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 1,261 records · Page 70

Temporal Forecasting of Distributed Temperature Sensing in a Thermal Hydraulic System With Machine Learning and Statistical Models

We benchmark performance of long-short term memory (LSTM) network machine learning model and autoregressive integrated moving average (ARIMA) statistical model in temporal forecasting of distributed temperature sensing (DTS). Data in this study consists of fluid temperature transient measured with two co-located Rayleigh scattering fiber optic sensors (FOS) in a forced convection mixing zone of a thermal tee. We treat each gauge of a FOS as an independent temperature sensor. We first study prediction of DTS time series using Vanilla LSTM and ARIMA models trained on prior history of the same FOS that is used for testing. The results yield maximum absolute percentage error (MaxAPE) and root mean squared percentage error (RMSPE) of 1.58% and 0.06% for ARIMA, and 3.14% and 0.44% for LSTM, respectively. Next, we investigate zero-shot forecasting (ZSF) with LSTM and ARIMA trained on history of the co-located FOS only, which is advantageous when limited training data is available. The ZSF MaxAPE and RMSPE values for ARIMA are comparable to those of the Vanilla use case, while the error values for LSTM increase. We show that in ZSF, performance of LSTM network can be improved by training on most correlated gauges between the two FOS, which are identified by calculating the Pearson correlation coefficient. The improved ZSF MaxAPE and RMSPE for LSTM are 4.4% and 0.33%, respectively. Performance of ZSF LSTM can be further enhanced through transfer learning (TL), where LSTM is re-trained on a subset of the FOS that is the target of forecasting. We show that LSTM pre-trained on correlated dataset and re-trained on 30% of testing target dataset achieves MaxAPE and RMSPE values of 2.32% and 0.28%, respectively.

ARIMA↗

Conditional distribution estimation of building characteristics with diffusion models for urban energy modeling

Understanding current energy consumption behavior in communities is critical for informing future energy use decisions and enabling efficient energy management. Urban energy models, which are used to simulate these energy use patterns, require large datasets with detailed building characteristics for accurate outcomes. However, such detailed characteristics at the individual building level are often unknown and costly to acquire, or unavailable. Through this work, we propose using a generative modeling approach to generate realistic building attributes to fill in the data gaps and finally provide complete characteristics as inputs to energy models. Our model learns complex, building-level patterns from training on a large-scale residential building stock model containing 2.2 million buildings. We employ a tabular diffusion-based framework that is designed to handle heterogeneous (discrete and continuous) features in tabular building data, such as occupancy, floor area, heating, cooling, and other equipment details. We develop a capability for conditional diffusion, enabling the imputation of missing building characteristics conditioned on known attributes. We conduct a comprehensive validation of our conditional diffusion model, firstly by comparing the generated conditional distributions against the underlying data distribution, and secondly, by performing a case study for a Baltimore residential region, showing the practical utility of our approach. Our work is one of the first to demonstrate the potential of generative modeling to accelerate building energy modeling workflows.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Genarris 2.0: A Random Structure Generator for Molecular Crystals

Genarris is an open source Python package for generating random molecular crystal structures with physical constraints for seeding crystal structure prediction algorithms and training machine learning models. Here we present a new version of the code, containing several major improvements. A MPI-based parallelization scheme has been implemented, which facilitates the seamless sequential execution of user-defined workflows. A new method for estimating the unit cell volume based on the single molecule structure has been developed using a machine-learned model trained on experimental structures. A new algorithm has been implemented for generating crystal structures with molecules occupying special Wyckoff positions. A new hierarchical structure check procedure has been developed to detect unphysical close contacts efficiently and accurately. New intermolecular distance settings have been implemented for strong hydrogen bonds. To demonstrate these new features, we study two specific cases: benzene and glycine. Genarris finds the experimental structures of the two polymorphs of benzene and the three polymorphs of glycine. Program summary Program Title: Genarris 2.0 Program Files doi: http://dx.doi.org/10.17632/grx6mz4pjn.1 Licensing provisions: BSD-3 Clause Programming language: Python, C External routines/libraries: Spglib, ASE, pymatgen, SciPy, mpi4py, scikit-learn, PyTorch, FHI-aims. Nature of problem: Molecular crystal structure prediction. Solution method: Genarris 2.0 generates molecular crystal structures over the 230 space groups, on general and special Wyckoff positions, using physical constraints. Down-sampling of the generated structures may be performed subsequently, based on molecular crystal packing descriptors and an unsupervised machine learning algorithm. Lastly, ab initio structure relaxation may be performed for the final pool. Depending on the user-defined workflow implemented, Genarris may be used to generate diverse molecular crystal datasets to seed evolutionary algorithms or to train machine learning algorithms or as a standalone crystal structure prediction method. Restrictions: For crystal structure generation, the molecule of interest must be semi-rigid with no bond rotational degrees of freedom. Unusual features: Genarris 2.0 is a highly distributed program, making use of MPI for Python parallelization. The user has the ability to design and implement workflows by executing a user-defined list of procedures. Genarris 2.0 offers new features including a machine learning model for estimating the molecular volume in the solid state from the single molecule structure, structure generation in special Wyckoff positions of space groups, hierarchical structure checks including rigorous treatment of non-orthogonal structures, and clustering and down-selection workflows combining first principles simulations with machine learning. (C) 2020 Elsevier B.V. All rights reserved.

