Search NASA⌕ Search

SEARCH · Search NASA

Results for “Computational 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 1,135 records · Page 63

Machine learning framework for quantum sampling of highly constrained, continuous optimization problems

In recent years, there is growing interest in using quantum computers for solving combinatorial optimization problems. In this work, we developed a generic, machine learning-based framework for mapping continuous-space inverse design problems into surrogate quadratic unconstrained binary optimization (QUBO) problems by employing a binary variational autoencoder and a factorization machine. The factorization machine is trained as a low-dimensional, binary surrogate model for the continuous design space and sampled using various QUBO samplers. Using the D-Wave Advantage hybrid sampler and simulated annealing, we demonstrate that by repeated resampling and retraining of the factorization machine, our framework finds designs that exhibit figures of merit exceeding those of its training set. We showcase the framework’s performance on two inverse design problems by optimizing (i) thermal emitter topologies for thermophotovoltaic applications and (ii) diffractive meta-gratings for highly efficient beam steering. This technique can be further scaled to leverage future developments in quantum optimization to solve advanced inverse design problems for science and engineering applications.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Bayesian optimization to design a novel x-ray shaping device

In radiation therapy, x-ray dose must be precisely sculpted to the tumor, while simultaneously avoiding surrounding organs at risk. This requires modulation of x-ray intensity in space and/or time. Typically, this is achieved using a multi leaf collimator (MLC) - a complex mechatronic device comprising over one hundred individually powered tungsten ‘leaves’ that move in or out of the radiation field as required. Here, an all-electronic x-ray collimation concept with no moving parts is presented, termed “SPHINX”: Scanning Pencil-beam High-speed Intensity-modulated X-ray source. SPHINX utilizes a spatially distributed bremsstrahlung target and collimator array in conjunction with magnetic scanning of a high energy electron beam to generate a plurality of small x-ray “beamlets.” A simulation framework was developed in Topas Monte Carlo incorporating a phase space electron source, transport through user defined magnetic fields, bremsstrahlung x-ray production, transport through a SPHINX collimator, and dose in water. This framework was completely parametric, meaning a simulation could be built and run for any supplied geometric parameters. This functionality was coupled with Bayesian optimization to find the best parameter set based on an objective function which included terms to maximize dose rate for a user defined beamlet width while constraining inter-channel cross talk and electron contamination. Designs for beamlet widths of 5, 7, and 10 mm 2 were generated. Each optimization was run for 300 iterations and took approximately 40 h on a 24-core computer. For the optimized 7-mm model, a simulation of all beamlets in water was carried out including a linear scanning magnet calibration simulation. Finally, a back-of-envelope dose rate formalism was developed and used to estimate dose rate under various conditions. The optimized 5–, 7–, and 10-mm models had beamlet widths of 5.1 , 7.2 , and 10.1 mm 2 and dose rates of 3574, 6351, and 10 015 Gy/C, respectively. The reduction in dose rate for smaller beamlet widths is a result of both increased collimation and source occlusion. For the simulation of all beamlets in water, the scanning magnet calibration reduced the offset between the collimator channels and beam centroids from 2.9 ±1.9 mm to 0.01 ±0.03 mm. A slight reduction in dose rate of approximately 2% per degree of scanning angle was observed. Based on a back-of-envelope dose rate formalism, SPHINX in conjunction with next-generation linear accelerators has the potential to achieve substantially higher dose rates than conventional MLC-based delivery, with delivery of an intensity modulated 100 x 100 mm 2 field achievable in 0.9 to 10.6 s depending on the beamlet widths used. Bayesian optimization was coupled with Monte Carlo modeling to generate SPHINX geometries for various beamlet widths. A complete Monte Carlo simulation for one of these designs was developed, including electron beam transport of all beamlets through scanning magnets, x-ray production and collimation, and dose in water. These results demonstrate that SPHINX is a promising candidate for sculpting radiation dose with no moving parts, and has the potential to vastly improve both the speed and robustness of radiotherapy delivery. A multi-beam SPHINX system may be a candidate for delivering magavoltage FLASH RT in humans.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

New Tabu Algorithm for Positioning Mining Drillholes with Blocks Uncertainty

This paper proposes a new approach to the mining exploration drillholes positioning problem (DPP) that incorporates both geostatistical and optimization techniques. A metaheuristic was developed to solve the DPP taking into account an uncertainty index that quantifies the reliability of the current interpretation of the mineral deposit. The uncertainty index was calculated from multiple deposit realizations obtained by truncated Gaussian simulations conditional to the available drillholes samplings. A linear programming model was defined to select the subset of future drillholes that maximizes coverage of the uncertainty. A Tabu Search algorithm was developed to solve large instances of this set partitioning problem. The proposed Tabu Search algorithm is shown to provide good quality solutions approaching 95% of the optimal solution in a reasonable computing time, allowing close to optimal coverage of uncertainty for a fixed investment in drilling.

