Search NASA⌕ Search

SEARCH · Search NASA

Results for “Distributed training”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

FedOSAA: Improving Federated Learning with One-Step Anderson Acceleration

Federated learning (FL) is a distributed machine learning approach that enables multiple local clients and a central server to collaboratively train a model while keeping the data on their own devices. First-order methods, particularly those incorporating variance reduction techniques, are the most widely used FL algorithms due to their simple implementation and stable performance. However, these methods tend to be slow and require a large number of communication rounds to reach the global minimizer. We propose FedOSAA, a novel approach that preserves the simplicity of first-order methods while achieving the rapid convergence typically associated with second-order methods. Our approach applies one Anderson acceleration (AA) step following classical local updates based on first-order methods with variance reduction, such as FedSVRG and SCAFFOLD, during local training. This AA step is able to leverage curvature information from the history points and gives a new update that approximates the Newton-GMRES direction, thereby significantly improving the convergence. We establish a local linear convergence rate to the global minimizer of FedOSAA for smooth and strongly convex loss functions. Numerical comparisons show that FedOSAA substantially improves the communication and computation efficiency of the original first-order methods, achieving performance comparable to second-order methods like GIANT.

Feng, Xue [University of California, Davis]↗
DOE OSTI2026

ML-Based Pebble Power Reconstruction for Pebble Bed Reactor Analysis

Pebble power reconstruction has been explored to complement the conventional homogenized modeling approach in pebble bed reactor (PBR) analysis, as detailed heterogeneous geometry calculations are computationally expensive. The random distribution of pebble fuels within the core challenges the application of conventional pin power reconstruction methods. To address this, we introduce a machine learning approach based on the transformer model, composed of encoder and decoder layers, to estimate the flux and power form functions for reconstructing individual pebble neutron fluxes and powers. The homogeneous neutron flux distribution within each spectral zone (SZ) is obtained from finite element solutions of global diffusion or transport calculations. Verification tests demonstrate that the trained transformer model accurately predicts power form functions over a range of conditions, including variations in pebble enrichment, location, type, SZ size, and burnup. In particular, verification using a three-dimensional PBR benchmark with burned pebbles shows good agreement in heterogeneous pebble power distributions between Griffin and Serpent. These results highlight the potential of applying conventional pin power reconstruction approaches to PBR cores with randomly distributed pebbles.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Code Description for "Brief Communication: Monitoring snow depth using small, cheap, and easy-to-deploy ground surface temperature sensors"

Temporally continuous snow depth estimates are vital for understanding changing snow patterns and impacts on permafrost in the Arctic. We train a random forest machine learning model to predict snow depth from variability in ground surface temperature. To our knowledge, this is the first time that small ground surface temperature sensors have been used to estimate snow depth. The model performs well at sites where the model was trained and at pan-arctic evaluation sites (RMSE <= 0.15 m). Small temperature sensors are cheap and easy-to-deploy, so this technique enables spatially distributed and temporally continuous snowpack monitoring to an extent previously infeasible. The model is flexible and can be applied to datasets retroactively to retrieve snow depth estimates at additional sites. This code package includes a *.joblib file of the trained random forest model and a *.ipynb file showing how to clean input data, train the random forest model, and apply the model.

Bachand, Claire↗
DOE OSTI2024

EV Forecasting-Based Model Predictive Control for Distribution System Congestion Mitigation

The uncoordinated charging of electric vehicles (EVs) in time and space brings congestion issues to the distribution network. This paper proposes an EV charging demand forecasting-based model predictive control (MPC) method for distribution system congestion management. To effectively forecast the time-series EV station charging demand, a hybrid forecasting model that integrates the long short-term memory network (LSTM) and Transformer is proposed. The Transformer-LSTM model is trained using a one-year real historical charging dataset of EV stations to forecast future charging demand in 15-minute intervals. This informs the MPC for distribution network congestion management and minimization of PV curtailment. Numerical results carried out on the modified IEEE 123-bus distribution system demonstrate that the proposed method can effectively resolve line congestion issues through EV smart charging and PV curtailment while outperforming other benchmarks.

