Search NASA⌕ Search

SEARCH · Search NASA

Results for “Large load”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Blue Rock CropSpanPV - Low Cost Racking for Agricultural Solar Photovoltaics

In this work, a total of four solar panel layouts, applied to Agrivoltaics, are investigated. This study presents two hypotheses: 1. Prefabrication - Lower total cost can be achieved by constructing a pre-assembled, pre-wired solar array in a factory, with automation, then rapidly deploying it in the field. 2. Tension Structures - Using tension structures in a way to suspend an overhead solar array also reduces costs when compared to conventional support methods. Designs are presented for frames to hold the panels. Configurations are advanced to connect the frames into arrays. Supports and foundations to hold the arrays in the desired layout are calculated, selected and presented. Cost models are prepared for the four layouts, which include definition of the business enterprise, the manufacturing operation, and the required facility for production at scale. Cost models include, detail material and task/labor take-offs for both manufacturing and installation. Comparisons with conventional system adapted to Agrivoltaics reveal that the two hypotheses do not hold true. A prefabricated array requires additional material that is not needed in the base system and the cost of this additional material is not outweighed by the cost savings of reduced field installation. The cost of wire rope for a tension structure does result in an economy of material, however, the cost of end connections and tensioning devices adds significantly to the cost of the overall tension structure. Foundation loads with tension structures, especially the large wind uplift seen with a fixed tilt solar array, impose significant constraints and high costs. The following conclusions are recommended for further consideration: 1. Best Case Agrivoltaic Configuration - The best value/lowest cost configuration for agrivoltaics is the installation of solar panels mounted on a single axis tracker in either a 2-in-portrait or 2-in landscape arrangement. This is useful for grazing lands and for staple crops which require high light levels. These applications represent the vast majority of the potential agrivoltaic market. Important elements in this configuration are the specific geometric layout coordinated with the various field operations and a control system linked between the farm equipment and single axis tracker. The control system is to monitor location and orientation, then actively tilt the solar panels or brake the field equipment to provide clearance and avoid collisions. 2. Long Span Applications - The use of tension structures for an overhead fixed tilt array is beneficial where long spans are a necessity, especially where the base support in the project location has consolidated rock near the surface or some other solid structure. One application is spanning wide irrigation canals. Another may be as an installation at the top deck of parking structures. 3. Direct Ground Mount Configuration – Having skids at the base of a pre-fabricated array that extend when the array is deployed, could be a successful design. This could be a useful strategy for dense, flat, ground-mount configurations like that employed by a couple of successful commercial operations (5G Maverick and Erthos). Arrays would be orientated north and south, deployed very low on prepared grade. Arrays would be aligned closely next to each other and anchored with small ground screws or simply with ballast. This same product arrangement could be very useful for rapid deployment and set-up of solar arrays for temporary deployments, disaster response and the military operations

14 SOLAR ENERGY↗
DOE OSTI2024

Report on technical development in ANL for innovative creep testing and measurement

This report summarizes a research effort undertaken by Argonne National Laboratory to accelerate the development of creep data package for ASME code qualification. Progress on the fabrication of a 3-sample load train and the application of a direct current potential drop (DCPD) technique to measure large creep strains are reported. Specifically: (1) The 3-sample load train, with three sets of Linear Variable Differential Transformers (LVDT), has been fabricated and assembled. (2) A DCPD method for creep-strain measurement up to 50% has been demonstrated. The DCPD data matches very well to LVDT results, suggesting its potential application in future creep testing. The technical developments contribute towards shortening the time needed to generate a creep data package for an ASME code case of the laser powder bed fusion 316H stainless steel, an effort under the US Department of Energy’s Advanced Materials and Manufacturing Technologies (AMMT) program.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

IDAES Enterprise: Generation Expansion Planning with Enhanced Requirements for Capacity Adequacy Under Renewable Intermittency

