Search NASA⌕ Search

SEARCH · Search NASA

Results for “variational 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 1,369 records · Page 76

Influence of silicon infiltration conditions on microstructure and mechanical properties in binder jet 3D printed SiC

Atmosphere, temperature, and hold time were varied for silicon infiltrated binder jet 3D printed SiC preforms. A two-step infiltration process optimized infiltration. The first step at 1450 °C under vacuum facilitated improved wetting. The second step at elevated temperature in argon advanced infiltration while suppressing vaporization. Variation of the second-step temperature (1550–1750 °C) and hold time revealed that 1550 °C for 1 h in argon following initial vacuum treatment yielded the highest silicon uptake ratio (∼96 %) with minimal residual porosity and strengths of 150–180 MPa. Flexural strength was either stable or improved with elevated temperatures up to 1000°C. In contrast, higher temperatures or extended hold times led to increased silicon vaporization and pores. The optimized process was applied to formulations with extra carbon to enhance SiC content and interconnectedness through reaction bonding, which corresponded to increased flexural strength. These findings provide guidance for silicon infiltration in additively manufactured SiC.

Yoon, Bola [ORNL] (ORCID:0000000260875373)↗

Analysis on burnup step effect for evaluating reactor criticality and fuel breeding ratio

Criticality condition of the reactors is one of the important factors for evaluating reactor operation and nuclear fuel breeding ratio is another factor to show nuclear fuel sustainability. This study analyzes the effect of burnup steps and cycle operation step for evaluating the criticality condition of the reactor as well as the performance of nuclear fuel breeding or breeding ratio (BR). Burnup step is performed based on a day step analysis which is varied from 10 days up to 800 days and for cycle operation from 1 cycle up to 8 cycles reactor operations. In addition, calculation efficiency based on the variation of computer processors to run the analysis in term of time (time efficiency in the calculation) have been also investigated. Optimization method for reactor design analysis which is used a large fast breeder reactor type as a reference case was performed by adopting an established reactor design code of JOINT-FR. The results show a criticality condition becomes higher for smaller burnup step (day) and for breeding ratio becomes less for smaller burnup step (day). Some nuclides contribute to make better criticality when smaller burnup step due to individul nuclide half-live. Calculation time for different burnup step shows a correlation with the time consuming requirement for more details step calculation, although the consuming time is not directly equivalent with the how many time the burnup time step is divided.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Hall thruster with grooved walls

Axial-oriented and azimuthal-distributed grooves are formed on channel walls of a Hall thruster after the engine undergoes a long-term operation. Existing studies have demonstrated the relation between the grooves and the near-wall physics, such as sheath and electron near-wall transport. The idea to optimize the thruster performance with such grooves was also proposed. Therefore, this paper is devoted to explore the effects of wall grooves on the discharge characteristics of a Hall thruster. With experimental measurements, the variations on electron conductivity, ionization distribution, and integrated performance are obtained. The involved physical mechanisms are then analyzed and discussed. The findings help to not only better understand the working principle of Hall thruster discharge but also establish a physical fundamental for the subsequent optimization with artificial grooves.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Prompt Phrase Ordering Using Large Language Models in HPC: Evaluating Prompt Sensitivity

Large language models (LLMs) have demonstrated effective performance in domain-specific tasks, often requiring a well-designed prompt to guide their responses. However, optimizing the right prompt is challenging due to prompt sensitivity—the phenomenon where small changes in the prompt can lead to significant variations in performance. In this study, we evaluate prompt performance by examining all permutations of independent phrases to investigate prompt sensitivity and robustness. We used two datasets: the GSM8k dataset, which assesses mathematical reasoning, and a custom template prompt for summarizing database metadata. Our goal was to evaluate the performance across all permutations of a sequence of prompt phrases. The study was conducted using the llama3-instruct- 7B model hosted on Ollama, with computations parallelized in a high-performance computing environment. By comparing the average index of phrases in the best and worst-performing prompts, we found that the order of independent phrases within a prompt significantly impacts LLM performance. Additionally, we used Hamming distance to assess changes between phrase orderings, concluding that prompt modifications can dramatically affect scores, often by almost random chance. These findings support existing research on prompt sensitivity. We discuss the challenges of prompt optimization, noting that altering phrases in a successful prompt does not always result in another successful prompt.

