Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Performance optimization”

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 793 records · Page 44

Characterisation of signal-induced noise in Hamamatsu R11265 Multianode Photomultiplier Tubes

Signal-induced noise is observed in Hamamatsu R11265 Multianode Photomultiplier Tubes, manifesting up to several microseconds after the single photoelectron response signal and localised in specific anodes. The mean number of noise pulses varies between devices, and shows significant dependence on the applied high-voltage. The characterisation of this noise and the mitigation strategies to perform optimal single-photon counting at 40 MHz, as required by the LHCb Ring-Imaging Cherenkov detectors, are reported.

47 OTHER INSTRUMENTATION↗

Variable Sampling Mapping

The performance of an optical system (for example, a telescope) is limited by the misalignments and manufacturing imperfections of the optical elements in the system. The impact of these misalignments and imperfections can be quantified by the phase variations imparted on light traveling through the system. Phase retrieval is a methodology for determining these variations. Phase retrieval uses images taken with the optical system and using a light source of known shape and characteristics. Unlike interferometric methods, which require an optical reference for comparison, and unlike Shack-Hartmann wavefront sensors that require special optical hardware at the optical system's exit pupil, phase retrieval is an in situ, image-based method for determining the phase variations of light at the system s exit pupil. Phase retrieval can be used both as an optical metrology tool (during fabrication of optical surfaces and assembly of optical systems) and as a sensor used in active, closed-loop control of an optical system, to optimize performance. One class of phase-retrieval algorithms is the iterative transform algorithm (ITA). ITAs estimate the phase variations by iteratively enforcing known constraints in the exit pupil and at the detector, determined from modeled or measured data. The Variable Sampling Mapping (VSM) technique is a new method for enforcing these constraints in ITAs. VSM is an open framework for addressing a wide range of issues that have previously been considered detrimental to high-accuracy phase retrieval, including undersampled images, broadband illumination, images taken at or near best focus, chromatic aberrations, jitter or vibration of the optical system or detector, and dead or noisy detector pixels. The VSM is a model-to-data mapping procedure. In VSM, fully sampled electric fields at multiple wavelengths are modeled inside the phase-retrieval algorithm, and then these fields are mapped to intensities on the light detector, using the properties of the detector and optical system, for comparison with measured data. Ultimately, this model-to-data mapping procedure enables a more robust and accurate way of incorporating the exit-pupil and image detector constraints, which are fundamental to the general class of ITA phase retrieval algorithms.

Smith, Jeffrey, S.↗

Development of Advanced, Radiation Resistant, Optical-based Detector Technology for Future Experiments.

