Search NASA⌕ Search

SEARCH · Search NASA

Results for “Robust 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 649 records · Page 36

Duality-Free Decomposition Based Data-Driven Stochastic Security-Constrained Unit Commitment

To incorporate the superiority of both stochastic and robust approaches, a data-driven stochastic optimization is employed to solve the security-constrained unit commitment model. This approach makes the most use of the historical data to generate a set of possible probability distributions for wind power outputs and then it optimizes the unit commitment under the worst-case probability distribution. However, this model suffers from huge computational burden, as a large number of scenarios are considered. To tackle this issue, a duality-free decomposition method is proposed in this paper. This approach does not require doing duality, which can save a large set of dual variables and constraints, and therefore reduces the computational burden. In addition, the inner max-min problem has a special mathematical structure, where the scenarios have the similar constraint. Thus, the max-min problem can be decomposed into independent sub-problems to be solved in parallel, which further improves the computational efficiency. A numerical study on an IEEE 118-bus system with practical data of a wind power system has demonstrated the effectiveness of the proposal.

42 ENGINEERING↗

Reduced ecosystem resilience quantifies fine-scale heterogeneity in tropical forest mortality responses to drought

Sensitivity of forest mortality to drought in carbon-dense tropical forests remains fraught with uncertainty, while extreme droughts are predicted to be more frequent and intense. Here, in this study, the potential of temporal autocorrelation of high-frequency variability in Landsat Enhanced Vegetation Index (EVI), an indicator of ecosystem resilience, to predict spatial and temporal variations of forest biomass mortality is evaluated against in situ census observations for 64 site-year combinations in Costa Rican tropical dry forests during the 2015 ENSO drought. Temporal autocorrelation, within the optimal moving window of 24 months, demonstrated robust predictive power for in situ mortality (leave-one- out cross-validation R 2 = 0.54), which allows for estimates of annual biomass mortality patterns at 30 m resolution. Subsequent spatial analysis showed substantial fine-scale heterogeneity of forest mortality patterns, largely driven by drought intensity and ecosystem properties related to plant water use such as forest deciduousness and topography. Highly deciduous forest patches demonstrated much lower mortality sensitivity to drought stress than less deciduous forest patches after elevation was controlled. Our results highlight the potential of high-resolution remote sensing to “fingerprint” forest mortality and the significant role of ecosystem heterogeneity in forest biomass resistance to drought.

