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 1,225 records · Page 68

Unified, Geometric Framework for Nonequilibrium Protocol Optimization

Controlling thermodynamic cycles to minimize the dissipated heat is a long-standing goal in thermodynamics, and more recently, a central challenge in stochastic thermodynamics for nanoscale systems. Here, we introduce a theoretical and computational framework for optimizing nonequilibrium control protocols that can transform a system between two distributions in a minimally dissipative fashion. These protocols optimally transport a system along paths through the space of probability distributions that minimize the dissipative cost of a transformation. Furthermore, we show that the thermodynamic metric—determined via a linear response approach—can be directly derived from the same objective function that is optimized in the optimal transport problem, thus providing a unified perspective on thermodynamic geometries. As a result, we investigate this unified geometric framework in two model systems and observe that our procedure for optimizing control protocols is robust beyond linear response.

36 MATERIALS SCIENCE↗

A general-purpose method for Pareto optimal placement of flow rate and concentration sensors in networked systems – With application to wastewater treatment plants

The advent of affordable computing, low-cost sensor hardware, and high-speed and reliable communications have spurred installation of ubiquitous sensors in complex engineered systems. However, ensuring reliable data quality remains a challenge. Exploitation of redundancy among sensor signals can help improving the precision of measured variables, detecting the presence of gross errors, and identifying faulty sensors. The cost of sensor ownership, maintenance efforts in particular, can still be cost-prohibitive however. Maximizing the ability to assess and control data quality while minimizing the cost of ownership thus requires a careful sensor placement. To solve this challenge, in this work we develop a generally applicable method to solve the multi-objective sensor placement problem in systems governed by linear and bilinear balance equations. Importantly, the method computes all Pareto-optimal sensor layouts with conventional computational resources and requires no information about the expected sensor quality.

42 ENGINEERING↗

Checkpoint-Restart and Platform Design Criteria

Long computations often use checkpoint/restart to survive the inevitable failures. Optimal checkpoint/restart strategy can be derived in terms of special functions, and approximated within the regime of interest. As a further step under nominal assumptions, platforms can be optimized for maximum progress rate, computed in terms of non-dimensional parameter JMTTI/Delta, where JMTTI is the Job Mean Time To Interrupt and Delta is the time it takes to write (or read) a checkpoint. Scaling to a larger platform of the same technology requires scaling checkpoint bandwidth quadratically to maintain the same progress rate. This was observed during the design of LANL’s Roadrunner in 2007, led to burst buffers proposed in 2009 and introduced on Trinity in 2015. The need to compute reliably continues to impact the design of future exascale platforms.

97 MATHEMATICS AND COMPUTING↗

Accelerating matrix-centric graph processing on GPUs through bit-level optimizations

Even though it is well known that binary values are common in graph applications (e.g., adjacency matrix), how to leverage the phenomenon for efficiency has not yet been adequately explored. This paper presents a systematic study on how to unlock the potential of the bit-level optimizations of graph computations that involve binary values. It proposes a two-level representation named Bit-Block Compressed Sparse Row (B2SR) and presents a series of optimizations to the graph operations on B2SR by the intrinsics of modern GPUs. It additionally introduces Deep Reinforcement Learning (DRL) as an efficient way to best configure the bit-level optimizations on the fly. Additionally, the DQN-based adaptive tile size selector with dedicated model training can reach 68% prediction accuracy. Evaluations on NVIDIA Pascal and Volta GPUs show that the optimizations bring up to 40× and 6555× for essential GraphBLAS kernels SpMV and SpGEMM, respectively, making GraphBLAS-based BFS accelerate up to 433×, SSSP, PR, and CC up to 35×, and TC up to 52×.

79 ASTRONOMY AND ASTROPHYSICS↗

Fundamental differences between optimization code test problems in engineering applications

The purpose here is to suggest that there is at least one fundamental difference between the problems used for testing optimization codes and the problems that engineers often need to solve; in particular, the level of precision that can be practically achieved in the numerical evaluation of the objective function, derivatives, and constraints. This difference affects the performance of optimization codes, as illustrated by two examples. Two classes of optimization problem were defined. Class One functions and constraints can be evaluated to a high precision that depends primarily on the word length of the computer. Class Two functions and/or constraints can only be evaluated to a moderate or a low level of precision for economic or modeling reasons, regardless of the computer word length. Optimization codes have not been adequately tested on Class Two problems. There are very few Class Two test problems in the literature, while there are literally hundreds of Class One test problems. The relative performance of two codes may be markedly different for Class One and Class Two problems. Less sophisticated direct search type codes may be less likely to be confused or to waste many function evaluations on Class Two problems. The analysis accuracy and minimization performance are related in a complex way that probably varies from code to code. On a problem where the analysis precision was varied over a range, the simple Hooke and Jeeves code was more efficient at low precision while the Powell code was more efficient at high precision.