The primary objective of this project has been to advance the design of high-performance electromagnetic (EM) calorimeters for future particle physics experiments, to identify and measure the timing, position and energy of electrons, positrons and gamma rays, particularly in high-luminosity environments with intense radiation and pileup conditions. To meet such challenges, the proposed research has focused on the development of ultra-compact, radiation-hard calorimeter modules, to provide excellent timing, spatial, and energy resolution. The work aligns with the DOE’s Basic Research Needs (BRN) for High Energy Physics (HEP) Instrumentation and the research team contributes actively to the Coordinating Panel on Advanced Detectors (CPAD) RDC9 calorimetry collaboration in the USA and the European Committee on Future Accelerators (ECFA) DRD-CALO calorimetry collaboration at CERN, the European Laboratory for Particle Physics located in Geneva, Switzerland. The research builds on the RADiCAL (radiation-hard, ultra-compact) modular sampling calorimeter approach, developed by the research team, which employs dense and very bright optical materials such as LYSO:Ce scintillator plates that are interleaved with very dense tungsten plates to minimize detector size while optimizing performance. The modules are comparable in size to a human index finger, dimensionally 14 mm x 14 mm in cross section and 135 mm in length. And despite the small size, the structure is capable of providing excellent timing and energy resolution. This is facilitated through the use of specialized quartz capillaries filled with wavelength-shifting filaments, positioned at various depths along the length of a module, to collect and guide light signals to silicon photomultipliers (SiPMs) which detect and convert the optical signals to electronic signals for analysis. The primary goals of this project have been: (1) Achieve a timing resolution to σ t ≤ 30 ps for high-energy electrons and photons, important for their association with specific events produced in colliding-beam experiments and for the detection of decays-in-flight of long-lived particles. The project has achieved this goal in beam tests of a single RADiCAL module at CERN, during which a timing resolution of σ t = 27 ps was measured for electrons of energy E = 150 GeV. Based upon a mathematical fit to the data measured over a broad energy range from low energy to high energy, a resolution of σ t ≤ 18 ps has been estimated for electrons of very high (TeV) energy. From these measurements and with further expected technical improvements, the timing resolution should reach σ t ≤ 10 ps, important for searches for discovery physics in upcoming and future experiments. (2) Achieve an energy resolution of σ E / E ≤ 10% / $\sqrt{E}$. The project has yet to achieve this goal, but is close to it, having measured a value of σ E / E ≤ 15.9% / $\sqrt{E}$ using a modular array. Ultimately, the resolution goal is expected to be reached by adjustments to material thicknesses within the modules, which will improve the sampling fraction to measure more precisely the shower energy for lower energy particles. The versatility of the modular RADiCAL approach enables the testing of advanced materials, photosensors and electronics, developed in collaboration with CPAD RDC and ECFA DRD-CALO groups. The structure can distinguish electrons, positrons and gamma rays from hadrons and muons and beam-induced backgrounds, making it a valuable tool in a variety of detector environments, including future circular colliders (FCC-ee, FCC-hh) proposed for the European Laboratory for Particle Physics (CERN), the muon-collider proposed for Fermi National Accelerator Laboratory (Fermilab), and searches for new physics in beam-dump, fixed target and forward-physics experiments. And, while designed with particle physics applications in mind, the technologies developed in this project have the potential for application more broadly in particle and nuclear physics, materials science, and medical physics, underscoring the far-reaching potential of this line of instrumentation research and development.

47 OTHER INSTRUMENTATION↗

Innovating the next generation of commercial smart building software

Nearly 30% of commercial building energy use is wasted due to equipment faults and HVAC controls problems. The result is increased emissions, compromised comfort and productivity, and less reliable coordination of building power needs with a clean grid. The energy impact alone represents $17 billion in potential savings. Today’s smart building software provides a robust solution to address these operational deficiencies. Energy management and information systems (EMIS) are saving up to 9% on average, with two-year paybacks. They are being incorporated into energy management processes, commissioning services, and utility programs. As effective as they are, two barriers prevent even deeper benefits; limited personnel to fix problems once they are identified, and the expense and time to manually implement changes in control systems. In partnership with the research community, the EMIS industry is developing new capabilities to overcome these barriers. Moving beyond siloed products for either fault detection and diagnostics, or optimal control, these new capabilities empower users to not only automatically identify faults, but also to push corrective action, and control improvements to their buildings. In this paper, several areas for enhancements are documented: ‘one-time’ correction of faults such as setpoints, schedules, and economizer lockouts; short-term active testing for automated proportional integral derivative (PID) loop tuning and functional testing; and continuous supervisory control for demand flexibility and year-round efficiency. Results are presented from a pair of partner implementations out of a dozen providers integrating these enhancements into their products, including field tests from across the country, and insights into operator acceptance and integration into operations and maintenance practices.