Crystal structure prediction↗

Three-dimensional reconstruction of inertial confinement fusion hot-spot plasma from x-ray and nuclear diagnostics on OMEGA

Multidimensional effects degrade the neutron yield and the compressed areal density of laser-direct-drive inertial confinement fusion implosions of layered deuterium–tritium cryogenic targets on the OMEGA Laser System with respect to 1D radiation-hydrodynamic simulation predictions. A comprehensive physics-informed 3D reconstruction effort is under way to infer hot-spot and shell conditions at stagnation from four x-ray and seven neutron detectors distributed around the OMEGA target chamber. Neutron diagnostics, providing measurements of the neutron yield, hot-spot flow velocity, and apparent ion-temperature distribution, are used to infer the mode-1 perturbation at stagnation. The x-ray imagers record the shape of the hot-spot plasma to diagnose mode-1 and mode-2 perturbations. A deep-learning convolutional neural network trained on an extensive set of 3D radiation-hydrodynamic simulations is used to interpret the x-ray and nuclear measurements to infer the 3D profiles of the hot-spot plasma conditions and the amount of laser energy coupled to the hot-spot plasma. A 3D simulation database shows that larger mode-1 asymmetries are correlated with higher hot-spot flow velocities and reduced laser-energy coupling and neutron yield. Three-dimensional hot-spot reconstructions from x-ray measurements indicate that higher amounts of residual kinetic energy are correlated with higher measured hot-spot flow velocities, consistent with 3D simulations.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Robust anomaly detection for particle physics using multi-background representation learning

Abstract Anomaly, or out-of-distribution, detection is a promising tool for aiding discoveries of new particles or processes in particle physics. In this work, we identify and address two overlooked opportunities to improve anomaly detection (AD) for high-energy physics. First, rather than train a generative model on the single most dominant background process, we build detection algorithms using representation learning from multiple background types, thus taking advantage of more information to improve estimation of what is relevant for detection. Second, we generalize decorrelation to the multi-background setting, thus directly enforcing a more complete definition of robustness for AD. We demonstrate the benefit of the proposed robust multi-background AD algorithms on a high-dimensional dataset of particle decays at the Large Hadron Collider.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Flow annealed importance sampling bootstrap meets differentiable particle physics

High-energy physics requires the generation of large numbers of simulated data samples from complex but analytically tractable distributions called matrix elements. Surrogate models, such as normalizing flows, are gaining popularity for this task due to their computational efficiency. We adopt an approach based on flow annealed importance sampling bootstrap (FAB) that evaluates the differentiable target density during training and helps avoid the costly generation of training data in advance. We show that FAB reaches higher sampling efficiency with fewer target evaluations in high dimensions in comparison to other methods.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

The stellar parameters and elemental abundances from low-resolution spectra – I. 1.2 million giants from LAMOST DR8