58 GEOSCIENCES↗

Empirical performance bounds for quantum approximate optimization

The quantum approximate optimization algorithm (QAOA) has been put forth as a method for near-term quantum computers to solve optimization problems. However, assessments of QAOA performance have mostly focused on small structured problem instances while performance on more general instances is less clear. Here, we numerically simulate QAOA pure state dynamics for every instance of MaxCut on non-isomorphic unweighted graphs with nine or fewer vertices with depth parameters p ≤ 3. We find the approximation ratios and optimized circuit parameters concentrate across graphs of a given size and empirically show increases in concentration as graph size increases. The parameter concentration leads to two median-angle heuristics that overcome difficulties in QAOA parameter optimization and obtain mean approximation ratios within 3% and 0.2% of the optimal. We also analyze the probability to measure an optimal solution and find increasing variations between graphs as depth increases, in stark contrast to the approximation ratios which concentrate as depth increases. Furthermore, the resulting benchmark data set gives empirical bounds for on-going experimental realizations and lays groundwork for theoretical extensions to greater problem sizes and depths where QAOA may prove important for practically relevant problems.

79 ASTRONOMY AND ASTROPHYSICS↗

Livermore Big Artificial Neural Network Toolkit

LBANN is a toolkit that is designed to train artificial neural networks efficiently on high performance computing architectures. It is optimized to take advantages of key High Performance Computing features to accelerate neural network training. Specifically it is optimized for low-latency, high bandwidth interconnects, node-local NVRAM, node-local GPU accelerators, and high bandwidth parallel file systems. It is built on top of the open source Elemental distributed-memory dense and spars-direct linear algebra and optimization library that is released under the BSD license. The algorithms contained within LBANN are drawn from the academic literature and implemented to work within a distributed-memory framework.

Essen, BrianVan↗

Optimal aircraft performance during microburst encounter

The effects of microburst characteristics on the optimal penetration performance of jet transport and general aviation aircraft are presented. The purpose is to determine the best possible performance that can be achieved in a broad range of microbursts. A secondary goal is to illustrate good strategies for dealing with a range of microbursts during takeoff and landing. Over 1100 optimal trajectories were computed for two aircraft types flying through idealized microbursts using a Successive Quadratic Programs trajectory optimization algorithm. Contours of safety metrics are plotted as functions of the length scales, magnitudes, and locations of horizontal wind shears and vertical downdrafts. These performance contours show three length-scale regimes for optimal microburst penetration. At short length scales, hazards usually associated with gustiness predominate (e.g., high normal load factor, rotational upset). At intermediate length scales, a degraded ability to maintain flight path and/or vertical velocity poses the most serious threat. At very long microburst length scales, excessive touchdown velocities may result. The ability to transit a microburst successfully also varies strongly with microburst location. The results show that both aircraft types could penetrate some very severe microbursts if optimal control histories were followed. Nevertheless, these control strategies assume perfect prior knowledge of the wind, and practical limits to successful encounter with real-time control capabilities would be lower. The optimally controlled jet transport can successfully penetrate higher intensity microbursts than can the general aviation aircraft.

Psiaki, Mark L.↗

Multi-level optimization of a beam-like space truss utilizing a continuum model

A continuous beam model is developed for approximate analysis of a large, slender, beam-like truss. The model is incorporated in a multi-level optimization scheme for the weight minimization of such trusses. This scheme is tested against traditional optimization procedures for savings in computational cost. Results from both optimization methods are presented for comparison.

Yates, K.↗

Multilevel optimization using a continuum model for structures

The new concept of using a continuum model in the multilevel decomposition of structural optimization problems is presented. The practicality of this concept is demonstrated by application in a scheme for the optimization of beam-like space structures. This scheme is tested against traditional optimization procedures for savings in computational cost. Results from both optimization methods are presented for comparison.

