Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed memory”

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 1,243 records · Page 69

A nonvoxel-based dose convolution/superposition algorithm optimized for scalable GPU architectures

Purpose: Real-time adaptive planning and treatment has been infeasible due in part to its high computational complexity. There have been many recent efforts to utilize graphics processing units (GPUs) to accelerate the computational performance and dose accuracy in radiation therapy. Data structure and memory access patterns are the key GPU factors that determine the computational performance and accuracy. In this paper, the authors present a nonvoxel-based (NVB) approach to maximize computational and memory access efficiency and throughput on the GPU. Methods: The proposed algorithm employs a ray-tracing mechanism to restructure the 3D data sets computed from the CT anatomy into a nonvoxel-based framework. In a process that takes only a few milliseconds of computing time, the algorithm restructured the data sets by ray-tracing through precalculated CT volumes to realign the coordinate system along the convolution direction, as defined by zenithal and azimuthal angles. During the ray-tracing step, the data were resampled according to radial sampling and parallel ray-spacing parameters making the algorithm independent of the original CT resolution. The nonvoxel-based algorithm presented in this paper also demonstrated a trade-off in computational performance and dose accuracy for different coordinate system configurations. In order to find the best balance between the computed speedup and the accuracy, the authors employed an exhaustive parameter search on all sampling parameters that defined the coordinate system configuration: zenithal, azimuthal, and radial sampling of the convolution algorithm, as well as the parallel ray spacing during ray tracing. The angular sampling parameters were varied between 4 and 48 discrete angles, while both radial sampling and parallel ray spacing were varied from 0.5 to 10 mm. The gamma distribution analysis method (γ) was used to compare the dose distributions using 2% and 2 mm dose difference and distance-to-agreement criteria, respectively. Accuracy was investigated using three distinct phantoms with varied geometries and heterogeneities and on a series of 14 segmented lung CT data sets. Performance gains were calculated using three 256 mm cube homogenous water phantoms, with isotropic voxel dimensions of 1, 2, and 4 mm. Results: The nonvoxel-based GPU algorithm was independent of the data size and provided significant computational gains over the CPU algorithm for large CT data sizes. The parameter search analysis also showed that the ray combination of 8 zenithal and 8 azimuthal angles along with 1 mm radial sampling and 2 mm parallel ray spacing maintained dose accuracy with greater than 99% of voxels passing the γ test. Combining the acceleration obtained from GPU parallelization with the sampling optimization, the authors achieved a total performance improvement factor of >175 000 when compared to our voxel-based ground truth CPU benchmark and a factor of 20 compared with a voxel-based GPU dose convolution method. Conclusions: The nonvoxel-based convolution method yielded substantial performance improvements over a generic GPU implementation, while maintaining accuracy as compared to a CPU computed ground truth dose distribution. Such an algorithm can be a key contribution toward developing tools for adaptive radiation therapy systems.

60 APPLIED LIFE SCIENCES↗

High-Resolution Synthetic Solar Irradiance Sequence Generation: An LSTM-Based Generative Adversarial Network

The rapid growth of renewable energy resources penetration is bringing more challenges to power system planning and operation. Relevant renewable energy integration studies, such as the capability and dynamic performance of inverter-based resources' primary frequency response and fast frequency response, require high-resolution renewable generation output data that are representative of renewable energy resources. This paper focuses on creating synthetic but realistic solar irradiance data and proposes a long short-term memory-based generative adversarial network to generate high-resolution (second-level) solar irradiance sequences from low-resolution (minute-level) measurements. Combined with a classifier to recognize the solar irradiance patterns, the proposed model is trained using multi-loss functions to accurately capture the temporal correlations among both high-resolution and low-resolution sequences. Verification of the proposed approach is performed on the data set of the Oahu Solar Measurement Grid collected through the National Renewable Energy Laboratory. The results of the case studies demonstrate the proposed approach's capability to capture the statistical characteristics of different solar irradiance patterns and to generate high-quality synthetic solar irradiance sequences in high resolution.

dynamic scheduling↗

Deep Cellular Recurrent Network for Efficient Analysis of Time-Series Data With Spatial Information