Casillas, Armando↗

Protecting and Defending against Autonomous Control Systems and Digital Twin Cyber Attacks: Response Strategy for Hyperparameter attacks of Digital Twin Machine Learning Models in Nuclear Power Plants (Final)

Navigating through the complex tapestry of technological advancements, "Response Strategy for Hyperparameter attacks of Digital Twin Machine Learning Model in Nuclear Power Plants" stands at the intersection of cybersecurity and nuclear power plant operations, embarking on a journey through the intricacies of securing digital twins against malicious cyber activities. As nuclear power plants progressively integrate digital twin technology and machine learning models to optimize operations and ensure system reliability, they inadvertently expose themselves to a new spectrum of vulnerabilities, notably in the realm of hyperparameter attacks. Hyperparameters, integral in machine learning model tuning and optimal performance of digital twins, have emerged as a target for adversaries aiming to destabilize the predictive capabilities and therefore, the operational accuracy of these digital entities within critical infrastructures like nuclear plants. This paper, therefore, meticulously threads the needle through the development of a robust response strategy, poised to shield these digital reflections against calculated hyperparameter manipulations, ensuring that the digital twin can effectively and securely function as a reliable proxy for its physical counterpart. The ensuing sections delve into the orchestrated maelstrom of multi-rate time-changing intelligent coordinated hyperparameter attacks and the implementation of event-triggered predictive control, laying down a structured, predictive, and responsive framework that safeguards the nexus where the digital and physical realms of nuclear power plants coalesce. The operational integrity of digital twins in nuclear power plants depends critically on the security of machine learning hyperparameters. This study makes two different contributions. First, a decision-based idea known as a multi-rate time changing intelligent coordinated hyperparameter attack is put forth. In this attack, many hyperparameters are repeatedly changed using both random and intelligent optimal techniques by the attacker. These assaults introduce varied rates at different attack steps, compromise various amounts of hyperparameters, and improve stealth and flexibility. Second, a technique is developed for event triggered predictive control to rapidly respond to potential hyperparameter attacks. This control integrates a sliding window framework, retaining a history of previous data points and employing linear regression to predict the next data point from the current dataset. The control gain K is determined using the Lyapunov-Krasovskii method, and subsequently, an action is developed. Finally, the outcome of the simulation demonstrates the viability of the proposed method for defending nuclear power plant digital twins from hyperparameter attacks.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Current Status of High Performance Large Format Si:As IBC Detectors for SOFIA-like Applications

Airborne and space telescope astronomical observations in the 5-25 micrometer wavelength region, are critical for understanding the physical conditions, composition, chemistry, and excitation of many environments in the interstellar medium, external galaxies, solar system objects, extra-solar systems, and stars. The scientific impact is particularly unique in the 5-8 micrometer and 14-25 micrometer regions which are inaccessible or poorly observed from ground-based observatories. Large-format area mid-infrared (MIR) detectors sensitive over these wavelengths and operable under moderate backgrounds (greater than 10(circumflex)6 e-/s at R=2000, 10(micrometers)) are essential for efficient large-area survey imaging and for taking moderate resolution spectra over a large spectral range. Both SOFIA and passively cooled Explorer observatories could benefit from this technology. Current first-light SOFIA instruments use small-format MIR FPA's of sizes 128x128 and 256x256, With the collaboration of Raytheon Infrared Operations (RIO), NASA-ARC has been testing the first device of this kind, a combination of 1) the ALADDIN III readout multiplexer, cryo-processed for 6K operation, with 2) SIRTF IRAC-like Sides IBC detector material. designed for higher QE and low dark current. Here we report requirements. design, and initial fabrication and testing of the first 1024x1024 Si:As IBC mid-IR detector array for these conditions. This device has exhibited low dark current, moderate noise levels, and greater than 200,000 electron linear well size at 6K operation. We conclude with suggestions for future device development for optimal performance under moderate background, SOFIA- and low Earth orbit observing conditions.

