Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Deep”

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 613 records · Page 34
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Designing Experiments for SpinSat, A Novel Variable-Gravity-and-Radiation Platform for Deep-Space Science

Conducting experiments to measure the effects of deep-space radiation and reduced gravity on biological and physical systems remains challenging. The result is a substantial knowledge gap that poses risks to our ability to sustain life and conduct critical operations in deep space. The SpinSat spacecraft platform is designed to bridge such gaps by providing low-cost, reliable, and frequent access to deep space. A disk-shaped rotating satellite that can provide artificial gravity and exposure to space radiation simultaneously, SpinSat is designed to accommodate payloads in a CubeSat form factor (with at least 48 “U” volume), providing power, communications, and a benign thermal environment. It is orbit-agnostic, enabling access to a variety of radiation environments (Van Allen belts, deep space, cis-lunar); and can be equipped with shielding to mimic planetary radiation environments, for both short- and long-term experiments. Because of its versatility and prioritization of late loading for biological payloads, it is well suited to host a wide range of ranging from human tissues and organoids to microorganisms, plants, chemistry, and regolith. Here, we present examples of potential experiment concepts for SpinSat, and discuss the details of how experimental designs could interact with the platform. Potential SpinSat studies have diverse applications, including fundamental radiation biology and DNA repair; cancer biology and countermeasure development; space agriculture; bioproduction of nutrients and pharmaceuticals; understanding regolith dynamics in low gravity; prebiotic chemistry and panspermia. We will further highlight ideas for SpinSat-compatible experimental hardware, existing and in development, and experiment-relevant details on SpinSat capabilities including artificial gravity, potential radiation environments, data, and power. This presentation will aim to provide investigators with the high-level technical information necessary to inspire experiments for SpinSat. We also seek to stimulate conversation and to gain community input on accommodations needs to help guide the evolving design of this platform.

experiment design↗
NASA NTRSDate not supplied

Developing Autonomous Technologies for Biological Missions to Deep Space

In upcoming biological missions beyond low Earth orbit (LEO), the use of autonomous instrumentation will allow scientists to perform a variety of experiments, including the characterization of the response to different space environments (Moon, Mars, interplanetary space) using biological models like microbes, plants, organoids, and tissue chips. BioSentinel is an ongoing deep space mission, currently at over 50 million kilometers from Earth and the first instrument developed to perform biological experiments beyond LEO. Even though the primary objective of this CubeSat mission was to investigate the effects of the deep space radiation environment on budding yeast, the spacecraft bus (i.e., all the subsystems that support the biological payload like power, thermal, data telemetry, navigation, etc.) can accommodate a variety of biological (and physical) experiments and model organisms. LEIA, an upcoming CLPS mission to the lunar surface, uses a microfluidic and optical instrument based on BioSentinel to study the effects of the lunar environment on different cellular processes and on bioproduction of antioxidants. A new series of science mission concepts are being proposed to be accommodated into platforms like BioSentinel. These missions will investigate the response of a variety of organisms to the deep space environment, including but not limited to single-cell eukaryotes, cyanobacteria, plants (including crops), organoids, and tissue chips. In addition to optical absorbance measurements like the ones performed in BioSentinel (and LEIA), we are investigating the use of fluorescence detection, microscopy, sequencing devices, etc. Thus, instruments like the ones proposed here can be adapted to a variety of platforms like free-flyers, deployable payloads, landers, rovers, and the lunar Gateway. These technologies can be used as steppingstones for establishing a sustained human presence on the Moon and in deep space while providing knowledge for the development of potential countermeasures.

Sergio R Santa Maria↗
DOE OSTI2016

An analysis of image storage systems for scalable training of deep neural networks

This study presents a principled empirical evaluation of image storage systems for training deep neural networks. We employ the Caffe deep learning framework to train neural network models for three different data sets, MNIST, CIFAR-10, and ImageNet. While training the models, we evaluate five different options to retrieve training image data: (1) PNG-formatted image files on local file system; (2) pushing pixel arrays from image files into a single HDF5 file on local file system; (3) in-memory arrays to hold the pixel arrays in Python and C++; (4) loading the training data into LevelDB, a log-structured merge tree based key-value storage; and (5) loading the training data into LMDB, a B+tree based key-value storage. The experimental results quantitatively highlight the disadvantage of using normal image files on local file systems to train deep neural networks and demonstrate reliable performance with key-value storage based storage systems. When training a model on the ImageNet dataset, the image file option was more than 17 times slower than the key-value storage option. Along with measurements on training time, this study provides in-depth analysis on the cause of performance advantages/disadvantages of each back-end to train deep neural networks. We envision the provided measurements and analysis will shed light on the optimal way to architect systems for training neural networks in a scalable manner.

