Search NASA⌕ Search

SEARCH · Search NASA

Results for “power prediction”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Effect of Heave Plate Hydrodynamic Force Parameterization on a Two-Body Wave Energy Converter

Heave plates are one approach to generating the reaction force necessary to harvest energy from ocean waves. In a Morison equation description of the hydrodynamic force, the components of drag and added mass depend primarily on the heave plate oscillation. These terms may be parameterized in three ways: (1) as a single coefficient invariant across sea state, most accurate at the reference sea state, (2) coefficients dependent on the oscillation amplitude, but invariant in phase, that are most accurate for relatively small amplitude motions, and (3) coefficients dependent on both oscillation amplitude and phase, which are accurate for all oscillation amplitudes. We validate a MATLAB model for a two-body point absorber wave energy converter against field data and a dynamical model constructed in ProteusDS. We then use the MATLAB model to evaluate the effect of these parameterizations on estimates of heave plate motion, tension between the float and heave plate, and wave energy converter electrical power output. We find that power predictions using amplitude-dependent coefficients differ by up to 30% from models using invariant coefficients for regular waves ranging in height from 0.5 to 1.9 m. Amplitude- and phase-dependent coefficients, however, yield less than a 5% change when compared with coefficients dependent on amplitude only. This suggests that amplitude-dependent coefficients can be important for accurate wave energy converter modeling, but the added complexity of phase-dependent coefficients yields little further benefit. We show similar, though less pronounced, trends in maximum tether tension, but note that heave plate motion has only a weak dependence on coefficient fidelity. Finally, we emphasize the importance of using experimentally derived added mass over that calculated from boundary element methods, which can lead to substantial under-prediction of power output and peak tether tension.

dynamical model↗
DOE OSTI2021

Design and analysis of a wake model for spatially heterogeneous flow

Abstract. Methods of turbine wake modeling are being developed to more accurately account for spatially variant atmospheric conditions within wind farms. Most current wake modeling utilities are designed to apply a uniform flow field to the entire domain of a wind farm. When this method is used, the accuracy of power prediction and wind farm controls can be compromised depending on the flow-field characteristics of a particular area. In an effort to improve strategies of wind farm wake modeling and power prediction, FLOw Redirection and Induction in Steady State (FLORIS) was developed to implement sophisticated methods of atmospheric characterization and power output calculation. In this paper, we describe an adapted FLORIS model that features spatial heterogeneity in flow-field characterization. This model approximates an observed flow field by interpolating from a set of atmospheric measurements that represent local weather conditions. The objective of this method is to capture heterogeneous atmospheric effects caused by site-specific terrain features, without explicitly modeling the geometry of the wind farm terrain. The implemented adaptations were validated by comparing the simulated power predictions generated from FLORIS to the actual recorded wind farm output from the supervisory control and data acquisition (SCADA) recordings and large eddy simulations (LESs). When comparing the performance of the proposed heterogeneous model to homogeneous FLORIS simulations, the results show a 14.6 % decrease for mean absolute error (MAE) in wind farm power output predictions for cases using wind farm SCADA data and a 18.9 % decrease in LES case studies. The results of these studies also indicate that the efficacy of the proposed modeling techniques may vary with differing site-specific operational conditions. This work quantifies the accuracy of wind plant power predictions under heterogeneous flow conditions and establishes best practices for atmospheric surveying for wake modeling.

17 WIND ENERGY↗
DOE OSTI2022

Deep Active Learning for Solvability Prediction in Power Systems

Traditional methods for solvability region analysis can only have inner approximations with inconclusive conservatism and handle limited types of power flow models. In this letter, we propose a deep active learning framework for solvability prediction in power systems. Compared with passive learning where the training is performed after all instances are labeled, active learning selects most informative instances to be labeled and therefore significantly reduces the size of the labeled dataset for training. In the active learning framework, the acquisition functions, which correspond to different sampling strategies, are defined in terms of the on-the-fly posterior probability from the classifier. First, the IEEE 39-bus system is employed to validate the proposed framework, where a two-dimensional case is illustrated to visualize the effectiveness of the sampling method followed by the high-dimensional numerical experiments. Then, the Northeast Power Coordinating Council (NPCC) 140-bus system is used to validate the performance on large-scale power systems.