97 MATHEMATICS AND COMPUTING↗

Formation of Plastic Inclusions in U71Mnk High-Speed Heavy-Rail Steel Refined by CaO-SiO{sub 2}-Al{sub 2}O{sub 3}-MgO Slag

To precisely control the characteristics of nonmetallic inclusions with high plasticity in U71Mnk high-speed heavy-rail steel deoxidized by Si and Mn, the chemical composition of CaO-SiO{sub 2}-Al{sub 2}O{sub 3}-MgO slag in the ladle furnace (LF) refining process was optimized by thermodynamic calculations. Relationships among the refining slag, nonmetallic inclusions, and elemental concentrations in the steels were analyzed to obtain the conditions for determining the optimal composition of refining slag. The optimal compositions and the component activities of CaO-SiO{sub 2}-Al{sub 2}O{sub 3}-MgO refining slag were determined and discussed according to calculations using FactSage 7.0. Finally, ten industrial experiments were performed. Variations in the compositions of LF refining slag, inclusions, and steel, as well as the sizes of the inclusions, and metallographic classification assessment of the U71Mnk steel product, were investigated and discussed to verify the feasibility and effectiveness of the optimized LF refining slag. Kinetics of steel–slag and steel-inclusion reactions during LF refining was studied and compared. The results show that to obtain low [O], [Al], and [S] contents and inclusions with high plasticity in molten steel, the optimal composition of the LF refining slag should be CaO: 44.2 to 51.7 wt pct, SiO{sub 2}: 41.4 to 47.0 wt pct, Al{sub 2}O{sub 3}: 0 to 4.7 wt pct and MgO 5.0 to 7.0 wt pct, where the basicity R is above 1.00 and the C/A ratio is above 9.00. After the LF refining process, the contents of total oxygen (T.O.), [Al], and [S] gradually stabilized in the range of 0.0008 to 0.0012 wt pct, 0.002 to 0.0028 wt pct and 0.0026 to 0.0037 wt pct, respectively. The densities of the three kinds of typical inclusions (CaO-SiO{sub 2}-Al{sub 2}O{sub 3}-MgO, MnS, and CaO-SiO{sub 2}-Al{sub 2}O{sub 3}-MgO-MnS) also decreased and reached approximately 0.25/mm{sup 2}, and no inclusion larger than 6.0 μm was found. The composition of the inclusions gradually changed into the optimal range in the CaO-SiO{sub 2}-Al{sub 2}O{sub 3}-MgO slag system during the LF refining process.

36 MATERIALS SCIENCE↗

Computational Optimization of a Rotary Valved Pulse Combustor Concept

A resonant pulse combustor valve concept is introduced that utilizes two slotted, coaxial counterrotating discs for mechanical actuation at the combustor inlet. The intended test article for prototype demonstration is a small, 22 in. long, propane fueled laboratory combustor, flowing approximately 0.006 lbm/s of air. The objective is to develop an externally actuated (i.e., active) valve that yields better performance and longer life than the traditional internally actuated (i.e., passive), reed-type valve found on most pulse combustors. The rotary valve motion is optimized using an axisymmetric, two-dimensional computational fluid dynamic simulation with a domain that includes the valve as a moveable interior wall. Parameters such as slew rate, dwell period in the open position, and total closed period are varied using fuel specific impulse as the figure of merit. Variations in fuel injector location and air fuel ratio are also examined. Additionally, the performance impact of leakage from the rotary valve is quantified since leakage is endemic to the design. The optimized simulation results indicate that the rotary valve concept can deliver the desired performance attributes using disc rotational speeds and stresses that are well within the realm of modern materials. A preliminary mechanical valve design is included in the report.