ADVANCED PROPULSION SYSTEMS,SOLAR ENERGY↗
DOE OSTI2025

Enhancing synchrotron radiation micro-CT images using deep learning: an application of Noise2Inverse on bone imaging

In bone-imaging research, in situ synchrotron radiation micro-computed tomography (SRµCT) mechanical tests are used to investigate the mechanical properties of bone in relation to its microstructure. Low-dose computed tomography (CT) is used to preserve bone's mechanical properties from radiation damage, though it increases noise. To reduce this noise, the self-supervised deep learning method Noise2Inverse was used on low-dose SRµCT images where segmentation using traditional thresholding techniques was not possible. Simulated-dose datasets were created by sampling projection data at full, one-half, one-third, one-fourth and one-sixth frequencies of an in situ SRµCT mechanical test. After convolutional neural networks were trained, Noise2Inverse performance on all dose simulations was assessed visually and by analyzing bone microstructural features. Visually, high image quality was recovered for each simulated dose. Lacunae volume, lacunae aspect ratio and mineralization distributions shifted slightly in full, one-half and one-third dose network results, but were distorted in one-fourth and one-sixth dose network results. Following this, new models were trained using a larger dataset to determine differences between full dose and one-third dose simulations. Significant changes were found for all parameters of bone microstructure, indicating that a separate validation scan may be necessary to apply this technique for microstructure quantification. Noise present during data acquisition from the testing setup was determined to be the primary source of concern for Noise2Inverse viability. While these limitations exist, incorporating dose calculations and optimal imaging parameters enables self-supervised deep learning methods such as Noise2Inverse to be integrated into existing experiments to decrease radiation dose.

Obata, Yoshihiro (ORCID:0000000303659129)↗
DOE OSTI2024

CIRCLEZ : Reliable photometric redshifts for active galactic nuclei computed solely using photometry from Legacy Survey Imaging for DESI

Photometric redshifts for galaxies hosting an accreting supermassive black hole in their center, known as active galactic nuclei (AGNs), are notoriously challenging. At present, they are most optimally computed via spectral energy distribution (SED) fittings, assuming that deep photometry for many wavelengths is available. However, for AGNs detected from all-sky surveys, the photometry is limited and provided by a range of instruments and studies. This makes the task of homogenizing the data challenging, presenting a dramatic drawback for the millions of AGNs that wide surveys such as SRG/eROSITA are poised to detect. This work aims to compute reliable photometric redshifts for X-ray-detected AGNs using only one dataset that covers a large area: the tenth data release of the Imaging Legacy Survey (LS10) for DESI. LS10 provides deep grizW1-W4 forced photometry within various apertures over the footprint of the eROSITA-DE survey, which avoids issues related to the cross-calibration of surveys. We present the results from CIRCLEZ, a machine-learning algorithm based on a fully connected neural network. CIRCLEZ is built on a training sample of 14 000 X-ray-detected AGNs and utilizes multi-aperture photometry, mapping the light distribution of the sources. The accuracy (σNMAD) and the fraction of outliers (η) reached in a test sample of 2913 AGNs are equal to 0.067 and 11.6%, respectively. The results are comparable to (or even better than) what was previously obtained for the same field, but with much less effort in this instance. We further tested the stability of the results by computing the photometric redshifts for the sources detected in CSC2 and Chandra-COSMOS Legacy, reaching a comparable accuracy as in eFEDS when limiting the magnitude of the counterparts to the depth of LS10. The method can be applied to fainter samples of AGNs using deeper optical data from future surveys (for example, LSST, Euclid), granting LS10-like information on the light distribution beyond the morphological type. Along with this paper, we have released an updated version of the photometric redshifts (including errors and probability distribution functions) for eROSITA/eFEDS.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Short-Term Probabilistic Solar Forecasting via Reinforcement Learning over ECMWF

