SEARCH · Engineering Papers

Results for “Growth efficiency”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 379 records · Page 21

All sources NASA NTRS DOE OSTI

NASA NTRS1995

A bacteria antibiotic system in space (23-F ANTIBIO)

In order to evaluate the effects of weightlessness and cosmic radiations on the bacteria resistance to antibiotics, the Antibio 23F experiment was undertaken onboard Discovery during the 1st International Microgravity Laboratory (IML-1) mission. The effects of various antibiotic concentrations (dihydrostreptomycin) on Escherichia coli growth and cell division behavior were studied. The antibiotic binding was investigated using a radioactive tracer (tritium). The results showed that microgravity did not affect E. coli cells in regards the growth and the cell division. The antibiotic added to the culture medium induced an inhibition of the cultures both in the flight and ground controls. However, the antibiotic was less efficient in flight. The behavior of bacteria was modified, and the exponential growth rate was increased in flight. The incorporation of radioactive antibiotics in flight was comparatively different to ground incorporation, which indicated some perturbations in antibiotic binding. The experiments performed in the 1 g centrifuge did not show any difference in the cultures developed on the static rack, and could support a radiative effect of cosmic radiation to explain the results.

Tixador, Rene↗

DOE OSTI2022

Quantum-size-tuned heterostructures enable efficient and stable inverted perovskite solar cells

The energy landscape of reduced-dimensional perovskites (RDPs) can be tailored by adjusting their layer width (n). Recently, 2D/3D heterostructures containing n = 1 and 2 RDPs have produced PSCs with > 25% power conversion efficiency (PCE). Unfortunately, this method does not translate to inverted PSCs due to electron blocking at the 2D/3D interface. Here we report a method to increase the layer width of RDPs in 2D/3D heterostructures in order to address this problem. We discover that bulkier organics form 2D heterostructures more slowly, resulting in wider RDPs; and that small modifications to ligand design induce preferential growth of n ≥ 3 RDPs. Levering these insights, we developed efficient inverted PSCs (certified 23.91% quasi-steady state efficiency). Furthermore, unencapsulated devices operate at room temperature and ~50% relative humidity for over 1000 hrs without loss of PCE; and, when subjected to ISOS-L3 accelerated aging encapsulated devices retain 92% of PCE after 500 hrs.

14 SOLAR ENERGY↗

NASA NTRS1979

Silicon web process development

Silicon dendritic web, a single crystal ribbon shaped during growth by crystallographic forces and surface tension (rather than dies), is a highly promising base material for efficient low cost solar cells. The form of the product smooth, flexible strips 100 to 200 microns thick, conserves expensive silicon and facilitates automation of crystal growth and the subsequent manufacturing of solar cells. These characteristics, coupled with the highest demonstrated ribbon solar cell efficiency-15.5%-make silicon web a leading candidate to achieve, or better, the 1986 Low Cost Solar Array (LSA) Project cost objective of 50 cents per peak watt of photovoltaic output power. The main objective of the Web Program, technology development to significantly increase web output rate, and to show the feasibility for simultaneous melt replenishment and growth, have largely been accomplished. Recently, web output rates of 23.6 sq cm/min, nearly three times the 8 sq cm/min maximum rate of a year ago, were achieved. Webs 4 cm wide or greater were grown on a number of occassions.

Duncan, C. S.↗

DOE OSTI2025

TGCM: (T)rait, (G)ene, and (C)rop Growth (M)odel Directed Targeted Gene Characterization in Sorghum (Final Technical Report)

Understanding which genes control important crop traits could help scientists develop better bioenergy and food crops more efficiently. However, plant genomes contain tens of thousands of genes, and testing each one individually is expensive and time-consuming. This project developed computational tools to predict which genes are most likely to matter, allowing researchers to focus their efforts where they will have the greatest impact. This project developed and validated integrated approaches combining machine learning, quantitative genetics, and crop growth modeling to improve the efficiency of functional gene characterization in sorghum (Sorghum bicolor), a critical bioenergy and food security crop. The research addressed a fundamental challenge in plant biology: the majority of genes in plant genomes lack experimentally validated functions, making it difficult to prioritize which genes to study using resource-intensive reverse genetics approaches.