As a typical data-driven method, deep learning becomes a natural choice for analysing astronomical data. In this study, we built a deep convolutional neural network (NN) to estimate basic stellar parameters $T\rm {_{eff}}$, log g , metallicity ([M/H] and [Fe/H]) and [α/M] along with nine individual elemental abundances ([C/Fe], [N/Fe], [O/Fe], [Mg/Fe], [Al/Fe], [Si/Fe], [Ca/Fe], [Mn/Fe], and [Ni/Fe]). The NN is trained using common stars between the APOGEE survey and the LAMOST survey. We used low-resolution spectra from LAMOST survey as input, and measurements from APOGEE as labels. For stellar spectra with the signal-to-noise ratio in g band larger than 10 in the test set, the mean absolute error (MAE) is 29 K for $T\rm {_{eff}}$, 0.07 dex for log g , 0.03 dex for both [Fe/H] and [M/H], and 0.02 dex for [α/M]. The MAE of most elements is between 0.02 and 0.04 dex. The trained NN was applied to 1210 145 giants, including sub-giants, from LAMOST DR8 within the range of stellar parameters 3500 K < $T\rm {_{eff}}$ < 5500 K, 0.0 dex < log g < 4.0 dex, −2.5 dex < [Fe/H] < 0.5 dex. The distribution of our results in the chemical spaces is highly consistent with APOGEE labels and stellar parameters show consistency with external high-resolution measurements from GALAH. The results in this study allow us to further studies based on LAMOST data and deepen our understanding of the accretion and evolution history of the Milky Way. The electronic version of the value added catalog is available at http://www.lamost.org/dr8/v1.1/doc/vac.

79 ASTRONOMY AND ASTROPHYSICS↗

Continuous snow depth, ground interface temperature and shallow soil temperature measurements from 2021-10-1 to 2022-6-14, Seward Peninsula, Alaska

The dataset contains co-located snow depth, ground interface temperature, and shallow soil temperature measured at 98 discrete locations in a watershed located along the Nome-Teller road at mile marker 27 (referred to as T27), and at 53 discrete locations on a hillslope located along the Kougarok road at mile marker 64 (referred to as K64), in Seward Peninsula, Alaska. The dataset aims to understand the local heterogeneity of snow depth, snow temperature, and soil temperature dynamics and their interactions in a discontinuous permafrost region. The dataset is also valuable to train and evaluate machine learning and physical models to predict snow depth or the impact of snow depth on ground surface temperature. At each location, temperatures above and below the ground surface were measured by a pair of vertically deployed distributed temperature profiling probes designed based on Dafflon et al (2022). The probes have high precision temperature sensors spaced at 5 or 10 cm. The mean daily snow depth was estimated by identifying the pair of consecutive sensors with maximum drop of daily temperature high-frequency fluctuations. The ground interface temperature was measured by the sensor located 1-5 cm above the ground surface at 15-minute intervals. The shallow soil temperature was measured by the sensor located 1-5 cm below the ground surface at 15-minute intervals. The dataset includes a description of the probe locations in the "Probe_locations_*.csv" file and the 3 data files (Snow_depths_*.csv, Ground_interface_temperatures_*.csv, Shallow_soil_temperatures_*.csv). * is either T27 or K64, which are the two study sites. In each data file, each column corresponds to a measurement location. Metadata files include data descriptions (_dd.csv) for tabular data. All included files are listed and described in xxxx_flmd.csv. A more detailed description of data processing, along with an updated dataset incorporating multiple seasons and improved snow depth estimation is available at https://doi.org/10.15485/2480365 (Wang et al., 2025a, Wang et al., 2025b).The Next-Generation Ecosystem Experiments: Arctic (NGEE Arctic), was a research effort to reduce uncertainty in Earth System Models by developing a predictive understanding of carbon-rich Arctic ecosystems and feedbacks to climate. NGEE Arctic was supported by the Department of Energy's Office of Biological and Environmental Research.The NGEE Arctic project had two field research sites: 1) located within the Arctic polygonal tundra coastal region on the Barrow Environmental Observatory (BEO) and the North Slope near Utqiagvik (Barrow), Alaska and 2) multiple areas on the discontinuous permafrost region of the Seward Peninsula north of Nome, Alaska.Through observations, experiments, and synthesis with existing datasets, NGEE Arctic provided an enhanced knowledge base for multi-scale modeling and contributed to improved process representation at global pan-Arctic scales within the Department of Energy's Earth system Model (the Energy Exascale Earth System Model, or E3SM), and specifically within the E3SM Land Model component (ELM).

54 ENVIRONMENTAL SCIENCES↗

Toward Uncertainty Quantification for Supervised Classification

Our goal is to develop a general theoretical basis for quantifying uncertainty in supervised machine learning models. Current machine learning accuracy-based validation metrics indicate how well a classifier performs on a given data set as a whole. However, these metrics do not tell us a model's efficacy in predicting particular samples. We quantify uncertainty by constructing probability distributions of the predictions made by an ensemble of classifiers. This report details our initial investigations into uncertainty quantification for supervised machine learning. We apply an uncertainty analysis to the problem of malicious website detection. Machine learning models can be trained to find suspicious characteristics in the text of a website's Uniform Resource Locator (URL). However, given the vast numbers of URLs and the ever changing tactics of malicious actors, it will always be possible to find sets of websites which are outliers with respect to a model's hypothesis. Therefore, we seek to understand a model's per-sample reliability when classifying URL data. Acknowledgements This work was funded by the Sandia National Laboratories Laboratory Directed Research and Development (LDRD) program.

