Search NASA⌕ Search

SEARCH · Search NASA

Results for “ensemble optimization”

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 181 records · Page 10

Application-specific optimal model weighting of global climate models: A red tide example

Global climate models (GCMs) and Earth system models (ESMs) provide many climate services with environmental relevance. The High Resolution Model Inter-comparison Project (HighResMIP) of the Coupled Model Intercomparison Project Phase 6 (CMIP6) provides model runs of GCMs and ESMs to address regional phenomena. Developing a parsimonious ensemble of CMIP6 requires multiple ensemble methods such as independent-model subset selection, prescreening-based subset selection, and model weighting. The work presented here focuses on application-specific optimal model weighting, with prescreening-based subset selection. As such, independent ensemble members are categorized, selected, and weighted based on their ability to reproduce physically-interpretable features of interest that are problem-specific. We discuss the strengths and caveats of optimal model weighting using a case study of red tide prediction in the Gulf of Mexico along the West Florida Shelf. Red tide is a common name of specific harmful algal blooms that occur worldwide, causing adverse socioeconomic and environmental impacts. Our results indicate the importance of prescreening-based subset selection as optimal model weighting can underplay robust ensemble members by optimizing error cancellation. Prescreening-based subset selection also provides insights about the validity of the model weights. By illustrating the caveats of using non-representative models when optimal model weighting is used, the findings and discussion of this study are pertinent to many other climate services.

54 ENVIRONMENTAL SCIENCES↗

Evolving Ensembles of Spiking Neural Networks for Neuromorphic Systems

Evolutionary algorithms have been proposed as a solution to overcome many of the challenges associated with training spiking neural networks. While evolutionary optimization for spiking neural networks is very flexible, its performance has difficulty scaling to complex tasks and correspondingly complex network structures. Here we propose a method for evolving ensembles of spiking neural networks. By using ensemble learning, the flexibility of evolutionary optimization is fully preserved while scaling to more challenging tasks. We test the performance of the proposed method using handwritten digit classification. We investigate multiple strategies for constructing ensembles of spiking neural networks, and demonstrate that evolving ensembles of SNNs offers significant performance advantages over evolutionary optimization.

Elbrecht, Daniel↗

Resilience-based Recovery Scheduling of Transportation Network in Mixed Traffic Environment: A Deep-Ensemble-Assisted Active Learning Approach

Devising effective post-hazard recovery strategies is critical in enhancing the resilience of transportation networks (TNs). However, existing work does not consider the multiclass users’ travel behavior in network functionality quantification and the metaheuristic solution procedures often suffer from extensive computational burden due to the exploration need in large solution space and the expensive functionality quantification. Here, this study develops a bilevel decision-making framework for the resilience-based recovery scheduling of the TN in a mixed traffic environment with connected and autonomous vehicles (CAVs) and human-driven vehicles (HDVs). The lower level quantifies the TN's functionality over time considering different travel behavior of CAV and HDV users arisen from their different levels of information perception. The upper level presents a novel deep-ensemble-assisted active learning approach to balance optimization performance and computational cost. This framework can help decision makers better quantify the TN's functionality to support effective recovery scheduling of TN with different mixed traffic scenarios ranging from HDV-only to future CAV-dominant traffic. The optimization approach bears the potential to be extended to solving general large-scale network recovery scheduling problems effectively and efficiently. The proposed methodology is demonstrated using a real-world traffic network in Southern California under earthquake considering deterministic and stochastic repair durations.