Efficient processing of large-scale time series data is an intricate problem in machine learning. Conventional sensor signal processing pipelines with hand engineered feature extraction often involve huge computational cost with high dimensional data. Deep recurrent neural networks have shown promise in automated feature learning for improved time-series processing. However, generic deep recurrent models grow in scale and depth with increased complexity of the data. This is particularly challenging in presence of high dimensional data with temporal and spatial characteristics. Consequently, this work proposes a novel deep cellular recurrent neural network (DCRNN) architecture to efficiently process complex multi-dimensional time series data with spatial information. Here, the cellular recurrent architecture in the proposed model allows for location-aware synchronous processing of time series data from spatially distributed sensor signal sources. Extensive trainable parameter sharing due to cellularity in the proposed architecture ensures efficiency in the use of recurrent processing units with high-dimensional inputs. This study also investigates the versatility of the proposed DCRNN model for classification of multi-class time series data from different application domains. Consequently, the proposed DCRNN architecture is evaluated using two time-series datasets: a multichannel scalp EEG dataset for seizure detection, and a machine fault detection dataset obtained in-house. The results suggest that the proposed architecture achieves state-of-the-art performance while utilizing substantially less trainable parameters when compared to comparable methods in the literature.

60 APPLIED LIFE SCIENCES↗

DNS and LES of turbulent flow in a closed channel featuring a pattern of hemispherical roughness elements

Three mesh adaptivity algorithms were developed to facilitate and expedite the use of the CADIS and FW-CADIS hybrid Monte Carlo/deterministic techniques in accurate full-scale neutronics simulations of fusion energy systems with immense sizes and complicated geometries. First, a macromaterial approach enhances the fidelity of the deterministic models without changing the mesh. Second, a deterministic mesh refinement algorithm generates meshes that capture as much geometric detail as possible without exceeding a specified maximum number of mesh elements. Finally, a weight window coarsening algorithm decouples the weight window mesh and energy bins from the mesh and energy group structure of the deterministic calculations in order to remove the memory constraint of the weight window map from the deterministic mesh resolution. The three algorithms were used to enhance an FW-CADIS calculation of the prompt dose rate throughout the ITER experimental facility and resulted in a 23.3% increase in the number of mesh tally elements in which the dose rates were calculated in a 10-day Monte Carlo calculation. Additionally, because of the significant increase in the efficiency of FW-CADIS simulations, the three algorithms enabled this difficult calculation to be accurately solved on a regular computer cluster, eliminating the need for a world-class super computer.

Nuclear Science & Technology↗

LAF-Net: A Deep Residual and Cross-Attention Framework for Day-Ahead Load Forecasting: Preprint

Accurate day-ahead load forecasting is essential for reliable power system operations and market efficiency. System operators such as the Midcontinent Independent System Operator (MISO) rely on forecasts from multiple vendors, yet combining them effectively remains a persistent challenge due to vendor-specific biases. This paper presents a novel LSTM-Attention Fusion Network with Error Representation (LAF-Net) that enhances day-ahead hourly load forecasting through deep residual learning and multi-modal cross-attention. The proposed model builds a historical error memory from past vendor performance and dynamically queries it with future hour context to generate adaptive, hour-specific trust weights for each vendor. A bounded residual correction further refines forecasts by mitigating systematic and temporally localized errors. Tested on real MISO LBA data with multi-vendor forecasts, LAF-Net consistently outperforms the best vendor baseline across all 38 LBAs, achieving more than a 40% reduction in system-level mean absolute error (MAE) during peak load hours relative to the best vendor baseline.

24 POWER TRANSMISSION AND DISTRIBUTION↗

3-D decoupled inversion of complex conductivity data in the real number domain

Complex conductivity inversion, aslo called spectral induced polarization inversion, involves estimating the frequency dependent complex electrical conductivity distribution of the subsurface. The superior diagnostic capabilities provided by complex conductivity spectra have driven advancements in mechanistic understanding of complex conductivity as well as modelling and inversion appraoches over the past several decades. In this work, we demonstrate the theory and application for an approach to three-dimensional modeling and inversion of complex conductivity data in the real number domain. Beginning from first principles, we demonstrate how the equations for the real and imaginary components of the complex potential may be decoupled. This leads to a description of the real and imaginary source current terms, and a corresponding assessment of error arising from an assumption necessary complete the decoupled modeling. We show that for most earth materials, which exhibit relatively small phases (e.g. less than 0.2 radians) in complex conductivity, these errors become insignificant. For higher phase materials, the errors may be quantified and corrected through an iterative procedure. We demonstrate the accuracy of numerical forward solutions by direct comparison to corresponding analytic solutions. We demonstrate the inversion using both synthetic and field examples with data collected over a waste infiltration trench, at frequencies ranging from 0.5 to 7.5 Hz. In addition to insight provided by the decoupled equations, we see three primary advantages of the decoupled inversion in comparison to fully coupled inversion in complex number domain. First, memory requirements are reduced by a factor of two because it is unnecessary to simultaneously store both the real and complex components and associated Jacobian matrices. Second, data noise is specified separately for the real and imaginary components of potential (or magnitude and phase), which removes the ambiguity in data misfit residuals that exists in the complex number domain, thereby enabling the inversion to appropriately fit the data. Final, except for the source current terms, the modeling and inversion algorithms are equivalent to the direct current electrical resistivity tomography counterpart. In conclusion, this enables direct application of advancements and approaches developed originally for electrical resistivity tomography.