Lim, Seung-Hwan↗
DOE OSTI2016

A Study of Complex Deep Learning Networks on High Performance, Neuromorphic, and Quantum Computers

Current Deep Learning models use highly optimized convolutional neural networks (CNN) trained on large graphical processing units (GPU)-based computers with a fairly simple layered network topology, i.e., highly connected layers, without intra-layer connections. Complex topologies have been proposed, but are intractable to train on current systems. Building the topologies of the deep learning network requires hand tuning, and implementing the network in hardware is expensive in both cost and power. In this paper, we evaluate deep learning models using three different computing architectures to address these problems: quantum computing to train complex topologies, high performance computing (HPC) to automatically determine network topology, and neuromorphic computing for a low-power hardware implementation. Due to input size limitations of current quantum computers we use the MNIST dataset for our evaluation. The results show the possibility of using the three architectures in tandem to explore complex deep learning networks that are untrainable using a von Neumann architecture. We show that a quantum computer can find high quality values of intra-layer connections and weights, while yielding a tractable time result as the complexity of the network increases; a high performance computer can find optimal layer-based topologies; and a neuromorphic computer can represent the complex topology and weights derived from the other architectures in low power memristive hardware. This represents a new capability that is not feasible with current von Neumann architecture. It potentially enables the ability to solve very complicated problems unsolvable with current computing technologies.

Potok, Thomas E↗
DOE OSTI2017

Process-based approach for the detection of deep gas invading the surface

The present invention includes a method for determining the level of deep gas in a near surface formation that includes: measuring CO.sub.2, O.sub.2, CH.sub.4, and N.sub.2 levels in percent by volume from one or more surface or near surface geological samples; adding the water vapor content to the measured CO.sub.2, O.sub.2, CH.sub.4, and N.sub.2 levels in percent by volume; normalizing the gas mixture to 100% by volume or 1 atmospheric total pressure; and determining the ratios of: O.sub.2 versus CO.sub.2 to distinguish in-situ vadose zone CO.sub.2 from exogenous deep leakage CO.sub.2; CO.sub.2 versus N.sub.2 to distinguish whether CO.sub.2 is being removed from the near surface formation or CO.sub.2 is added from an exogenous deep leakage input; or CO.sub.2 versus N.sub.2/O.sub.2 to determine the degree of oxygen influx, consumption, or both; wherein the ratios are indicative of natural in situ CO.sub.2 or CO.sub.2 from the exogenous deep leakage input.

47 OTHER INSTRUMENTATION↗
DOE OSTI2015

Field Test to Evaluate Deep Borehole Disposal.

The U.S. Department of Energy (DOE) has embarked on the Deep Borehole Field Test (DBFT), which will investigate whether conditions suitable for disposal of radioactive waste can be found at a depth of up to 5 km in the earth’s crust. As planned, the DBFT will demonstrate drilling and construction of two boreholes, one for initial scientific characterization, and the other at a larger diameter such as could be appropriate for waste disposal (the DBFT will not involve radioactive waste). A wide range of geoscience activities is planned for the Characterization Borehole, and an engineering demonstration of test package emplacement and retrieval is planned for the larger Field Test Borehole. Characterization activities will focus on measurements and samples that are important for evaluating the long-term isolation capability of the Deep Borehole Disposal (DBD) concept. Engineering demonstration activities will focus on providing data to evaluate the concept’s operational safety and practicality. Procurement of a scientifically acceptable DBFT site and a site management contractor is now underway. The concept of deep borehole disposal (DBD) for radioactive wastes is not new. It was considered by the National Academy of Science (NAS 1957) for liquid waste, studied in the 1980’s in the U.S. (Woodward–Clyde 1983), and has been evaluated by European waste disposal R&D programs in the past few decades (for example, Grundfelt and Crawford 2014; Grundfelt 2010). Deep injection of wastewater including hazardous wastes is ongoing in the U.S. and regulated by the Environmental Protection Agency (EPA 2001). The DBFT is being conducted with a view to use the DBD concept for future disposal of smaller-quantity, DOE-managed wastes from nuclear weapons production (i.e., Cs/Sr capsules and granular solid wastes). However, the concept may also have broader applicability for nations that have a need to dispose of limited amounts of spent fuel from nuclear power reactors. For such nations the cost for disposing of volumetrically limited waste streams could be lower than mined geologic repositories.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2019