42 ENGINEERING↗

Strain-Modulated Platinum–Palladium Nanowires for Oxygen Reduction Reaction

Electrocatalytic activity of alloy nanocatalytsts can be manipulated effectively by tuning their physical properties (ensemble, geometric, and ligand effects) to afford optimal surface structure and compositions for proton exchange membrane fuel cell (PEMFC) application. Herein, highly catalytic platinum-palladium nanowires (Pt n Pd 100-n NWs) with a subtle lattice strain and Boerdijk-Coxeter helix type morphology are synthesized through a surfactant-free, thermal single phase solvent method. X-ray diffraction results show that Pt n Pd 100-n NWs are exposed through the (111) facets and their shrinking or expanding lattice parameters can be modulated by the alloy compositions. Electrochemical results reveal that their high catalytic activity correlates with the lattice shrinking, facets, and bimetallic compositions, showing higher activity when the ratio of Pt and Pd is similar to ~78:22, which is further supported by DFT results. Compared to the nanoparticle type platinum-palladium alloyed catalysts with similar metal compositions (Pt n Pd 100-n NPs), the Pt n Pd 100-n NWs exhibit significantly improved electrocatalytic activity and stability for the oxygen reduction reaction. Our findings open new strategies to design the highly active and stable alloy nanocatalysts with controllable compositions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Models, data, and scripts associated with “Prediction of Distributed River Sediment Respiration Rates using Community-Generated Data and Machine Learning”

This data package is associated with the publication “Prediction of Distributed River Sediment Respiration Rates using Community-Generated Data and Machine Learning’’ submitted to the Journal of Geophysical Research: Machine Learning and Computation (Scheibe et al. 2024). River sediment respiration observations are expensive and labor intensive to obtain and there is no physical model for predicting this quantity. The Worldwide Hydrobiogeochemisty Observation Network for Dynamic River Systems (WHONDRS) observational data set (Goldman et al.; 2020) is used to train machine learning (ML) models to predict respiration rates at unsampled sites. This repository archives training data, ML models, predictions, and model evaluation results for the purposes of reproducibility of the results in the associated manuscript and community reuse of the ML models trained in this project. One of the key challenges in this work was to find an optimum configuration for machine learning models to work with this feature-rich (i.e. 100+ possible input variables) data set. Here, we used a two-tiered approach to managing the analysis of this complex data set: 1) a stacked ensemble of ML models that can automatically optimize hyperparameters to accelerate the process of model selection and tuning and 2) feature permutation importance to iteratively select the most important features (i.e. inputs) to the ML models. The major elements of this ML workflow are modular, portable, open, and cloud-based, thus making this implementation a potential template for other applications. This data package is associated with the GitHub repository found at Please see the file level metadata (flmd; “sl-archive-whondrs_flmd.csv”) for a list of all files contained in this data package and descriptions for each. Please see the data dictionary (dd; “sl-archive-whondrs_dd.csv”) for a list of all column headers contained within comma separated value (csv) files in this data package and descriptions for each. The GitHub repository is organized into five top-level directories: (1) “input_data” holds the training data for the ML models; (2) “ml_models” holds machine learning models trained on the data in “input_data”; (3) “scripts” contains data preprocessing and postprocessing scripts and intermediate results specific to this data set that bookend the ML workflow; (4) “examples” contains the visualization of the results in this repository including plotting scripts for the manuscript (e.g., model evaluation, FPI results) and scripts for running predictions with the ML models (i.e., reusing the trained ML models); (5) “output_data” holds the overall results of the ML model on that branch. Each trained ML model resides on its own branch in the repository; this means that inputs and outputs can be different branch-to-branch. Furthermore, depending on the number of features used to train the ML models, the preprocessing and postprocessing scripts, and their intermediate results, can also be different branch-to-branch. The “main-*” branches are meant to be starting points (i.e. trunks) for each model branch (i.e. sprouts). Please see the Branch Navigation section in the top-level README.md in the GitHub repository for more details. There is also one hidden directory “.github/workflows”. This hidden directory contains information for how to run the ML workflow as an end-to-end automated GitHub Action but it is not needed for reusing the ML models archived here. Please the top-level README.md in the GitHub repository for more details on the automation.