59 BASIC BIOLOGICAL SCIENCES↗

Membrane Purification Cell for Aluminum Recycling

Recycling mixed aluminum scrap usually requires adding primary aluminum to the scrap stream as a diluent to reduce the concentration of non-aluminum constituents used in aluminum alloys. Since primary aluminum production requires approximately 10 times more energy than melting scrap, the bulk of the energy and carbon dioxide emissions for recycling are associated with using primary aluminum as a diluent. Eliminating the need for using primary aluminum as a diluent would dramatically reduce energy requirements, decrease carbon dioxide emissions, and increase scrap utilization in recycling. Electrorefining can be used to extract pure aluminum from mixed scrap. Some example applications include producing primary grade aluminum from specific scrap streams such as consumer packaging and mixed alloy saw chips, and recycling multi-alloy products such as brazing sheet. Electrorefining can also be used to extract valuable alloying elements such as Li from Al-Li mixed scrap. This project was aimed at developing an electrorefining process for purifying aluminum to reduce energy consumption and emissions by 75% compared to conventional technology. An electrolytic molten aluminum purification process, utilizing a horizontal membrane cell anode, was designed, constructed, operated and validated. The electrorefining technology could also be used to produce ultra-high purity aluminum for advanced materials applications. The technical objectives for this project were to: - Validate the membrane cell concept with a lab-scale electrorefining cell; - Determine if previously identified voltage increase issue for chloride electrolytes holds for a fluoride-based electrolyte system; - Assess the probability that voltage change issues can be solved; and - Conduct a market and economic analysis to assess commercial feasibility. The process was tested using three different binary alloy compositions (Al-2.0 wt.% Cu, Al-4.7 wt.% Si, Al-0.6 wt.% Fe) and a brazing sheet scrap composition (Al-2.8 wt.% Si-0.7 wt.% Fe-0.8 wt.% Mn),. Purification factors (defined as the initial impurity concentration divided by the final impurity concentration) of greater than 20 were achieved for silicon, iron, copper, and manganese. Cell performance was measured using its current and voltage characteristics and composition analysis of the anode, cathode, and electrolytes. The various cells were autopsied as part of the study. Three electrolyte systems tested were: LiCl-10 wt. % AlCl3, LiCl-10 wt. % AlCl3-5 wt.% AlF3 and LiF-10 wt.% AlF3. An extended four-day run with the LiCl-10 wt.% AlCl3-5 wt.% AlF3 electrolyte system was stable for the entire duration of the experiment, running at energy requirements about one third of the Hoopes and the conventional Hall-Heroult process. Three different anode membranes were investigated with respect to their purification performance and survivability: a woven graphite cloth with 0.05 cm nominal thickness & > 90 % porosity, a drilled rigid membrane with nominal porosity of 33%, and another drilled rigid graphite membrane with increased thickness. The latter rigid drilled graphite was selected as the most promising membrane design. The economic viability of the membrane cell to purify scrap is sensitive to primary & scrap aluminum prices, and the cost of electricity. In particular, it is sensitive to the differential between scrap and primary aluminum price which is highly variable and dependent on the scrap source. In order to be economically viable, any scrap post-processing technology in the U.S. market must have a total operating cost well below the scrap price differential of $0.20-$0.40 per lb to the London Metal Exchange (LME), a margin of 65%-85% of the LME price. The cost to operate the membrane cell is estimated to be < $0.24/lb of purified aluminum. The energy cost is estimated to be $0.05/lb of purified aluminum with the remaining costs being repair and maintenance, electrolyte, labor, taxes and depreciation. The bench-scale work on membrane purification cell process has demonstrated technological advantages and substantial energy and investment savings against other electrolytic processes. However, in order to realize commercial reality, the following items need to be fully investigated: 1. Further evaluation of a pure fluoride electrolyte. 2. Investigate alternative non conductive, more mechanically robust and chemically inert membrane candidates. 3. Optimized membrane cell design to understand contribution of fluid flow patterns and the mass transfer conditions. 4. Improve current efficiency and total metallic aluminum recovery from the cell. All Tasks and Milestones were completed successfully.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Steady-State Gyrokinetics Transport Code (SSGKT), A Scientific Application Partnership with the Framework Application for Core-Edge Transport Simulations, Final Report

This project initiated the development of TGYRO - a steady-state Gyrokinetic transport code (SSGKT) that integrates micro-scale GYRO turbulence simulations into a framework for practical multi-scale simulation of conventional tokamaks as well as future reactors. Using a lightweight master transport code, multiple independent (each massively parallel) gyrokinetic simulations are coordinated. The capability to evolve profiles using the TGLF model was also added to TGYRO and represents a more typical use-case for TGYRO. The goal of the project was to develop a steady-state Gyrokinetic transport code (SSGKT) that integrates micro-scale gyrokinetic turbulence simulations into a framework for practical multi-scale simulation of a burning plasma core ? the International Thermonuclear Experimental Reactor (ITER) in particular. This multi-scale simulation capability will be used to predict the performance (the fusion energy gain, Q) given the H-mode pedestal temperature and density. At present, projections of this type rely on transport models like GLF23, which are based on rather approximate fits to the results of linear and nonlinear simulations. Our goal is to make these performance projections with precise nonlinear gyrokinetic simulations. The method of approach is to use a lightweight master transport code to coordinate multiple independent (each massively parallel) gyrokinetic simulations using the GYRO code. This project targets the practical multi-scale simulation of a reactor core plasma in order to predict the core temperature and density profiles given the H-mode pedestal temperature and density. A master transport code will provide feedback to O(16) independent gyrokinetic simulations (each massively parallel). A successful feedback scheme offers a novel approach to predictive modeling of an important national and international problem. Success in this area of fusion simulations will allow US scientists to direct the research path of ITER over the next two decades. The design of an efficient feedback algorithm is a serious numerical challenge. Although the power source and transport balance coding in the master are standard, it is nontrivial to design a feedback loop that can cope with outputs that are both intermittent and extremely expensive. A prototypical feedback scheme has already been successfully demonstrated for a single global GYRO simulation, although the robustness and efficiency are likely far from optimal. Once the transport feedback scheme is perfected, it could, in principle, be embedded into any of the more elaborate transport codes (ONETWO, TRANSP, and CORSICA), or adopted by other FSP-related multi-scale projects.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Material Discovery and Design Principles of Perovskite Oxides for Reversible Solid Oxide Cells (R-SOC)

