SEARCH · Search NASA
Results for “FORECASTS”
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.
Analyzing inference workloads for spatiotemporal modeling
Ensuring power grid resiliency, forecasting climate conditions, and optimization of transportation infrastructure are some of the many application areas where data is collected in both space and time. Spatiotemporal modeling is about modeling those patterns for forecasting future trends and carrying out critical decision-making by leveraging machine learning/deep learning. Once trained offline, field deployment of trained models for near real-time inference could be challenging because performance can vary significantly depending on the environment, available compute resources and tolerance to ambiguity in results. Users deploying spatiotemporal models for solving complex problems can benefit from analytical studies considering a plethora of system adaptations to understand the associated performance-quality trade-offs. To facilitate the co-design of next-generation hardware architectures for field deployment of trained models, it is critical to characterize the workloads of these deep learning (DL) applications during inference and assess their computational patterns at different levels of the execution stack. In this paper, we develop several variants of deep learning applications that use spatiotemporal data from dynamical systems. We study the associated computational patterns for inference workloads at different levels, considering relevant models (Long short-term Memory, Convolutional Neural Network and Spatio-Temporal Graph Convolution Network), DL frameworks (Tensorflow and PyTorch), precision (FP16, FP32, AMP, INT16 and INT8), inference runtime (ONNX and AI Template), post-training quantization (TensorRT) and platforms (Nvidia DGX A100 and Sambanova SN10 RDU). Overall, our findings indicate that although there is potential in mixed-precision models and post-training quantization for spatiotemporal modeling, extracting efficiency from contemporary GPU systems might be challenging. Instead, co-designing custom accelerators by leveraging optimized High Level Synthesis frameworks (such as SODA High-Level Synthesizer for customized FPGA/ASIC targets) can make workload-specific adjustments to enhance the efficiency.
Assignment and calibration of relative permeability by hydrostratigraphic units for multiphase flow analysis, case study: CO 2 -EOR operations at the Farnsworth Unit, Texas
Included in the most critical factors for geological CO 2 storage site screening, selection, and operation is effective simulations of multiphase flow and transport. Relative permeability is probably the greatest source of potential uncertainty in multiphase flow simulation, second only to intrinsic permeability heterogeneity. The specific relative permeability relationship assigned greatly impacts forecasts of CO 2 trapping mechanisms, phase behavior, and long-term plume movement. A primary goal of this study is to evaluate the impacts and implications of different methods of assigning relative permeability relationships for CO 2 -EOR model forecasts. Most simulation studies published in the literature base selection of relative permeability functions on the geologic formation or rock type alone. In this study, we initially implemented reservoir model grids with previously-identified hydrostratigraphic units based on porosity and permeability relationship of the Morrow ‘B’ Sandstone, then assigned relative permeability functions for those hydrostratigraphic units. Specific, constrained relative permeability relationships were created and assigned to each hydrostratigraphic unit using petrophysical data and Mercury Intrusion Capillary Pressure (MICP) measurements, from core samples of each hydrostratigraphic unit. Results of forward simulations with the newly-calibrated models will be compared to those of previous models as well as to simulation results for a range of different relative permeability relationships. The study site is the Farnsworth Unit (FWU) in the northeast Texas Panhandle, an active CO 2 -EOR operation. The target formation is the Morrow ‘B’ Sandstone, a clastic formation composed of medium to course sands.