97 MATHEMATICS AND COMPUTING↗

Neural MUSE Analysis

Researchers at Oak Ridge National Laboratory (ORNL) created data as part of the MUSE (Multi-Agency Urban Search Experiment Detector and Algorithm Test Bed) project simulating illicit nuclear materials located in various buildings along a road. In the simulation, a truck containing a radiation detector drives down the road gathering listmode data (counting the and energy of incident gamma radiation). Building materials, source shielding, driving speed, truck direction, truck location on the road, source type, and source placement are all varied between runs of the data set. This data was created using deterministic neutron transport and Monte Carlo methods through a combination of SCALE, MAVRIC, MCNP, and GADRAS. As part of a follow-on NA-22 project, two Kaggle competitions were created to determine the best algorithms for finding and identifying gamma sources in this simulated urban environment. The winning algorithm was neural network-based and had a test accuracy of 76.4% accuracy for source identification. This work seeks to build upon this work and improve the results through the application of novel machine learning techniques. As a first step, the data was classified by a simple Convolutional Neural Network (CNN) To accomplish this, the data was first preprocessed into “waterfall plots.” These plots are composed of energy vs count plots that are stacked vertically to show progression in time. The horizontal axis indicating the particle energy incorporated user defined bin spacing with options for in linear-, logarithmic-, square root-, and user-spaced bins. The z or color dimension showed the number of counts corresponding the energy-time combination. This data was then used to generate more data, by generating a local estimate of the mean of the distribution for a bin and then randomly re-sampling that bin from a Poisson distribution. Once all of this data was generated, it was fed into a well-known CNN architecture, ResNet50. The output layer of this model was removed and replaced with layers corresponding to the shape desired isotope outputs. The provided training data was used to train the classifier and the remaining testing data was used to evaluate the model. Results are soon to be forthcoming.

61 RADIATION PROTECTION AND DOSIMETRY↗

Dynamic Topology Reconfiguration of Boltzmann Machines on Quantum Annealers

Boltzmann machines have useful roles in deep learning applications, such as generative data modeling, initializing weights for other types of networks, or extracting efficient representations from high-dimensional data. Most Boltzmann machines use restricted topologies that exclude looping connectivity, as such connectivity creates complex distributions that are difficult to sample. We have used an open-system quantum annealer to sample from complex distributions and implement Boltzmann machines with looping connectivity. Further, we have created policies mapping Boltzmann machine variables to the quantum bits of an annealer. These policies, based on correlation and entropy metrics, dynamically reconfigure the topology of Boltzmann machines during training and improve performance.

Liu, Jeremy (ORCID:0000000172299417)↗

PV inverter experimental data

The increase in power electronic based generation sources require accurate modeling of inverters. Accurate modeling requires experimental data over wider operation range. We used 20 kW off-the-shelf grid following PV inverter in the experiments. We used controllable AC supply and controllable DC supply to emulate AC and DC side characteristics. The experiments were performed at NREL's Energy Systems Integration Facility. Due to the limitations of the DC supply used, inverter is tested under 75%, 50%, 25% load conditions (This dataset does not contain 100% load condition). In the first dataset, for each operating condition, controllable AC source voltage is varied from 0.88 to 1.09 per unit (p.u) with a step value of 0.025 p.u while keeping the frequency at 60 Hz. In the second dataset, under similar load conditions (75%, 50%, 25% ), the frequency of the controllable AC source voltage was varied from 59.4 Hz to 60.45 Hz with a step value of 0.2 Hz. Voltage and frequency range is chosen based on inverter protection. Voltages and currents on DC and AC side are included in the dataset.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Statistical methods and neural network approaches for classification of data from multiple sources

Statistical methods for classification of data from multiple data sources are investigated and compared to neural network models. A problem with using conventional multivariate statistical approaches for classification of data of multiple types is in general that a multivariate distribution cannot be assumed for the classes in the data sources. Another common problem with statistical classification methods is that the data sources are not equally reliable. This means that the data sources need to be weighted according to their reliability but most statistical classification methods do not have a mechanism for this. This research focuses on statistical methods which can overcome these problems: a method of statistical multisource analysis and consensus theory. Reliability measures for weighting the data sources in these methods are suggested and investigated. Secondly, this research focuses on neural network models. The neural networks are distribution free since no prior knowledge of the statistical distribution of the data is needed. This is an obvious advantage over most statistical classification methods. The neural networks also automatically take care of the problem involving how much weight each data source should have. On the other hand, their training process is iterative and can take a very long time. Methods to speed up the training procedure are introduced and investigated. Experimental results of classification using both neural network models and statistical methods are given, and the approaches are compared based on these results.