13C↗

HAIMOS Ensemble Forecasts for Intra-day and Day- Ahead GHI, DNI and Ramps

The objective of this research is to develop a hybrid physics-based/data-driven forecast model to improve direct normal and global horizontal irradiance (DNI and GHI) prediction for horizons ranging from 1 to 72 hours. Project objectives also address key gaps in state-of-the-art solar forecasting: accurate probabilistic solar forecasts and the forecasting of large irradiance ramps (ramp onset and magnitude). The proposed model ensembles Numerical Weather Prediction (NWP) forecasts, determinist physics-based algorithms, and new-generation cloud cover products (high-resolution rapid refresh satellite images and Large Eddy Simulations). The result is the Hybrid Adaptive Input Model Objective Selection (HAIMOS) ensemble model. HAIMOS blends state of the art machine learning methodologies with physics-based models for cloud cover and cloud optical depth forecasts. The technical activities followed a two-pronged strategy. First, the preprocessing of data, the selection of inputs to the nonlinear approximators, the type of approximator and objective functions, and post-processing ensembling techniques included in HAIMOS were all optimized adaptively to find the best model for a specific goal (reduce DNI/GHI forecast error, improve the prediction of ramp onset, etc.). Second, a large effort was put in improving cloud identification and the forecast of cloud cover and cloud optical depth. To this end, new-generation cloud parametrization products were developed in this work. These include improved algorithms to assist in cloud identification, cloud classification and cloud parametrization from satellite images – three key factors in the accuracy of 1 to 6-hours irradiance forecasts and prediction of ramp onset. Furthermore, we also included cloud information extracted from high resolution rapid refresh satellite images (GOES-16) and Large Eddy Simulations (LES). LES was used to model the atmosphere in detail over locations of interest and produce cloud optical depth forecasts. Once these data streams were validated, they were used as input data to the HAIMOS forecast. The model was developed using data from several climatologically distinct locations with potential for high solar penetration. In the last year of the project, we conducted a validation campaign according to the guidelines stipulated by the Topic Area 1 project as described in the FOA. This effort brings, for the first time, proven machine-learning methodologies for generating state-of-the-art solar forecasts interweaved with detailed physics-based models for cloud detection, and cloud optical depth forecasts. HAIMOS will generate accurate irradiance probabilistic forecast to assist in reducing solar generation prediction error. Globally optimized solar forecast models are more likely to impact solar energy stakeholders. The goal of this project was to increase the state-of-the-art forecast skill from their present values of 10 to 35%. At the end of the project, we achieved between 30% and 50% forecast skill across a wide range of horizons for both GHI and DNI.

14 SOLAR ENERGY↗

Forecasting Lightning Threat Using WRF Proxy Fields

Objectives: Given that high-resolution WRF forecasts can capture the character of convective outbreaks, we seek to: 1. Create WRF forecasts of LTG threat (1-24 h), based on 2 proxy fields from explicitly simulated convection: - graupel flux near -15 C (captures LTG time variability) - vertically integrated ice (captures LTG threat area). 2. Calibrate each threat to yield accurate quantitative peak flash rate densities. 3. Also evaluate threats for areal coverage, time variability. 4. Blend threats to optimize results. 5. Examine sensitivity to model mesh, microphysics. Methods: 1. Use high-resolution 2-km WRF simulations to prognose convection for a diverse series of selected case studies. 2. Evaluate graupel fluxes; vertically integrated ice (VII). 3. Calibrate WRF LTG proxies using peak total LTG flash rate densities from NALMA; relationships look linear, with regression line passing through origin. 4. Truncate low threat values to make threat areal coverage match NALMA flash extent density obs. 5. Blend proxies to achieve optimal performance 6. Study CAPS 4-km ensembles to evaluate sensitivities.

McCaul, E. W., Jr.↗

Reply to Comment by Laprise on 'the Added Value to Global Model Projections of Climate Change by Dynamical Downscaling: a Case Study over the Continental U.S. Using the GISS-ModelE2 and WRF Models'

In his comment, Laprise raises several points that we agree merit consideration. His primary critique is that our study [Racherla et al., 2012] tested the ability of the WRF regional climate model to reproduce historical temperature and precipitation change relative to the driving global climate model (GCM) using only a single simulation rather than an ensemble. He asserts that the observed changes are smaller than the internal variability in the climate system (i.e., not statistically significant) and that thus a single simulation should not necessarily be able to capture the observations. Laprise points out that the statistical signal is reduced for a multi-decadal trend such as the one we analyzed in comparison with mean climatology and cites two studies showing that for particular climate parameters it can take any years for a signal to be discerned over internal variability. He states that The results of theexperiment as designed were strongly influenced by the presence of internal variability and sampling errors,which masked the rather small climate changes that may have occurred as a consequence of changes inforcing during the period considered. While Laprise discusses statistics in general terms at some length, for the actual climate trends examined in our study, he offers no evidence that the forced signal was smallcompared with internal variability. The two studies he cites [de Ela et al., 2013; Maraun, 2013] do not provide convincing evidence as they concern climate variables averaged over different times and areas. One in fact examines extreme precipitation events, which by definition are rare and thus have a lower significance level. We accept the general point that it is important to consider internal variability, and as noted in our paper we agree that an ensemble of simulations is in principle an optimal, though computationally expensive, approach. While we did not present the statistical significance of the observations in our original paper, we have now evaluated those for the regional temperature trends used in our study to evaluate the added value of WRF and thus can analyze data as to the magnitude of the trends with respect to internal variability.

