from environmental fate and drug delivery. Currently, partition coefficients are either experimentally measured or predicted as a function of structural fragments, topological descriptors, or thermodynamic properties known or calculated from precise molecular structures. The MF-LOGP method presented here differs from classical methods as it does not require any structural information and uses molecular formula as the sole model input. MF-LOGP is therefore useful for situations in which the structure is unknown or where the use of a low dimensional, easily automatable, and computationally inexpensive calculations is required. MF-LOGP is a random forest algorithm that is trained and tested on 15,377 data points, using 10 features derived from the molecular formula to make $$LogP$$ LogP predictions. Using an independent validation set of 2713 data points, MF-LOGP was found to have an average $$RMSE$$ RMSE = 0.77 ± 0.007, $$MAE$$ MAE = 0.52 ± 0.003, and $${R}^{2}$$ R 2 = 0.83 ± 0.003. This performance fell within the spectrum of performances reported in the published literature for conventional higher dimensional models ( $$RMSE$$ RMSE = 0.42–1.54, $$MAE$$ MAE = 0.09–1.07, and $${R}^{2}$$ R 2 = 0.32–0.95). Compared with existing models, MF-LOGP requires a maximum of ten features and no structural information, thereby providing a practical and yet predictive tool. The development of MF-LOGP provides the groundwork for development of more physical prediction models leveraging big data analytical methods or complex multicomponent mixtures. Graphical Abstract

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1972

A model to predict the degradation of a SiGe thermopile.

Discussion of the various phenomena which affect the available output power of a typical radioisotope thermoelectric generator (RTG), including estimates of the magnitude of these effects. Power degradation, as well as the corresponding changes in operating temperature, are calculated by means of a computer code which systematically accounts for the time- and temperature-dependent variables. A model for predicting available output power of an RTG is developed which includes changes in performance due to fuel decay, alteration of bulk thermoelectric properties, changes in electrode contacts, geometry changes due to material sublimation, and degradation of the thermal insulation.

Stapfer, G.↗
DOE OSTI2026

Generative large language models for predictive maintenance planning

Maintenance planning and the generation of necessary components for tasks can prove time-consuming and complex. Automating the creation of recurring or similar tasks by leveraging previous planning packages and data, while uncovering insights to automate planning package generation, presents an opportunity to conserve valuable time and resources. This work aims to harness the textual and probabilistic capabilities of large language models (LLMs) to automate the generation of planning packages. Utilizing diverse data sources ranging from raw data to handwritten text, both singular and collaborative LLMs are trained and tested. Results demonstrate their capability to generate essential planning package components, effectively replicating the statistical patterns in the data. This demonstrates the use of these tools inside a digital asset for automated planning. This work outlines a methodology for constructing datasets, a training suite, and evaluation methods for LLM-based textual and conversational planning tools utilized in an asset digital twin. Results indicate that the fine-tuned models generate estimated planning information within the statistical ranges observed in real maintenance data. The models achieve high accuracy (>90%) in document question-answering and instruction generation tasks. Furthermore, the conversational retrieval-augmented generation (RAG) assistant system achieves 100% document retrieval accuracy, while conversational information capture exceeds 98% across the majority of work-package assistant modules.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2006

A Mathematical Model to Predict and Maintain the Neutral Buoyancy of Suited Astronauts

A previous study reported that inadequate weigh outs of suited subjects contribute to fatigue and the risk of injury during training in the Neutral Buoyancy Laboratory (NBL). Another study suggested that shoulder injuries observed in suited subjects who train in the NBL may be attributed to excessive righting moments caused by a non-optimal weigh out. The purpose of this study was to develop a mathematical model to predict and maintain the neutral buoyancy of suited subjects during training operations at the NBL. Due to time constraints, one certified NBL support diver served as a subject (height: 66.54 in; weight: 182 lbs) for this study and only one complete test was conducted. The study was divided into two runs for which the first run required the NBL divers to perform a weigh out similar to a suited astronaut on a scuba diver wearing a mock Portable Life Support System and a Displays and Control Module. For the second run, the same subject and equipment were weighed out according to the mathematical model. The objective of each run was to achieve a neutrally buoyant subject floating 450 to the pool floor. Motion data was collected using two underwater cameras and analyzed using Dartfish video analysis software while force and moment data were recorded using an AMTI force plate. The results from the NBL divers visual run indicate that the subject was floating at an angle of 29.50 while the resultant force and moment data were 1.139 lb and 1.125 ft-lb respectively. The mathematical model s weigh out resulted in the subject floating at an angle of 37.40 and a resultant force of 0.765 lb and resultant moment of 1.248 ft-lb. The mathematical model was better able to orient the subject and reduce resultant moment and force as compared to the NBL divers.