Statistical analysis of historic hydrocarbon production data from Gulf of Mexico oil and gas fields and application to dynamic capacity assessment in CO 2 storage
Numerical modeling of CO 2 injection and reservoir flow is typically performed to forecast the number of wells, sustainable injection rates, and total storage volume (or mass). A critical determination for CO 2 storage in depleted oil and gas reservoirs is characterization of reservoir compartmentalization which informs boundary conditions in simulating injection scenarios. Constraining boundary conditions during CO 2 injection into geological formations is a key factor for feasible deployment project. Production history data from 616 wells in 100 oil and gas fields from Gulf of Mexico (GOM) offshore basin can be used to evaluate boundary conditions and total production (oil and gas) and therefor constrain the capacity for potential carbon storage. In general, the combination of decline curve analysis and statistical analysis (to specify constraining boundary conditions) allows the determination of the range of reservoir performance if existing inactive production wells in GOM area of study are used reversibly for CO 2 injection. To constrain the mass of CO 2 which can be injected, it is useful to consider estimates of cumulative bulk (hydrocarbon + brine) production (CBP), which can be converted to equivalent CO 2 mass considering reservoir conditions. Summary CBP statistics are presented as a probability of non-exceedance (PNE), providing a forecast of likely injection rates and masses for other located CO 2 storage projects with similar geology and boundary conditions in the future. The 50% PNE for Equivalent CO 2 is a novel quantitative approach to investigate the possible injection capacity in CO 2 storage projects. The PNE sensitivity analysis shows that reservoir age, drive mechanism, reservoir trap, and reservoir porosity are the key controlling parameters for productivity and consequently optimum CO 2 storage capacity. Furthermore another key finding is the negligible correlation between CBP with reservoir transmissivity and porosity, which implies that other factors than just petrophysical parameters should be studied as constraining factors for CO 2 storage statistical analysis.
Investigations of vertical wind variations at a mountain top in the Himalaya using Doppler Lidar observations and model simulations
Hight-resolved observations of vertical winds remain nearly non-existing over the Himalayas, despite of anticipated crucial role of vertical motions in transporting pollution across the Himalayan hms. The present study analyze the vertical wind observations from surface to 1 km above ground level over Manora Peak (29.4° N; 79.5° E; 1958 m ams!) in the Himalaya performed using a Doppler Lidar during the Ganges Valley Aerosol Experiment (GVAX). Vertical wind exhibited a pronounced diurnal variability at Manora Peak comprising of upward motions during the daytime (05-10 UT) and downward motions during nighttime typical of a mountainvalley system. Mean vertical wind speeds are observed to be varying from -0.8 to +0.8 ms- 1 during the study period with a variance of 0.1-1.5 m2s·2, which is attributed to the thermally driven turbulence. Mean vertical winds are observed to be stronger in the Doppler Lidar profiles above Manora Peak (-0_8 to 0.8 ms- 1) as compared to near surface measurements at this station using an ultrasonic anemometer ( -0.4 to 0.4 ms - 1 ), and low altitude stations in India. Daytime vertical wind speeds are observed to be higher during pre-monsoon (0.81 ms- 1), as compared to post-monsoon (0.24 ms- 1) and winter (0.33 ms- 1)_ Average Black Carbon (BC) concentrations are significantly higher during strong upward vertical winds, which indicates efficient transport of polluted air mass from low-altitude regions to the Himalaya. Weather Research and Forecasting (WRF) model reproduces the observed diurnal pattern in the vertical wind at the observation site however the model underestimates the variability.
EPIsembleVis: A geo-visual analysis and comparison of the prediction ensembles of multiple COVID-19 models
In this work, we present EPIsembleVis, a web-based comparative visual analysis tool for evaluating the consistency of multiple COVID-19 prediction models. Our approach analyzes a collection of COVID-19 predictions from different epidemiological models as an ensemble and utilizes two metrics to quantify model performance. These metrics include (a) prediction uncertainty (represented as the dispersion of predictions in each ensemble) and (b) prediction error (calculated by comparing individual model predictions with the recorded data). Through an interactive visual interface, our approach provides a data-driven workflow for (a) selecting and constructing the COVID-19 model prediction ensemble based on the spatiotemporal overlap of available predictions of multiple epidemiological models, (b) quantifying the model performance using both the uncertainty of each model prediction ensemble, and the error of each ensemble member that represents individual model predictions, and (c) visualizing the spatiotemporal variability in the projection performance of individual models using a suite of novel ensemble visualization techniques, such as the data availability map, a spatiotemporal textured-tile calendar, multivariate rose chart, and time-series leaflet glyph. We demonstrate the capability of our ensemble visual interface through a case study that investigates the performance of weekly COVID-19 predictions, which are provided through the COVID-19 Forecast Hub UMass-Amherst Influenza Forecasting Center of Excellence [47] for the United States and United States Territories. The EPIsembleVis tool is implemented using open-source web technologies and adaptive system design, rendering it interoperable with Elasticsearch and Kibana for automatically ingesting COVID-19 predictions from online repositories, and it is generalizable for analyzing worldwide projections from more epidemiological models.