In this paper, we present an innovative reinforcement learning approach for short-term solar forecasting, leveraging data from the European Centre for Medium-Range Weather Forecasts (ECMWF). The methodology begins with the application of the System Advisor Model (SAM) to transform various ECMWF numerical weather prediction members into predictive photovoltaic power generation. To enhance the precision of deterministic forecasting, we introduce a dynamic model selection algorithm based on Q-learning. This algorithm dynamically identifies and utilizes the most accurate ensemble member for forecasting purposes. Furthermore, we employ a support vector regression surrogate model with a Gaussian distribution to generate probabilistic forecasts, providing a holistic view of solar energy generation uncertainty. To expedite the training process and make it more practical for real-world applications, we integrate a rolling update workflow. This innovative workflow reduces the training period from months to a mere 19 days, making our method highly efficient. Numerical results of the case study show that in comparison to benchmark models, the proposed method improves the deterministic and probabilistic solar forecasting accuracy by up to 40.84% and 48.42%, respectively.

ensemble forecasting↗
DOE OSTI2025

A Weakly Supervised Machine Learning Procedure for Magnet Quench Diagnostics

Voltage taps remain the standard and reliable diagnostic tool for detecting quenches in superconducting magnets. However, they identify a quench only at the time of voltage rise and do not provide information on earlier physical precursors. In this work, we investigate whether acoustic emission data can reveal precursor activity that occurs before conventional voltage detection using machine learning techniques. We introduce an event selection method and a weakly supervised machine learning procedure to learn data-driven criteria for identifying potential acoustic precursors to quenches. Two Convolutional Neural Network (CNN) architectures are trained: one on acoustic sensor events from our selection procedure and one on the Fast Fourier Transforms (FFTs) of these events. Both networks are trained iteratively using confidence-weighted loss functions to associate certain subsets of training data with a precursor label. We evaluate the performance of these models by examining the time distribution of events classified as potential precursors relative to the quench onset. Results indicate that the proposed approach can possibly distinguish acoustic emission events occurring closer to the quench from earlier acoustic activity during ramping, suggesting the potential for flagging quench precursors in acoustic data.

Khan, Maira [Fermilab] (ORCID:0009000891602387)↗
DOE OSTI2026

The LCLStream Ecosystem for Multi-Institutional Dataset Exploration

We describe a new end-to-end experimental data streaming framework designed from the ground up to support new types of applications – AI training, extremely high-rate X-ray time-of-flight analysis, crystal structure determination with distributed processing, and custom data science applications and visualizers yet to be created. Throughout, we use design choices merging cloud microservices with traditional HPC batch execution models for security and flexibility. This project makes a unique contribution to the DOE Integrated Research Infrastructure (IRI) landscape. By creating a flexible, API-driven data request service, we address a significant need for high-speed data streaming sources for the X-ray science data analysis community. With the combination of data request API, mutual authentication web security framework, job queue system, high-rate data buffer, and complementary nature to facility infrastructure, the LCLStreamer framework has prototyped and implemented several new paradigms critical for future generation experiments.

Rogers, David [ORNL] (ORCID:0000000251871768)↗
DOE OSTI2025

Stochastic Ensemble Generation for Improved Characterization of Representing Geologic Variability in a Reservoir: IBDP Case Study for SMART Initiative

This document is a poster covering the findings from activities on training data generation, specifically geologic ensemble generation. The generated geologic realizations captured the range of possible permeability distributions of the subsurface at the Illinois Basin - Decatur Project (IBDP) site, based on available well log variabilities. The percentages of reservoirs and baffles in the injection zone and a truncation of baffle permeability led to more variance in the simulations. This will be used to build forward modeling, history matching, and optimization workflows. The geologic realizations were also ranked according to dynamic measures of hydraulic diffusivity, and simulations confirm a greater contrast between the reservoir and the baffles during injection.

stochastic ensemble generation↗
DOE OSTI2026

Towards Scaling Law Analysis For Spatiotemporal Weather Data

Compute-optimal scaling laws are relatively well studied for NLP and CV, where objectives are typically single-step and targets are comparatively homogeneous. Weather forecasting is harder to characterize in the same framework: autoregressive rollouts compound errors over long horizons, outputs couple many physical channels with disparate scales and predictability, and globally pooled test metrics can disagree sharply with per-channel, late-lead behavior implied by short-horizon training. We extend neural scaling analysis for autoregressive weather forecasting from single-step training loss to long rollouts and per-channel metrics. We quantify (1) how prediction error is distributed across channels and how its growth rate evolves with forecast horizon, (2) if power law scaling holds for test error, relative to rollout length when error is pooled globally, and (3) how that fit varies jointly with horizon and channel for parameter, data, and compute-based scaling axes. We find strong cross-channel and cross-horizon heterogeneity: pooled scaling can look favorable while many channels degrade at late leads. We discuss implications for weighted objectives, horizon-aware curricula, and resource allocation across outputs.