Ennico, Kimberly↗

Efficient exascale discretizations: High-order finite element methods

Efficient exploitation of exascale architectures requires rethinking of the numerical algorithms used in many large-scale applications. These architectures favor algorithms that expose ultra fine-grain parallelism and maximize the ratio of floating point operations to energy intensive data movement. One of the few viable approaches to achieve high efficiency in the area of PDE discretizations on unstructured grids is to use matrix-free/partially assembled high-order finite element methods, since these methods can increase the accuracy and/or lower the computational time due to reduced data motion. In this paper we provide an overview of the research and development activities in the Center for Efficient Exascale Discretizations (CEED), a co-design center in the Exascale Computing Project that is focused on the development of next-generation discretization software and algorithms to enable a wide range of finite element applications to run efficiently on future hardware. CEED is a research partnership involving more than 30 computational scientists from two US national labs and five universities, including members of the Nek5000, MFEM, MAGMA and PETSc projects. We discuss the CEED co-design activities based on targeted benchmarks, miniapps and discretization libraries and our work on performance optimizations for large-scale GPU architectures. We also provide a broad overview of research and development activities in areas such as unstructured adaptive mesh refinement algorithms, matrix-free linear solvers, high-order data visualization, and list examples of collaborations with several ECP and external applications.

97 MATHEMATICS AND COMPUTING↗

Prevention efficacy of the broadly neutralizing antibody VRC01 depends on HIV-1 envelope sequence features

In the Antibody Mediated Prevention (AMP) trials (HVTN 704/HPTN 085 and HVTN 703/HPTN 081), prevention efficacy (PE) of the monoclonal broadly neutralizing antibody (bnAb) VRC01 (vs. placebo) against HIV-1 acquisition diagnosis varied according to the HIV-1 Envelope (Env) neutralization sensitivity to VRC01, as measured by 80% inhibitory concentration (IC80). Here, we performed a genotypic sieve analysis, a complementary approach to gaining insight into correlates of protection that assesses how PE varies with HIV-1 sequence features. We analyzed HIV-1 Env amino acid (AA) sequences from the earliest available HIV-1 RNA-positive plasma samples from AMP participants diagnosed with HIV-1 and identified Env sequence features that associated with PE. The strongest Env AA sequence correlate in both trials was VRC01 epitope distance that quantifies the divergence of the VRC01 epitope in an acquired HIV-1 isolate from the VRC01 epitope of reference HIV-1 strains that were most sensitive to VRC01-mediated neutralization. In HVTN 704/HPTN 085, the Env sequence-based predicted probability that VRC01 IC80 against the acquired isolate exceeded 1 µg/mL also significantly associated with PE. In HVTN 703/HPTN 081, a physicochemical-weighted Hamming distance across 50 VRC01 binding-associated Env AA positions of the acquired isolate from the most VRC01-sensitive HIV-1 strain significantly associated with PE. These results suggest that incorporating mutation scoring by BLOSUM62 and weighting by the strength of interactions at AA positions in the epitope:VRC01 interface can optimize performance of an Env sequence-based biomarker of VRC01 prevention efficacy. Future work could determine whether these results extend to other bnAbs and bnAb combinations.

60 APPLIED LIFE SCIENCES↗

Unifying thermochemistry concepts in computational heterogeneous catalysis

Thermophysical properties of adsorbates and gas-phase species define the free energy landscape of heterogeneously catalyzed processes and are pivotal for an atomistic understanding of the catalyst performance. These thermophysical properties, such as the free energy or the enthalpy, are typically derived from density functional theory (DFT) calculations. Enthalpies are species-interdependent properties that are only meaningful when referenced to other species. The widespread use of DFT has led to a proliferation of new energetic data in the literature and databases. However, there is a lack of consistency in how DFT data is referenced and how the associated enthalpies or free energies are stored and reported, leading to challenges in reproducing or utilizing the results of prior work. Additionally, DFT suffers from exchange–correlation errors that often require corrections to align the data with other global thermochemical networks, which are not always clearly documented or explained. In this review, we introduce a set of consistent terminology and definitions, review existing approaches, and unify the techniques using the framework of linear algebra. This set of terminology and tools facilitates the correction and alignment of energies between different data formats and sources, promoting the sharing and reuse of ab initio data. Standardization of thermochemistry concepts in computational heterogeneous catalysis reduces computational cost and enhances fundamental understanding of catalytic processes, which will accelerate the computational design of optimally performing catalysts.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