Numerical simulation of seawater intrusion to coastal aquifers and brine water/freshwater interaction in south coast of Laizhou Bay, China
Seawater intrusion and brine water/freshwater interaction have significantly affected agriculture, industry and public water supply at Laizhou Bay, Shandong Province, China. A two-dimensional SEAWAT model is developed to simulate the seawater intrusion to coastal aquifers and brine water/fresh water interaction in the south of Laizhou Bay. This model is applied to predict the seawater intrusion and brine water/freshwater interface development in the coming years. The model profile is perpendicular to the coastal line with two interfaces, freshwater-saline water interface near the shore and inland brine water-saline water-seawater interface. The hydrogeological parameters in the SEAWAT-2000 model are calibrated by the head and salinity measurements. The precipitation infiltration coefficient, boundary conditions and thicknesses of aquifers are studied in a sensitivity analysis. The predicted results indicate that equivalent freshwater head in shallow freshwater-saline water area will decline 2.0 m by the end of the forecasting period, caused by groundwater over-pumping for farmland irrigation. The groundwater head in the brine-saline water area will also decrease about 1.8 m by the end of forecasting period, caused by excessive brine mining. Salinity decreases below 105 g/L in the brine area, but increases in other areas and contaminates fresh groundwater resources.
Sensitivity and history match analysis of a carbon dioxide “huff-and-puff” injection test in a horizontal shale gas well in Tennessee
Due to improvements in well development, shale gas production has gained importance recently, especially in the United States. To improve gas production and develop a better understanding of shale reservoirs, researchers are conducting field tests to monitor CO 2 storage and enhanced gas recovery. Reservoir simulations can then utilize these field results for sensitivity studies, uncertainty analysis, history matching, gas production forecasts, and CO 2 storage capacity estimations. One such field test was performed for the Chattanooga Shale formation in Morgan County, Tennessee. Approximately 463 tonnes (510 tons) of CO 2 was successfully injected into a hydraulically fractured horizontal well over a thirteen-day period in March 2014. The injection test was achieved in four steps: pre-injection, injection, soaking, and flowback. In this paper, those steps were modeled with a reservoir simulator to match the historic production and injection rates with gas component compositions and forecast the production for five years. The reservoir fluid was modeled as a multi-component gas, including CH 4 , C 2 H 6 , C 3 H 8 , N 2 , and CO 2 . Langmuir constants, reservoir pressure, and fracture network volume were adjusted to match simulation results with observed production rates and gas composition. Results showed that, under the same reservoir conditions, each gas component behaves differently by way of compositional changes in production. CH4 behaves like N2, while CO 2 behaves like heavier hydrocarbons such as C 2 H 6 and C 3 H 8 . CO 2 plume results showed that, after injection, the produced CO 2 is mostly derived from fractured limestone (i.e., Fort Payne Formation) because the injected CO 2 is not adsorbed in limestone but rather by shale formations.