58 GEOSCIENCES↗

Design of a V–Ti–Ni alloy with superelastic nano-precipitates

Stress-induced martensitic transformations enable metastable alloys to exhibit enhanced strain hardening capacity, leading to improved formability and toughness. As is well-known from transformation-induced plasticity (TRIP) steels, however, the resulting martensite can limit ductility and fatigue life due to its intrinsic brittleness. In this work, we explore an alloy design strategy that utilizes stress-induced martensitic transformations but does not retain the martensite phase. This strategy is based on the introduction of superelastic nano-precipitates, which exhibit reverse transformation after initial stress-induced forward transformation. To this end, utilizing ab-initio simulations and thermodynamic calculations we designed and produced a V 45 Ti 30 Ni 25 (at%) alloy. In this alloy, TiNi is present as nano-precipitates uniformly distributed within a ductile V-rich base-centered cubic (bcc) β matrix, as well as being present as a larger matrix phase. We characterized the microstructure of the produced alloy using various scanning electron microscopy (SEM) and transmission electron microscopy (TEM) methods. Furthermore, the bulk mechanical properties of the alloy are demonstrated through tensile tests, and the reversible transformation in each of the TiNi morphologies were confirmed by in-situ TEM micro-pillar compression experiments, in-situ high-energy diffraction synchrotron cyclic tensile tests, indentation experiments, and differential scanning calorimetry experiments. The observed transformation pathways and variables impacting phase stability are critically discussed.

59 BASIC BIOLOGICAL SCIENCES↗

Whole-brain hippocampal sparing radiation therapy: Volume-modulated arc therapy vs intensity-modulated radiation therapy case study

The hippocampus is responsible for memory and cognitive function. An ongoing phase II clinical trial suggests that sparing dose to the hippocampus during whole-brain radiation therapy can help preserve a patient's neurocognitive function. Progressive research and advancements in treatment techniques have made treatment planning more sophisticated but beneficial for patients undergoing treatment. The aim of this study is to evaluate and compare hippocampal sparing whole-brain (HS-WB) radiation therapy treatment planning techniques using volume-modulated arc therapy (VMAT) and intensity-modulated radiation therapy (IMRT). We randomly selected 3 patients to compare different treatment techniques that could be used for reducing dose to the hippocampal region. We created 2 treatment plans, a VMAT and an IMRT, from each patient's data set and planned on the Eclipse 11.0 treatment planning system (TPS). A total of 6 plans (3 IMRT and 3 VMAT) were created and evaluated for this case study. The physician contoured the hippocampus as per the Radiation Therapy Oncology Group (RTOG) 0933 protocol atlas. The organs at risk (OR) were contoured and evaluated for the plan comparison, which included the spinal cord, optic chiasm, the right and left eyes, lenses, and optic nerves. Both treatment plans produced adequate coverage on the planning target volume (PTV) while significantly reducing dose to the hippocampal region. The VMAT treatment plans produced a more homogenous dose distribution throughout the PTV while decreasing the maximum point dose to the target. However, both treatment techniques demonstrated hippocampal sparing when irradiating the whole brain.