TU-EF-304-03: 4D Monte Carlo Robustness Test for Proton Therapy

Purpose: Breathing motion and approximate dose calculation engines may increase proton range uncertainties. We address these two issues using a comprehensive 4D robustness evaluation tool based on an efficient Monte Carlo (MC) engine, which can simulate breathing with no significant increase in computation time. Methods: To assess the robustness of the treatment plan, multiple scenarios of uncertainties are simulated, taking into account the systematic and random setup errors, range uncertainties, and organ motion. Our fast MC dose engine, called MCsquare, implements optimized models on a massively-parallel computation architecture and allows us to accurately simulate a scenario in less than one minute. The deviations of the uncertainty scenarios are then reported on a DVH-band and compared to the nominal plan.The robustness evaluation tool is illustrated in a lung case by comparing three 60Gy treatment plans. First, a plan is optimized on a PTV obtained by extending the CTV with an 8mm margin, in order to take into account systematic geometrical uncertainties, like in our current practice in radiotherapy. No specific strategy is employed to correct for tumor and organ motions. The second plan involves a PTV generated from the ITV, which encompasses the tumor volume in all breathing phases. The last plan results from robust optimization performed on the ITV, with robustness parameters of 3% for tissue density and 8 mm for positioning errors. Results: The robustness test revealed that the first two plans could not properly cover the target in the presence of uncertainties. CTV-coverage (D95) in the three plans ranged respectively between 39.4–55.5Gy, 50.2–57.5Gy, and 55.1–58.6Gy. Conclusion: A realistic robustness verification tool based on a fast MC dose engine has been developed. This test is essential to assess the quality of proton therapy plan and very useful to study various planning strategies for mobile tumors. This work is partly funded by IBA (Louvain-la-Neuve, Belgium)

60 APPLIED LIFE SCIENCES↗

Wave Prediction using X-Band Radar (Final Summary Report)

Model Predictive Control (MPC) is the best controls framework available to wave energy today. It allows for the maximization of energy yield from a WEC system while respecting various constraints and losses in the system. This type of constrained optimization is a critical capability for the techno-economic optimization of WEC devices and is difficult to do effectively with causal controls frameworks. MPC requires a future prediction of wave excitation forces, requiring wave prediction on a future time horizon of up to 30 seconds. The future horizon required is highly dependent on the WEC and PTO type available. To our knowledge no-one has been able to implement MPC on a WEC device at sea, due to the fact that phase-resolved wave-prediction is not a capability that has been sufficiently developed to date. The key objectives of this project are focused on developing “industry-ready” wave prediction technology building blocks that can be applied to a wide range of different wave energy conversion topologies. Our collaborative work with device developers has demonstrated that the level of improvement attainable for any particular device is heavily dependent on the device and PTO topology chosen. The range of annual average power capture improvements is on the order of 10% to over 100%. However, the important aspect is that we can get within about 10% of theoretical upper limits even when considering errors in the wave prediction and more importantly, we can optimize performance while respecting various constraints such as motion amplitude or peak structural loads. The combination of improved performance while limiting structural loads has a net effect of reducing the levelized cost of energy from these emerging technologies and allows us to design control laws that meet an economic optimum. Efforts under a previous project on controls has focused on developing a set of wave-prediction algorithms that work by leveraging a network of wave sensing measurement buoys that are equipped with real-time telemetry. The present work is focusing primarily on using X-band radar to measure the wave-field around the WEC to predict the future excitation forces on the structure and combining the measurements of these different sources to create improvements in the wave-prediction accuracy.