42 ENGINEERING↗
DOE OSTI2020

Predicting the proximity to macroscopic failure using local strain populations from dynamic in situ X-ray tomography triaxial compression experiments on rocks

Predicting the proximity of large-scale dynamic failure is a critical concern in the engineering and geophysical sciences. Here we use evolving contractive, dilatational, and shear strain deformation preceding failure in dynamic X-ray tomography experiments to examine which strain components best predict the proximity to failure. We develop machine learning models to predict the proximity to failure using time series of three-dimensional local incremental strain tensor fields acquired in rock deformation experiments under stress conditions of the upper crust. Three-dimensional scans acquired in situ throughout triaxial compression experiments provide a distribution of density contrasts from which we estimate the three-dimensional incremental strain that accumulates between each scan acquisition. Training machine learning models on multiple experiments of six rock types provides suites of feature importance that indicate the predictive power of each feature. Comparing the average importance of groups of features that include information about each strain component quantifies the ability of the contractive, dilatational and shear strain to predict the proximity of macroscopic failure. A total of 24 models of four machine learning algorithms with six rock types indicate that 1) the dilatational strain provides the best predictive power of the strain components, and 2) the intermediate values (25th-75th percentile) of the strain population provide the best predictive power of the statistics of the strain populations. In addition, the success of the predictions of models trained on one rock type and tested on other rock types quantifies the similarities and differences of the precursory strain accumulation process in the six rock types. These similarities suggest the potential existence of a unified theory of brittle rock deformation for a range of rock types.

58 GEOSCIENCES↗
DOE OSTI2024

Human limits in machine learning: prediction of potato yield and disease using soil microbiome data

Abstract Background The preservation of soil health is a critical challenge in the 21st century due to its significant impact on agriculture, human health, and biodiversity. We provide one of the first comprehensive investigations into the predictive potential of machine learning models for understanding the connections between soil and biological phenotypes. We investigate an integrative framework performing accurate machine learning-based prediction of plant performance from biological, chemical, and physical properties of the soil via two models: random forest and Bayesian neural network. Results Prediction improves when we add environmental features, such as soil properties and microbial density, along with microbiome data. Different preprocessing strategies show that human decisions significantly impact predictive performance. We show that the naive total sum scaling normalization that is commonly used in microbiome research is one of the optimal strategies to maximize predictive power. Also, we find that accurately defined labels are more important than normalization, taxonomic level, or model characteristics. ML performance is limited when humans can’t classify samples accurately. Lastly, we provide domain scientists via a full model selection decision tree to identify the human choices that optimize model prediction power. Conclusions Our study highlights the importance of incorporating diverse environmental features and careful data preprocessing in enhancing the predictive power of machine learning models for soil and biological phenotype connections. This approach can significantly contribute to advancing agricultural practices and soil health management.

Aghdam, Rosa↗
DOE OSTI2024

Capturing Infrastructure Interdependencies for Power Outages Prediction During Extreme Events

As extreme weather events such as hurricanes, severe thunderstorms, and floods grow in frequency and intensity, the disruption of power grid systems poses significant challenges, including widespread electrical outages, economic losses, and threats to public safety. This paper presents a forward-looking approach that leverages geographical graph-based machine learning models to predict county-level maximum power outages during such events. By capturing the intricate interdependencies within power system networks, our approach aims to provide precise and actionable predictions that can optimize emergency response efforts and enhance grid resilience. Through the integration of real-world data, including hurricane advisories and power outage records, we have trained and benchmarked multiple machine learning models, demonstrating the feasibility and potential of this method. While our initial results are promising, this paper also charts a course for advancing these models, addressing the remaining challenges, and ultimately transforming how we anticipate and respond to the impacts of extreme weather on power systems.