On the Susceptibility of Deep Neural Networks to Natural Perturbations

Deep learning systems are increasingly being adopted for safety critical tasks such as autonomous driving. These systems can be exposed to adverse weather conditions such as fog, rain and snow. Vulnerability of deep learning systems to synthetic adversarial attacks has been extensively studied and demonstrated, but the impact of natural weather conditions on these systems has not been studied in detail. In this paper, we study the effects of fog on classification accuracy of the popular Inception deep learning model. We use stereo images from the Cityscapes dataset and computer graphics techniques to mimic realistic naturally occurring fog. We show that the Inception deep learning model is vulnerable to the addition of fog in images.

Ozdag, Mesut↗
DOE OSTI2020

MEASUREMENT OF Fn 2 /Fp 2 FROM DEEP INELASTIC ELECTRON SCATTERING OFF A=3 MIRROR NUCLEI AT JEFFERSON LAB

This research is part of the Je?erson Lab MARATHON experiment (MeAsurement of the Fn 2 /Fp 2 , d/u RAtios and A = 3 EMC E?ect in Deep Inelastic Electron Scattering Off the Tritium and Helium MirrOr Nuclei). MARATHON is an experiment on electron deep inelastic scattering off the 3H, 3He mirror nuclei. It has extracted, from the measured deep inelastic scattering cross section ratio o? 3H and 3He, the ratio of the neutron to proton inelastic structure functions Fn 2 /Fp 2 . The extraction has taken advantage of the isospin symmetry of the A = 3 mirror nuclei within a novel technique, which avoids the theoretical uncertainties that are present in the SLAC experiments of the 1970s, which used hydrogen and deuterium targets. The experiment was performed using the upgraded 11 GeV beam of the Jefferson Lab electron accelerator, the Hall A Experimental Facility, and a newly designed cryogenic, high-pressure gas target system containing 3H and 3He cells. Electrons scattered o? the gas targets were detected in the two superconducting, high-resolution magnetic spectrometers of Hall A. The cross section data cover the four-momentum transfer squared range 2.0 < Q2 < 14.0 (GeV/c)2, and the Bjorken x scaling variable range 0.19 < x < 0.83. The experiment also measured, for calibration purposes, electron deep inelastic scattering off hydrogen and deuterium at selected kinematics.

Su, Tong↗
DOE OSTI2021

Towards Efficient Uncertainty estimation in deep learning for robust energy prediction in crystal materials

In material science, recent studies have started to explore the potential of using deep learning to improve property prediction from high-fidelity simulations, e.g, density functional theory (DFT). However, the design spaces are sometimes too large and intractable to sample completely. This results in a critical question that is how to evaluate the confidence and robustness of the prediction. In this paper, we propose an efficient approach to estimate uncertainty in deep learning using a single forward pass and then apply it for robust prediction of the total energy in crystal lattice structures. Our approach is built upon the deep kernel learning (DKL) that originally introduces to leverage the expressiveness of deep neural networks as input with a probabilistic prediction of Gaussian processes (GPs) as output. Existing DKL methods have difficulties in the accuracy of predictive uncertainty, training stability, and scaling to large datasets, which lead to significant barriers in real-world applications. We propose to address these challenges by using an inducing point approximate GP in feature space combined with spectral normalization as a regularization. We finally demonstrate our robust performance on an artificial example and a real-world application from materials chemistry.