Feedforward-feedback ammonia control at a water resource recovery facility based on a digital twin with hybrid model
Ammonia-based aeration control (ABAC) at full-scale Water Resource Recovery Facilities (WRRFs) can be challenged by diurnal loading and transport delays. This work addressed these challenges using a hybrid feedforward–feedback controller built on Activated Sludge Model 1 (ASM1), marking the first full-scale deployment to pair a mechanistic feedforward core with data-driven corrections. The objectives were to improve ammonia setpoint tracking, assess performance of the mechanistic model when enhanced with data-driven corrections, and document full-scale operation. The hybrid model incorporates two data-driven components: (1) a Mechanistic Error Forecasting Engine (MEFE), consisting of a multivariate linear regressor and a long short-term memory (LSTM) ensemble. Defying expectations, low-parameter models outperformed more complex alternatives, reducing the mechanistic error by 71%. (2) A Residual Oscillation Forecasting Engine (ROFE), based on Fast Fourier Transform, reduced the remaining error by another 35%. Two proportional–integral (PI) feedback loops further (i) trim the feedforward output and (ii) eliminate residual controller error in the final aerobic zone. In full-scale operation, the controller reduced mean-squared error (MSE) by 94% over the baseline and produced more stable dissolved oxygen (DO) setpoints. Overall, it was proven that layering multi-timescale data-driven models on a mechanistic core can yield reliable ABAC performance at WRRFs.
Microstructure and mechanical properties of a novel rapidly solidified, high-temperature Al-alloy
Rapid solidification (RS) processing, as a production method, offers a variety of unique properties based on far-from-equilibrium microstructures obtained through rapid cooling rates. In this study, we seek to investigate the microstructures and properties of a novel Al-alloy specifically designed for high temperature mechanical stability. Synthesis of, AlFe{sub 11.4}Si{sub 1.8}V{sub 1.6}Mn{sub 0.9} (wt.%), was performed by two approaches: rotating cup atomization (“shot”) and melt spinning (“flake”). These methods were chosen because of their ability to produce alloys with tailored microstructures due to their inherent differences in cooling rate. The as-solidified precursor materials were microstructurally characterized with electron microscopy. The results show that the higher cooling rate flake material exhibited the formation of nanocrystalline regions as well additional phase morphologies not seen in the shot material. Secondary dendritic branching in the flake material was on the order of 0.1–0.25 μm whereas branching in the shot material was 0.5–1.0 μm. Consolidated and extruded material from both precursor materials was mechanically evaluated at both ambient and high (300 °C) temperature. The consolidated RS flake material is shown to exhibit higher strengths than the shot material. The ultimate tensile strength of the melt spun flake was reported as 544.2 MPa at room temperature and 298.0 MPa at 300 °C. These results forecast the ability to design alloys and processing approaches with unique non-equilibrium microstructures with robust mechanical properties at elevated temperatures. - Highlights: • A novel alloy, AlFe{sub 11.4}Si{sub 1.8}V{sub 1.6}Mn{sub 0.9} was fabricated by rapid solidification. • Room temperature yield strength exceeded 500 MPa. • Elevated temperature (300 °C) yield strength exceeded 275 MPa. • Forging, after extrusion of the alloy resulted in microstructural coarsening. • Decreased strength and ductility was identified as a result of forging.
Confidence-weighted integration of human and machine judgments for superior decision-making
Large language models (LLMs) can surpass humans in certain forecasting tasks. What role does this leave for humans in the overall decision process? One possibility is that humans, despite performing worse than LLMs, can still add value when teamed with them. A human and machine team can surpass each individual teammate when team members’ confidence is well calibrated and team members diverge in which tasks they find difficult (i.e., calibration and diversity are needed). We simplified and extended a Bayesian approach to combining judgments using a logistic regression framework that integrates confidence-weighted judgments for any number of team members. Using this straightforward method, we demonstrated its effectiveness in both image classification and neuroscience forecasting tasks. Combining human judgments with one or more machines consistently improved overall team performance. Our hope is that this simple and effective strategy for integrating the judgments of humans and machines will lead to productive collaborations.
