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 1,657 records · Page 92

Towards reverse mode automatic differentiation of Kokkos-based codes

Derivative computation is a key component of optimization, sensitivity analysis, uncertainty quantification, and the solving of nonlinear problems. Automatic differentiation (AD) is a powerful technique for evaluating such derivatives, and in recent years, has been integrated into programming environments such as Jax, PyTorch, and TensorFlow to support derivative computations needed for training of machine learning models, facilitating wide-spread use of these technologies. The C++ language has become the de facto standard for scientific computing due to numerous factors, yet language complexity has made the wide-spread adoption of AD technologies for C++ difficult, hampering the incorporation of powerful differentiable programming approaches into C++ scientific simulations. This is exacerbated by the increasing emergence of architectures, such as GPUs, with limited memory capabilities and requiring massive thread-level concurrency. C++ AD tools must effectively use these environments to bring novel scientific simulations to next-generation DOE experimental and observational facilities. In this project, we investigated source transformation-based automatic differentiation using LLVM compiler infrastructure to automatically generate portable and efficient gradient computations of Kokkos-based code. We have demonstrated that our proposed strategy is feasible by investigating the usage of a prototype LLVM-based source transformation tool to generate gradients of simple functions made of sequences of simple Kokkos parallel regions. Speedups of up to 500x compared to Sacado were observed on NVIDIA V100 GPU.

97 MATHEMATICS AND COMPUTING↗

Efficient Reformulation and Optimization for SC-ACOPF with Line Switching

This project aims to develop efficient and robust computational methods for solving the security-constrained alternating current optimal power flow problem (SC-ACOPF). The SC-ACOPF problem is a central problem in operating the electric power grids in the United States. It determines the most economically efficient way to operate the generation and transmission system to meet daily electricity demand. The solution found by solving an SC-ACOPF problem must satisfy the physics of the alternating current (AC) power flows, various generator and network operational constraints, and must maintain secure operation under various contingency scenarios, where a generator, a transmission branch, or a transformer may unexpectedly trip offline.

97 MATHEMATICS AND COMPUTING↗

Do optimization methods in deep learning applications matter?

With advances in deep learning, exponential data growth and increasing model complexity, developing efficient optimization methods are attracting much research attention. Several implementations favor the use of Conjugate Gradient (CG) and Stochastic Gradient Descent (SGD) as being practical and elegant solutions to achieve quick convergence, however, these optimization processes also present many limitations in learning across deep learning applications. Recent research is exploring higher-order optimization functions as better approaches, but these present very complex computational challenges for practical use. Comparing first and higher-order optimization functions, in this paper, our experiments reveal that Levemberg-Marquardt (LM) significantly supersedes optimal convergence but suffers from very large processing time increasing the training complexity of both, classification and reinforcement learning problems. Our experiments compare off-the-shelf optimization functions(CG, SGD, LM and L-BFGS) in standard CIFAR, MNIST, CartPole and FlappyBird experiments.The paper presents arguments on which optimization functions to use and further, which functions would benefit from parallelization efforts to improve pretraining time and learning rate convergence.

Ozyildirim, Buse Melis↗

Selecting Observation Platforms for Optimized Anomaly Detectability under Unreliable Partial Observations

Diagnosers for keeping track on the occurrences of special events in the framework of unreliable partially observed discrete-event dynamical systems were developed in previous work. This paper considers observation platforms consisting of sensors that provide partial and unreliable observations and of diagnosers that analyze them. Diagnosers in observation platforms typically perform better as sensors providing the observations become more costly or increase in number. This paper proposes a methodology for finding an observation platform that achieves an optimal balance between cost and performance, while satisfying given observability requirements and constraints. Since this problem is generally computational hard in the framework considered, an observation platform optimization algorithm is utilized that uses two greedy heuristics, one myopic and another based on projected performances. These heuristics are sequentially executed in order to find best observation platforms. The developed algorithm is then applied to an observation platform optimization problem for a multi-unit-operation system. Results show that improved observation platforms can be found that may significantly reduce the observation platform cost but still yield acceptable performance for correctly inferring the occurrences of special events.