Yates, K.↗

Robust Airfoil Optimization in High Resolution Design Space

The robust airfoil shape optimization is a direct method for drag reduction over a given range of operating conditions and has three advantages: (1) it prevents severe degradation in the off-design performance by using a smart descent direction in each optimization iteration, (2) it uses a large number of B-spline control points as design variables yet the resulting airfoil shape is fairly smooth, and (3) it allows the user to make a trade-off between the level of optimization and the amount of computing time consumed. The robust optimization method is demonstrated by solving a lift-constrained drag minimization problem for a two-dimensional airfoil in viscous flow with a large number of geometric design variables. Our experience with robust optimization indicates that our strategy produces reasonable airfoil shapes that are similar to the original airfoils, but these new shapes provide drag reduction over the specified range of Mach numbers. We have tested this strategy on a number of advanced airfoil models produced by knowledgeable aerodynamic design team members and found that our strategy produces airfoils better or equal to any designs produced by traditional design methods.

Li, Wu↗

Aeroelastic Optimization Study Based on the X-56A Model

One way to increase the aircraft fuel efficiency is to reduce structural weight while maintaining adequate structural airworthiness, both statically and aeroelastically. A design process which incorporates the object-oriented multidisciplinary design, analysis, and optimization (MDAO) tool and the aeroelastic effects of high fidelity finite element models to characterize the design space was successfully developed and established. This paper presents two multidisciplinary design optimization studies using an object-oriented MDAO tool developed at NASA Armstrong Flight Research Center. The first study demonstrates the use of aeroelastic tailoring concepts to minimize the structural weight while meeting the design requirements including strength, buckling, and flutter. Such an approach exploits the anisotropic capabilities of the fiber composite materials chosen for this analytical exercise with ply stacking sequence. A hybrid and discretization optimization approach improves accuracy and computational efficiency of a global optimization algorithm. The second study presents a flutter mass balancing optimization study for the fabricated flexible wing of the X-56A model since a desired flutter speed band is required for the active flutter suppression demonstration during flight testing. The results of the second study provide guidance to modify the wing design and move the design flutter speeds back into the flight envelope so that the original objective of X-56A flight test can be accomplished successfully. The second case also demonstrates that the object-oriented MDAO tool can handle multiple analytical configurations in a single optimization run.

flutter constraints↗

Livermore Big Artificial Neural Network Toolkit

LBANN is a toolkit that is designed to train artificial neural networks efficiently on high performance computing architectures. It is optimized to take advantages of key High Performance Computing features to accelerate neural network training. Specifically it is optimized for low-latency, high bandwidth interconnects, node-local NVRAM, node-local GPU accelerators, and high bandwidth parallel file systems. It is built on top of the open source Elemental distributed-memory dense and spars-direct linear algebra and optimization library that is released under the BSD license. The algorithms contained within LBANN are drawn from the academic literature and implemented to work within a distributed-memory framework.

Essen, Brian Van↗

Black-box optimization of CT acquisition and reconstruction parameters: a reinforcement learning approach

Protocol optimization is critical in Computed Tomography (CT) for achieving desired diagnostic image quality while minimizing radiation dose. Due to the inter-effect of influencing CT parameters, traditional optimization methods rely on the testing of exhaustive combinations of these parameters. This poses a notable limitation due to the impracticality of exhaustive parameter testing. This study introduces a novel methodology leveraging Virtual Imaging Trials (VITs) and reinforcement learning to more efficiently optimize CT protocols. Computational phantoms with liver lesions were imaged using a validated CT simulator and reconstructed with a novel CT reconstruction Toolkit. The optimization parameter space included tube voltage, tube current, reconstruction kernel, slice thickness, and pixel size. The optimization process was done using a Proximal Policy Optimization (PPO) agent which was trained to maximize the Detectability Index (d’) of the liver lesion for each reconstructed image. Results showed that our reinforcement learning approach found the absolute maximum d’ across the test cases while requiring 79.7% fewer steps compared to an exhaustive search, demonstrating both accuracy and computational efficiency, offering a efficient and robust framework for CT protocol optimization. The flexibility of the proposed technique allows for use of varying image quality metrics as the objective metric to maximize for. Our findings highlight the advantages of combining VIT and reinforcement learning for CT protocol management.

Fenwick, David [Duke University Medical Center]↗

Exploiting Quantum Resonance to Solve Combinatorial Problems

Quantum resonance would be exploited in a proposed quantum-computing approach to the solution of combinatorial optimization problems. In quantum computing in general, one takes advantage of the fact that an algorithm cannot be decoupled from the physical effects available to implement it. Prior approaches to quantum computing have involved exploitation of only a subset of known quantum physical effects, notably including parallelism and entanglement, but not including resonance. In the proposed approach, one would utilize the combinatorial properties of tensor-product decomposability of unitary evolution of many-particle quantum systems for physically simulating solutions to NP-complete problems (a class of problems that are intractable with respect to classical methods of computation). In this approach, reinforcement and selection of a desired solution would be executed by means of quantum resonance. Classes of NP-complete problems that are important in practice and could be solved by the proposed approach include planning, scheduling, search, and optimal design.