Benediktsson, Jon Atli↗

A Probabilistic Mass Estimation Algorithm for a Novel 7- Channel Capacitive Sample Verification Sensor

A document describes an algorithm created to estimate the mass placed on a sample verification sensor (SVS) designed for lunar or planetary robotic sample return missions. A novel SVS measures the capacitance between a rigid bottom plate and an elastic top membrane in seven locations. As additional sample material (soil and/or small rocks) is placed on the top membrane, the deformation of the membrane increases the capacitance. The mass estimation algorithm addresses both the calibration of each SVS channel, and also addresses how to combine the capacitances read from each of the seven channels into a single mass estimate. The probabilistic approach combines the channels according to the variance observed during the training phase, and provides not only the mass estimate, but also a value for the certainty of the estimate. SVS capacitance data is collected for known masses under a wide variety of possible loading scenarios, though in all cases, the distribution of sample within the canister is expected to be approximately uniform. A capacitance-vs-mass curve is fitted to this data, and is subsequently used to determine the mass estimate for the single channel s capacitance reading during the measurement phase. This results in seven different mass estimates, one for each SVS channel. Moreover, the variance of the calibration data is used to place a Gaussian probability distribution function (pdf) around this mass estimate. To blend these seven estimates, the seven pdfs are combined into a single Gaussian distribution function, providing the final mean and variance of the estimate. This blending technique essentially takes the final estimate as an average of the estimates of the seven channels, weighted by the inverse of the channel s variance.

Wolf, Michael↗

Type Ia Supernova Hubble Residuals and Host-Galaxy Properties

Kim et al. (2013) [K13] introduced a new methodology for determining peak- brightness absolute magnitudes of type Ia supernovae from multi-band light curves. We examine the relation between their parameterization of light curves and Hubble residuals, based on photometry synthesized from the Nearby Supernova Factory spec- trophotometric time series, with global host-galaxy properties. The K13 Hubble residual step with host mass is 0.013 ? 0.031 mag for a supernova subsample with data coverage corresponding to the K13 training; at ? 1?, the step is not significant and lower than previous measurements. Relaxing the data coverage requirement the Hubble residual step with host mass is 0.045 ? 0.026 mag for the larger sample; a calculation using the modes of the distributions, less sensitive to outliers, yields a step of 0.019 mag. The analysis of this article uses K13 inferred luminosities, as distinguished from previous works that use magnitude corrections as a function of SALT2 color and stretch param- eters: Steps at> 2? significance are found in SALT2 Hubble residuals in samples split by the values of their K13 x(1) and x(2) light-curve parameters. x(1) affects the light- curve width and color around peak (similar to the∆m15 and stretch parameters), and x(2) affects colors, the near-UV light-curve width, and the light-curve decline 20 to 30 days after peak brightness. The novel light-curve analysis, increased parameter set, and magnitude corrections of K13 may be capturing features of SN Ia diversity arising from progenitor stellar evolution.

79 ASTRONOMY AND ASTROPHYSICS↗

ASGS - An Innovative Approach for Waste Characterization - From the First Study to Experimental Validation - 20260

For disposal of nuclear waste a clear identification of the nuclear inventory is mandatory. Since this waste is usually filled in drums, nuclear specific characterization methods using HPGe detectors usually require the understanding of the limitations of the method assumptions (such as the homogeneity criteria), and often results in applying high uncertainties to the result in waste that is not well known a priori. There are several methods and approaches such as Far-Field, Segmented Gamma Scans (SGS) or Tomographic Gamma Scans (TGS) available. The SGS allows measuring the waste drum in different segments (or 'layers') which allows getting more detailed information about the activity distribution inside a waste drum. In addition, with the transmission source used in tomographic scans, it is possible to also determine the material (and its homogeneity) composition inside a drum. Mirion Technologies developed in cooperation with Aachen Institute for Nuclear Training (AiNT) an innovative measurement approach which allows more accurate results and therefore applying lower uncertainties. This new measurement approach extends the concept of SGS by partitioning the drum into single drum sectors instead of segments. The goal is to perform measurements with high throughput and an accurate analysis based on the reconstruction of activities for each sector. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Simulation of the Aerosol Size Distribution Using a Neural Network Surrogate for the Modal Aerosol Module (MAM7)