Bi, Sirui↗
DOE OSTI2013

Diffusion Dominant Solute Transport Modelling In Deep Repository Under The Effect of Emplacement Media Degradation - 13285

Deep geologic disposal of high activity and long-lived radioactive waste is being actively considered and pursued in many countries, where low permeability geological formations are used to provide long term waste contaminant with minimum impact to the environment and risk to the biosphere. A multi-barrier approach that makes use of both engineered and natural barriers (i.e. geological formations) is often used to further enhance the containment performance of the repository. As the deep repository system subjects to a variety of thermo-hydro-chemo-mechanical (THCM) effects over its long 'operational' lifespan (e.g. 0.1 to 1.0 million years, the integrity of the barrier system will decrease over time (e.g. fracturing in rock or clay)). This is broadly referred as media degradation in the present study. This modelling study examines the effects of media degradation on diffusion dominant solute transport in fractured media that are typical of deep geological environment. In particular, reactive solute transport through fractured media is studied using a 2-D model, that considers advection and diffusion, to explore the coupled effects of kinetic and equilibrium chemical processes, while the effects of degradation is studied using a pore network model that considers the media diffusivity and network changes. Model results are presented to demonstrate the use of a 3D pore-network model, using a novel architecture, to calculate macroscopic properties of the medium such as diffusivity, subject to pore space changes as the media degrade. Results from a reactive transport model of a representative geological waste disposal package are also presented to demonstrate the effect of media property change on the solute migration behaviour, illustrating the complex interplay between kinetic biogeochemical processes and diffusion dominant transport. The initial modelling results demonstrate the feasibility of a coupled modelling approach (using pore-network model and reactive transport model) to examine the long term behaviour of deep geological repositories with media property change under complex geochemical conditions. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2013

Development of an Integrated Natural Barrier Database System for Site Evaluation of a Deep Geologic Repository in Korea - 13527

Korea Radioactive-waste Management Corporation (KRMC) established in 2009 has started a new project to collect information on long-term stability of deep geological environments on the Korean Peninsula. The information has been built up in the integrated natural barrier database system available on web (www.deepgeodisposal.kr). The database system also includes socially and economically important information, such as land use, mining area, natural conservation area, population density, and industrial complex, because some of this information is used as exclusionary criteria during the site selection process for a deep geological repository for safe and secure containment and isolation of spent nuclear fuel and other long-lived radioactive waste in Korea. Although the official site selection process has not been started yet in Korea, current integrated natural barrier database system and socio-economic database is believed that the database system will be effectively utilized to narrow down the number of sites where future investigation is most promising in the site selection process for a deep geological repository and to enhance public acceptance by providing readily-available relevant scientific information on deep geological environments in Korea. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2017

High-Resolution Characterization of Comingled Contaminants in the Deep Vadose Zone, Hanford Site, Washington - 17387

The contaminated deep vadose zone in the Central Plateau of the Hanford Site in Washington state presents a unique challenge to remediation. The Comprehensive Environmental Response, Compensation, and Liability Act, commonly known as the CERCLA project responsible for addressing these deep contaminants is designated as the 200-Deep Vadose Zone Operable Unit (DV-1 project). The vadose zone is 55-82 meters (180-270 feet) thick and is contaminated with both radiological and chemical contaminants from multiple historical waste streams. The mobile contaminants from this waste are still present in the deep vadose zone and pose a potential threat to the groundwater. Characterizing the nature and extent of the comingled mobile contaminants is an important step before assessing possible remedial alternatives. One factor controlling the migration and distribution of subsurface contaminants is the lithologic heterogeneity of the vadose zone sediments. A sampling plan for the DV-1 project includes continuous sediment core collection and analysis to identify key lithologic factors (i.e., small-scale bedding and structural features) that act as barriers, preferential pathways or conduits for mobile contaminants migrating through the vadose zone. Over six hundred meters (2,000 feet) of intact sediment core from 26 boreholes were collected in 1.5 meters (5 feet) long, 10 centimeter (4-inch) diameter segments. These intact sediment samples provide significant improvement in the level of detail and understanding of mechanisms and features controlling contaminant distribution that is not possible with traditional homogenized drill cuttings or conventional split-spoon samples. Detailed geological descriptions and photographs of the intact core are correlated to high-resolution geophysical logs which aid in the development of a comprehensive contaminant distribution profile of the borehole. These lithologic interpretations are used to define specific depth intervals for sampling based on changes in moisture, lithology, and elevated gamma radiation. The selected samples are analyzed for contaminants of concern, geochemical parameters, and hydraulic properties. Significant lithologic features observed in the intact cores include sharp contact boundaries to graded bedding, iron-oxide and carbonate cementation, silty laminations, clastic dike intrusions, and soft sediment deformation. Preliminary geologic logs integrated with the sample analysis indicate features that correlate to the variable distribution of contaminants both laterally and vertically. Integration of measured hydraulic parameters to lithologic units in individual borehole profiles provide key inputs for more precise geo-framework and vadose zone flux models supporting the DV-1 project. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2015