Zak, Michail↗

3-regular three-XORSAT planted solutions benchmark of classical and quantum heuristic optimizers

With current semiconductor technology reaching its physical limits, special-purpose hardware has emerged as an option to tackle specific computing-intensive challenges. Optimization in the form of solving quadratic unconstrained binary optimization problems, or equivalently Ising spin glasses, has been the focus of several new dedicated hardware platforms. These platforms come in many different flavors, from highly-efficient hardware implementations on digital-logic of established algorithms to proposals of analog hardware implementing new algorithms. In this work, we use a mapping of a specific class of linear equations whose solutions can be found efficiently, to a hard constraint satisfaction problem (three-regular three-XORSAT, or an Ising spin glass) with a 'golf-course' shaped energy landscape, to benchmark several of these different approaches. We perform a scaling and prefactor analysis of the performance of Fujitsu's digital annealer unit (DAU), the D-Wave advantage quantum annealer, a virtual MemComputing machine, Toshiba's simulated bifurcation machine (SBM), the SATonGPU algorithm from Bernaschi et al, and our implementation of parallel tempering. We identify the SATonGPU and DAU as currently having the smallest scaling exponent for this benchmark, with SATonGPU having a small scaling advantage and in addition having by far the smallest prefactor thanks to its use of massive parallelism. Furthermore, our work provides an objective assessment and a snapshot of the promise and limitations of dedicated optimization hardware relative to a particular class of optimization problems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Optimal multiple-pass aeroassisted plane change

This paper presents the exact dimensionless equation of motion and the necessary conditions for the computation of the optimal trajectories of a hypervelocity vehicle flying through a non-rotating spherical planetary atmosphere. Numerical solution is then presented for the case when the vehicle makes several passages through the atmosphere near the perigee of its orbit. While the orbit is slowly contracting, aerodynamic maneuver is performed to obtain the maximum plane change. Several plots were presented to show the optimal variations of the lift coefficient and the bank angle and the various elements of the orbit.

Vinh, Nguyen X.↗

Decentralized Distributed Proximal Policy Optimization (DD-PPO) for High Performance Computing Scheduling on Multi-User Systems

Resource allocation in High Performance Computing (HPC) environments presents a complex and multifaceted challenge for job scheduling algorithms. Beyond the efficient allocation of system resources, schedulers must account for and optimize multiple performance metrics, including job wait time and system throughput. Traditional heuristic-based scheduling algorithms increasingly struggle and lack the efficiency needed to meet the demands and address the complexity and scale of modern HPC systems. Consequently, recent research efforts have focused on leveraging advancements in Artificial Intelligence (AI) and Deep Learning (DL), particularly Reinforcement Learning (RL), to develop more adaptable and intelligent scheduling strategies. Previous RL-based scheduling approaches have explored a range of algorithms, from Deep Q-Networks (DQN) to Proximal Policy Optimization (PPO), and more recently, hybrid methods that integrate Graph Neural Networks (GNNs) with RL techniques. However, a common limitation across these methods is their reliance on relatively small datasets, with few methods being evaluated using large-scale, multi-million-job trace datasets representative of real-world HPC workloads. Moreover, existing RL schedulers face scalability issues due to centralized policy updates, which hinder training efficiency and performance when applied to large datasets. This study introduces a novel RL-based scheduler utilizing Decentralized Distributed Proximal Policy Optimization (DD-PPO) algorithm, which supports large-scale distributed training across multiple workers without requiring parameter synchronization at every step. By eliminating reliance on centralized updates to a shared policy, the DD-PPO scheduler enhances scalability, training efficiency, and sample utilization. Experimental validation using a large real-world dataset containing over 11.5 million job traces collected from petascale HPC systems over six years assesses the influence of dataset scale on training effectiveness and compares DD-PPO performance to traditional and advanced scheduling approaches. The experimental results demonstrate improved scheduling performance in comparison to both heuristic-based schedulers and existing RL-based scheduling algorithms.

AI↗

Reduction method with system analysis for multiobjective optimization-based design

An approach for reducing the number of variables and constraints, which is combined with System Analysis Equations (SAE), for multiobjective optimization-based design is presented. In order to develop a simplified analysis model, the SAE is computed outside an optimization loop and then approximated for use by an operator. Two examples are presented to demonstrate the approach.

Azarm, S.↗