Achieving net zero carbon emissions likely requires future power systems to integrate new, flexible energy technologies to accommodate higher levels of capacity from variable renewable energy sources. To determine the optimal deployment of new electricity capacity and to study the likelihood of deployments of new energy technologies, an expansion planning model has been developed as part of the IDAES-Enterprise suite of grid models. The Generation Expansion Planning (GEP) model is a multi-period model in which investment decisions occur yearly, and a Unit Commitment (UC) problem is examined on an hourly timescale. To reduce computational complexity of the GEP model, the UC problem is solved for average “representative days” which leaves out extreme, but relatively common, scenarios in which low renewable generation occurs, leaving the system with inadequacy in capacity. The IDAES-Enterprise GEP model has been modified to include these extreme scenarios while keeping the model reasonably tractable. Specifically, a lazy constraint technique was implemented to check for capacity adequacy on an hourly basis over a large data set of aligned load-wind-solar profiles. As a vast majority of the capacity constraints will not be violated, the technique lowers computational expense by searching for violated capacity constraints over an “iterative manner,” adding those infeasible constraints back into the model. Results on a test case of the Southwest Power Pool shows that the lazy constraint technique significantly reduces retirements and increases installments of natural gas combined cycles and flexible natural gas units. It also reduces some retirements of coal units. These modifications provide a more reasonable estimation of required dispatchable power generation capacity to ensure feasibility during peak net load.

Liu, Peng↗
DOE OSTI2026

Electromagnetic Transient Modeling of Large Data Centers for Grid-Level Studies

The magnitude and complexity of electricity usage patterns from large data centers are having significant impacts on the operation and dynamics of the power grid; grid operators and planners require a range of specialized data center models to properly evaluate these impacts and specify technical solutions as needed. Towards addressing this need, Pacific Northwest National Laboratory (PNNL) has developed a library of electromagnetic transient (EMT) models for grid-level studies of data centers called the data center model library (DML). This report describes how the DML was created and how it may properly be used. The models present in the DML are generic models; subject matter expertise and additional technical data are needed to modify these models before they can represent any real data center. However, they will significantly reduce the level of effort required to develop site-specific models and can serve as a common starting point to guide industry towards a more refined consensus. Most of the models within DML are dedicated to representing the power electronics interfaces commonly used in modern data centers, such as double-conversion uninterruptible power supplies and single-phase power factor correction converters. These models are intended for use in grid-level studies and are a simplified aggregation of many small components. That said, background material on the physical and electrical design of large data centers is provided as companion material so that users can be aware of many of the details which have been omitted or streamlined as a matter of practical necessity. Additionally, guidance on the application of EMT analysis for data center interconnection studies is provided, which aids users in identifying when the DML is necessary and what sort of additional model development may be necessary for conducting real-world studies.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

Electromagnetic Transient Modeling of Large Data Centers for Grid-Level Studies: Beta Release

The magnitude and complexity of electricity usage patterns from large data centers are having significant impacts on the operation and dynamics of the power grid; grid operators and planners require a range of specialized data center models to properly evaluate these impacts and specify technical solutions as needed. Towards addressing this need, Pacific Northwest National Laboratory (PNNL) has developed a library of electromagnetic transient (EMT) models for grid-level studies of data centers called the data center model library (DML). This report describes how the DML was created and how it may properly be used. This report details the DML’s beta release, completed in July 2026. This is a revision and expansion of the alpha release, which was made available in January 2026 The models present in the DML are generic models; subject matter expertise and additional technical data are needed to modify these models before they can represent any real data center. However, they will significantly reduce the level of effort required to develop site-specific models and can serve as a common starting point to guide industry towards a more refined consensus. Most of the models within DML are dedicated to representing the power electronics interfaces commonly used in modern data centers, such as double-conversion uninterruptible power supplies and single-phase power factor correction converters. These models are intended for use in grid-level studies and are a simplified aggregation of many small components. That said, background material on the physical and electrical design of large data centers is provided as companion material so that users can be aware of many of the details which have been omitted or streamlined as a matter of practical necessity. Additionally, guidance on the application of EMT analysis for data center interconnection studies is provided, which aids users in identifying when the DML is necessary and what sort of additional model development may be necessary for conducting real-world studies.

electromagnetic transients↗
DOE OSTI2022

Comparative critical mass calculations for NNL and ENDF/B-VIII.0 - Zirconium hydride thermal neutron scattering laws