Kiefer Jr, Alexander [ORNL] (ORCID:000000025398874↗
DOE OSTI2025

Brief communication: Monitoring snow depth using small, cheap, and easy-to-deploy snow–ground interface temperature sensors

Abstract. Temporally continuous snow depth estimates are vital for understanding changing snow patterns and impacts on permafrost in the Arctic. We trained a random forest machine learning model to predict snow depth from variability in snow–ground interface temperature. The model performed well on Alaska's Seward Peninsula where it was trained and at Arctic evaluation sites (RMSE ≤ 0.15 m). It performed poorly at temperate sites with deeper snowpacks, partially due to training data limitations. Small temperature sensors are cheap and easy to deploy, so this technique enables spatially distributed and temporally continuous snowpack monitoring at high latitudes to an extent previously infeasible.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Fault-Tolerant Deep Learning Cache with Hash Ring for Load Balancing in HPC Systems

Large-scale DL on HPC systems like Frontier and Summit uses distributed node-local caching to address scalability and performance challenges. However, as these systems grow more complex, the risk of node failures increases, and current caching approaches lack fault tolerance, jeopardizing large-scale training jobs. We analyzed six months of SLURM job logs from Frontier and found that over 30% of jobs failed after an average of 75 minutes. To address this, we propose fault-tolerance strategies that recache data lost from failed nodes using a hash ring technique for balanced data recaching in the distributed node-local caching, reducing reliance on the PFS. Our extensive evaluations on Frontier showed that the hash ring-based recaching approach reduced training time by approximately 25% compared to the approach that redirects I/O to the PFS after node failures and demonstrated effective load balancing of training data across nodes.

Lee, Seoyeong↗
DOE OSTI2025

Demonstrate new plasticity models for doped UO 2 that capture dislocation mechanisms