60 APPLIED LIFE SCIENCES↗

NASA NTRS1990

Data Management System (DMS) Evolution Analysis

The all encompassing goal for the Data Management System (DMS) Evolution Analysis task is to develop an advocacy for ensuring that growth and technology insertion issues are properly and adequately addressed during DMS requirements specification, design, and development. The most efficient methods of addressing those issues are via planned and graceful evolution, technology transparency, and system growth margins. It is necessary that provisions, such as those previously mentioned, are made to accommodate advanced missions requirements (e.g., Human Space Exploration Programs) in addition to evolving Space Station Freedom operations and user requirements .

Douglas, Katherine↗

NASA NTRS1995

Displacement based multilevel structural optimization

Multidisciplinary design optimization (MDO) is expected to play a major role in the competitive transportation industries of tomorrow, i.e., in the design of aircraft and spacecraft, of high speed trains, boats, and automobiles. All of these vehicles require maximum performance at minimum weight to keep fuel consumption low and conserve resources. Here, MDO can deliver mathematically based design tools to create systems with optimum performance subject to the constraints of disciplines such as structures, aerodynamics, controls, etc. Although some applications of MDO are beginning to surface, the key to a widespread use of this technology lies in the improvement of its efficiency. This aspect is investigated here for the MDO subset of structural optimization, i.e., for the weight minimization of a given structure under size, strength, and displacement constraints. Specifically, finite element based multilevel optimization of structures (here, statically indeterminate trusses and beams for proof of concept) is performed. In the system level optimization, the design variables are the coefficients of assumed displacement functions, and the load unbalance resulting from the solution of the stiffness equations is minimized. Constraints are placed on the deflection amplitudes and the weight of the structure. In the subsystems level optimizations, the weight of each element is minimized under the action of stress constraints, with the cross sectional dimensions as design variables. This approach is expected to prove very efficient, especially for complex structures, since the design task is broken down into a large number of small and efficiently handled subtasks, each with only a small number of variables. This partitioning will also allow for the use of parallel computing, first, by sending the system and subsystems level computations to two different processors, ultimately, by performing all subsystems level optimizations in a massively parallel manner on separate processors. It is expected that the subsystems level optimizations can be further improved through the use of controlled growth, a method which reduces an optimization to a more efficient analysis with only a slight degradation in accuracy. The efficiency of all proposed techniques is being evaluated relative to the performance of the standard single level optimization approach where the complete structure is weight minimized under the action of all given constraints by one processor and to the performance of simultaneous analysis and design which combines analysis and optimization into a single step. It is expected that the present approach can be expanded to include additional structural constraints (buckling, free and forced vibration, etc.) or other disciplines (passive and active controls, aerodynamics, etc.) for true MDO.

Striz, Alfred G.↗

DOE OSTI2024

Heterogeneous Solid Electrolyte Interphase Interactions Dictate Interface Instability in Sodium Metal Electrodes

Sodium (Na) metal batteries have attracted recent attention due to their low cost and high abundance of Na. However, the advancement of Na metal batteries is impeded due to key challenges such as dendrite growth, solid electrolyte interphase (SEI) fracture, and low Coulombic efficiency. This study examines the coupled electro‐chemo‐mechanical interactions governing the electrodeposition stability and morphological evolution at the Na/electrolyte interface. The SEI heterogeneities influence transport and reaction kinetics leading to the formation of current and stress hotspots during Na plating. Further, it is demonstrated that the heterogeneity‐induced Na metal evolution and its influence on the stress distribution critically affect the mechanical overpotential, contributing to a faster SEI failure. The analysis reveals three distinct failure mechanisms—mechanical, transport, and kinetic—that govern the onset of instabilities at the interface. Finally, a comprehensive comparative study of SEI failure in Na and lithium (Li) metal anodes illustrates that the electrochemical and mechanical characteristics of the SEI are crucial in tailoring the anode morphology and interface stability. This work delineates mechanistic stability regimes cognizant of the SEI attributes and underlying failure modes and offers important guidelines for the design of artificial SEI layers for stable Na metal electrodes.