Zirconium hydride (ZrH{sub x}) is a moderator material for TRIGA reactors and historical space reactor systems, such as SNAP-10A. Thermal neutron scattering laws (TSL) for two phases of this material, δ and ε, have been previously evaluated by Naval Nuclear Laboratory (NNL) and submitted to the National Nuclear Data Center (NNDC) for inclusion in the US national ENDF/B-VIII.1 nuclear data library. In contrast to the current ENDF/B-VIII.0 TSL evaluations, which consider only a single phase, the new evaluations are derived from separate ab initio calculations for both phases and include coherent elastic effects of the zirconium sublattice. To estimate the impact of these changes to the TSL evaluation of this material, comparative critical mass calculations were performed with MC21 for homogenous mixtures of high-enriched uranium (HEU) and ZrH{sub x} in bare and water reflected sphere configurations. These calculations yield an impact on the estimated critical mass as a function of {sup 235}U loading density with maximum differences as large as 1% - 5% for over-moderated thermal spectrum systems. Consequently, the NNL TSL evaluations are anticipated to have a small impact on criticality calculations of thermal reactor systems regardless of the material phase. Nevertheless, characteristic differences exist in the predicted thermal spectra as function of energy for the two sets of TSL evaluations, which are attributed to difference in the underlying phonon density of states of hydrogen bound in ZrH{sub x}. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

Energy Arbitrage: Comparison of Options for use with LWR Nuclear Power Plants

Arbitrage is the opportunistic buying and selling of a commodity during local pricing valleys and peaks respectively to maximize economic value. This report evaluates options for energy arbitrage integrated with existing light water reactor (LWR) nuclear power plants (NPPs) where nuclear energy could be stored in a variety of forms and later recovered to generate electrical power during periods when grid electricity demand and pricing are high. The forms of energy storage examined in this report include the potential value of batteries, hydrogen, and thermal energy storage for coupling with nuclear power. Various large demand response options are also analyzed, including the production of liquid nitrogen via air separation and liquefaction, liquefaction of hydrogen, compressed hydrogen, and the cryogenic capture of CO2. Demand response refers to dispatchable loads that can cycle up or down depending on-grid electricity demand to aid in balancing the grid. Large demand response options could dispatch to aid nuclear power stations in avoiding power turndowns by providing an alternate disposition for electrical energy by producing marketable products (e.g., liquid nitrogen, hydrogen, or captured CO2). Static conditions were chosen and analyzed in this report for each option. Dynamic operation or optimization of energy arbitrage or demand response are out of scope for this report. The analysis is based on storage systems with discharge capacities of 500 MW for which various durations of storage and costs of charging (electricity cost) are examined. While the value of thermal energy to an industrial user for flexible plant operations has been previously proven as a business case, this report evaluates costs of hydrogen energy storage and leading thermal energy storage options, and large demand response loads that could be integrated with LWRs in comparison to utility-scale battery storage for use of off-peak nuclear energy. Compilation of this information will be used by the Idaho National Laboratory (INL) RAVEN/HERON systems integration and economics tool to evaluate thermal energy dispatch to industrial users. Relative ranking of energy storage options was done using a levelized cost of storage (LCOS) metric which calculates a rough breakeven cost for the system, taking into account the capital and operating costs as well as the revenue from arbitrage. Table ES1 below shows the LCOS for each of the energy storage options considered. First, in the table, lithium iron (Fe) phosphate batteries are listed as the base case for comparison against the other options. Next is hydrogen storage where most of the hydrogen analyses assumed the hydrogen to be produced using solid oxide electrolytic cell (SOEC) high temperature steam electrolysis (HTSE). The others used existing models of polymer electrolyte membrane (PEM) low temperature electrolysis to produce hydrogen. HTSE performance parameters and costs were taken from existing INL models. Various means were assumed to convert the hydrogen to electricity, including PEM fuel cells (FCs) and a gas turbine mixed in a 30 vol% mixture with natural gas. Physical storage (pressure vessels) and geological storage (natural underground features) were used to store the hydrogen as noted. Geological storage is more economical, but the locations are limited because of the requirement for pre-existing geological formations that will support storage. Thermal energy storage (TES) options were also analyzed including electro-thermal energy storage (ETES) and four different liquid sensible heat TES storage media as noted (Hitec, Hitec XL, Therminol-66, and Dowtherm A). The ETES process considered was modified using existing public documentation on an Echogen process and uses a separate supercritical CO2 charge and discharge cycle with sand as the heat storage media.

25 ENERGY STORAGE↗
DOE OSTI2022

Wind farm response to mesoscale-driven coastal low level jets: a multiscale large eddy simulation study

Realistic atmospheric turbulence–wind farm interactions during coastal low-level jet (LLJ) events are captured using high-fidelity, mesoscale-driven large eddy simulations (LES) to understand wind turbine loads, wakes and overall performance. The simulation has been carried out using the ExaWind aeroelastic solver, AMR-Wind. The simulations have been compared against a baseline unstable case matching the wind speed, wind direction and TI at hub-height location. Results indicate that the LLJ has negative impacts on the turbine hub and tower loads, and opens up potential avenues for design load mitigation strategies.