Discussion on the merits of deep borehole disposal, especially with recycling - 15389

With the continued global expansion of the nuclear industry, disposal of spent nuclear fuel (SNF) and high level waste (HLW) continues to be a largely unimplemented (but extensively studied) activity throughout the world and can continue to do so due to the safe practices of 'interim' SNF/used nuclear fuel (UNF) and HLW storage. However as the industry continues to operate and expand in both existing and new countries, the accumulated SNF/UNF and HLW will continue to grow in both quantity and locations across the globe; hence, requiring more effort to control and monitor. Countries with nuclear power programs, or are considering starting nuclear power programs, must consider the potential financial and the engineering burden associated with both the interim storage and the disposal of SNF/UNF and HLW produced by these programs: can the cost of disposal be reasonably captured in the cost of electricity produced from nuclear programs and are there suitable sites for disposal in their country, such as a mined-geologic site, for the expected accumulated quantity of SNF and HLW? Answering these questions may be particularly vexing for countries with small nuclear programs where establishing viable disposal sites for SNF and HLW and funding of the development of these sites may be a challenge. In an effort to establish an alternative to traditional, expensive, mined-geologic disposal sites, the deep borehole option has been (re-)considered by multiple countries and entities to be a relatively simple solution to disposal of SNF and HLW. However, disposal of SNF in deep boreholes must overcome several issues including: (1) limiting-sized borehole diameters; (2) limited ability for retrieval; and (3) potential hang-up issues associated with lowering longer (e.g., > 180 inches) canisters down into a deep borehole. Since most SNF is a resource when recycled (i.e., UNF) and to dispose of it in an irretrievable manner would be wasteful, a more optimal alternative that could overcome these issues and maximize the benefits of the UNF is to recycle the UNF and dispose of the resulting HLW into deep boreholes. The multiple and mutual benefits of this approach are elucidated in this paper. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2015

Siting considerations for a deep borehole disposal facility - 15403

This paper describes considerations for siting and designing a facility for deep borehole disposal of spent nuclear fuel (SNF) and/or high-level radioactive waste (HLW). Technical considerations include geological, hydrogeochemical, and geophysical characteristics that are related to the suitability of the site for deep drilling and borehole construction, waste emplacement activities, waste isolation (seals and surrounding geology), and long-term safety of the deep borehole disposal system. Logistical factors to be considered include: local or regional availability of drilling contractors (equipment, services, and materials) capable of drilling a large-diameter hole to approximately 5,000 m depth; construction of surface facilities and waste handling and emplacement equipment; legal and regulatory requirements associated with drilling, operation, and post-closure; and access to transportation systems. Social and political considerations include the distance from population centers and the support or opposition of local and state entities and other stakeholders to the facility and its operations. These technical, logistical, and socio-political factors are examined in the context of a consent-based siting process for a deep disposal borehole, which would be the first phase in developing an operating facility. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2019