One objective of atmospheric simulations is to quantify the distribution of aerosols and their properties. Accurate parameterizations of the processes governing aerosol mass, particle number, and particle size distribution are important for predicting the Earth’s net radiative balance and aerosol-cloud interactions. The Modal Aerosol Module (MAM7) is a two-moment aerosol model that simulates mass, number, and size distribution of seven modes comprised of internally mixed aerosol species. The two-moment scheme adds significant computational expense but allows for the prediction of varying particle size distribution relative to the bulk method which predicts only total mass. In this work, we developed a neural network surrogate model for MAM7 (MAMnet) to predict the aerosol number concentration in NASA’s Global Earth Observing System (GEOS) without adding prohibitive computational expense. MAMnet, can be driven by output from a single moment, mass-based, aerosol scheme (Goddard Chemistry Aerosol and Radiation model (GOCART)) or from reanalysis products (Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2)). MAMnet was trained using number concentrations from a 5-year GEOS/MAM7 simulation at 1-degree horizontal resolution and using the total mass calculated across modes as inputs, as well as temperature and air density. The model architecture for MAMnet was based on AlexNet, the 2012 winner of the ImageNet Large Scale Visual Recognition Challenge. While some modifications were necessary to accommodate our problem, important aspects of the network were preserved. MAMnet was able to reproduce zonal dynamics and spatial distributions of the aerosol number concentration however predictability in the upper troposphere was poor.

Katherine H Breen↗

Hydropower Black Start: A Guidebook for Retrofitting Grid Dependent Hydropower

Not all United States (US) hydropower plants were designed to provide black start, but they are increasingly needed to uphold resilience in the evolving electric grid. This guidance is designed to help understand the minimal retrofits required for grid dependent hydropower (GDH) plants behind the point of interconnection (POI). For distribution connected hydropower plants or those with dedicated cranking paths, such upgrades can be sufficient for the plant to provide black start. For others, more coordination with the transmission system operator will be needed. This guidebook answers a number of questions relevant to retrofitting hydropower plants with black start capabilities. For example, the guidebook answers: • How flexible do the wicket gate controls need to be? • Who needs to do hydro governor model validation, why, and how? • How robust and flexible do the excitation and AVR controls need to be? • What protection settings need to be adjusted? • What relay(s) will need to be bypassed or overridden and at what risk? • What is the electrical energy demand of the station load or auxiliary power systems? • What should the strategy to energize transformer(s) along cranking path to address inrush currents be? • How should the critical load restoration be sequenced? In addition to outlining the specifications that hydropower plants need to meet for each component to be able to perform black start, this guidebook provides a set of case studies for specific upgrades needed at actual plants. Between the case studies of plants that have already performed black start retrofits and the examples of how this guidebook can be applied to scope future retrofits, five key themes have been identified for retrofit needs. 1. Protection needs “black start” mode: hydropower plants that are not designed with black start capabilities will have protections that prevent them from interconnecting to a “dead bus.” These protections will need to be overridden in every retrofit case and a separate black start mode should be established so that operators can safely switch between black start and grid connected modes, minimizing the risk to the plant. 2. Wicket gates need modern controls: digital governors accelerate the parameter tuning process and gate position sensors improve controllability, so plants with mechanical governors should be upgraded. Furthermore, a black start and islanding mode should be established for controls to maximize plant performance. 3. Robust excitation support: the DC system or excitation generator needs to be reliable enough to form and sustain the rotor electromagnetic field. These systems are typically undersized in plants that were not designed for black start, so they will need to be upgraded. 4. Turbine-governor model validation and operator training: validation of a standard hydro governor model is needed to characterize the dynamic response (i.e., inertial and primary frequency response) of the GDH. This is required for control development and old hydropower plants often have outdated or incorrect models. Operator training is also typically required to ensure the hardware retrofits are utilized correctly during the black start process. 5. Transformer and cranking path energization: any upgradation and control adjustment in front of the POI will depend upon the existing interconnection. Coordination with the transmission or distribution operator may be required.

13 HYDRO ENERGY↗