25 ENERGY STORAGE↗

DOE OSTI2024

Efficient Anomaly Detection Driven By Different Machine Learning Architectures And Models

The rapid growth and ubiquitous adoption of the internet and cyber-physical systems (CPS) have fundamentally transformed modern communication, work, and human-system interactions. While networks now form the backbone of critical digital ecosystems, enabling seamless data transmission across diverse, interconnected systems, this increased connectivity also expands the attack surface, making real-time detection of network intrusions and anomalies a pressing challenge. Detecting unusual activities within network infrastructure requires advanced data traffic analysis to differentiate between legitimate and malicious interactions. Traditional approaches to network anomaly detectionâ??such as rule-based and signature-based systemsâ??often depend on predefined patterns to identify known anomalies, limiting their effectiveness against emerging, stealthy, or previously unseen threats. These conventional methods suffer from high false alarm rates and fail to adapt to the ever-evolving nature of network traffic, particularly in large-scale, decentralized environments where data volume, velocity, and variety are constantly increasing. This dissertation presents artificial intelligence (AI)-driven approaches to anomaly detection that leverage graphics processing unit (GPU)-enabled high-performance computing (HPC) platforms for processing massive network traffic data and monitoring the components of cyber-physical systems (CPS) for potentially hazardous conditions. The research advances several key contributions: (1) Designing efficient machine learning techniques for CPS condition monitoring and anomaly detection; (2) enabling federated learning (FL) frameworks that enable distributed detection while preserving data privacy and system resilience; (3) exploring graph-based methodologies combining graph neural networks (GNN) and graph machine learning (ML) approaches for the Internet of Things (IoT) and automotive network security, and (4) performing distributed edge computing optimizations that integrate FL with scalable technologies for reduced communication overhead. Through extensive experiments, these methodologies demonstrate that complex anomaly detection and condition monitoring tasks can be achieved while balancing computational efficiency and detection accuracy through fine-grained network information processing. The frameworks developed in this research establish a robust foundation for network anomaly detection, providing scalable, adaptive, and privacy-preserving solutions for safeguarding CPS and IoT networks in an increasingly interconnected digital landscape. The practical implications of these research findings are significant, as they can inform the development of next-generation network security systems and contribute to the protection of critical infrastructure against sophisticated cyber attacks.

Marfo, William↗

NASA NTRS1995

A Model for Particle Microphysics, Turbulent Mixing, and Radiative Transfer in the Stratocumulus-Topped Marine Boundary Layer and Comparisons with Measurements

A detailed 1D model of the stratocumulus-topped marine boundary layer is described. The model has three coupled components: a microphysics module that resolves the size distributions of aerosols and cloud droplets, a turbulence module that treats vertical mixing between layers, and a multiple wavelength radiative transfer module that calculates radiative heating rates and cloud optical properties. The results of a 12-h model simulation reproduce reasonably well the bulk thermodynamics, microphysical properties, and radiative fluxes measured in an approx. 500-m thick, summertime marine stratocumulus cloud layer by Nicholls. However, in this case, the model predictions of turbulent fluxes between the cloud and subcloud layers exceed the measurements. Results of model simulations are also compared to measurements of a marine stratus layer made under gale conditions and with measurements of a high, thin marine stratocumulus layer. The variations in cloud properties are generally reproduced by the model, although it underpredicts the entrainment of overlying air at cloud top under gale conditions. Sensitivities of the model results are explored. The vertical profile of cloud droplet concentration is sensitive to the lower size cutoff of the droplet size distribution due to the presence of unactivated haze particles in the lower region of the modeled cloud. Increases in total droplet concentrations do not always produce less drizzle and more cloud water in the model. The radius of the mean droplet volume does not correlate consistently with drizzle, but the effective droplet radius does. The greatest impacts on cloud properties predicted by the model are produced by halving the width of the size distribution of input condensation nuclei and by omitting the effect of cloud-top radiative cooling on the condensational growth of cloud droplets. The omission of infrared scattering produces noticeable changes in cloud properties. The collection efficiencies for droplets less than 30-micrometers radius, and the value of the accommodation coefficient for condensational droplet growth, have noticeable effects on cloud properties. The divergence of the horizontal wind also has a significant effect on a 12-h model simulation of cloud structure. Conclusions drawn from the model are tentative because of the limitations of the 1D model framework. A principal simplification is that the model assumes horizontal homogeneity, and, therefore, does not resolve updrafts and downdrafts. Likely consequences of this simplification include overprediction of the growth of droplets by condensation in the upper region of the cloud, underprediction of droplet condensational growth in the lower region of the cloud, and underprediction of peak supersaturations.