Lee, Sangkeun (Matt) [ORNL] (ORCID:000000021317511↗
DOE OSTI2020

Progress in DIII-D towards validating divertor power exhaust predictions

In this study, UEDGE simulations highlight the role of cross-field drifts on the onset of detached conditions, and new calibrated divertor vacuum ultra violet (VUV) spectroscopy is used to challenge the predictions of radiative constituents in these simulations. UEDGE simulations for DIII-D H-mode plasmas with the open divertor with the ion ∇B-drift towards the X-point show a bifurcated onset of the low field side (LFS) divertor detachment, consistent with experimentally observed step-like detachment onset (Jaervinen A.E. et al 2018 Phys. Rev. Lett. 121 075001). The divertor plasma in the simulations exhibits hysteresis in upstream separatrix density between attached and detached solution branches. Reducing the drift magnitude by a factor of 3 eliminates the step-like detachment onset in the simulations, confirming the strong role of drifts in the bifurcated detachment onset. When measured local plasma densities and temperatures are within proximity of predicted values in the simulations, there is no shortfall of the local emission of the dominant resonant radiating lines. However, the simulations systematically predict a factor of two lower total integrated radiated power than measured by the bolometer with the difference lost through radial heat flow out of the computational domain. Even though there is no shortfall in the emission of the dominant lines, a shortfall of total radiated power can be caused by underpredicted spatial extent of the radiation front, indicating a potential upstream or divertor transport physics origin for the radiation shortfall, or shortfall of radiated power in the spectrum between the dominant lines. In addition to the underpredicted spatial extent, in detached conditions, the simulations overpredict the peak radiation and dominant carbon lines near the X-point, which can be alleviated by manually increasing divertor diffusivity in the simulations, highlighting the ad hoc cross-field transport as one of the key limitations of the predictive capability of these divertor fluid codes.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Current Observer Based Predictive Decoupled Power Control Grid-Interactive Inverter

This paper presents a sensor-less current model predictive control (MPC) scheme via a full state observer based current estimator. The grid interactive inverters’ control schemes require current and voltage sensors. Elimination of current sensor enhances the inverter reliability. This paper leverages the inherent characteristics of MPC towards robust current sensor-less grid interactive inverter with LC filter. The observer for inductor current is developed based on the existing capacitor voltage measurement. The estimator dynamic state-space matrices are then obtained through reconstruction of the inverter model with the capacitor voltage and inductor current being the state variables. The controller objectives are to regulate active and reactive power in a decoupled manner. The theoretical expectation, controller performance, and accuracy of the current estimation are verified by conducting a real-time simulation via Typhoon HIL.

Zhang, Zhen↗
DOE OSTI2026

NGEE Arctic Authorship Guidelines

Authorship Guidelines were developed to help facilitate trust among team members as we span multiple institutions, scientific disciplines, and career stages. NGEE Arctic was built on a foundation of open science, data sharing, and collaboration. In Phase 4 of the project, it was particularly important to keep this foundation in mind as we develop new collaborations across the Arctic. Included in this package is one *.pdf. The Next-Generation Ecosystem Experiments in the Arctic (NGEE Arctic) project is a research effort to reduce uncertainty in the Department of Energy’s Energy Exascale Earth System Model (E3SM) by developing a predictive understanding of Arctic tundra ecosystems underlain by permafrost and to quantify feedbacks from the Arctic tundra to the Earth system. NGEE Arctic is supported by the Department of Energy's Office of Biological and Environmental Research. Over Phases 1–3, observations made by the NGEE Arctic team across a gradient of permafrost landscapes in Arctic Alaska improved the representation of tundra processes in the land surface component of E3SM (the E3SM Land Model, ELM). Model improvements emphasized unique aspects of permafrost environments and explored reductions in model complexity while retaining predictive power. The Arctic-informed ELM developed by NGEE Arctic has been used to make novel predictions on processes ranging from permafrost thaw to soil biogeochemical cycling to Earth system feedbacks associated with the unique characteristics of tundra plants. In Phase 4, the NGEE Arctic team is evaluating our new predictive understanding under novel conditions across the Arctic domain. In collaboration with partners at long-term pan-Arctic research sites we are examining whether an Arctic-informed ELM can faithfully simulate interactions among surface and subsurface processes at site, regional, and pan-Arctic scales. In turn, we are using variety of tools to dynamically extend and evaluate ELM inference, with an emphasis on data synthesis and pan-Arctic model evaluation, reintegration of code with an evolving E3SM, scaling across heterogeneous Arctic landscapes, and the appropriate representation of the impacts of increasingly frequent Arctic disturbances.

Iversen, Colleen [ORNL] (ORCID:0000000182933450)↗
DOE OSTI2023

Extrapolation Domains for Aggregating Environmental Outcomes from Local to Regional Levels