Reversible solid oxide cells (R-SOCs) are highly efficient devices for energy conversion and storage, capable of operating for both hydrogen utilization and production. In fuel cell mode, an R-SOC consumes hydrogen or natural gas to generate electricity, while in electrolysis mode, it produces hydrogen from steam. The discover of new materials with rapid oxygen surface exchange kinetics and enduring stability is crucial for the economically viable commercialization of R-SOCs. To facilitate this pursuit, we conducted extensive Density Functional Theory (DFT) calculations and developed Machine Learning (ML) models to predict critical catalytic properties essential for R-SOCs, such as oxygen surface exchange/diffusivity, and area-specific resistance (ASR). BaCoxFeyZrzO3-d(BFCZ)(x+y+z=1) emerged as a promising family of electrode materials with high activity and stability, validated through systematic experimental study. Moreover, a robust numerical multiphysics model was developed to optimize materials and microstructure parameters, providing the ability to predict the performance of functional R-SOCs.

Liu, Jian↗

A Theoretical Analysis on the Dynamic Behavior and Electromechanical Response of PZT

Piezoelectric materials are used as a power source in stockpile components for safety and reliability assurances. The objective of this project is to gain insights into the fundamental pressure-induced behavior and electromechanical response of lead zirconate titanate (PZT). Specifically, to establish the basis for an accurate, physics-based material model for robust model-based design to rapidly optimize PZT-based materials and components for performance studies. Established ab-initio methods are used to interrogate and understand the dynamic behavior of PZT as a function of composition (50/50, 65/35, 80/20, 95/5) and dopants (La, Nb) to overcome the costly and time-consuming experimental methodologies. New cold curves for pure and doped single-crystal PZT are obtained as the reference equation-of-state (EOS). A negligible change in the pressure responses was observed for the systems and strains studied. Dielectric and piezoelectric responses of pure and doped single-crystal PZT were also calculated as a function of pressure. For undoped PZT, there is a clear and orderly pressure and composition dependence. For doped PZT, there is a significant increase in the responses, but the behavior is very disordered and inconclusive.

36 MATERIALS SCIENCE↗

Optimization of Superconducting Linac for Proton Improvement Plan-II (PIP-II)

PIP-II is an essential upgrade of the Fermilab complex that will enable the worlds most intense high-energy beam of neutrinos for the international Deep Underground Neutrino Experiment at LBNF and support a broad physics program at Fermilab. Ultimately, the PIP-II superconducting linac will be capable of accelerating the $H-$ CW beam to 800 MeV with an average power of 1.6 MW. To operate the linac with such high power, beam losses and beam emittance growth must be tightly controlled. In this paper, we present the results of global optimization of the Linac options towards a robust and efficient physics design for the superconducting section of the PIP-II linac. We also investigate the impact of the nonlinear field of the dipole correctors on the beam quality and derive the requirement on the field quality using statistical analysis. Finally, we assess the need to correct the quadrupole focusing produced by Half Wave, and Single Spoke accelerating cavities. We assess the feasibility of controlling the beam coupling in the machine by changing the polarity of the field of linac focusing solenoids