Application of predictive data analytics to model daily hydrocarbon production using petrophysical, geomechanical, fiber-optic, completions, and surface data: A case study from the Marcellus Shale, North America
Predicting gas production from stimulated unconventional reservoirs has always been a challenge to oil and gas companies because the physics of fluid flow through such reservoirs is not well understood. On the other hand, modeling gas production from shale reservoirs is a complex multi-variate problem that requires proper integration of data from multiple disciplines, such as geology, petrophysics, geomechanics, completions, and surface-based measurements, etc. Here this study demonstrates the application of data-driven machine learning algorithms, integrating geoscientific, distributed acoustic sensing (DAS), distributed temperature sensing (DTS) fiber-optic, completions, flow scanner production log, and surface data to model daily gas production from a 28-stage stimulated horizontal well drilled in the Marcellus Shale of the Appalachian basin in North America. In addition, this study aims to utilize the data from a fiber-optic monitoring system, such as DAS and DTS to evaluate the well performance. We build supervised data-driven machine learning models using Random Forest (RF), Artificial Neural Network (ANN), and Support Vector Machine (SVM) algorithms to understand the well performance and forecast the daily gas production. The study compares the forecasted results, tests the level of accuracy, and addresses different issues with these machine learning algorithms. A spatio-temporal database is constructed and used to perform sensitivity analyses to identify the key drivers for gas production. The results show that RF and ANN algorithms can be used to predict daily gas production with significantly high accuracy; however, RF algorithm is the best predictor in terms of highest accuracy (96%), less computational time and cost. Based on sensitivity analysis, it appears that Poisson's ratio, minimum horizontal stress, DAS, casing pressure, gamma log are the most important parameters to predict gas production.
The global energy footprint of information and communication technology electronics in connected Internet-of-Things devices
Information and communication technology (ICT) has been critical to the emergence of a new Internet-of-Things (IoT) era, as a rapidly increasing number and variety of IoT electronic devices in a wide range of end-use sectors require massive data communication in terms of volume, velocity/speed, and variety. Additionally, semiconductor technology advancements have played a key role in ICT and unlike the past ICT energy footprint analysis, which focused on the final electronic product, an attempt has been made to examine the global energy impacts for the period 2016–2025 based on the market forecasts of three major IoT IC types, i.e., sensors and actuators, connectivity, and processor ICs distributed among four major IoT device components.Total global IoT semiconductor primary energy demand is projected to increase from 2 EJ in 2016 to 35 EJ by 2025, resulting mainly from a substantial projected increase energy needed in the manufacturing of ICs, with growth in the use of sensors and energy-intensive next fab generation manufacturing. Unlike the trend in manufacturing energy, total global operational energy use is projected to decrease significantly with the development of smaller transistor size, low-power devices, and faster wireless data communication technology. Total operational energy is projected to significantly decrease from 118 TWh/year (1,2 EJ/year) to less than 1 TWh/year (0.003 EJ/year) in spite of a significantly higher anticipated growth in its market size during the 2016–2025 forecast period. Its annual share of total primary energy use is projected to reach an insignificant level of 0.01% by the end of 2025, consistent with the past limited electronic product life cycle analyses besides the analysis focus here limited to ICs in IoT devices.
GenAI4UQ: A software for forward and inverse uncertainty quantification using conditional generative AI
We introduce GenAI4UQ, a software package for forward and inverse uncertainty quantification in model calibration, parameter estimation, and ensemble forecasting. GenAI4UQ leverages a generative AI-based conditional modeling framework to address limitations of traditional inverse modeling techniques, such as Markov Chain Monte Carlo (MCMC) methods. By replacing computationally intensive iterative processes with a direct, learned mapping, GenAI4UQ enables efficient calibration of input parameters and generation of predictions directly from observations. The software supports rapid ensemble forecasting with robust uncertainty quantification while maintaining computational and storage efficiency. Built-in auto-tuning of hyperparameters simplifies model training, ensuring accessibility for users with varying expertise. Its versatile conditional generative framework is applicable across diverse scientific domains. While GenAI4UQ offers significant advantages in flexibility and efficiency, users should interpret its uncertainty estimates with caution in data-sparse scenarios, as the model may overestimate uncertainty—an effect common to all surrogate-based approaches including MCMC with surrogate models. Despite this, GenAI4UQ transforms inverse modeling by providing a fast, reliable, and user-friendly solution. It empowers researchers and practitioners to quickly estimate parameter distributions and generate model predictions for new observations, facilitating efficient decision-making and advancing the state of uncertainty quantification in computational modeling.