Billions of dollars are invested every year to run field experiments to quantify the response of crops to new technologies and associated environmental outcomes. Given the diversity of environments where crop production occurs, conducting these research studies without a robust framework for research site selection and upscaling results to larger spatial scales is inefficient. In this project, researchers from the University of Nebraska-Lincoln used their unique Technology Extrapolation Domains framework (TED) as a tool to guide the selection of experimental sites and as basis for aggregating and validating environmental outcomes from local to regional levels. In the present project, we first used the TED framework to evaluate the current distribution and area coverage of the current SMARTFARM sites. The purpose was to showcase how it is possible to use the TED framework to guide site selection and extrapolate results over space. Additionally, the original TED framework was expanded to account for other factors influencing environmental outcomes by inclusion of three additional variables: soil organic matter, soil texture, and topographic wetness index. The resulting expanded framework (TED-E) was validated using nitrogen (N) losses from corn in the United States as a case study. To do so, we used N balance as a proxy to N losses and we evaluated the capacity of the framework to explain variation in N balance across fields and across countries. We found that the TED-E has substantially higher predictive power than the original TED framework to explain spatial variation in N balance. However, improvements in predictive power with the TED-E tool come at the expense of a higher number of TEDs needed to achieve a given crop area coverage compared with the original TED framework. We conclude that the new TED-E framework can help aggregate and extrapolate environmental outcomes from research sites to regional levels and improve the visualization of their spatial patterns across the United States. An online version of the framework is available at: https://www.toolted.org/

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Hybrid Natural Gas Geothermal Combined Cycle Power Plant Analysis

Low temperature geothermal resources, including those associated with oil and gas production, are an underutilized source of low carbon energy. The present work investigates coupling of low-temperature geothermal resources with concentrating solar and/or natural gas energy sources to increase the number of locations at which power generation from low temperature geothermal resources would be technically and economically viable. Stand-alone and hybrid geothermal power cycles are simulated using SimTech IPSEpro process modeling software. Design point strategies for a hybrid power cycle that may operate with either a single heat source or two simultaneous heat input sources are considered. Additionally, off-design power plant operation is investigated to consider the impacts of the heat source availability and ambient temperature variations. The off-design analysis utilizes a modeling tool that predicts power plant performance at each time step as a function of topping cycle heat input (from gas or solar), bottoming cycle heat input (from geothermal), and ambient temperature. Addition of a steam topping cycle to an organic Rankine cycle geothermal power plant provides opportunities to increase the efficiency and power output relative to a stand-alone geothermal power plant. Additionally, use of the waste heat from gas turbine power generation in a geothermal bottoming cycle provides opportunities to increase the amount of power generation associated with each unit of carbon dioxide emitted. This paper will describe the hybrid plant configuration evaluated, discuss the predicted power cycle performance, and compare with stand-alone natural gas and geothermal power generation cases. The power cycle investigated is expected to be applicable for use with conventional hydrothermal resources as well as with geologic thermal energy storage applications and/or enhanced geothermal systems. The steam topping cycle could use a concentrated solar heat source for fully renewable hybrid plant configuration. A plant initially constructed with a natural gas topping cycle heat source could be converted to a solar heat source part way though the power cycle operational life to achieve life cycle carbon emission reductions.

concentrating solar power↗
DOE OSTI2020

Predicting Solar Plant Generation with Machine Learning Techniques

Accurately predicting power generation for PV sites is critical for prioritizing relevant operations & maintenance activities, thereby extending the lifetime of a system and improving profit margins. A number of factors influence power generation at PV sites, including local weather, shading and soiling losses, design of modules, DC mismatches, and degradation over time. Other external factors such as curtailment and grid outages can also have a notable impact on power generation. Machine learning techniques can be used to provide more accurate predictions of PV power production by accounting for important weather and climate information neglected by current industry methods. This article will cover the deficiencies of those methods and will show how machine learning can dramatically improve power generation predictions.

14 SOLAR ENERGY↗
DOE OSTI2025

Uncertainty quantification for nuclear forensics with population analyses

Although neural networks offer cutting-edge predictive power, their deployment in high-consequence nuclear forensic applications is limited, partly because of their black-box nature. Incorporating robust uncertainty quantification methods into the predictive frameworks of neural networks is progress towards their future deployment in such scenarios. This work integrates uncertainty quantification into neural networks for nuclear reactor core-average burnup estimation from simulated environmental samples. We test two regimes (homogeneous and heterogeneous events) on DeepSets and Set Transformer architectures, we find both quantify predictive uncertainty effectively, but Set Transformer excels in partitioning latent events, offering superior predictive power and more informative uncertainty estimates.