pressure gain combustion↗

Computational Optimization of a Rotary Valved Pulse Combustor Concept

A resonant pulse combustor valve concept is introduced that utilizes two slotted, coaxial counterrotating discs for mechanical actuation at the combustor inlet. The intended test article for prototype demonstration is a small, 22 in. long, propane fueled laboratory combustor, flowing approximately 0.006 lbm/s of air. The objective is to develop an externally actuated (i.e., active) valve that yields better performance and longer life than the traditional internally actuated (i.e., passive), reed-type valve found on most pulse combustors. The rotary valve motion is optimized using an axisymmetric, two-dimensional computational fluid dynamic simulation with a domain that includes the valve as a moveable interior wall. Parameters such as slew rate, dwell period in the open position, and total closed period are varied using fuel specific impulse as the figure of merit. Variations in fuel injector location and air fuel ratio are also examined. Additionally, the performance impact of leakage from the rotary valve is quantified since leakage is endemic to the design. The optimized simulation results indicate that the rotary valve concept can deliver the desired performance attributes using disc rotational speeds and stresses that are well within the realm of modern materials. A preliminary mechanical valve design is included in the report.

pressure gain combustion↗

Combining Observations in the Reflective Solar and Thermal Domains for Improved Mapping of Carbon, Water and Energy FLuxes

This study investigates the utility of integrating remotely sensed estimates of leaf chlorophyll (C(sub ab)) into a thermal-based Two-Source Energy Balance (TSEB) model that estimates land-surface CO2 and energy fluxes using an analytical, light-use-efficiency (LUE) based model of canopy resistance. Day to day variations in nominal LUE (LUE(sub n)) were assessed for a corn crop field in Maryland U.S.A. through model calibration with CO2 flux tower observations. The optimized daily LUE(sub n) values were then compared to estimates of C(sub ab) integrated from gridded maps of chlorophyll content weighted over the tower flux source area. Changes in Cab exhibited a curvilinear relationship with corresponding changes in daily calibrated LUE(sub n) values derived from the tower flux data, and hourly water, energy and carbon flux estimation accuracies from TSEB were significantly improved when using C(sub ab) for delineating spatio-temporal variations in LUE(sub n). The results demonstrate the synergy between thermal infrared and shortwave reflective wavebands in producing valuable remote sensing data for monitoring of carbon and water fluxes.

Houborg, Rasmus↗

Reactions induced by {sup 7}Li beam and optimization of silicon detector telescope

Focus of this article is put on a method of compensating for non-uniformity of 50{mu}m thin silicon detectors, rather than measurement results. The same high sensitivity of particle identification on thin-detector thickness enables to accurately model thickness variation using the measurement data, rather than measuring the thickness separately.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

WE-AB-209-12: Quasi Constrained Multi-Criteria Optimization for Automated Radiation Therapy Treatment Planning