17 WIND ENERGY↗
DOE OSTI2022

Potential bill impacts of dynamic electricity pricing on California utility customers

The rapid growth of renewable generation is creating challenges for the California grid in the form of the “duck curve,” with increasingly steep ramping required for conventional generation resources in the morning and evening, and growing curtailment of solar resources in midday periods. Time-varying electricity tariffs have received considerable attention as a tool to address these challenges, with a renewed recent focus on the potential for dynamic tariffs that vary to reflect conditions on the grid in near-real time. Consideration of dynamic tariffs may raise concerns about the financial impact on utility customers, especially for those who have limited flexibility to modify their electricity consumption in response. Specific areas of concern include electricity bills, bill volatility, and equity implications related to cost shifting among customer groups. In this paper we leverage smart meter data for more than 400,000 California utility customers, spanning residential, commercial, industrial, and agricultural customers, to assess potential customer bill impacts arising from a multi-component dynamic tariff . Specifically, we compute impacts on customer bills and bill volatility under the assumption of fully inelastic demand, i.e., where customers do not change their consumption patterns in response to the tariff. We also assess various approaches designing subscription load shapes that customers can pre-purchase as a hedge that may provide a measure of protection against large negative impacts, while still incentivizing the modification of loads on the margin. We compare and contrast the relative impacts on different customer classes and discuss benefits and pitfalls of different dynamic tariff structures and subscription load shapes.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2022

Influence of sampling frequency and estimation method on phosphorus load uncertainty in the Western Lake Erie Basin, Ohio, USA

Accurate estimates of nutrient loads are necessary to identify critical source areas and quantify the impact of management practices on pollutant export. Previous studies have investigated nutrient load estimate uncertainty, but they often focus on nutrient loads estimated using an interpolation method for large-scale watersheds with short-term datasets. The study objective was to quantify uncertainty in soluble reactive phosphorus (SRP), total phosphorus (TP), and suspended solids (SS) load estimates from two small (<10 3 km 2 ) agricultural watersheds in the western Lake Erie Basin resulting from different sampling frequencies. Each watershed had high temporal resolution datasets of discharge (15 min) and nutrient concentration (1 to 3 samples per day) collected over a 30-year period (1990–2020). Firstly, SRP, TP, and SS loads were calculated using the high temporal resolution datasets, which was assumed as “true loads”. Secondly, the high temporal concentration data were decomposed to semiweekly, weekly, biweekly, and monthly sampling and annual loads were estimated using four common load estimation methods to assess the effect of sampling frequency and load estimation method on load estimate error. Across the four different methods, the composite method had the lowest relative root mean square and absolute bias, but the rectangular interpolation method was the most precise. Furthermore, even with semiweekly sampling, the composite method resulted in an unacceptable level of precision (average imprecision = 39 %), while the interpolation method resulted in an unacceptable bias (average absolute bias = 16 %). Because neither method could provide acceptable accuracy and precision at the lowest decrease in sampling (e.t. semiweekly sampling), continued daily sampling is recommended in these watersheds.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Near wake evolution of a tidal stream turbine due to asymmetric sheared turbulent inflow with different integral length scales

Tidal stream turbines deployed at highly energetic open water sites are subjected to sheared inflow in the rotor plane. The inflow shear is expected to cause asymmetric loading on the rotor blades and affect the downstream wake. In the current study, two different turbulent inflow conditions, static-high shear and dynamic shear, were generated via an active-grid turbulence generator. A 1:20 scaled three-bladed horizontal axis tidal turbine model was tested in those conditions. The results were compared to a quasi-laminar case with no imposed turbulence or shear. The results show that the high shear reduces the average performance, with a drop of up to 16% in the optimal power coefficient. Besides, the shear profiles increase torque fluctuations and induce significant differences in wake hydrodynamics between the high-speed (upper) and low-speed (lower) regions. The large integral length scales further enhance the load fluctuations perceived by the rotor but have a negligible effect on the mean wake field quantities and the wake recovery. The lower half region featured a faster breakdown of tip vortex structure and a rapid drop of swirl number, a phenomenon conjectured to be a consequence of the strong turbulence intensities and Reynolds stresses in the lower half region. Furthermore, the sheared turbulent inflow also results in a very intensive energy redistribution process towards large-scale, low-frequency motions, which is important to the downstream turbines.