climate↗

Spin squeezing of macroscopic nuclear spin ensembles

Spin squeezing has been explored in atomic systems as a tool for quantum sensing, improving experimental sensitivity beyond the spin standard quantum limit for certain measurements. To optimize absolute metrological sensitivity, it is beneficial to consider macroscopic spin ensembles, such as nuclear spins in solids and liquids. Coupling a macroscopic spin ensemble to a parametrically-modulated resonant circuit can create collective spin squeezing by generating spin correlations mediated by the circuit. We analyze the squeezing dynamics in the presence of decoherence and finite spin polarization, showing that achieving 7 dB spin squeezing is feasible in several nuclear spin systems. The metrological benefit of squeezing a macroscopic spin ensemble lies in the suppression of technical noise sources in the spin detection system relative to the spin projection noise. This expands the experimental sensitivity bandwidth when searching for signals of unknown frequency and can improve the resonant signal-to-noise ratio. Squeezing macroscopic spin ensembles may prove to be a useful technique for fundamental physics experiments aimed at detecting spin interactions with oscillating background fields, such as ultralight dark matter. Published by the American Physical Society 2025

Boyers, Eric↗

Pairing a Global Optimization Algorithm with EXAFS to Characterize Lanthanide Structure in Solution

Ensemble-average sampling of structures from ab initio molecular dynamics (AIMD) simulations can be used to predict theoretical extended X-ray absorption fine structure (EXAFS) signals that closely match experimental spectra. However, AIMD simulations are time-consuming and resource-intensive, particularly for solvated lanthanide ions, which often form multiple nonrigid geometries with high coordination numbers. Here, to accelerate the characterization of lanthanide structures in solution, we employed the Northwest Potential Energy Surface Search Engine (NWPEsSe), an adaptive-learning global optimization algorithm, to efficiently screen first-shell structures. As case studies, we examine two systems: Eu(NO 3 ) 3 dissolved in acetonitrile with a terpyridine ligand (terpyNO 2 ), and Nd(NO 3 ) 3 dissolved in acetonitrile. The theoretical spectra for structures identified by NWPEsSe were compared to both experimental and AIMD-derived EXAFS spectra. The NWPEsSe algorithm successfully identified the proper solvation structure for both Eu(NO 3 ) 3 (terpyNO 2 ) and Nd(NO 3 )(acetonitrile) 3 , with the calculated EXAFS signals closely matching the experimental spectra for the Eu-ligand complex and showing good similarity for the Nd salt; the better agreement with the ligand-containing structure is attributed to a less dynamic coordination environment due to the rigid ligand. The key advantage of the global optimization algorithm lies in its ability to sample the coordination environment across the potential energy surface and reduce the time required to identify structures from generally a month to within a week. Additionally, this approach is versatile and can be adapted to characterize main-group metal complexes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Novel Photonic & Electronic BioMolecular Materials