Clowers, Kurt↗
DOE OSTI2016

Analyses of Model-Based Predictive Control for a S-CO 2 Brayton Cycle Power Converter

The supercritical carbon dioxide (S-CO 2 ) Brayton cycle is a promising alternative to steam cycles as an energy converter for advanced nuclear reactors as well as fossil-fired and concentrated solar power plants. For advanced reactors, such as Sodium-cooled Fast Reactors (SFRs), the S-CO 2 cycle offers the benefits of high cycle efficiency, very compact turbomachinery, and elimination of sodium-water reactions, all of which may enable improved economics. The efficiency gains calculated for the S-CO 2 cycle are particularly significant if advantage can be taken of the sharp changes in the carbon dioxide thermo-physical properties near the critical point. On the other hand, the same property variations give rise to challenges in the cycle design, analysis, and control. The onset of two-phase flow (where the minimum cycle temperature and pressure drop drop below the critical value) can degrade compressor performance and potentially cause damage to the compressor blades. In previous work, the feasibility of the Model-based Predictive Control (MPC) approach was assessed for improving regulation of the precooler outlet temperature. A linear reduced-order model for the precooler was generated and for this simplified standalone system operating with an MPC controller the response of outlet temperature was evaluated through simulations. These evaluations included the presence of temperature measurement time delay representing a degraded sensor. It was demonstrated that the MPC controller maintained stable precooler temperature for those delay cases where a proportional-integral controller became unstable. In the present work the simulations have been extended to evaluate the MPC controller operating under more realistic conditions, namely in the split-flow recompression cycle non-linear plant. This compares with the previous standalone reduced-order linear system described above. In the present work the MPC-based control scheme was implemented in a GPASS 1-D systems code representation of the split-flow recompression cycle. The performance of the controller was again investigated for degraded sensor conditions and compared to that of a proportional-integral controller. The results were qualitatively similar to the previous results with the MPC controller and again providing superior stability characteristics. This work provides a sound starting point for the development of model-based control systems. In particular, the MPC control approach applied to Multiple Input Multiple Output systems can be used to synchronize the operation of the different system components to improve control system dynamic performance.

20 FOSSIL-FUELED POWER PLANTS↗
NASA NTRS2000

Laser-Induced Fluorescence Measurements and Modeling of Nitric Oxide in Counterflow Diffusion Flames

The feasibility of making quantitative nonintrusive NO concentration ([NO]) measurements in nonpremixed flames has been assessed by obtaining laser-induced fluorescence (LIF) measurements of [NO] in counterflow diffusion flames at atmospheric and higher pressures. Comparisons at atmospheric pressure between laser-saturated fluorescence (LSF) and linear LIF measurements in four diluted ethane-air counterflow diffusion flames with strain rates from 5 to 48/s yielded excellent agreement from fuel-lean to moderately fuel-rich conditions, thus indicating the utility of a model-based quenching correction technique, which was then extended to higher pressures. Quantitative LIF measurements of [NO] in three diluted methane-air counterflow diffusion flames with strain rates from 5 to 35/s were compared with OPPDIF model predictions using the GRI (version 2.11) chemical kinetic mechanism. The comparisons revealed that the GRI mechanism underpredicts prompt-NO by 30-50% at atmospheric pressure. Based on these measurements, a modified reaction rate coefficient for the prompt-NO initiation reaction was proposed which causes the predictions to match experimental data. Temperature measurements using thin filament pyrometry (TFP) in conjunction with a new calibration method utilizing a near-adiabatic H2-air Hencken burner gave very good comparisons with model predictions in these counterflow diffusion flames. Quantitative LIF measurements of [NO] were also obtained in four methane-air counterflow partially-premixed flames with fuel-side equivalence ratios (phi(sub B)) of 1.45, 1.6, 1.8 and 2.0. The measurements were in excellent agreement with model predictions when accounting for radiative heat loss. Spatial separation between regions dominated by the prompt and thermal NO mechanisms was observed in the phi(sub B) = 1.45 flame. The modified rate coefficient proposed earlier for the prompt-NO initiation reaction improved agreement between code predictions and measurements in the region where prompt-NO dominates. Finally, LIF measurements of NO were obtained in counterflow diffusion flames at 2 to 5 atm. Comparisons between [NO] measurements and predictions show that the GRI mechanism underpredicts prompt-NO by a factor of two to three at all pressures. In general, the results indicate a need for refinement of the CH chemistry, especially the pressure-dependent CH formation and destruction reactions.