Hatton, Conner [ORNL] (ORCID:0009000804970959)↗
DOE OSTI2023

Model certainty in cellular network-driven processes with missing data

Mathematical models are often used to explore network-driven cellular processes from a systems perspective. However, a dearth of quantitative data suitable for model calibration leads to models with parameter unidentifiability and questionable predictive power. Here we introduce a combined Bayesian and Machine Learning Measurement Model approach to explore how quantitative and non-quantitative data constrain models of apoptosis execution within a missing data context. We find model prediction accuracy and certainty strongly depend on rigorous data-driven formulations of the measurement, and the size and make-up of the datasets. For instance, two orders of magnitude more ordinal (e.g., immunoblot) data are necessary to achieve accuracy comparable to quantitative (e.g., fluorescence) data for calibration of an apoptosis execution model. Notably, ordinal and nominal (e.g., cell fate observations) non-quantitative data synergize to reduce model uncertainty and improve accuracy. Finally, we demonstrate the potential of a data-driven Measurement Model approach to identify model features that could lead to informative experimental measurements and improve model predictive power.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Dynamics modeling of molten salt reactor with reduced and expanded representations of delayed neutron precursors

Molten salt reactors (MSRs) present unique challenges in dynamic behavior due to the mobility of their fuel. In these reactors, delayed neutron precursors (DNPs) drift with the fuel circulation through the primary loop. As a result, a fraction of DNPs decays outside the core, effectively reducing the available delayed neutron population for reactivity control. Consequently, precise modeling of the distribution and behavior of DNPs is critical for accurate reactor dynamics simulations. In this study, the System Dynamics Analysis Tool (SDAT) was used to simulate a thermal-spectrum MSR under steady-state conditions and following transients. The effects of using reduced and expanded representations of DNPs with fewer or more groups than the conventional 6-group model were investigated. Their impact on the simulated distribution of precursors in the primary loop, reactivity loss value, and reactor response to transients was analyzed. Simulation results showed that reduced models lead to the loss of the actual DNPs distribution data, resulting in less accurate estimates of reactivity loss. Reactor power predictions using these reduced models showed significant deviations compared to those using the conventional 6-group model in transient simulations. Expanded models offered a more accurate representation of the distribution of DNPs and reactivity loss estimates. Reactor power predictions using expanded models showed minimal deviation from the conventional 6-group model during the simulated transients.

analysis↗
DOE OSTI2022

Structure-Informed Graph Learning of Networked Dependencies for Online Prediction of Power System Transient Dynamics

Online transient analysis plays an increasingly important role in dynamic power grids as the renewable generation continues growing. Traditional numerical methods for transient analysis not only are computationally intensive but also require precise contingency information as input, and therefore, are not suitable for online applications. Existing online transient assessment studies focus on the determination of post-contingency system stability or stability margin. Here, this paper develops a novel graph-learning framework, Deep-learning Neural Representation or DNR, for online prediction, of the time-series trajectories of the system states using initial system responses that can be measured by phasor measurement units (PMUs). The proposed DNR framework consists of two sequential modules: a Network Constructor that captures network dependencies among generators, and a Dynamics Predictor that predicts the system trajectories. The key to improved prediction performance is the introduction of the spatio-temporal message-passing operations into graph neural networks with structural knowledge. Its effectiveness and scalability are validated through comparative studies, demonstrating the prediction performance under different contingency scenarios for systems of different sizes. This framework provides a solution to online predicting post-fault system dynamics based on real-time PMU measurements. Additionally, it can also be applied to facilitate the offline transient simulation without simulating the entire trajectories.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Prediction of Power Measurements Using Adaptive Filters

With the advent of smart grid concept, Internet of Things (IoT) and the deployment of smart meters, the cyberattack threats on power networks have increased due to the use of communication systems that can be accessed by adversaries. Attackers will have the ability to manipulate the outcomes of smart meters which in turn influence the core application of Energy Management System 9EMS): State Estimation (SE). Bad data analytic tools may fail to detect some attacks into measurements. Meanwhile, Machine Learning (ML) solutions have been proposed for detecting False Data Injection (FDI) attacks. However, there is a lack of ML time-series solutions presented in the state-of-the-art that is yet to be complex. In signal processing, time-series solutions do not only consider the signal, but also the statistics of the signal over time. Therefore, in this paper, a machine learning for time-series solutions is presented as an application to model the measurements of the power grid that are used in SE. The presented model takes into account adaptive linear and non-linear filters: Finite Impulse Response (FIR), and Infinite Impulse Response (IIR). The presented models are implemented and performed on the IEEE-118 bus system. The results indicate the advantage of applying those filters over the state-of-the-art machine learning solutions.

Hamad, Khaled↗