61 RADIATION PROTECTION AND DOSIMETRY↗

FORCE-DISPATCHES Integration - Initial Demonstration

Integrated energy systems (IES) combine, in mutually beneficial ways, power from variable renewable energy sources and nuclear power plants (NPP) to improve economic viability under uncertain market and weather conditions. The open-source Framework for Optimization of Resources and Economics (FORCE) tool suite, developed at Idaho National Laboratory (INL), has enabled comprehensive modeling and simulation of IES. The capabilities within FORCE include grid portfolio optimization through the Holistic Energy Resource Optimization Network (HERON) and the transient process model analysis library HYBRID, among others. Continuous efforts and investments from the IES programs have been made to expand and improve the versatility of the FORCE toolset in fiscal year 2022. Code-coupling and cross-tool communication have been important methods for improving this versatility. This report focuses on an additional workflow in the HERON tool for capacity and dispatch stochastic optimization through integration with the external tool Design Integration and Synthesis Platform to Advance Tightly Coupled Hybrid Energy Systems (DISPATCHES). DISPATCHES was primarily developed by the National Energy Technology Laboratory, in collaboration with other national laboratories, which included INL, universities, and industry partners. It is coupled to a library of algebraic models for specific plant components, and to a framework for stochastic optimization different from that provided in the current Risk Analysis Virtual Environment (RAVEN)-running-RAVEN algorithm in HERON. HERON currently conducts stochastic optimization via an outer-inner loop: it optimizes over variable capacity on the outer loop, and at each step within the capacity parameter space, conducts an inner optimization over scenarios (of market signals, demand, and/or weather patterns) and hourly dispatch throughout a user-specified number of years. On the other hand, DISPATCHES conducts stochastic optimization via an “all-at-once” strategy in which capacity variables are optimized at the same level as dispatch variables, as all scenarios are considered at once. The latter method works especially well for projects of limited size and project length, as the necessary computational power and memory increases with the number of variables and scenarios. The new capability to use the DISPATCHES workflow in HERON enhances standalone simulations by leveraging FORCE tools—namely, the economic metrics from the Tool for Economic Analysis (TEAL) and reduced-order model (ROM) sampling from RAVEN. The initial demonstration of the DISPATCHES workflow simulates an existing nuclear-case flowsheet within the DISPATCHES repository—this models a NPP with a secondary revenue stream for hydrogen production. Electrical output from the plant is converted to hydrogen via a proton-exchange membrane (PEM) electrolyzer, hydrogen tanks are used for storage, and an additional turbine is added for hydrogen combustion. Continued work regarding this FORCE-DISPATCHES integration will include automatic generation of DISPATCHES models from HERON inputs, offering analysts the option of using either the RAVEN-runsRAVEN or DISPATCHES workflow to solve technoeconomic optimization problems.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Proximity Portability and in Transit , M-to-N Data Partitioning and Movement in SENSEI [Book Chapter]

In high-performance parallel in situ processing, the term in transit processing refers to those configurations where data must move from a producer to a consumer that runs on separate resources. In the context of parallel and distributed computing on an HPC platform one of the central challenges is to determine a mapping of data from producer ranks to consumer ranks. This problem is complicated by the heterogeneity that arises in producer-consumer pairs, such as when producer and consumer codes have different levels of concurrency, different scaling characteristics, or different data models. The resulting mapping and movement of data from M producer to N consumer ranks can have a significant impact on aggregate application performance, particularly when the data consumer requires only a subset of the overall data for its task. This chapter focuses on the design considerations that underlie SENSEI’s implementation to this challenging problem. These design considerations extend the core SENSEI architecture and include ideas like the need to accommodate flexibility in the choice of different partitioning methods, the ability for a data consumer to request and receive only the subset of data needed for its particular operation, and the ability to leverage any of several different data transport tools. The idea of proximity portability, being able to use different data transport methods as part of an in transit workflow, is illustrated through the use of three different transport layers where switching from one transport tool to another is accomplished with only a configuration file change. Here, the chapter also includes a performance analysis summary showing the performance gains that are possible in terms of multiple metrics, such as memory footprint, time to solution, and amount of data moved, when using optimized partitioners in an in transit setting, gains that are made possible by the implementation shaped by specific design considerations.