The overall objective is to create robust artificial protein modules as scaffolds to control both (a) the conformation of novel cofactors incorporated into the modules thereby making the modules possess a desired functionality and (b) the organization of these functional modules into ordered macroscopic ensembles, whose macroscopic materials properties derive from the designed microscopic function of the modules. We focus on two specific types of cofactors for imparting functionality in this project; primarily nonlinear optical (NLO) chromophores designed to exhibit extraordinary molecular hyperpolarizabilities, as well as donor-bridge-acceptor cofactors designed to exhibit highly efficient, 'through-bonds' light-induced electron transfer (LIET) over nano-scale distances. The ensembles range from 2-D to 3-D, designed to possess the degree of orientational and positional order necessary to optimize their macroscopic response, the latter ranging from liquid-crystalline or glass-like to long-range periodic. Computational techniques, firmly based in statistical thermodynamics, are utilized for the design the artificial protein modules, based on robust {alpha}-helical bundle motifs, necessarily incorporating the desired conformation, location, and environment of the cofactor. Importantly, this design approach also includes optimization of the interactions between the modules to promote their organization into ordered macroscopic ensembles in 2-D and 3-D via either directed-assembly or self-assembly. When long-range periodic order is required, the design can be optimized to result a specified lattice symmetry. The structure and functionality of the individual modules are fully characterized at the microscopic level, as well as that of the ensembles at the macroscopic level, employing modern experimental physical-chemical and computational techniques. These include, for example, multi-dimensional NMR, various pump-probe transient spectroscopies to ultrafast time-scales, and hyper-Rayleigh scattering at the microscopic level, and synchrotron radiation-based x-ray scattering and x-ray spectroscopy, cold neutron scattering, molecular dynamics simulation, and optical harmonic generation at the macroscopic level. This overall approach has some distinct advantages, compared to more traditional approaches, for example, those based on organic polymers, biopolymers or undressed cofactors. The resulting functional ensembles thereby have potential for important device applications in the areas of optical communications and photovoltaics. The approach also has an absolute requirement for a tightly coupled collaborative effort necessary to span the rigorous demands for the design, synthesis and characterization of such novel photonic and electronic biomolecular materials.

36 MATERIALS SCIENCE↗

Machine Learning with Gradient-Based Optimization of Nuclear Waste Vitrification with Uncertainties and Constraints

Gekko is an optimization suite in Python that solves optimization problems involving mixed-integer, nonlinear, and differential equations. The purpose of this study is to integrate common Machine Learning (ML) algorithms such as Gaussian Process Regression (GPR), support vector regression (SVR), and artificial neural network (ANN) models into Gekko to solve data based optimization problems. Uncertainty quantification (UQ) is used alongside ML for better decision making. These methods include ensemble methods, model-specific methods, conformal predictions, and the delta method. An optimization problem involving nuclear waste vitrification is presented to demonstrate the benefit of ML in this field. ML models are compared against the current partial quadratic mixture (PQM) model in an optimization problem in Gekko. GPR with conformal uncertainty was chosen as the best substitute model as it had a lower mean squared error of 0.0025 compared to 0.018 and more confidently predicted a higher waste loading of 37.5 wt% compared to 34 wt%. The example problem shows that these tools can be used in similar industry settings where easier use and better performance is needed over classical approaches. Future works with these tools include expanding them with other regression models and UQ methods, and exploration into other optimization problems or dynamic control.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Optimization of magnetic switches for single particle and cell transport

The ability to manipulate an ensemble of single particles and cells is a key aim of lab-on-a-chip research; however, the control mechanisms must be optimized for minimal power consumption to enable future large-scale implementation. Recently, we demonstrated a matter transport platform, which uses overlaid patterns of magnetic films and metallic current lines to control magnetic particles and magnetic-nanoparticle-labeled cells; however, we have made no prior attempts to optimize the device geometry and power consumption. Here, we provide an optimization analysis of particle-switching devices based on stochastic variation in the particle's size and magnetic content. These results are immediately applicable to the design of robust, multiplexed platforms capable of transporting, sorting, and storing single cells in large arrays with low power and high efficiency.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Multi-Parameter Scattering Sensor and Methods

Methods, detectors and systems detect particles and/or measure particle properties. According to one embodiment, a detector for detecting particles comprises: a sensor for receiving radiation scattered by an ensemble of particles; and a processor for determining a physical parameter for the detector, or an optimal detection angle or a bound for an optimal detection angle, for measuring at least one moment or integrated moment of the ensemble of particles, the physical parameter, or detection angle, or detection angle bound being determined based on one or more of properties (a) and/or (b) and/or (c) and/or (d) or ranges for one or more of properties (a) and/or (b) and/or (c) and/or (d), wherein (a)-(d) are the following: (a) is a wavelength of light incident on the particles, (b) is a count median diameter or other characteristic size parameter of the particle size distribution, (c) is a standard deviation or other characteristic width parameter of the particle size distribution, and (d) is a refractive index of particles.