Purpose: To introduce quasi-constrained Multi-Criteria Optimization (qcMCO) for unsupervised radiation therapy optimization which generates alternative patient-specific plans emphasizing dosimetric tradeoffs and conformance to clinical constraints for multiple delivery techniques. Methods: For N Organs At Risk (OARs) and M delivery techniques, qcMCO generates M(N+1) alternative treatment plans per patient. Objective weight variations for OARs and targets are used to generate alternative qcMCO plans. For 30 locally advanced lung cancer patients, qcMCO plans were generated for dosimetric tradeoffs to four OARs: each lung, heart, and esophagus (N=4) and 4 delivery techniques (simple 4-field arrangements, 9-field coplanar IMRT, 27-field non-coplanar IMRT, and non-coplanar Arc IMRT). Quasi-constrained objectives included target prescription isodose to 95% (PTV-D95), maximum PTV dose (PTV-Dmax)< 110% of prescription, and spinal cord Dmax<45 Gy. The algorithm’s ability to meet these constraints while simultaneously revealing dosimetric tradeoffs was investigated. Statistically significant dosimetric tradeoffs were defined such that the coefficient of determination between dosimetric indices which varied by at least 5 Gy between different plans was >0.8. Results: The qcMCO plans varied mean dose by >5 Gy to ipsilateral lung for 24/30 patients, contralateral lung for 29/30 patients, esophagus for 29/30 patients, and heart for 19/30 patients. In the 600 plans computed without human interaction, average PTV-D95=67.4±3.3 Gy, PTV-Dmax=79.2±5.3 Gy, and spinal cord Dmax was >45 Gy in 93 plans (>50 Gy in 2/600 plans). Statistically significant dosimetric tradeoffs were evident in 19/30 plans, including multiple tradeoffs of at least 5 Gy between multiple OARs in 7/30 cases. The most common statistically significant tradeoff was increasing PTV-Dmax to reduce OAR dose (15/30 patients). Conclusion: The qcMCO method can conform to quasi-constrained objectives while revealing significant variations in OAR doses including mean dose reductions >5 Gy. Clinical implementation will facilitate patient-specific decision making based on achievable dosimetry as opposed to accept/reject models based on population derived objectives.

60 APPLIED LIFE SCIENCES↗

Optimization of Distribution Feeder Topology: A Differential Programming Learning Approach

This paper presents a gradient based method for optimizing distribution feeder network topology under load un- certainty. We recast the optimal network reconfiguration problem as a learning problem where edge weights of a graph are learned to produce an optimized spanning tree for a distribution network. Using recent methods published on differentiable programming, we provide a data driven method for learning these weights. We test our method on 100 variations of an IEEE 15-bus test system. Our results show that our method outperforms more traditional mathematical programming-based approaches.

differentiable programming↗

Photon optimizer (PO) vs progressive resolution optimizer (PRO): a conformality- and complexity-based comparison for intensity-modulated arc therapy plans

This study aimed to provide guidance on the advantages and limitations of a new optimizer, “photon optimizer” (PO), when compared with its predecessor, “progressive resolution optimizer” (PRO), for intensity-modulated arc therapy (IMAT) plans. Eleven study plans that included a cohort of prostate, head and neck, and brain treatment sites were optimized using both PRO and PO algorithms. A plan template using the same objectives for the same number of iterations was used for each optimized plan to obtain hypothetical treatment plans that would be comparable with a clinical plan. Analysis was performed using plan conformity-based parameters such as target volume coverage factor, conformation number and homogeneity indices, and plan complexity assessment parameters such as small aperture score, modulation indices, and monitor unit variation with arc angle for prostate, brain and head, and neck IMAT treatment plans. Plan conformality analysis demonstrated that conformation numbers, target volume coverage factors, and homogeneity indices produced by the 2 optimizers were comparable for most anatomic sites. IMAT treatment plans produced using the PRO optimizer were found to be less complex than plans produced using the PO optimizer, in terms of multileaf collimator (MLC) leaf position variability and modulation complexity scores. Similarly, the PRO optimizer was shown to produce treatment plans that used fewer monitor units (and generally fewer monitor unit per degree of arc rotation) than PO optimizer. This study demonstrated that the PO optimizer can produce IMAT treatment plans with a similar degree of dose conformity to the target volume and generally improved organ at risk sparing, compared with the PRO optimizer. Better coverage to organs at risk produced by plans optimized using PO was observed to have higher MLC variability and monitor units. Therefore, careful evaluation of treatment plan conformity and complexity before assessing its deliverability is recommended when implementing the routine use of PO optimizer.