97 MATHEMATICS AND COMPUTING↗

A comparison of surrogate constitutive models for viscoplastic creep simulation of HT-9 steel

Mechanistic microstructure-informed constitutive models for the mechanical response of polycrystals are a cornerstone of computational materials science. However, as these models become increasingly more complex – often involving coupled differential equations describing the effect of specific deformation modes – their associated computational costs can become prohibitive, particularly in optimization or uncertainty quantification tasks that require numerous model evaluations. To address this challenge, surrogate constitutive models that balance accuracy and computational efficiency are highly desirable. Data-driven surrogate models, that learn the constitutive relation directly from data, have emerged as a promising solution. In this work, we develop two local surrogate models for the viscoplastic response of a steel: a piecewise response surface method and a mixture of experts model. These surrogates are designed to adapt to complex material behavior, which may vary with material parameters or operating conditions. The surrogate constitutive models are applied to creep simulations of HT-9 steel, an alloy of considerable interest to the nuclear energy sector due to its high tolerance to radiation damage, using training data generated from viscoplastic self-consistent (VPSC) simulations. In conclusion, we define a set of test metrics to numerically assess the accuracy of our surrogate models for predicting viscoplastic material behavior, and show that the mixture of experts model outperforms the piecewise response surface method in terms of accuracy.

36 MATERIALS SCIENCE↗

A review of computational methods for studying oscillating water columns – the Navier-Stokes based equation approach

This review evaluates the state-of-the-practice numerical tools used to predict the performance of Oscillating Water Column (OWC). The OWC is a widely studied Wave Energy Converter that provides a reliable form of renewable form of electricity that can potentially meet global energy needs. However, the fluid-flow phenomena affecting its hydrodynamic performance are not fully understood. While there has been the successful full-scale deployment of OWCs, various computational methods are being explored to optimize this technology. Potential flow theory is commonly used to evaluate the efficiency of OWCs; however, this assumption tends to over-predict the hydrodynamic performance. Recently, numerical studies using a diverse set of commercial, open-source, or in-house Computational Fluid Dynamics (CFD) using Reynolds Averaged Navier Stokes (RANS) and Large-Eddy Simulation codes show a better comparison to available experimental results but are computationally expensive. ANSYS Fluent was found to be the most widely used CFD code applied to the study of the OWCs, with a high degree of accuracy in terms of experimental validation of numerical results.

16 TIDAL AND WAVE POWER↗

Optimal explicit strong-stability-preserving general linear methods.

This paper constructs strong-stability-preserving general linear time-stepping methods that are well suited for hyperbolic PDEs discretized by the method of lines. These methods generalize both Runge-Kutta (RK) and linear multistep schemes. They have high stage orders and hence are less susceptible than RK methods to order reduction from source terms or nonhomogeneous boundary conditions. A global optimization strategy is used to find the most efficient schemes that have low storage requirements. Numerical results illustrate the theoretical findings.

97 MATHEMATICS AND COMPUTING↗

Machine Learning-Based Model Predictive Control of Two-Time-Scale Systems

In this study, we present a general form of nonlinear two-time-scale systems, where singular perturbation analysis is used to separate the dynamics of the slow and fast subsystems. Machine learning techniques are utilized to approximate the dynamics of both subsystems. Specifically, a recurrent neural network (RNN) and a feedforward neural network (FNN) are used to predict the slow and fast state vectors, respectively. Moreover, we investigate the generalization error bounds for these machine learning models approximating the dynamics of two-time-scale systems. Next, under the assumption that the fast states are asymptotically stable, our focus shifts toward designing a Lyapunov-based model predictive control (LMPC) scheme that exclusively employs the RNN to predict the dynamics of the slow states. Additionally, we derive sufficient conditions to guarantee the closed-loop stability of the system under the sample-and-hold implementation of the controller. A nonlinear chemical process example is used to demonstrate the theory. In particular, two RNN models are constructed: one to model the full two-time-scale system and the other to predict solely the slow state vector. Both models are integrated within the LMPC scheme, and we compare their closed-loop performance while assessing the computational time required to execute the LMPC optimization problem.