Greenberg, Paul S.↗

An Ensemble Neural Network Model for Predicting Rare-Earth Oxide and Silicate Heat Capacities at High Temperature

In this work, a neural network model was developed to predict the constant pressure heat capacity for materials in the rare-earth oxide—silica material space. Several model architectures were trained and tested on heat capacity data generated from first-principles density functional theory calculations. Hyperparameter optimization was performed, and the optimal model was selected for heat capacity predictions. The optimal model architecture was found to have a root-mean-squared error of 5.12 ± 3.37 J/mol-K. The optimal model architecture was then used in a bagging ensemble model trained using the leave-one-group-out method to provide error estimates for model predictions. The out-of-bag score for the ensemble model was 0.997. The predicted heat capacities agree well with the DFT and experimental results and were computed orders of magnitude faster than DFT simulations. Machine learning shows the potential to provide a suitable surrogate model for thermochemical property predictions for candidate environmental barrier coating materials but refining of input material features and model architectures could further improve accuracy for these models.

environmental barrier coatings↗

A two-step short-term probabilistic wind forecasting methodology based on predictive distribution optimization

With increasing wind penetrations into electric power systems, probabilistic wind forecasting becomes more critical to power system operations because of its capability of quantifying wind uncertainties. In this paper, a two-step probabilistic wind forecasting approach based on pinball loss optimization is developed. First, a multimodel machine learning-based ensemble deterministic forecasting framework is adopted to generate deterministic forecasts. The deterministic forecast is assumed to be the mean value of the predictive distribution at each forecasting time stamp. Then, the optimal unknown parameter (i.e., standard deviation) of the predictive distribution is estimated by a support vector regression surrogate model based on the deterministic forecasts. Lastly, probabilistic forecasts are generated from the predictive distribution. Numerical results of case studies at eight locations show that the developed two-step probabilistic forecasting methodology has improved the pinball loss metric score by up to 35% compared to a baseline quantile regression forecasting model.

17 WIND ENERGY↗

Statistical mechanics and invariant perception

The problem of discrimination among ensembles of images generated by distortions of a prototype and the addition of noise is considered. As the noise level is increased, the discrimination task becomes qualitatively more difficult in that optimal discrimination requires the computation of increasingly longer-ranged correlations, or the solution of increasingly difficult optimization problems. These results suggest the use of such image ensembles in probing the computational abilities of the human visual system, and possible theoretical implications of such experiments are discussed.

Bialek, William↗

GrainNN: A neighbor-aware long short-term memory network for predicting microstructure evolution during polycrystalline grain formation

High fidelity simulations of grain formation in alloys are an indispensable tool for process-to-mechanical-properties characterization. Such simulations, however, can be computationally expensive as they require fine spatial and temporal discretizations. Their cost becomes an obstacle to parametric studies and ensemble runs and ultimately makes downstream tasks like optimal control and uncertainty quantification challenging. To enable such downstream tasks, we introduce GrainNN, an efficient and accurate reduced-order model for epitaxial grain growth in additive manufacturing conditions. GrainNN is a sequence-to-sequence long-short-term-memory (LSTM) deep neural network that evolves the dynamics of manually crafted features. Its innovations are (1) an attention mechanism with grain-microstructure-specific transformer architecture; and (2) an overlapping combination of several clones of the network to generalize to grain configurations that are different from those used for training. This design enables GrainNN to predict grain formation for unseen physical parameters, grain number, domain size and geometry. Furthermore, GrainNN not only reconstructs the quantities of interest but also can be pointwise accurate. In our numerical experiments, we use a polycrystalline phase field method to both generate the training data and assess GrainNN. For multiparametric, ensemble simulations with many grains, GrainNN can be orders of magnitude faster than phase field simulations, while delivering 5%–15% pointwise error. Additionally, this speedup includes the cost of the phase field simulations for generating training data.

36 MATERIALS SCIENCE↗