43 PARTICLE ACCELERATORS↗

Using Rose and Compass for Authentication

Many recent non-proliferation software projects include a software authentication component. In this context, 'authentication' is defined as determining that a software package performs only its intended purpose and performs that purpose correctly and reliably over many years. In addition to visual inspection by knowledgeable computer scientists, automated tools are needed to highlight suspicious code constructs both to aid the visual inspection and to guide program development. While many commercial tools are available for portions of the authentication task, they are proprietary, and have limited extensibility. An open-source, extensible tool can be customized to the unique needs of each project. ROSE is an LLNL-developed robust source-to-source analysis and optimization infrastructure currently addressing large, million-line DOE applications in C, C++, and FORTRAN. It continues to be extended to support the automated analysis of binaries (x86, ARM, and PowerPC). We continue to extend ROSE to address a number of security specific requirements and apply it to software authentication for non-proliferation projects. We will give an update on the status of our work.

99 GENERAL AND MISCELLANEOUS↗

Super Resolving Unrolled Neural Networks for Remote Sensing

In remote sensing systems, the capabilities of the system are constrained by the complex interactions between size, weight, and power (SWAP) of potential designs. In electro-optical (EO) systems, examples of these critical parameters include the system’s sensitivity and resolution. Those parameters can be increased by ever larger optical apertures and focal planes but at the cost of more SWAP. Multi-image super resolution (MISR) techniques allow resolution to be enhanced via computation rather than more sophisticated optical hardware. These algorithms combine multiple images together into a single, higher resolution image, trading temporal resolution and computation for spatial resolution. Fielded MISR techniques, such as Drizzle, can require several hundred images to create a single super resolved image, implying reduced temporal resolution, increased data acquisition load, and limiting mission applications. Iterative techniques, such as model-based image reconstruction and compressive sensing, have been shown to create super resolved images using fewer images than Drizzle. They do this by posing an optimization problem that balances accuracy between a highly accurate physical model and an image model. In the case of super resolution, the physical model is defined by the relation between low resolution input images and the desired high resolution output image. The image model encodes some assumptions about the super resolved image. These assumptions are meant to suppress reconstruction artifacts that arise due to deterministic physical model error, stochastic measurement noise, and potential undersampling. In practice, the performance of iterative methods are limited by imaging models compatible with optimization. Deep learning-based methods can effectively learn image models of arbitrary complexity, but lack the theoretical explainability and robustness of iterative techniques. Consensus equilibrium (CE) generalizes the iterative techniques beyond optimization, enabling blackbox algorithms such as traditional and neural image denoisers to be used as the image model. CE-based approaches retain much of the explainability and robustness of iterative techniques while allowing the expressiveness of machine learning image models to be used. Additionally, by unrolling iterations of CE with an embedded image denoiser, the image denoiser can be further trained and specialized to the specific application with potentially higher quality reconstructions. Under this project, we demonstrated the feasibility of training an unrolled neural network based upon CE. While we didn’t train one, we showed that the CE process is differentiable and its gradient can be tractably computed. We also explored the usage of a variants of CE akin to generative neural works. Most importantly, we applied the CE framework to a number of problems including non-blind deconvolution, upsampling, single-image super resolution, MISR, event-based sensing, and saturated deconvolution. Our MISR prototype creates high quality reconstructions with an order of magnitude fewer images than previous approaches and, critically, produces these reconstructions fast enough for practical usage.

47 OTHER INSTRUMENTATION↗

Convolutional Neural Network Surrogate Models for the Mechanical Properties of Periodic Structures

The work in this paper describes neural network surrogate models for calculating the effective mechanical properties of a periodic composites. The models achieve good accuracy even when only provided with training data sampling a small portion of the design space. As an example, the surrogate models are applied to solving the inverse design problem of finding structures with optimal mechanical properties. The surrogate models are sufficiently accurate to recover optimal solutions in general agreement with established topology optimization methods. However, improvements will be required to develop robust, efficient neural network-based surrogate models and several directions for future research are highlighted here.