97 MATHEMATICS AND COMPUTING↗

Compact tunable Compton x-ray source from laser-plasma accelerator and plasma mirror

We present an in-depth experimental-computational study of the parameters necessary to optimize a tunable, quasi-monoenergetic, efficient, low-background Compton backscattering (CBS) x-ray source that is based on the self-aligned combination of a laser-plasma accelerator (LPA) and a plasma mirror (PM). The main findings are (1) an LPA driven in the blowout regime by 30 TW, 30 fs laser pulses produce not only a high-quality, tunable, quasi-monoenergetic electron beam, but also a high-quality, relativistically intense (a{sub 0} ∼ 1) spent drive pulse that remains stable in profile and intensity over the LPA tuning range. (2) A thin plastic film near the gas jet exit retro-reflects the spent drive pulse efficiently into oncoming electrons to produce CBS x-rays without detectable bremsstrahlung background. Meanwhile, anomalous far-field divergence of the retro-reflected light demonstrates relativistic “denting” of the PM. Exploiting these optimized LPA and PM conditions, we demonstrate quasi-monoenergetic (50% FWHM energy spread), tunable (75–200 KeV) CBS x-rays, characteristics previously achieved only on more powerful laser systems by CBS of a split-off, counter-propagating pulse. Moreover, laser-to-x-ray photon conversion efficiency (∼6 × 10{sup −12}) exceeds that of any previous LPA-based quasi-monoenergetic Compton source. Particle-in-cell simulations agree well with the measurements.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A Highly Efficient Control Framework for Centralized Residential Charging Coordination of Large Electric Vehicle Populations

The potential for widespread adoption of plug-in electric vehicles (PEVs) brings with it the potential for negative impacts on the electric grid from electric vehicle charging. Uncoordinated PEV charging may increase electricity demand during peak hours, which could create concerns for the grid as the number of PEVs increases. Therefore, it is important to coordinate charging to alleviate potential negative impacts. Centralized charging coordination is preferred by grid operator over distributed control strategies because it can systematically allocate energy across a large population of PEVs and achieve global coordination benefits. However, centralized methods presented in the literature to date require prohibitively expensive computational resources and lack realistic PEV charging models. As a result, they cannot achieve the efficiency and accuracy required to implement charging coordination in the real world. This paper introduces a highly efficient receding horizon control framework that enables dynamic charging coordination for large PEV populations. A two-stage hierarchical optimization routine is proposed that aggregates individual PEV charging flexibility to reduce the computational complexity of the optimization process. The control framework is based on high-fidelity, validated charging system models and charging behavior models derived from a real-world data set of residential charging activities collected from thousands of charging stations over multiple years. Case studies illustrate that the proposed charging control framework is capable of effectively coordinating the charging of millions of PEVs using a standard desktop computer.

30 DIRECT ENERGY CONVERSION↗

DAKOTA Design Analysis Kit for Optimization and Terascale

The DAKOTA (Design Analysis Kit for Optimization and Terascale Applications) toolkit provides a flexible and extensible interface between simulation codes (computational models) and iterative analysis methods. By employing object-oriented design to implement abstractions of the key components required for iterative systems analyses, the DAKOTA toolkit provides a flexible and extensible problem-solving environment for design and analysis of computational models on high performance computers.A user provides a set of DAKOTA commands in an input file and launches DAKOTA. DAKOTA invokes instances of the computational models, collects their results, and performs systems analyses. DAKOTA contains algorithms for optimization with gradient and nongradient-based methods; uncertainty quantification with sampling, reliability, polynomial chaos, stochastic collocation, and epistemic methods; parameter estimation with nonlinear least squares methods; and sensitivity/variance analysis with design of experiments and parameter study methods. These capabilities may be used on their own or as components within advanced strategies such as hybrid optimization, surrogate-based optimization, mixed integer nonlinear programming, or optimization under uncertainty. Services for parallel computing, simulation interfacing, approximation modeling, fault tolerance, restart, and graphics are also included.