Panel Session 62: Progress on Deep Repository Programs Around the World

This panel session focused on the progress of deep repository programs, worldwide. While efforts on a deep geological repository for SNF/UNF and HLW in the USA were terminated in 2010, several other repository programs continue to make considerable progress and deserve to be recognized. Panelists were invited from national, multi-national, and international repository programs and organizations to report upon the status and future as well as the perceived reasons underlying the success and the failure of components of their respective repository program. This panel provided a forum for interested parties to present, exchange, and take advantage of, as appropriate, repository progress, challenges, concepts/designs and other lessons learned around the world, including site characterization, site selection, research and development, licensing, construction, operation, and public acceptance and participation. Presentations proposed include: 1 - Cigeo (French Deep Geological Repository Project) update 2018 (Daniel Delort, Andra (France)); 2 - Progress on Deep Geological Repository Programs Around the World (Bruce McKirdy, Radioactive Waste Management (United Kingdom)); 3 - Geological disposal facility in Belgium (Didier Leonard, ONDRAF/NIRAS (Belgium))

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2019

Application of a Deep Learning Network to 3D Modeling of Nuclear Facilities - 19614

The ultimate goal of this study is to minimize (or eliminate) human intervention in the scanning and 3D CAD modeling process in the generation of 3D as-built model of nuclear facilities. This paper describes the application of the state-of-the-art deep learning technology in 3D point cloud data segmentation which is a part of the 3D modeling process. The baseline deep learning network chosen in this study is PointNet. We compare the deep learning approach with a conventional filtering approach, and show the feasibility and potential of deep learning technology. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2017

Applications of data mining technology to enhance O and M: automatic planning of electrical isolation with deep learning - 255

Data mining technology is now readily available, thanks to high powered, low cost computing resource and useful coding tools. We study several applications of data mining to enhance operation and maintenance (O and M), such as abnormality prediction, efficiency diagnosis, document traceability, image recognition, etc. In field support, we try automatic planning of electrical isolation with deep learning. Currently, a skilled engineer plans electrical isolation procedure with hundreds of the circuit diagrams and the related documents, taking man-hours. If this task becomes automatic, it is very efficient. The automatic planning has two issues. One is much calculation time of electrical circuit simulator, searching billions of electrical conducting paths. The other is interpretation of human fuzzy information, such as ease of work. Deep learning is helpful to resolve them. We performed a model case study of the electrical isolation. We applied a deep neural network (DNN) for dropping in the calculation time. We trained the circuit diagrams and the responses of the circuit simulator to the DNN, constructing an efficient path search algorithm in the DNN. The calculation time of the DNN was shorter by a factor of 32 for the model case, compared with that of the electrical circuit simulator. We applied a deep Q-network (DQN) for the interpretation of fuzzy information. The DQN optimized isolation procedure from the viewpoint of the work load with the equipment layout, related doses, and values of wet bulb globe temperature (WBGT) to prevent heatstroke. (authors)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

Deep reinforcement learning assisted co-optimization of Volt-VAR grid service in distribution networks

With the increasing penetration of distributed energy resources in distribution networks, Volt-VAR control and optimization (VVC/VVO) have become very important to ensure an acceptable quality of service to all customers. System operators can rely on slow-responding utility devices, including capacitor banks and on-load tap changing transformers, along with fast-responding battery and photovoltaic (PV) inverters for the VVC/VVO implementation. Because of variations in response time of these two classes of devices, and different control actions (discrete versus continuous), coordinated and optimal scheduling and operation have become of utmost importance. Here, this paper develops a look-ahead deep reinforcement learning (DRL)-based multi-objective VVO technique to improve the voltage profile of active distribution networks, decrease network and inverter power loss, and save the operational cost of the grid. It proposes a deep deterministic policy gradient (DDPG)-based approach to schedule the optimal reactive and/or active power set-points of fast-responding inverters, and a deep Q-network (DQN)-based DRL agent to schedule the discrete decisions variables of slow-responding assets. The reactive power output of PV and battery smart inverters are scheduled at 30-minute intervals and the capacitors’ commitment status is scheduled with several hour intervals. The proposed framework is validated on the modified IEEE 34-bus and 123-bus test cases with embedded PV and PV-plus-storage. To validate the efficacy of the proposed VVO, it is compared with several scenarios, including the base case without VVO, localized droop control of DERs, DDPG-only, and twin delayed DDPG (TD3) agent-based DRL techniques. The results justify the superior performance of the proposed method to improve the voltage profile, reduce network power loss, and minimize the look-ahead grid operational cost while minimizing the undesirable power losses in inverters as a result of power factor adjustments.

24 POWER TRANSMISSION AND DISTRIBUTION↗