42 ENGINEERING↗

Enhancing Power Resilience for Remote Communities: A Comprehensive Renewable Energy Solution for Itbayat Island

The island of Itbayat, Philippines, faces significant challenges in maintaining a reliable and resilient power supply due to its current reliance on a vulnerable power distribution system managed by a local electric cooperative. The existing infrastructure, which includes diesel generators and a radial network configuration with some above-ground lines, is highly susceptible to frequent typhoons and adverse weather conditions. These factors, combined with inadequate staffing and high operational costs, result in frequent power outages that disrupt daily life and hinder economic development. This white paper proposes a comprehensive solution to enhance the resilience and reliability of Itbayat's power system by integrating renewable energy sources, specifically solar photovoltaic (PV) systems, battery storage, and a microgrid controller. The proposed solution aims to reduce dependency on diesel fuel, optimize energy use, and provide a sustainable and robust power supply for the island. Key components of the solution include: 1. Solar PV Installation: Deploying solar PV panels to harness abundant solar energy, reducing reliance on diesel fuel. 2. Battery Storage Systems: Installing battery storage to store excess solar energy and ensure a continuous power supply during low solar generation periods. 3. Microgrid Controller: Implementing a microgrid controller to manage and optimize the integration of solar PV, battery storage, and existing diesel generators. The proposed solution addresses several critical issues, including system vulnerability, generator dependency, and operational inefficiencies. By adopting this innovative approach, Itbayat Island can achieve a more resilient, efficient, and sustainable energy infrastructure, ensuring a stable power supply for its residents and enhancing overall energy security.

14 SOLAR ENERGY↗

A metamodeling approach for studying ignition target robustness in a highly dimensional parameter space

Inertial confinement fusion targets must be carefully designed to ignite their central hot spots and burn. Changes in the optimal implosion could reduce the fusion energy or even prevent ignition. Since there are unavoidable uncertainties due to technological defects and not perfect reproducibility from shot to shot, the fusion energy will remain uncertain. The degree with which a target can tolerate larger specifications than specified, and the probability with which a particular yield is exceeded, are possible measures of the robustness of that design. This robustness must be assessed in a very high-dimensional parameter space whose variables include every characteristics of the given target and of the associated laser pulse shape, using high-fidelity simulations. Therefore, these studies would remain computationally very intensive. In this paper we propose an approach which consist first of constructing an accurate metamodel of the yield on the whole parameter space with a reasonable data set of simulations. Then the robustness is very quickly assessed for any set of specifications with this surrogate. The yield is approximated by a neural network, and an iterative method adds new points in the data set by means of D-optimal experimental designs. The robustness study of the baseline Laser Megajoule target against one-dimensional defects illustrates this approach. A set of 2000 simulations is sufficient to metamodel the fusion energy on a large 22-dimensional parameter space around the nominal point. Furthermore, a metamodel of the robustness margin against all specifications has been obtained, providing guidance for target fabrication research and development.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Leveraging visible and near-infrared spectroelectrochemistry to calibrate a robust model for Vanadium(IV/V) in varying nitric acid and temperature levels

Spectroelectrochemistry and optimal design of experiments can be used to rapidly build accurate models for species quantification and enable a greater level of process awareness. Optical spectroscopy can provide vital elemental and molecular information, but several hurdles must be overcome before it can become a widely adopted analytical method for remote analysis in the nuclear field. Analytes with varying oxidation state, acid concentration, and fluctuating temperature must be efficiently accounted for to minimize time and resources in restrictive hot cell environments. The classic one-factor-at-a-time approach is not suitable for frequent calibration/maintenance operations in this setting. Therefore, a novel alternative was developed to characterize a system containing vanadium(IV/V) (0.01–0.1 M), nitric acid (0.1–4 M), and varying temperatures (20–45 °C). Here, spectroelectrochemistry methods were used to acquire a sample set selected by optimal design of experiments. This new approach allows for the accurate analysis of vanadium and HNO 3 concentration by leveraging UV–Vis–NIR absorption spectroscopy with robust and accurate chemometric models. The top model's root mean squared error of prediction percent values were 3.47%, 4.06%, 3.40%, and 10.9% for V(IV), V(V), HNO 3 , and temperature, respectively. These models, efficiently developed using the designed approach, exhibited strong predictive accuracy for vanadium and acid with varying oxidation states and temperature using only spectrophotometry, which advances current technology for real-world hot cell applications. Additionally, Nernstian analysis of the V(IV/V) standard potential was performed using traditional absorbance methods and multivariate curve resolution (MCR). The successful tests demonstrated that MCR Nernst tests may be valuable in highly convoluted spectral systems to better understand the redox processes' behavior.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Carbon fiber design improvements based on economic models