16 TIDAL AND WAVE POWER↗

Green’s functions on a renormalized lattice: An improved method for the integer quantum Hall transition

Highlights: • Speed up of numerical calculations by exact renormalization group steps. • Practical approach to handle delicate scaling analysis. • Absence of evidence to prefer marginal corrections over power-law corrections. We introduce a performance-optimized method to simulate localization problems on bipartite tight-binding lattices. It combines an exact renormalization group step to reduce the sparseness of the original problem with the recursive Green’s function method. We apply this framework to investigate the critical behavior of the integer quantum Hall transition of a tight-binding Hamiltonian defined on a simple square lattice. In addition, we employ an improved scaling analysis that includes two irrelevant exponents to characterize the shift of the critical energy as well as the corrections to the dimensionless Lyapunov exponent. We compare our findings with the results of a conventional implementation of the recursive Green’s function method, and we put them into broader perspective in view of recent development in this field.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Vacancy-Dependent Diffusion Mechanism in Oxygen-Defective SrFeO 3 Perovskite Materials: First-Principles Density Functional Theory and Experimental Approach

Understanding oxygen diffusion at the atomic scale in SrFeO 3−δ perovskites is crucial for developing oxygen storage materials with optimal performance. Such materials are required to have high stability, corrosion resistance, and acceptable oxygen storage capacity at moderate operating temperatures and pressures. Here, in this study, we used first-principles density functional theory and thermogravimetric analysis to study the vacancy-dependent oxygen diffusion in oxygen-deficient SrFeO 3−δ (δ = 0, 0.065, 0.125, 0.25, 0.5) perovskites. The electronic structures, including the partial- and spin-resolved density of states, for different SrFeO 3−δ phases were calculated and compared with available experimental and theoretical results. By mapping the migration pathways, we investigated diffusion mechanisms and calculated the energy barriers for oxygen diffusion in cubic, orthorhombic, and brownmillerite phases of SrFeO 3−δ perovskites. Using the calculated energy barriers, we deduced the diffusion time scales and diffusion coefficients within SrFeO 3−δ . A diffusion coefficient on the order of 10 –8 m 2 /s was obtained for SrFeO 2.875 . We experimentally investigated the roles of temperature and oxygen partial pressures on the redox kinetics and deduced the kinetics rate and diffusion density, which agreed well with the calculated values for the density of diffusing oxygen vacancy in the lattice. Our results showed that the energy barrier tends to reduce at higher oxygen concentrations. Our results serve as an important guideline for designing oxygen storage materials with optimal redox kinetics.

chemical looping with oxygen uncoupling (CLOU)↗

Dynamic PET Measures of Tau Accumulation in Cognitively Normal Older Adults and Alzheimer’s Disease Patients Measured Using [18F] THK-5351

Background [ 18F]THK5351, a recently-developed positron emission tomography (PET) tracer for measuring tau neurofibrillary tangle accumulation, may help researchers examine aging, disease, and tau pathology in living human brains. We examined THK5351 tracer pharmacokinetics to define an optimal acquisition time for static late images. Methods Primary measurements were calculation of regional values of distribution volume ratios (DVR) and standardized uptake value ratios (SUVR) in 6 healthy older control and 10 Alzheimer’s disease (AD) participants. We examined associations between DVR and SUVR, searching for a 20 min SUVR time window that was stable and comparable to DVR. We additionally examined diagnostic group differences in this 20 min SUVR. Results In healthy controls, [18F]THK5351 uptake was low, with increased temporal relative to frontal binding. In AD, regional uptake was substantially higher than in healthy controls, with temporal exceeding frontal binding. Retention in cerebellar gray matter, which was used as the reference region, was low compared to other regions. Both DVR and SUVR values showed minimal change over time after 40 min. SUVR 20–40, 30–50, and 40–60 min were most consistently correlated with DVR; SUVR 40–60 min, the most stable time window, was used in further analyses. Significant (AD > healthy control) group differences existed in temporoparietal regions, with marginal medial temporal differences. We found high basal ganglia SUVR 40–60 min signal, with no group differences. Conclusions We examined THK5351, a new PET tracer for measuring tau accumulation, and compared multiple analysis methods for quantifying regional tracer uptake. SUVR 40–60 min performed optimally when examining 20 min SUVR windows, and appears to be a practical method for quantifying relative regional tracer retention. The results of this study offer clinical potential, given the usefulness of THK5351-PET as a biomarker of tau pathology in aging and disease.