16 TIDAL AND WAVE POWER↗
DOE OSTI2021

Pseudoelasticity of SrNi2P2 Micropillar via Double Lattice Collapse and Expansion

The maximum recoverable strain of most crystalline solids is less than 1% because plastic deformation or fracture usually occurs at a small strain. In this work, we show that a SrNi 2 P 2 micropillar exhibits pseudoelasticity with a large maximum recoverable strain of ~14% under uniaxial compression via unique reversible structural transformation, double lattice collapse–expansion that is repeatable under cyclic loading. Its high yield strength (~3.8 ± 0.5 GPa) and large maximum recoverable strain bring out the ultrahigh modulus of resilience (~146 ± 19 MJ/m 3 ), a few orders of magnitude higher than that of most engineering materials. The double lattice collapse–expansion mechanism shows stress–strain behaviors similar to that of conventional shape-memory alloys, such as hysteresis and thermo-mechanical actuation, even though the structural changes involved are completely different. Our work suggests that the discovery of a new class of high-performance ThCr 2 Si 2 -structured materials will open new research opportunities in the field of pseudoelasticity.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2022

Generalizability analysis of tool condition monitoring ensemble machine learning models

Tool condition monitoring (TCM) is an essential research area for the optimization and automation of metal machining processes, and could help manufacturers reduce costs, production time, machine downtime, energy use, and part scrappage. However, TCM systems developed in prior studies have struggled to reach the high level of generalizability which is necessary for industrial applications. This study addresses TCM system generalizability to new machining conditions, how variations in machining and environmental conditions may be used to improve model generalizability, and ensemble machine learning techniques for TCM. Further, milling tool life experiments were conducted using various machining conditions, and the processes' sound, spindle power, and axial load signals were collected. Different machine learning models were evaluated for the prediction of tool wear levels, including four individual models and five ensemble models. Changes in cutting speed were found to display a large effect on model performance, while the chip load showed some effect, and the feed rate had little effect. A simulated noise data augmentation technique for model improvement is applied within TCM for the first time, and resulted in increased model generalizability and reduced overfitting. Across several performance metrics the extremely randomized trees ensemble machine learning model generally performed the best for this application, achieving a leave-one-group-out cross validation accuracy score of 92.4 %, a 10-fold cross validation score of 98.9 %, and an averaged accuracy across 11 generalizability tests of 87.3 %.

42 ENGINEERING↗
DOE OSTI2020

National Spherical Torus eXperiment - Upgrade NSTX-U: Structural Analysis of Passive Plates Support Brackets

The objective of this analysis is to estimate and assess the stresses in the passive plates caused by a limiting plasma disruption. The objective of this calculation is also to address observed damage to mounting hardware after completion of NSTX operation. These observations initiated diagnostic efforts during the Upgrade project to monitor the severity of the disruptions that were causing the damage, and planned instrumentation to trend the behavior of the mounting hardware. The Recovery Project mission for the passive plate hardware is to appropriately reinforce the passive plates with stainless steel ribs mounted on the back of the plate to reduce their flexure and stress and reinforcements of the “biscuit” components of the bracketry to transfer loads without damage to bolts, biscuits, and bracket plates and welds. Mid-plane disruptions and quenches were manageable with NSTX vintage hardware. The Vertical Displacement Event (VDE) disruptions position the plasma closer to the passive plates and are more severe than the centered plasma. Earlier work pointed to the P1-P5 VDE as limiting. This has been confirmed by A. Brooks work in support of the load generation effort in reference. During vertical displacement large counter currents are generated in the plate as the plasma approaches it. - as would be expected from passive plates. When the plasma disrupts the currents in the plates are reversed, which also reverses the loading on the plates. This report documents the methodology and results of transient structural analysis on the plasma, passive plates support brackets. The loads induced by eddy and halo currents have been considered, and were generated using a separate electromagnetic analysis (NSTXU_1_11_2_1_CALC_051). The structural model also incorporates the effects from preload. Calculation was performed with the assumption that existing biscuits are not participating in the load bearing except the biscuits connected to jumper bolts, which are fully engaged. Results show that with the planned reinforcements, Passive plates and mounting structures satisfy static and fatigue criteria in all areas except welds, where separate analysis was performed. Both symmetric and asymmetric (with locating block) designs were analyzed, and although stresses were higher without locating block they still satisfy NSTX design criteria.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Scalable probabilistic estimates of electric vehicle charging given observed driver behavior