Circular fiber geometries are predominant in commercial carbon fiber material systems, but the use of this fiber shape has numerous limitations. Circular geometries have continuous symmetry, which is helpful for various processing considerations, but also have the largest possible maximum diffusion thickness for a given fiber area. This characteristic means that circular carbon fibers always have the highest material processing cost and lowest production throughput compared to any other fiber shape with the same area and tow count. To quantify material cost and other benefits for non-circular carbon fiber geometries, process models for polyacrylonitrile based carbon fiber production are developed in relationship to the carbon fiber shape, size, and tow count. For a given fiber shape, precursor production costs are shown to favor maximizing fiber size while conversion costs are minimized by the smallest fiber size. These competing cost trends result in a numerically optimal fiber size for a given shape and tow count, while both cost components are decreased by increasing tow count. Cost–performance tradeoffs for three lobe fiber geometries are studied by supplementing the cost trends with a numerical failure model to predict compressive strength for discrete shape variants. The shape selection is shown to be more sensitive to variations in cost than compressive strength while suboptimal shape designs can improve manufacturing robustness and achievable fiber volume fractions. Finally, an optimal three lobe carbon fiber is identified that balances the set of considerations while reducing costs and embodied energy and increasing production throughput compared to a commercial carbon fiber.

Carbon fiber↗

Analysis of the SiMPL Method for Density-Based Topology Optimization

We present a rigorous convergence analysis of a new method for density-based topology optimization that provides pointwise bound-preserving design updates and faster convergence than other popular first-order topology optimization methods. Due to its strong bound preservation, the method is exceptionally robust, as demonstrated in numerous examples here and in the companion article [D. Kim et al., Struct. Multidiscip. Optim., 68 (2025), 74]. Furthermore, it is easy to implement with clear structure and analytical expressions for the updates. Our analysis covers two versions of the method, characterized by the employed line search strategies. We consider a modified Armijo backtracking line search and a Bregman backtracking line search. For both line search algorithms, our algorithm delivers a strict monotone decrease in the objective function and further intuitive convergence properties, e.g., strong and pointwise convergence of the density variables on the active sets, norm convergence to zero of the increments, convergence of the Lagrange multipliers, and more. In addition, the numerical experiments demonstrate apparent mesh-independent convergence of the algorithm. Here, we refer to the new algorithm as the SiMPL method (pronounced “simple”), which stands for Sigmoidal Mirror descent with a Projected Latent variable.

97 MATHEMATICS AND COMPUTING↗

XploreNAS : Explore Adversarially Robust and Hardware-efficient Neural Architectures for Non-ideal Xbars

Compute In-Memory platforms such as memristive crossbars are gaining focus as they facilitate acceleration of Deep Neural Networks (DNNs) with high area and compute efficiencies. However, the intrinsic non-idealities associated with the analog nature of computing in crossbars limits the performance of the deployed DNNs. Furthermore, DNNs are shown to be vulnerable to adversarial attacks leading to severe security threats in their large-scale deployment. Thus, finding adversarially robust DNN architectures for non-ideal crossbars is critical to the safe and secure deployment of DNNs on the edge. This work proposes a two-phase algorithm-hardware co-optimization approach called XploreNAS that searches for hardware efficient and adversarially robust neural architectures for non-ideal crossbar platforms. We use the one-shot Neural Architecture Search approach to train a large Supernet with crossbar-awareness and sample adversarially robust Subnets therefrom, maintaining competitive hardware efficiency. Our experiments on crossbars with benchmark datasets (SVHN, CIFAR10, CIFAR100) show up to ~8–16% improvement in the adversarial robustness of the searched Subnets against a baseline ResNet-18 model subjected to crossbar-aware adversarial training. We benchmark our robust Subnets for Energy-Delay-Area-Products (EDAPs) using the Neurosim tool and find that with additional hardware efficiency–driven optimizations, the Subnets attain ~1.5–1.6× lower EDAPs than ResNet-18 baseline.