Ackerman, Andrew S.↗

NASA NTRS1995

A Model for Particle Microphysics,Turbulent Mixing, and Radiative Transfer in the Stratocumulus-Topped Marine Boundary Layer and Comparisons with Measurements

A detailed 1D model of the stratocumulus-topped marine boundary layer is described. The model has three coupled components: a microphysics module that resolves the size distributions of aerosols and cloud droplets, a turbulence module that treats vertical mixing between layers, and a multiple wavelength radiative transfer module that calculates radiative heating rates and cloud optical properties. The results of a 12-h model simulation reproduce reasonably well the bulk thermodynamics, microphysical properties, and radiative fluxes measured in an approx. 500-m thick, summertime marine stratocumulus cloud layer by Nicholls. However, in this case, the model predictions of turbulent fluxes between the cloud and subcloud layers exceed the measurements. Results of model simulations are also compared to measurements of a marine stratus layer made under gate conditions and with measurements of a high, thin marine stratocumulus layer. The variations in cloud properties are generally reproduced by the model, although it underpredicts the entrainment of overlying air at cloud top under gale conditions. Sensitivities of the model results are explored. The vertical profile of cloud droplet concentration is sensitive to the lower size cutoff of the droplet size distribution due to the presence of unactivated haze particles in the lower region of the modeled cloud. Increases in total droplet concentrations do not always produce less drizzle and more cloud water in the model. The radius of the mean droplet volume does not correlate consistently with drizzle, but the effective droplet radius does. The greatest impacts on cloud properties predicted by the model are produced by halving the width of the size distribution of input condensation nuclei and by omitting the effect of cloud-top radiative cooling on the condensational growth of cloud droplets. The omission of infrared scattering produces noticeable changes in cloud properties. The collection efficiencies for droplets less than 30-micron radius, and the value of the accommodation coefficient for condensational droplet growth, have noticeable effects on cloud properties. The divergence of the horizontal wind also has a significant effect on a 12-h model simulation of cloud structure. Conclusions drawn from the model are tentative because of the limitations of the 1D model framework. A principal simplification is that the model assumes horizontal homogeneity, and, therefore, does not resolve updrafts and downdrafts. Likely consequences of this simplification include overprediction of the growth of droplets by condensation in the upper region of the cloud, underprediction of droplet condensational growth in the lower region of the cloud, and under-prediction of peak supersaturations.

Ackerman, Andrew S.↗

DOE OSTI2020

Crystal-growth-dominated fabrication of metal-organic frameworks with orderly distributed hierarchical porosity

Constructing architectures with hierarchical porosity has been widely considered as the most efficient way to bypass the problems related to slow mass transfer and inaccessibility of internal space in MOFs. Now, a crystal-growth-dominated strategy is proposed to fabricate hierarchically porous MOFs (HP-MOFs). When the crystal growth is dominated by the monomer attachment, the aggregation of nonionic surfactant or polymer can be easily captured and released during the crystal growth process, resulting in the formation and ordering hierarchical pores along the radial direction. Furthermore, owing to the accelerated mass diffusion and more exposed active sites of this design, HP-MOFs exhibited an enhanced catalytic efficiency in styrene oxidation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS1980

High efficiency compound semiconductor concentrator photovoltaics

Special emphasis was given to the high yield pilot production of packaged AlGaAs/GaAs concentrator solar cells, using organometallic VPE for materials growth, the demonstration of a concentrator module using 12 of these cells which achieved 16.4 percent conversion efficiency at 50 C coolant inlet temperature, and the demonstration of a spectral splitting converter module that achieved in excess of 20 percent efficiency. This converter employed ten silicon and ten AlGaAs cells with a dichroic filter functioning as the beam splitter. A monolithic array of AlGaAs/GaAs solar cells is described.