Bethel, E. Wes↗

Chiral THz metamaterial with tunable optical activity

Optical activity in chiral metamaterials is demonstrated in simulation and shows actively tunable giant polarization rotation at THz frequencies. Electric current distributions show that pure chirality is achieved by our bi-Iayer chiral metamaterial design. The chirality can be optically controlled by illumination with near-infrared light. Optical activity, occurring in chiral materials such as DNA, sugar and many other bio-molecules, is a phenomenon of great importance to many areas of science including molecular biology, analytical chemistry, optoelectronics and display applications. This phenomenon is well understood at an effective medium level as a magnetic/electric moment excited by the electric/magnetic field of the incident electromagnetic (EM) wave. Usually, natural chiral materials exhibit very weak optical activity e.g. a gyrotropic quartz crystal. The optical activity of chiral metamaterials, however, can be five orders of magnitude stronger. Chiral metamaterials are made of sub-wavelength resonators lacking symmetry planes. The asymmetry allows magnetic moments to be excited by the electric field of the incident EM wave and vice versa. Recently, chiral metamaterials have been demonstrated and lead to prospects in giant optical activity, circular dichroism, negative refraction and reversing the Casmir force. These fascinating optical properties require strong chirality, which may be designed through the microscopic structure of chiral metamaterials. However, these metamaterials have a fixed response function, defined by the geometric structuring, which limits their ability to manipulate EM waves. Active metamaterials realize dynamic control of response functions and have produced many influential applications such as ultra-fast switching devices, frequency and phase modulation and memory devices. Introducing active designs to chiral metamaterials will give additional freedom in controlling the optical activity, and therefore enable dynamic manipulation of electromagnetic waves. This should prove useful for many potential applications. In this work, we show an active chiral metamaterial design with tunable giant optical activity. Our numerical simulations show the proposed chiral metamaterial has exceptionally strong polarization rotation with low transmission losses at THz frequencies, which makes it a very efficient material for tunable polarization rotators. Our preliminary experimental results (now shown here) show good agreements with numerical simulations.

36 MATERIALS SCIENCE↗

Ab initio modeling of transport and thermodynamic stability for hafnia memristive devices

HfO 2 -based memristive switching devices are currently under intensive investigation due to their high performance and mature fabrication techniques. However, several critical issues have to be addressed to bring them from lab to market. We have recently looked into two important issues with the use of density functional theory methods. One is the wide distribution of device resistance in off-states. We have modeled the switching process of a Pt-HfO 2 -Pt structure for which quantized conductance was observed. Oxygen atoms moving inside a conductive oxygen vacancy filament divide the filament into several quantum wells. Device conductance changes exponentially when one oxygen atom moves away from interface into filament. We propose that the high sensitivity of device conductance to the position of oxygen atoms results in the large variation of device off-state resistance. Another issue that we have recently addressed is the poor switching performance of devices based on a TiN-HfO 2 -TiN structure. While recent experiments have shown that by inserting an "oxygen scavenger" metal between positive electrode and oxide significantly improves device performance, the fundamental understanding of the improvement is lacking.We provide detailed understanding how scavenger layers improve device performance. First, we show that Ta insertion facilitates formation of on-states by reducing the formation energy. Second, the inserted Ta layer reduces the Schottky barrier height in the off-states by changing interface electric dipole at the oxide electrode interface. Nevertheless, the device maintains a high on/off resistance ratio. Finally, with Ta insertion the on-state conductance becomes much less sensitive to the specific location from which the oxygen was removed from the oxide. In conclusion, our studies provide fundamental understanding needed for enabling realization of a non-volatile memory technology with reduced energy consumption.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Impact of zeolite aging in hot liquid water on activity for acid-catalyzed dehydration of alcohols

The catalytic performance of zeolite in aqueous medium depends on a multitude of factors, such as the concentration and distribution of active sites and framework integrity. Al K–edge extended X–ray absorption fine structure and 27Al MAS NMR spectroscopies in combination with DFT calculations are used to determine the distribution of tetrahedral Al sites both qualitatively and quantitatively for both parent and 48 h 160 ºC water treated HBEA catalysts. There is no evidence of Al coordination modification after aging in water. The distribution and concentration of Al T–sites, active centers for the dehydration of cyclohexanol, do not markedly impact the catalytic performance in water, because the Brønsted acidic protons are present in the form of hydrated hydronium ions and thus have very similar acid properties. The results suggest that all Brønsted acid sites are equally active in aqueous medium. The decrease of zeolite catalytic performance after water treatment is attributed to the reduced concentration of Brønsted acid sites. Increasing the stability of pore walls and decreasing the rate of Si–O–Si group hydrolysis may result in improved apparent zeolite catalytic performance in aqueous medium. Authors thank B. W. Arey (PNNL) for HIM measurements, T. Huthwelker for support during Al XAFS measurements at the Swiss Light Source (PSI, Switzerland), J. Z. Hu and S. D. Burton (PNNL) for support during NMR experiments. This work was supported by the U. S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences & Biosciences. MD acknowledges support by the Materials Synthesis and Simulation Across Scales (MS3 Initiative) conducted under Laboratory Directed Research & Development Program at PNNL. HIM imaging and NMR experiments were performed at the Environmental Molecular Sciences Laboratory, a national scientific user facility sponsored by the DOE Office of Science, Office of Biological and Environmental Research, located at Pacific Northwest National Laboratory (PNNL). PNNL is a multiprogram national laboratory operated for the DOE by Battelle Memorial Institute under contract # DE-AC05-76RL0-1830

alcohol dehydration↗

Estimating Flexibility Envelopes for Residential Customers From Utility Smart Meter Data: Preprint