61 RADIATION PROTECTION AND DOSIMETRY↗

Minimizing thickness variation in monolithic U-10Mo fuel foil and Zr interlayer during hot rolling: A microstructure-based finite element method analysis

Low-enriched uranium alloyed with 10 wt. % molybdenum (U-10Mo) has been identified as a promising alternative to highly enriched uranium fuel for the United States’ high performance research reactors. The monolithic U-10Mo fuel plate consists of a metallic U-10Mo fuel foil with a 25 µm Zr interlayer and a relatively thick cladding of aluminum alloy 6061. The Zr interlayer is typically applied during the hot co-rolling process, and this process dictates the uniformity of the Zr interlayer. Thickness variation observed in the U-10Mo and Zr interlayer has been attributed to several sources: the initial grain size of the U-10Mo castings, can materials, rolling temperature, inhomogeneous molybdenum content, and porosity in the cast U-10Mo. This thickness variation limits the ability to meet the dimensional specification; thus, a better understanding of the factors causing the nonuniform thickness is needed. In this work, we used a novel, microstructure-based finite element method to model the hot rolling process to address these concerns. Grain microstructures in U-10Mo were tessellated and explicitly considered in the finite element model. Each grain was assigned a random material property to mimic the grain strength variations induced by different grain orientations. Simulations were performed using six steel can thicknesses, four grain sizes, and with or without a Zr interlayer to investigate the influences of those variables on the thickness nonuniformity. The simulation results showed that a thinner steel can and finer U-10Mo grain size reduce thickness variations in both the U-10Mo fuel foil and Zr interlayer. The direct findings from the simulations and analysis can be used to optimize the hot rolling schedule, reduce fabrication defects, and meet the dimensional specifications. The proposed microstructure-based finite element model can be also coupled with experimental microstructure characterization data, images, and models to simulate multi-pass hot rolling.

36 MATERIALS SCIENCE↗

New approach to the design of Schottky barrier diodes for THz mixers

Near-ideal GaAs Schottky barrier diodes especially designed for mixing applications in the THz frequency range are presented. A diode fabrication process for submicron diodes with near-ideal electrical and noise characteristics is described. This process is based on the electrolytic pulse etching of GaAs in combination with an in-situ platinum plating for the formation of the Schottky contacts. Schottky barrier diodes with a diameter of 1 micron fabricated by the process have already shown excellent results in a 650 GHz waveguide mixer at room temperature. A conversion loss of 7.5 dB and a mixer noise temperature of less than 2000 K have been obtained at an intermediate frequency of 4 GHz. The optimization of the diode structure and the technology was possible due to the development of a generalized Schottky barrier diode model which is valid also at high current densities. The common diode design and optimization is discussed on the basis of the classical theory. However, the conventional fomulas are valid only in a limited forward bias range corresponding to currents much smaller than the operating currents under submillimeter mixing conditions. The generalized new model takes into account not only the phenomena occurring at the junction such as current dependent recombination and drift/diffusion velocities, but also mobility and electron temperature variations in the undepleted epi-layer. Calculated diode I/V and noise characteristics are in excellent agreement with the measured values. Thus, the model offers the possibility of optimizing the diode structure and predicting the diode performance under mixing conditions at THz frequencies.

Jelenski, A.↗

Inexact convex relaxations for AC optimal power flow: Towards AC feasibility

Convex relaxations of AC optimal power flow (AC-OPF) problems have attracted significant interest as in several instances they provably yield the global optimum to the original non-convex problem. If, however, the relaxation is inexact, the obtained solution is not AC-feasible. The quality of the obtained solution is essential for several practical applications of AC-OPF, but detailed analyses are lacking in existing literature. Here, this paper aims to cover this gap. We provide an in-depth investigation of the solution characteristics when convex relaxations are inexact, we assess the most promising AC feasibility recovery methods for large-scale systems, and we propose two new metrics that lead to a better understanding of the quality of the identified solutions. We perform a comprehensive assessment on 96 different test cases, ranging from 14 to 3120 buses, and we show the following: (i) Despite an optimality gap of less than 1%, several test cases still exhibit substantial distances to both AC feasibility and local optimality and the newly proposed metrics characterize these deviations. (ii) Penalization methods fail to recover an AC-feasible solution in 15 out of 45 test cases. (iii) The computational benefits of warm-starting non-convex solvers have significant variation, but a computational speedup exists in over 75% of cases.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Optimization study for high speed radial turbine with special reference to design variables