Ravikrishna, Rayavarapu V.↗
DOE OSTI2017

Numerical investigation of Lower Tuscaloosa Sandstone and Selma Chalk caprock under geological CO 2 sequestration conditions: mineral precipitation and permeability evolution

Abstract A numerical model was developed using CrunchFlow to simulate reactive transport and porosity and permeability changes of sandstone and carbonate rock samples taken from the Lower Tuscaloosa Formation and the Selma Chalk Formation, Jackson County, MS, USA. The model predicted a permeability decrease from 2190 mD to 2038 mD for the Lower Tuscaloosa Sandstone sample in a static batch reactor after 180 days of exposure to CO 2 ‐saturated brine, which is consistent with measured permeability results. The model predicted a negligible permeability change from 2.00 mD to 2.08 mD for the Selma Chalk carbonate sample after 180 days of exposure to CO 2 ‐saturated brine. Based on model prediction, key mineral dissolution and precipitation reactions in the Lower Tuscaloosa Sandstone sample include dissolution of quartz, chlorite, and feldspar, as well as precipitation of amorphous silica and kaolinite. For the Selma Chalk carbonate sample, key predicted reactions include dissolution of calcite, quartz and chlorite, and precipitation of kaolinite and amorphous silica. Initial porosity, initial feldspar content and the exponent n value (related to pore structure and tortuosity) used in permeability calculations were three important factors affecting permeability evolution of sandstone samples under CO 2 sequestration conditions. The small permeability change predicted for both the Lower Tuscaloosa Sandstone and the Selma Chalk caprock after exposure to CO 2 ‐saturated brine suggests that poro‐permeability changes during CO 2 injection into the Lower Tuscaloosa Formation are not likely to significantly affect reservoir and seal quality. © 2017 Society of Chemical Industry and John Wiley & Sons, Ltd.

Zhang, Liwei↗
DOE OSTI2020

Shell Predictive Battery Life Models (SPBLM) [SWR-20-35]

Battery life models to predict the capacity fade and resistance growth of lithium-ion batteries from various manufacturers with different chemistries. The battery life model is trained on experimental test data, and accounts for calendar and cycling aging mechanisms as functions of cell environmental conditions and use. The battery life model is then used to predict capacity fade and resistance growth under any arbitrary aging scenario.

Smith, Kandler↗
DOE OSTI2023

Bayesian-Motivated Probabilistic Model of Hurricane-Induced Multimechanism Flood Hazards

Multimechanism floods (MMFs) are caused by the simultaneous occurrence of more than one flood mechanism such as storm surge, precipitation, tides, and waves. MMFs can lead to more severe or differing impacts than single-mechanism floods. As a result, comprehensive risk assessments require the ability to assess the multivariate probabilistic behaviors of hazards from MMFs. Here this study introduces a novel Bayesian-motivated approach for the probabilistic assessment of hurricane-induced hazards from the combination of the surge, precipitation, tides, and river antecedent flow. A Bayesian network (BN) is developed to capture the physical (conditional) relationship between variables and facilitate the generation of a hazard curve for river discharge that captures the contributions from multiple flood drivers. A case study located along the Delaware River is used to illustrate the proposed approach. Five computationally efficient representative predictive models are developed to estimate the conditional distributions required for the BN as a means of demonstrating the overall framework. The predictive models used in this study act as placeholders and can be replaced with more sophisticated and high-fidelity models depending on the desired accuracy level. While the predictive models are intended to be representative and illustrative, the model performance is evaluated using three historical storms that affected the area. Overall, the proposed framework is shown to be transparent, effective, and adaptable.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Data to Accompany: PM2.5 is insufficient to explain personal PAH exposure