Demand response from residential customers has significant potential to support power system operations, but accurate flexibility estimation is challenging due to the limited resolution of advanced metering infrastructure (AMI) data. Most utility AMI measurements are recorded at hourly intervals, with only a small portion at higher resolutions, and even fewer households have appliance-level energy usage data. To address this issue, this paper proposes a two-stage long short-term memory (LSTM) framework for estimating household flexibility envelopes from low-resolution AMI data. In the first stage, the heating, ventilating, and air-conditioning (HVAC) load and non-HVAC loads are estimated by using a model trained on a small set of households with appliance-level profiles. These estimated data are then used to compute the upper- and lower-flexibility bounds, which are subsequently down-sampled to lower-resolution data. In the second stage, these flexibility bounds serve as training inputs for another LSTM model, enabling direct prediction of flexibility envelopes for households with only hourly AMI data. This method is validated using Pecan Street data from two different areas, and the results demonstrate its applicability and effectiveness.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Short-term apartment-level load forecasting using a modified neural network with selected auto-regressive features

Residential electricity load profiles and their diversity have become increasingly important to realize the benefits of Smart or Transactive Energy Networks (TENs). An important element of TENs will be practical, accurate, and implementable residential load forecasting techniques. While there have been many approaches to short-term load forecasting, few have included forecasting for individual households, partly because the high volatility and idiosyncrasies present in individual household load data can pose significant challenges. In this study, we develop a Convolutional Long Short-Term Memory-based neural network with Selected Autoregressive Features (termed a CLSAF model) to improve short-term household electricity load forecasting accuracy by employing three strategies: autoregressive features selection, exogenous features selection, and a “default” state to avoid overfitting at times of high load volatility. We include aggregations of apartments to floor and building level, because utilities may favor transactive approaches that rely on aggregator models, e.g., a cluster of consumers as opposed to an individual. We demonstrate that the CLSAF model, by virtue of its enhanced feature representation and modest computational resources, can accomplish load forecasting in a multi-family residential building across three spatial granularities (individual apartment/household, floor, and building levels), with an accuracy improvement of up to 25% compared to a persistence model. We propose a data screening technique to characterize time-series electricity-load data. This technique is suitable for integration into a TEN ecosystem and allows one to estimate confidence levels of the load forecasts to optimize computational resources and the risks associated with uncertain forecasts.

24 POWER TRANSMISSION AND DISTRIBUTION↗

VRN3P: Variational Recurrent Neural Network Based Net-Load Prediction under High Solar Penetration

This is the final technical report for the SETO-funded VRN3P project (PNNL# 76914). The goal of this project, led by Pacific Northwest National Laboratory (PNNL), in collaboration with Lawrence Livermore National Laboratory (LLNL) and Portland General Electric (PGE), was to develop and validate a deep variational recurrent neural network-based net-load prediction (VRN3P) framework for probabilistic time-series forecasting of day-ahead net-load under high solar penetration scenarios. The project team reports successful design of a novel probabilistic net-load forecasting architecture, comprising of a variational autoencoder and a recurrent neural network, which demonstrates 30% improvement in forecast performance, 60% improvement in training time, and consumes 44% less memory, when compared with conventional baseline models. The team tested the VRN3P model performance on GridLAB-D test-cases representing varying BTM solar penetration levels of 20%, 30%, and 50%, with integrated time-series net-load profiles provided by the utility partner (PGE). The VRN3P model demonstrate <2% hourly MAPE (averaged over the year) for day- ahead net-load forecast on the test scenario with 20% BTM solar. Transfer learning extension of the VRN3P model has demonstrated 8.33× speed-up in training, while still achieving acceptable forecast performance of 2.24% hourly MAPE on the 30% BTM solar penetration test-scenario. A preliminary version of the VRN3P GridAPPS-D™has been developed, along with a web-based interactive user-interface (named ‘Forte’) which has made available on GitHub for public use.

24 POWER TRANSMISSION AND DISTRIBUTION↗