Numerical results of a theoretical investigation are presented to provide information about the effect of variation of the different design and operating parameters on radial inflow turbine performance. The effects of variations in the mass flow rate, rotor tip Mach number, inlet flow angles, number of rotor blades and hub to shroud radius ratio, on the internal fluid dynamics of turbine rotors, was investigated. A procedure to estimate the flow deviation angles at the turbine exit is also presented and used to examine the influence of the operating conditions and the rotor geometrical configuration on these deviations. The significance of the results obtained is discussed with respect to improved turbine performance.

Khalil, I.↗

TH-C-BRD-11: Robustness of Pencil Beam Scanning Proton Therapy for Pelvic Cancer Under Anatomical Changes

Purpose: Pencil beam scanning (PBS) proton therapy provides excellent dosimetric benefits in pelvic cancer treatment, yet day-to-day anatomical variations in pelvic region tend to cause range uncertainties. This study evaluates the dosimetric robustness under anatomical changes for three PBS intensity-modulated proton therapy (IMPT), IMPT using worstcase robust optimization (thereafter ‘Robust IMPT’), and single-field uniform dose (SFUD), in cervical cancer treatment. Methods: IMPT, Robust IMPT, and SFUD plans using the same beam directions and the same prescription (Rx) were generated on computed tomography (CT) images acquired on the simulation day. The dose from each plan was then recomputed on CT images acquired in subsequent two to five weeks using the same protocol. The weekly CTs were registered to the planning CT based on bony anatomy. Target coverage was considered adequate on each weekly CT if dose to 99% of the internal target volume (D-ITV99%) reached at least 95% of the Rx dose. Statistical analysis was then performed on the 21 weekly CT images available for the 7 enrolled patients. Results: Statistically, IMPT was unable to maintain target coverage (mean D-ITV99% = 90.5% Rx, p = 0.004), and SFUD was able to maintain target coverage (mean D-ITV99% = 98.0% Rx, p = 0.0064), in the weeks following simulation. Robust IMPT was able to improve the robustness of IMPT significantly (p < 0.0001), though its maintenance of target coverage was not statistically significant by the 95% Rx criteria (mean D-ITV99% = 96.0%, p = 0.1677). Conclusion: During the multi-week treatment course with anatomical variations, SFUD is robust in terms of maintaining target coverage while IMPT is not. The worst-case optimized Robust IMPT, assuming ±3.5% range uncertainties, improves the robustness of IMPT under anatomical changes significantly, even though it was not designed to account for anatomical changes by mechanism.

60 APPLIED LIFE SCIENCES↗

A semi-agnostic ansatz with variable structure for variational quantum algorithms

Quantum machine learning—and specifically Variational Quantum Algorithms (VQAs)—offers a powerful, flexible paradigm for programming near-term quantum computers, with applications in chemistry, metrology, materials science, data science, and mathematics. Here, one trains an ansatz, in the form of a parameterized quantum circuit, to accomplish a task of interest. However, challenges have recently emerged suggesting that deep ansatzes are difficult to train, due to flat training landscapes caused by randomness or by hardware noise. This motivates our work, where we present a variable structure approach to build ansatzes for VQAs. Our approach, called VAns (Variable Ansatz), applies a set of rules to both grow and (crucially) remove quantum gates in an informed manner during the optimization. Consequently, VAns is ideally suited to mitigate trainability and noise-related issues by keeping the ansatz shallow. We employ VAns in the variational quantum eigensolver for condensed matter and quantum chemistry applications, in the quantum autoencoder for data compression and in unitary compilation problems showing successful results in all cases.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