Constructing probabilistic scenarios for wide-area solar power generation
Optimizing thermal generation commitments and dispatch in the presence of high penetrations of renewable resources such as solar energy requires a characterization of their stochastic properties. In this study, we describe novel methods designed to create day-ahead, wide-area probabilistic solar power scenarios based only on historical forecasts and associated observations of solar power production. Each scenario represents a possible trajectory for solar power in next-day operations with an associated probability computed by algorithms that use historical forecast errors. Scenarios are created by segmentation of historic data, fitting non-parametric error distributions using epi-splines, and then computing specific quantiles from these distributions. Additionally, we address the challenge of establishing an upper bound on solar power output. Our specific application driver is for use in stochastic variants of core power systems operations optimization problems, e.g., unit commitment and economic dispatch. These problems require as input a range of possible future realizations of renewables production. However, the utility of such probabilistic scenarios extends to other contexts, e.g., operator and trader situational awareness. Finally, we compare the performance of our approach to a recently proposed method based on quantile regression, and demonstrate that our method performs comparably to this approach in terms of two widely used methods for assessing the quality of probabilistic scenarios: the Energy score and the Variogram score.
Nowcasting, predictive control, and feedback control for temperature regulation in a novel hybrid solar-electric reactor for continuous solar-thermal chemical processing
A novel 15 kW reactor for the hybridization of concentrated solar and conventional electric heat was fabricated for renewable and continuous chemical processing at temperatures up to 1700 °C. Solar-electric controllers based on feedback or predictive linear models were used to regulate the device at 925 °C for the production of syngas via the gasification of carbon. The system was challenged with cloud transients programmed on a 45 kW high-flux solar simulator that approximated weather observed in the San Luis Valley (Colorado, USA). In experiment it was found that model predictive control with a 1 min ahead nowcast of incipient clouding best regulated solar-electric reactor temperatures, potentially averting thermal fatigue. Upon clouding, model predictive control with a nowcast yielded temperature disturbances of ±10 °C, whereas feedback control alone featured ±25 °C excursions. Overall, the performance of model predictive control with a nowcast was 75% better than feedback as indicated by the integral squared error of temperature excursion residuals. The importance of nowcast accuracy was explored, and although the model-based controller was robust to forecast amplitude inaccuracy, temporal forecast inaccuracy wholly negated the benefits of predictive control.
A Markov framework for generalized post-event systems recovery modeling: From single to multihazards
State-dependent models can be used to represent the system recovery process as a series of stochastic transitions from lower to higher functional states. However, the applications of these models have been limited in scope and there is a lack of a generalized recovery modeling framework. A generalized framework would permit a robust forecasting of systems and system-of-systems recovery under multiple hazards, and more broadly, would contribute to community disaster preparedness. This paper develops a generalized post hazard-event recovery modeling framework based on state-dependent Markov-type processes. We then apply the proposed framework to solve a spectrum of problems that range from hind-casting single-system recovery following a single hazard event to forecasting post-event trajectories under multiple hazards and modeling the recovery of a system-of-systems. First, Markov chains are used to hind-cast the observed recovery for a portfolio of buildings affected by the 2014 South Napa, California, earthquake. Next, Markov processes are used to formulate a parametric post hazard-event recovery model, which can be updated using Bayesian statistics when relevant datasets become available. Semi-Markov processes are then used to develop a more general model of single hazard recovery, which accounts for the intensity of the loading and level of damage caused by the event. Semi-Markov processes with non-renewal features are then used to account for multihazard interactions in a post-event recovery model, and applied to a case study that involves a community in Charleston, South Carolina. Lastly, Markov-type processes are combined with Bayesian networks to model the recovery of residential, commercial, educational, and industrial buildings (system-of-systems) following a hazard event. Overall, these applications demonstrate the versatility of the Markov framework towards handling recovery problems with varying levels of complexity.