Borden, P.↗

DOE OSTI2021

Novel usage of deep learning and high-performance computing in long-baseline neutrino oscillation experiments

Mención Internacional en el título de doctorDeep-learning methods are playing a crucial role in numerous scientific and industrialapplications. Over the past two decades, these techniques have helped in the collection,reconstruction, and analysis of large data samples in particle physics experiments. Themain topic of this PhD research is the study of deep-learning techniques in long-baselineneutrino oscillation experiments. Neutrinos are mysterious light elementary particles,and their investigation is essential to shed light on some of the remaining open questionsin physics. The work presented here describes an algorithm based on a convolutionalneural network developed to provide highly accurate and efficient selections of electronneutrino and muon neutrino interactions in the Deep Underground Neutrino Experiment(DUNE). With this algorithm, the electron neutrino (antineutrino) selection efficiencypeaks at 90% (94%) and exceeds 85% (90%) for reconstructed neutrino energies between2-5 GeV. The selection efficiency for muon neutrino (antineutrino) interactions is foundto have a maximum of 96% (97%) and exceeds 90% (95%) efficiency for reconstructedneutrino energies above 2 GeV. When considering all electron neutrino and antineutrinointeractions as signal (both those appearing from oscillations and those intrinsic tothe beam), a selection purity of 90% is achieved. These event selections are criticalto maximise the sensitivity of the experiment to CP-violating effects, key to furtherunderstand the matter-antimatter asymmetry of the Universe.In high-energy physics experiments, deep learning has also been explored for producingfast simulations and physically-motivated manipulations of simulated images. Some ofthose simulations, such as the light production and detection, are very computationallyexpensive and require novel methods to produce the necessary samples while controllingthe varied underlying physics model parameters. To do so, we invented the model-assistedgenerative adversarial network (MAGAN), first validated on simple generic case studiesand then successfully applied to the DUNE photon-detector simulation.Moreover, we also developed graph neural networks for 3D-voxel classification ofambiguities and optical crosstalk for a different particle physics experiment, most preciselyfor the proposed SuperFGD. This novel 3D-granular plastic-scintillator neutrino detectorwill be used to upgrade the near detector of the T2K neutrino oscillation experiment, and our method reports efficiencies and purities of 94-96% per event in the classificationof particle track voxels.Due to the growth and complexity of deep neural networks, researchers have beeninvestigating techniques to train those networks in a more computationally-efficient way.Many efforts have been made by the community to optimise deep-learning models byparallelising or distributing their training computation across multiple devices. In thisthesis, we study an approach based on data locality for those neural networks that cannotbenefit from scaling their computation due to a significant bottleneck in the data I/O.The research also includes a detailed study on the performance of deep neural networkson hardware accelerator boards.Los métodos de aprendizaje profundo son cada vez más utilizados en numerosas aplicacionescientíficas e industriales hoy en día. Durante las dos últimas décadas, estastécnicas se han empleado en la recolección, reconstrucción y análisis de la gran cantidadde datos generados por experimentos de física de partículas. El tema principal de estatesis doctoral es el uso de estos modelos de aprendizaje profundo en experimentos defísica de neutrinos, en concreto en los experimentos de larga distancia DUNE y T2K. Losneutrinos, partículas fundamentales neutras, de las más ligeras del Universo, pueden serclave para explicar algunas de las cuestiones todavía sin resolver en física fundamental.Entre las diferentes contribuciones que esta tesis ha hecho a su estudio, cabe destacar eldesarrollo de un algoritmo basado en una red de neuronas convolucional para seleccionarcon gran eficiencia y precisión las interacciones de neutrinos electrónicos y muónicos enel Deep Underground Neutrino Experiment (DUNE). La eficiencia de selección obtenidapara neutrinos (antineutrinos) electrónicos alcanza un máximo del 90% (94%) y supera el85% (90%) para neutrinos con energías reconstruidas en el rango 2-5 GeV. La selección deneutrinos (antineutrinos) muónicos tiene una eficiencia máxima del 96% (97%) y excedeel 90% (95%) para neutrinos con energías reconstruidas de más de 2 GeV. Considerandocomo señal todas las interacciones de neutrinos y antineutrinos electrónicos (procedentestanto de oscilaciones como intrínsecos en el haz inicial), se logra una pureza en la seleccióndel 90%. Dichas selecciones de eventos son fundamentales para maximizar la sensibilidaddel experimento a los efectos de violació...