60 APPLIED LIFE SCIENCES↗

TU-FG-201-01: 18-Month Clinical Experience of a Linac Daily Quality Assurance (QA) Solution Using Only EPID and OBI

Purpose: To describe the clinical use of a Linear Accelerator (Linac) DailyQA system with only EPID and OBI. To assess the reliability over an 18-month period and improve the robustness of this system based on QA failure analysis. Methods: A DailyQA solution utilizing an in-house designed phantom, combined EPID and OBI image acquisitions, and a web-based data analysis and reporting system was commissioned and used in our clinic to measure geometric, dosimetry and imaging components of a Varian Truebeam Linac. During an 18-month period (335 working days), the Daily QA results, including the output constancy, beam flatness and symmetry, uniformity, TPR20/10, MV and KV imaging quality, were collected and analyzed. For output constancy measurement, an independent monthly QA system with an ionization chamber (IC) and annual/incidental TG51 measurements with ADCL IC were performed and cross-compared to Daily QA system. Thorough analyses were performed on the recorded QA failures to evaluate the machine performance, optimize the data analysis algorithm, adjust the tolerance setting and improve the training procedure to prevent future failures. Results: A clinical workflow including beam delivery, data analysis, QA report generation and physics approval was established and optimized to suit daily clinical operation. The output tests over the 335 working day period cross-correlated with the monthly QA system within 1.3% and TG51 results within 1%. QA passed with one attempt on 236 days out of 335 days. Based on the QA failures analysis, the Gamma criteria is revised from (1%, 1mm) to (2%, 1mm) considering both QA accuracy and efficiency. Data analysis algorithm is improved to handle multiple entries for a repeating test. Conclusion: We described our 18-month clinical experience on a novel DailyQA system using only EPID and OBI. The long term data presented demonstrated the system is suitable and reliable for Linac daily QA.

60 APPLIED LIFE SCIENCES↗

Aircraft design optimization with multidisciplinary performance criteria

The method described here for aircraft design optimization with dynamic response considerations provides an inexpensive means of integrating dynamics into aircraft preliminary design. By defining a dynamic performance index that can be added to a conventional objective function, a designer can investigate the trade-off between performance and handling (as measured by the vehicle's unforced response). The procedure is formulated to permit the use of control system gains as design variables, but does not require full-state feedback. The examples discussed here show how such an approach can lead to significant improvements in the design as compared with the more common sequential design of system and control law.

Morris, Stephen↗

KEBLM: Knowledge-Enhanced Biomedical Language Models

Pretrained language models (PLMs) have demonstrated strong performance on many natural language processing (NLP) tasks. Despite their great success, these PLMs are typically pretrained only on unstructured free texts without leveraging existing structured knowledge bases that are readily available for many domains, especially scientific domains. As a result, these PLMs may not achieve satisfactory performance on knowledge-intensive tasks such as biomedical NLP. Comprehending a complex biomedical document without domain-specific knowledge is challenging, even for humans. Inspired by this observation, we propose a general framework for incorporating various types of domain knowledge from multiple sources into biomedical PLMs. We encode domain knowledge using lightweight adapter modules, bottleneck feed-forward networks that are inserted into different locations of a backbone PLM. For each knowledge source of interest, we pretrain an adapter module to capture the knowledge in a self-supervised way. We design a wide range of self-supervised objectives to accommodate diverse types of knowledge, ranging from entity relations to description sentences. Once a set of pretrained adapters is available, we employ fusion layers to combine the knowledge encoded within these adapters for downstream tasks. Each fusion layer is a parameterized mixer of the available trained adapters that can identify and activate the most useful adapters for a given input. Our method diverges from prior work by including a knowledge consolidation phase, during which we teach the fusion layers to effectively combine knowledge from both the original PLM and newly-acquired external knowledge using a large collection of unannotated texts. After the consolidation phase, the complete knowledge-enhanced model can be fine-tuned for any downstream task of interest to achieve optimal performance. Extensive experiments on many biomedical NLP datasets show that our proposed framework consistently improves the performance of the underlying PLMs on various downstream tasks such as natural language inference, question answering, and entity linking. These results demonstrate the benefits of using multiple sources of external knowledge to enhance PLMs and the effectiveness of the framework for incorporating knowledge into PLMs. Finally, while primarily focused on the biomedical domain in this work, our framework is highly adaptable and can be easily applied to other domains, such as the bioenergy sector.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Atomic-scale Imaging of PGM-free Catalyst Active Sites by 30 keV 4D-STEM