SciDAC↗

Geomancy: Automated Performance Enhancement through Data Layout Optimization

Large distributed storage systems such as high- performance computing (HPC) systems used by national or international laboratories require sufficient performance and scale for demanding scientific workloads and must handle shifting workloads with ease. Ideally, data is placed in locations to optimize performance, but the size and complexity of large storage systems inhibit rapid effective restructuring of data layouts to maintain performance as workloads shift. To address these issues, we have developed Geomancy, a tool that models the placement of data within a distributed storage system and reacts to drops in performance. Using a combination of machine learning techniques suitable for temporal modeling, Geomancy determines when and where a bottleneck may happen due to changing workloads and suggests changes in the layout that mitigate or prevent them. Our approach to optimizing throughput offers benefits for storage systems such as avoiding potential bottlenecks and increasing overall I/O throughput from 11% to 30%.

Bel, Oceane M.↗

Solving global optimization problems on GPU cluster

The paper contains the results of investigation of a parallel global optimization algorithm combined with a dimension reduction scheme. This allows solving multidimensional problems by means of reducing to data-independent subproblems with smaller dimension solved in parallel. The new element implemented in the research consists in using several graphic accelerators at different computing nodes. The paper also includes results of solving problems of well-known multiextremal test class GKLS on Lobachevsky supercomputer using tens of thousands of GPU cores.

97 MATHEMATICS AND COMPUTING↗

Addressing Load Imbalance in Bioinformatics and Biomedical Applications: Efficient Scheduling across Multiple GPUs

Computational bioinformatics and biomedical applications frequently contain heterogeneously sized units of work or tasks, for instance due to variability in the sizes of biological sequences and molecules. Variable-sized workloads lead to load imbalances in parallel implementations which detract from efficiency and performance. Many modern computing resources now have multiple graphics processing units(GPUs) per computer for acceleration. These multiple GPU resources need to be used efficiently through balancing of workloads across the GPUs. OpenMP is a portable directive-based parallel programming API used ubiquitously in bioscience applications to program CPUs; recently, the use of OpenMP directives for GPU acceleration has become possible. Here, motivated by experiences with imbalanced loads in GPU-accelerated bioinformatics applications, we address the load balancing problem using OpenMP task-to-GPU scheduling combined with OpenMP GPU offloading for multiply heterogeneous workloads – loads with both variable input sizes, and simultaneously, variable convergence rates for algorithms with a stochastic component – scheduled across multiple GPUs. We aim to develop strategies which are both easy to use and have lower overheads, and may be incorporated incrementally in existing programs which already make use of OpenMP for CPU-based threading in order to make use of multi-GPU computers. We test different combinations of input size variability and convergence rate variability, and characterize the effects of these different scenarios on the performance of scheduling strategies across multiple GPUs with OpenMP. We present several dynamic scheduling solutions for different parallel patterns, explore optimizations, and provide publicly available example computational kernels to make these strategies easy to use in programs. This work will enable application developers to efficiently and easily use multiple GPUs for imbalanced workloads found in bioinformatics and biomedical applications.

Thavappiragasam, Mathialakan↗

Serpentine Magnet Designs for the Interaction Region of the Electron-Ion Collider (EIC)

The Electron-Ion Collider (EIC), hosted by Brookhaven National Laboratory, is designed to deliver a peak luminosity of 1 × 10 34 cm −2 sec −1 . The interaction region (IR) of the EIC imposes several constraints in terms of field quality, aperture, and spatial layout, which necessitates the development of several unique superconducting serpentine direct wind magnets. These magnets are constructed using either a single strand or a small-diameter 6-around-1 NbTi cable, presenting unique challenges for design and optimization. This paper introduces a new computational code specifically developed to streamline and integrate the design process for these magnets, enabling faster design iterations while addressing their complex requirements. Here, in this paper, we first introduce the code, which builds on established electromagnetic fundamentals. The code incorporates tools for optimizing winding patterns and for correcting magnetic multipoles; additionally, it interfaces with established magnet design software. We also present the design of several serpentine magnets for the EIC IR, demonstrating the code’s capability to deliver precise and efficient solutions. These designs highlight the code’s ability to accelerate the development cycle, ensuring the serpentine magnets meet the demanding specifications of the EIC project.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