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2023

Wood‐density has no effect on stomatal control of leaf‐level water use efficiency in an Amazonian forest

Forest disturbances increase the proportion of fast-growing tree species compared to slow-growing ones. To understand their relative capacity for carbon uptake and their vulnerability to climate change, and to represent those differences in Earth system models, it is necessary to characterise the physiological differences in their leaf-level control of water use efficiency and carbon assimilation. We used wood density as a proxy for the fast-slow growth spectrum and tested the assumption that trees with a low wood density (LWD) have a lower water-use efficiency than trees with a high wood density (HWD). We selected 5 LWD tree species and 5 HWD tree species growing in the same location in an Amazonian tropical forest and measured in situ steady-state gas exchange on top-of-canopy leaves with parallel sampling and measurement of leaf mass area and leaf nitrogen content. We found that LWD species invested more nitrogen in photosynthetic capacity than HWD species, had higher photosynthetic rates and higher stomatal conductance. Furthermore, contrary to expectations, we showed that the stomatal control of the balance between transpiration and carbon assimilation was similar in LWD and HWD species and that they had the same dark respiration rates.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Potential use of engineered nanoparticles in ocean fertilization for large-scale atmospheric carbon dioxide removal

Artificial ocean fertilization (AOF) aims to safely stimulate phytoplankton growth in the ocean and enhance carbon sequestration. AOF carbon sequestration efficiency appears lower than natural ocean fertilization processes due mainly to the low bioavailability of added nutrients, along with low export rates of AOF-produced biomass to the deep ocean. Here we explore the potential application of engineered nanoparticles (ENPs) to overcome these issues. Data from 123 studies show that some ENPs may enhance phytoplankton growth at concentrations below those likely to be toxic in marine ecosystems. ENPs may also increase bloom lifetime, boost phytoplankton aggregation and carbon export, and address secondary limiting factors in AOF. Life-cycle assessment and cost analyses suggest that net CO 2 capture is possible for iron, SiO 2 and Al 2 O 3 ENPs with costs of 2–5 times that of conventional AOF, whereas boosting AOF efficiency by ENPs should substantially enhance net CO 2 capture and reduce these costs. Therefore, ENP-based AOF can be an important component of the mitigation strategy to limit global warming.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2020

High Performance Green LEDs for Solid State Lighting

The development of white LEDs for solid state lighting (SSL) has been driven in recent years by phosphor converted LEDs (pc-LEDs). However, losses (known as Stokes’ losses) between the blue pump LED and phosphor impose a fundamental efficiency limit of ~300 lm/W on pc-LEDs. White light can also be generated from color mixed LEDs (cm-LEDs), which employ red, green, blue, and amber LEDs and have a fundamental efficiency limit of ~400 lm/W. Efficient group III-nitride materials are used for the blue LED, while efficient group III-phosphide materials are used for the red LED component. Currently, the poor efficiency of green and amber LEDs (i.e. the “green gap”) is the primary limitation for cm-LEDs. Relative to nitride-based blue LEDs, green and amber nitride LEDs suffer from lower radiative recombination rates and higher nonradiative recombination rates, which ultimately lead to reduced internal quantum efficiency (IQE). The IQE represents the portion of all electron-hole recombination events that result in a photon. In addition, long-wavelength LEDs have lower electrical efficiency (EE) compared to their blue counterparts. Addressing the green gap, would ultimately enable cm-LEDs that rival or exceed the performance of pc-LEDs. Our project focused on III-nitride materials growth and characterization, device fabrication and testing, and semiconductor physics to understand efficiency limitations of green LEDs and develop solutions to these challenges. Insights gained during our research has led to novel long-wavelength LED designs which will enable efficient solid-state lighting.