In light water reactors, fuel vendors are investigating the use of dopants to modify the properties of UO 2 pellets, with the goal of improving pellet-cladding mechanical interactions during operation. Dopants are expected to ‘soften’ the pellets; that is, the doped pellets have higher plastic deformation than conventional UO 2 . This leads to a reduction in the severity of mechanical pellet-cladding interactions, helping to reduce the hoop strain on the cladding. By minimizing the strain exerted by the pellet on the cladding, it is anticipated that cladding performance under accident conditions can be enhanced (i.e., lowering the risk of burst during a LOCA). Dopants such as chromium (Cr) promote grain growth during pellet fabrication, leading to larger grains; therefore, understanding the link between chemistry, microstructure and mechanical deformation (enhanced creep rates) behavior of UO 2 is critical to helping operators further substantiate the benefits of doping UO 2 . Historically, the nuclear energy industry has relied on empirical models to make assessments of performance. Compared to empirical models, mechanistic physics-based models provide benefits, such as, fewer data points for validation and better extrapolation where experimental data is scarce or non-existent. In this report, Bayesian inference techniques have been applied to a previously developed lower length-scale-informed diffusional creep model. The objective is to i) infer lower-length-scale parameter distributions from available experiment and then ii) determine the uncertainties in the measurable quantity (in this case creep rates) after propagating the inferred lower length scale parameter uncertainties. The approach requires many evaluations of the model, which becomes computationally insurmountable; therefore, a neural-network model is trained to data obtained by sampling the full model over the most important parameters. This neural-network is then used in the Bayesian inference approach to determine probability distributions in the parameter values that represent the uncertainty in the model given what is known from the experiments (posterior). A significant reduction compared to conservative initial (prior) uncertainties is achieved through inference against the experimental data, demonstrating the efficacy of this approach. Furthermore, by accounting for uncertainties in the experimental conditions and sample non-stoichiometry, it is possible to resolve apparent discrepancies in experimental measurements within a self-consistent grain boundary (Coble) creep model that is sensitive to chemistry. This work has been written up and submitted to Nuclear Technology for a special issue on accelerated fuel qualification (AFQ). This uncertainty quantification (UQ) work not only improves the diffusional model, while accounting for uncertainty, but also establishes a framework which can readily be applied to the mechanistic models of dislocation deformation developed in this study. The most likely values from the Bayesian analysis are incorporated into our UO 2 diffusional creep model and a lower length scale-informed irradiation UO 2 creep mechanistic model to generate a dataset. This dataset has been provided to our INL collaborators for training an artificial neural network surrogate model, which will be implemented in the BISON fuel performance code to assess how the results differ from those currently obtained using a fully empirical model and that of using the nominal (uncalibrated) atomic scale parameters in our mechanistic model. Plastic deformation (creep and glide) in UO 2 is a complex phenomenon, governed by multiple underlying processes such as local defect concentrations, applied stresses, and microstructural characteristics. Consequently, there is a need for a meso-scale model with polycrystalline resolution capable of extrapolating to large grain sizes applicable to doped UO 2 , where data is limited and the model can help bridge the knowledge gap. By integrating atomistic data into the polycrystal LApx code, it becomes possible to predict dislocation climb and glide plasticity that simple analytical models cannot accurately represent. The application of atomic-scale data within LApx demonstrated the importance of climb and glide mechanisms in reproducing high-stress UO 2 behavior. Behaviors such as this are crucial to capture and implement in BISON, as parts of the fuel pellet can reach temperatures where glide can occur before pellet cracking. This model which captures dislocation based mechanisms for UO 2 is then used to stand up the doped model accounting for larger grain sizes. It was found that larger grain sizes can lead to enhanced deformation rates in the glide regime, and therefore can help with the pellet cladding mechanical interaction. Therefore if the fuel pellet reaches conditions (stress/temperature) where glide is active, the enhanced creep rates for larger grains in the glide regime (doped UO 2 ) can help with pellet cladding mechanical interactions. Plastic deformation in UO 2 involves multiple mechanisms, including diffusional creep, dislocation climb, and glide. This milestone contains two parts: (1) UQ of a pre-existing lower length scale informed mechanistic diffusional creep model, and (2) development of a new LApx based model for dislocation-mediated creep mechanisms in UO 2 , with application to large-grain doped UO 2 .

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Debiasing with Diffusion: Probabilistic Reconstruction of Dark Matter Fields from Galaxies with CAMELS

Abstract Galaxies are biased tracers of the underlying cosmic web, which is dominated by dark matter (DM) components that cannot be directly observed. Galaxy formation simulations can be used to study the relationship between DM density fields and galaxy distributions. However, this relationship can be sensitive to assumptions in cosmology and astrophysical processes embedded in galaxy formation models, which remain uncertain in many aspects. In this work, we develop a diffusion generative model to reconstruct DM fields from galaxies. The diffusion model is trained on the CAMELS simulation suite that contains thousands of state-of-the-art galaxy formation simulations with varying cosmological parameters and subgrid astrophysics. We demonstrate that the diffusion model can predict the unbiased posterior distribution of the underlying DM fields from the given stellar density fields while being able to marginalize over uncertainties in cosmological and astrophysical models. Interestingly, the model generalizes to simulation volumes ≈500 times larger than those it was trained on and across different galaxy formation models. The code for reproducing these results can be found athttps://github.com/victoriaono/variational-diffusion-cdm✎.

Astronomy & Astrophysics↗
DOE OSTI2024

A novel conditional generative model for efficient ensemble forecasts of state variables in large-scale geological carbon storage