97 MATHEMATICS AND COMPUTING↗

Analytical ab initio hessian from a deep learning potential for transition state optimization

Identifying transition states—saddle points on the potential energy surface connecting reactant and product minima—is central to predicting kinetic barriers and understanding chemical reaction mechanisms. In this work, we train a fully differentiable equivariant neural network potential, NewtonNet, on thousands of organic reactions and derive the analytical Hessians. By reducing the computational cost by several orders of magnitude relative to the density functional theory (DFT) ab initio source, we can afford to use the learned Hessians at every step for the saddle point optimizations. We show that the full machine learned (ML) Hessian robustly finds the transition states of 240 unseen organic reactions, even when the quality of the initial guess structures are degraded, while reducing the number of optimization steps to convergence by 2–3× compared to the quasi-Newton DFT and ML methods. All data generation, NewtonNet model, and ML transition state finding methods are available in an automated workflow.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

The dosimetric impact of inversely optimized arc radiotherapy plan modulation for real-time dynamic MLC tracking delivery

Purpose: Real-time dynamic multileaf collimator (MLC) tracking for management of intrafraction tumor motion can be challenging for highly modulated beams, as the leaves need to travel far to adjust for target motion perpendicular to the leaf travel direction. The plan modulation can be reduced by using a leaf position constraint (LPC) that reduces the difference in the position of adjacent MLC leaves in the plan. The purpose of this study was to investigate the impact of the LPC on the quality of inversely optimized arc radiotherapy plans and the effect of the MLC motion pattern on the dosimetric accuracy of MLC tracking delivery. Specifically, the possibility of predicting the accuracy of MLC tracking delivery based on the plan modulation was investigated. Methods: Inversely optimized arc radiotherapy plans were created on CT-data of three lung cancer patients. For each case, five plans with a single 358 deg. arc were generated with LPC priorities of 0 (no LPC), 0.25, 0.5, 0.75, and 1 (highest possible LPC), respectively. All the plans had a prescribed dose of 2 Gy x 30, used 6 MV, a maximum dose rate of 600 MU/min and a collimator angle of 45 deg. or 315 deg. To quantify the plan modulation, an average adjacent leaf distance (ALD) was calculated by averaging the mean adjacent leaf distance for each control point. The linear relationship between the plan quality [i.e., the calculated dose distributions and the number of monitor units (MU)] and the LPC was investigated, and the linear regression coefficient as well as a two tailed confidence level of 95% was used in the evaluation. The effect of the plan modulation on the performance of MLC tracking was tested by delivering the plans to a cylindrical diode array phantom moving with sinusoidal motion in the superior-inferior direction with a peak-to-peak displacement of 2 cm and a cycle time of 6 s. The delivery was adjusted to the target motion using MLC tracking, guided in real-time by an infrared optical system. The dosimetric results were evaluated using gamma index evaluation with static target measurements as reference. Results: The plan quality parameters did not depend significantly on the LPC (p {>=} 0.066), whereas the ALD depended significantly on the LPC (p < 0.001). The gamma index failure rate depended significantly on the ALD, weighted to the percentage of the beam delivered in each control point of the plan (ALD{sub w}) when MLC tracking was used (p < 0.001), but not for delivery without MLC tracking (p {>=} 0.342). The gamma index failure rate with the criteria of 2% and 2 mm was decreased from > 33.9% without MLC tracking to <31.4% (LPC 0) and <2.2% (LPC 1) with MLC tracking. Conclusions: The results indicate that the dosimetric robustness of MLC tracking delivery of an inversely optimized arc radiotherapy plan can be improved by incorporating leaf position constraints in the objective function without otherwise affecting the plan quality. The dosimetric robustness may be estimated prior to delivery by evaluating the ALD{sub w} of the plan.

62 RADIOLOGY AND NUCLEAR MEDICINE↗