30 DIRECT ENERGY CONVERSION↗

DOE OSTI2026

Towards Generalizable and Efficient Circuit Topology Design: A Graph-Transformer-based Surrogate Model with Curriculum Learning

Unlike circuit parameter and sizing optimizations, the automated design of analog circuit topologies poses significant challenges for learning-based approaches. One challenge arises from the combinatorial growth of the topology space with circuit size, which limits the topology optimization efficiency. Moreover, traditional circuit evaluation methods are time-consuming, while the presence of data discontinuity in the topology space makes the accurate prediction of circuit performance exceptionally difficult for unseen topologies. To tackle these challenges, we design a novel Graph-Transformer-based Network (GTN) as the surrogate model for circuit evaluation, offering a substantial acceleration in the speed of circuit topology optimization without sacrificing performance. Our GTN model architecture is designed to embed voltage changes in circuit loops and current flows in connected devices, enabling accurate performance predictions for circuits with unseen topologies. To address the cold start problem when scaling GTN to large-scale circuits, we further introduce a curriculum learning strategy that progressively trains GTN from small-scale to large-scale circuits. This approach enables the model to first learn fundamental physical principles from simpler topologies and gradually adapt to complex configurations, effectively bridging the circuit complexity gap and improving prediction accuracy. Taking the power converter circuit design as an experimental task, our GTN model significantly outperforms an analytical approach and baseline methods directly utilizing graph neural networks. Furthermore, GTN achieves less than 5% relative error and 196× speed-up compared with high-fidelity simulation. Notably, our GTN surrogate model empowers an automatic circuit design framework to discover circuits of comparable quality to those identified through high-fidelity simulation while reducing the time required by up to 98.2%. With curriculum learning, the enhanced GTN achieves a 51% improvement for performance prediction of large-scale circuits compared to the GTN model without this strategy. These advancements establish GTN as a scalable framework for automated analog circuit design across varying circuit complexity levels.

Lu, Haoshu [New Jersey Institute of Technology (NJ↗

DOE OSTI2023

acCRISPR: an activity-correction method for improving the accuracy of CRISPR screens

Abstract High throughput CRISPR screens are revolutionizing the way scientists unravel the genetic underpinnings of engineered and evolved phenotypes. One of the critical challenges in accurately assessing screening outcomes is accounting for the variability in sgRNA cutting efficiency. Poorly active guides targeting genes essential to screening conditions obscure the growth defects that are expected from disrupting them. Here, we develop acCRISPR, an end-to-end pipeline that identifies essential genes in pooled CRISPR screens using sgRNA read counts obtained from next-generation sequencing. acCRISPR uses experimentally determined cutting efficiencies for each guide in the library to provide an activity correction to the screening outcomes via calculation of an optimization metric, thus determining the fitness effect of disrupted genes. CRISPR-Cas9 and -Cas12a screens were carried out in the non-conventional oleaginous yeast Yarrowia lipolytica and acCRISPR was used to determine a high-confidence set of essential genes for growth under glucose, a common carbon source used for the industrial production of oleochemicals. acCRISPR was also used in screens quantifying relative cellular fitness under high salt conditions to identify genes that were related to salt tolerance. Collectively, this work presents an experimental-computational framework for CRISPR-based functional genomics studies that may be expanded to other non-conventional organisms of interest.

59 BASIC BIOLOGICAL SCIENCES↗