Integrating monitoring data to efficiently update reservoir pressure and CO 2 plume distribution forecasts presents a significant challenge in geological carbon storage (GCS) applications. Inverse modeling techniques are commonly used to fuse observational data and refine reservoir model parameters, thereby improving state variable forecasts. However, these techniques often rely on linear or Gaussian assumptions, which can limit their effectiveness in accurately predicting state variables. Moreover, simulating large-scale three-dimensional (3D) GCS problems is computationally expensive, making iterative runs in inverse problems prohibitive. To address these challenges, we propose a conditional generative model utilizing the score-based diffusion method for real-time 3D pressure and saturation field distribution predictions. Our approach involves solving the score function with a mini-batch-based Monte Carlo estimator to generate labeled data. This data is subsequently employed to train a fully connected neural network, enabling it to learn the conditional sample generator within a supervised learning framework. This method enables the rapid generation of a large ensemble of predictions, facilitating comprehensive uncertainty quantification of state variables. Here we applied our method to forecast the dynamic 3D distributions of pressure and saturation fields over a 30-year injection period. The statistical assessment with low root mean square error (RMSE) values demonstrates that our method can accurately predict the spatiotemporal distributions of both pressure and saturation fields. Moreover, the developed conditional generative model shows high computational efficiency by generating 100 ensemble forecasts of 3D state variables in less than 10 min. The consistency between ensemble averages and ground truth values further illustrates the model’s capability to capture state variable dynamics during the CO 2 plume injection process. Notably, the ground truth values fall within the ensemble forecasts, indicating that our uncertainty quantification effectively captures variability and potential noise in the observations. Thus, the developed conditional generative model proves to be a more efficient, accurate, and practical tool for GCS applications, facilitating timely risk analysis and informed decision-making.

58 GEOSCIENCES↗
DOE OSTI2024

Agent-Based Model of Combined Community- and Jail-Based Take-Home Naloxone Distribution

Importance Opioid-related overdose accounts for almost 80 000 deaths annually across the US. People who use drugs leaving jails are at particularly high risk for opioid-related overdose and may benefit from take-home naloxone (THN) distribution. Objective To estimate the population impact of THN distribution at jail release to reverse opioid-related overdose among people with opioid use disorders. Design, Setting, and Participants This study developed the agent-based Justice-Community Circulation Model (JCCM) to model a synthetic population of individuals with and without a history of opioid use. Epidemiological data from 2014 to 2020 for Cook County, Illinois, were used to identify parameters pertinent to the synthetic population. Twenty-seven experimental scenarios were examined to capture diverse strategies of THN distribution and use. Sensitivity analysis was performed to identify critical mediating and moderating variables associated with population impact and a proxy metric for cost-effectiveness (ie, the direct costs of THN kits distributed per death averted). Data were analyzed between February 2022 and March 2024. Intervention Modeled interventions included 3 THN distribution channels: community facilities and practitioners; jail, at release; and social network or peers of persons released from jail. Main Outcomes and Measures The primary outcome was the percentage of opioid-related overdose deaths averted with THN in the modeled population relative to a baseline scenario with no intervention. Results Take-home naloxone distribution at jail release had the highest median (IQR) percentage of averted deaths at 11.70% (6.57%-15.75%). The probability of bystander presence at an opioid overdose showed the greatest proportional contribution (27.15%) to the variance in deaths averted in persons released from jail. The estimated costs of distributed THN kits were less than $\$$15 000 per averted death in all 27 scenarios. Conclusions and Relevance This study found that THN distribution at jail release is an economical and feasible approach to substantially reducing opioid-related overdose mortality. Training and preparation of proficient and willing bystanders are central factors in reaching the full potential of this intervention.

Tatara, Eric [Argonne National Laboratory (ANL), A↗
DOE OSTI2024

Voltage Mining for (De)lithiation-Stabilized Cathodes and a Machine Learning Model for Li-Ion Cathode Voltage

Advances in lithium-metal anodes have inspired interest in discovery of Li-free cathodes, most of which are natively found in their charged state. This is in contrast to today's commercial lithium-ion battery cathodes, which are more stable in their discharged state. In this study, we combine calculated cathode voltage information from both categories of cathode materials, covering 5577 and 2423 total unique structure pairs, respectively. The resulting voltage distributions with respect to the redox pairs and anion types for both classes of compounds emphasize design principles for high-voltage cathodes, which favor later Period 4 transition metals in their higher oxidation states and more electronegative anions like fluorine or polyanion groups. Generally, cathodes that are found in their charged, delithiated state are shown to exhibit voltages lower than those that are most stable in their lithiated state, in agreement with thermodynamic expectations. Deviations from this trend are found to originate from different anion distributions between redox pairs. In addition, a machine learning model for voltage prediction based on chemical formulas is trained and shows state-of-the-art performance when compared to two established composition-based ML models for material properties predictions, Roost and CrabNet.

25 ENERGY STORAGE↗