Fine particulate matter (PM2.5) air quality index (AQI) data from outdoor stationary monitors and Hazard Mapping System (HMS) smoke density data from satellites are often used as proxies for personal chemical exposure. Silicone wristbands can quantify more individualized exposure data than stationary air monitors or smoke satellites. However, it is not understood how these proxy measurements compare to chemical data measured from wristbands. We hypothesized that predictive models for personal chemical exposure would be significantly improved by expanding beyond stationary PM2.5 AQI data or satellite HMS data to also include environmental and behavioral information. In Eugene, Oregon, participants wore daily wristbands, carried a phone that recorded locations, and answered daily questionnaires for a seven-day period in multiple seasons. We gathered publicly available daily PM2.5 AQI data and HMS data. We analyzed wristbands for 94 organic chemicals, including 53 polycyclic aromatic hydrocarbons (PAHs). Wristband chemical detections and concentrations, behavioral variables (e.g., time spent indoors), and environmental conditions (e.g., PM2.5 AQI) significantly differed between seasons. Machine learning models were fit to predict personal chemical exposure using PM2.5 AQI only, HMS only, and a multivariate feature set including PM2.5 AQI, HMS, and other environmental and behavioral information. On average, the multivariate models increased predictive accuracy by approximately 70% compared to either the AQI model or the HMS model for all chemicals modeled. This study provides evidence that PM2.5 AQI data alone or HMS data alone is insufficient to explain personal chemical exposures. Our results identify additional key predictors of personal chemical exposure.

Bramer, Lisa M↗
NASA NTRS1981

Durability/life of fiber composites in hygrothermomechanical environments

Statistical analysis and multiple regression were used to determine and quantify the significant hygrothermomechanical variables which infuence the tensile durability/life (cycle loading, fatigue) of boron-fiber/epoxy-matrix (B/E) and high-modulus-fiber/epoxy-matrix (HMS/E) composites. The use of the multiple regression analysis reduced the variables from fifteen, assumed initially, to six or less with a probability of greater than 0.999. The reduced variables were used to derive predictive models for compression an intralaminar shear durability/life of B/E and HMS/E composites assuming isoparametric fatigue behavior. The predictive models were subsequently generalized to predict the durability/life of graphite-fiber-r generalized model is of simple form, predicts conservative values compared with measured data and should be adequate for use in preliminary designs.

Chamis, C. C.↗
NASA NTRS1987

A low-Reynolds-number two-equation turbulence model for predicting heat transfer on turbine blades

A modified form of the Lam-Bremhorst low-Reynolds number kappa-epsilon turbulence model was developed for predicting transitional boundary layer flows under conditions characteristic of gas turbine blades. The application of the model to flows with pressure gradients is described. Tests against a number of turbine blade cascade data sets are included. Some additional refinements of the model that were made in recent months are explained.

Patankar, Suhas V.↗
DOE OSTI2017

Energy Storage Sizing Taking Into Account Forecast Uncertainties and Receding Horizon Operation

Energy storage systems (ESS) have the potential to be very beneficial for applications such as reducing the ramping of generators, peak shaving, and balancing not only the variability introduced by renewable energy sources, but also the uncertainty introduced by errors in their forecasts. Optimal usage of storage may result in reduced generation costs and an increased use of renewable energy. However, optimally sizing these devices is a challenging problem. This paper aims to provide the tools to optimally size an ESS under the assumption that it will be operated under a model predictive control scheme and that the forecast of the renewable energy resources include prediction errors. A two-stage stochastic model predictive control is formulated and solved, where the optimal usage of the storage is simultaneously determined along with the optimal generation outputs and size of the storage. Wind forecast errors are taken into account in the optimization problem via probabilistic constraints for which an analytical form is derived. This allows for the stochastic optimization problem to be solved directly, without using sampling-based approaches, and sizing the storage to account not only for a wide range of potential scenarios, but also for a wide range of potential forecast errors. In the proposed formulation, we account for the fact that errors in the forecast affect how the device is operated later in the horizon and that a receding horizon scheme is used in operation to optimally use the available storage.

25 ENERGY STORAGE↗
NASA NTRS1983

Development of a computational model for predicting solar wind flows past nonmagnetic terrestrial planets

A computational model for the determination of the detailed plasma and magnetic field properties of the global interaction of the solar wind with nonmagnetic terrestrial planetary obstacles is described. The theoretical method is based on an established single fluid, steady, dissipationless, magnetohydrodynamic continuum model, and is appropriate for the calculation of supersonic, super-Alfvenic solar wind flow past terrestrial ionospheres.

Stahara, S. S.↗