To prepare for rapid growth in global electric vehicle adoption, grid and policy planners depend on detailed forecasts of future charging demand. In this paper we propose a novel holistic, scalable, probabilistic framework to produce large-scale estimates of electric vehicle charging load for long-term planning that capture real drivers’ charging patterns. Our framework captures the uncertainty and stochasticity in charging demand by taking a graphical modeling approach. It has three core elements: driver groups, charging segment choices, and charging session time and energy requirements. The framework uses hierarchical clustering to group drivers by their charging histories, capturing their heterogeneous behaviors and preferences across different segments or types of charging. The framework uses probabilistic mixture models for each driver group’s sessions to identify the unique charging behaviors observed within each segment. We illustrate its application with a large data set from California, profiling the charging patterns and unique driver clusters it identifies. Using the model knobs representing drivers’ battery capacities, behavior, and segment access we present scenarios for California’s charging demand in 2030 with 8 million passenger electric vehicles. Peak charging demand ranged from 3.3 to 8.7 GW across scenarios. Furthermore, each was calculated in under 45 s on a laptop computer.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2023

Effect of the integral length scales of turbulent inflows on wind turbine loads

As wind turbines become larger, the fluctuations in the inflow become increasingly influential in the turbine structural loading. These fluctuations are characterized by the integral length scale, which represents the average size of the largest energy-containing turbulent eddies. Current design standards neglect the varying integral length scales that characterize inflows of wind turbines in operation. Using large-eddy simulations, we generate turbulent inflows of varying integral length scales and quantify the loads of the IEA 15-MW reference wind turbine. Results illustrate that the impact of turbulence on rotor and tower loads is up to 10 times greater than the impact of the mean shear profile. Increasing integral length scales from 0.3x to 0.5x the rotor diameter reduces blade root flapwise moments and rotor and tower loads. Increasing integral length scales from 0.5x to 1.7x the rotor diameter increases the rotor aerodynamic thrust force and the blade root flapwise shear loads and decreases the tower base torsional moment and the tilting and yawing rotor aerodynamic moments. Additionally, turbulence intensity has a greater impact on wind turbine loads than integral length scales. In conclusion, findings indicate that design standards should consider varying integral length scales for accurate wind turbine loading characterization in turbulent inflow conditions.

17 WIND ENERGY↗
DOE OSTI2024

Why is My Zero Energy Home Not a Zero Carbon Home?

For years, carbon calculations were done very simply. The method of calculation was to take annual totals of energy consumption and multiply by an average emission factor, either for the grid serving a project or for a larger region (e.g. an EPA eGRID sub region). The level of accuracy of this approximation was reasonably good, although the issue of accuracy was not, to our knowledge, tested. And the data required were minimal – just a year’s worth of bills for each fuel and one lookup factor. But this method assures that a net zero energy home is automatically a net zero carbon home because zero times any possible emission factor is still zero. Starting in the early 2010s, things changed – grids were starting to rely more and more heavily on renewables, and the difference was showing up on aggregate load curves. This was perhaps noticed first in California, where aggressive renewable policies led to significant renewable power generation large enough to affect the overall shape of the diurnal load curve for the Independent Systems Operator.

14 SOLAR ENERGY↗
DOE OSTI2020

Improving Primary Frequency Response in Networked Microgrid Operations using Multilayer Perceptron-Driven Reinforcement Learning

Individual microgrids can improve the reliability of power grids during extreme events, and networked microgrids can further improve efficiency through resource sharing and increase the resilience of critical end-use loads. However, networked microgrid operations are subject to large switching transients, which can cause dynamic instability and lead to system collapse. These transients are especially prevalent in microgrids with high penetrations of inverter-connected renewable energy resources, which do not provide the system inertia needed to mitigate the transients. Existing generator controls can be modified to invoke a drop in terminal voltage in response to a frequency deviation, thereby reducing load and improving frequency response. This paper investigates the use of a reinforcement-learning--based controller trained over several switching transient scenarios to modify generator controls during large frequency deviations. Compared to previously used proportional integral controllers, the proposed controller can improve primary frequency response while adapting to changes in system topologies and events.

Radhakrishnan, Nikitha↗