Platinum group metal (PGM)-free catalysts have attracted a large amount of attention due to their potential for enabling low-cost, commercially viable hydrogen fuel cells and electrolyzers [1]. Degradation of cell performance remains a significant challenge, however, currently limiting implementation of devices that utilize PGM-free materials [2]. Controlling degradation in these materials requires a better understanding of the atomic structure of active sites, thought to be FeN4 structures within a graphitic carbon lattice, which would enable more accurate prediction of the associated degradation pathways. The exact structural arrangement of active sites in these materials is still debated since a range of structures have been computationally predicted and methods to directly validate these models are still needed [1,3-5].While scanning transmission electron microscopy (STEM) has provided initial glimpses into the nature of the proposed active sites, detailed evaluation of the structure of these sites is challenging since they involve defects, edges, and nitrogen dopants in the graphitic lattice, which are susceptible to beam damage [6,7]. Conventional STEM imaging and spectroscopy methods exacerbate this problem with dose-inefficiency or nonideal contrast characteristics for imaging light and heavy elements simultaneously. A method that minimizes damage while generating dose-efficient, easily-interpretable contrast for both light and heavy elements is therefore needed to facilitate imaging of sensitive active site atomic structure.Here, we demonstrate direct atomic-scale imaging of PGM-free catalyst active sites by low-voltage four-dimensional (4D)-STEM. To accomplish this, we pair a 30 keV aberration-corrected probe, which minimizes knock-on damage, with a fast pixelated detector that has optimal performance at low beam energies [8]. We show how this setup enables relatively simple center-of-mass (CoM) techniques [9,10] to produce images with an increased signal-to-noise ratio (SNR) and light-element contrast over conventional imaging modes, allowing the entire active site structure to be imaged at the atomic scale. Moreover, we show how electron ptychography [11,12] enables images with further improved characteristics to be obtained, for example by minimizing residual aberrations, which provides a more accurate structural representation. The increased understanding that these low-voltage 4D-STEM techniques will provide about the atomic structure of PGM-free active sites and their associated degradation pathways will facilitate rational design of next-generation materials, promoting development of low-cost hydrogen fuel cells, electrolyzers, and other energy conversion devices [13]. References:[1] U Martinez et al., Adv. Mater. 31 (2019), p. 1806545.[2] Y Shao et al., Adv. Mater. 31 (2019), p. 1807615.[3] J Kneebone et al., J. Phys. Chem. C 121, (2017), p.16283.[4] T Mineva et al., ACS Catal. 9 (2019), p. 9359.[5] J Li et al., Nat. Catal. 4 (2021), p. 10.[6] T Susi et al., ACS Nano 6 (2012), p. 8837.[7] P Zelenay and DJ Myers, US Department of Energy Hydrogen and Fuel Cells Program 2017 Annual Merit Review and Peer Evaluation Meeting, Washington, DC (2017).[8] H Ryll et al., J. Inst. 11 (2016), p. P04006.[9] K Müller et al., Nat. Commun. 5 (2014), p. 5653.[10] I Lazic et al., Ultramicroscopy 160 (2016), p. 265.[11] H Yang et al., Nat. Commun. 7 (2016), p. 12532.[12] Y Jiang et al., Nature 559 (2018), p. 343. [13] Research sponsored by the Hydrogen and Fuel Cell Technologies Office, Office of Energy Efficiency and Renewable Energy, US Department of Energy (DOE). Research was conducted at the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility.

Zachman, Michael↗