Eason, E. D.↗

Gamma guidance of trajectories for coplanar, aeroassisted orbital transfer

The optimization and guidance of trajectories for coplaner, aeroassisted orbital transfer (AOT) from high Earth orbit (HEO) to low Earth orbit (LEO) are examined. In particular, HEO can be a geosynchronous Earth orbit (GEO). It is assumed that the initial and final orbits are circular, that the gravitational field is central and is governed by the inverse square law, and that at most three impulses are employed: one at HEO exit, one at atmospheric exit, and one at LEO entry. It is also assumed that, during the atmospheric pass, the trajectory is controlled via the lift coefficient. The presence of upper and lower bounds on the lift coefficient is considered. First, optimal trajectories are computed by minimizing the total velocity impulse (hence, the propellant consumption) required for AOT transfer. The sequential gradient-restoration algorithm (SGRA) is used for optimal control problems. The optimal trajectory is shown to include two branches: a relatively short descending flight branch (branch 1) and a long ascending flight branch (branch 2). Next, attention is focused on guidance trajectories capable of approximating the optimal trajectories in real time, while retaining the essential characteristics of simplicity, ease of implementation, and reliability. For the atmospheric pass, a feedback control scheme is employed and the lift coefficient is adjusted according to a two-stage gamma guidance law. Further improvements are possible via a modified gamma guidance which is more stable with respect to dispersion effects arising from navigation errors, variations of the atmospheric density, and uncertainties in the aerodynamic coefficients than gamma guidance trajectory. A byproduct of the studies on dispersion effects is the following design concept. For coplaner aeroassisted orbital transfer, the lift-range-to-weight ratio appears to play a more important role than the lift-to-drag ratio. This is because the lift-range-to-weight ratio controls mainly the minimum altitude (hence, the peak heating rate) of the guidance trajectory; on the other hand, the lift-to-drag ratio controls mainly the duration of the atmospheric pass of the guidance trajectory.

Miele, A.↗

Multi-GPU implementation of a VMAT treatment plan optimization algorithm

Purpose: Volumetric modulated arc therapy (VMAT) optimization is a computationally challenging problem due to its large data size, high degrees of freedom, and many hardware constraints. High-performance graphics processing units (GPUs) have been used to speed up the computations. However, GPU’s relatively small memory size cannot handle cases with a large dose-deposition coefficient (DDC) matrix in cases of, e.g., those with a large target size, multiple targets, multiple arcs, and/or small beamlet size. The main purpose of this paper is to report an implementation of a column-generation-based VMAT algorithm, previously developed in the authors’ group, on a multi-GPU platform to solve the memory limitation problem. While the column-generation-based VMAT algorithm has been previously developed, the GPU implementation details have not been reported. Hence, another purpose is to present detailed techniques employed for GPU implementation. The authors also would like to utilize this particular problem as an example problem to study the feasibility of using a multi-GPU platform to solve large-scale problems in medical physics. Methods: The column-generation approach generates VMAT apertures sequentially by solving a pricing problem (PP) and a master problem (MP) iteratively. In the authors’ method, the sparse DDC matrix is first stored on a CPU in coordinate list format (COO). On the GPU side, this matrix is split into four submatrices according to beam angles, which are stored on four GPUs in compressed sparse row format. Computation of beamlet price, the first step in PP, is accomplished using multi-GPUs. A fast inter-GPU data transfer scheme is accomplished using peer-to-peer access. The remaining steps of PP and MP problems are implemented on CPU or a single GPU due to their modest problem scale and computational loads. Barzilai and Borwein algorithm with a subspace step scheme is adopted here to solve the MP problem. A head and neck (H and N) cancer case is then used to validate the authors’ method. The authors also compare their multi-GPU implementation with three different single GPU implementation strategies, i.e., truncating DDC matrix (S1), repeatedly transferring DDC matrix between CPU and GPU (S2), and porting computations involving DDC matrix to CPU (S3), in terms of both plan quality and computational efficiency. Two more H and N patient cases and three prostate cases are used to demonstrate the advantages of the authors’ method. Results: The authors’ multi-GPU implementation can finish the optimization process within ∼1 min for the H and N patient case. S1 leads to an inferior plan quality although its total time was 10 s shorter than the multi-GPU implementation due to the reduced matrix size. S2 and S3 yield the same plan quality as the multi-GPU implementation but take ∼4 and ∼6 min, respectively. High computational efficiency was consistently achieved for the other five patient cases tested, with VMAT plans of clinically acceptable quality obtained within 23–46 s. Conversely, to obtain clinically comparable or acceptable plans for all six of these VMAT cases that the authors have tested in this paper, the optimization time needed in a commercial TPS system on CPU was found to be in an order of several minutes. Conclusions: The results demonstrate that the multi-GPU implementation of the authors’ column-generation-based VMAT optimization can handle the large-scale VMAT optimization problem efficiently without sacrificing plan quality. The authors’ study may serve as an example to shed some light on other large-scale medical physics problems that require multi-GPU techniques.

60 APPLIED LIFE SCIENCES↗

Nontangent, Developed Contour Bulkheads for a Single-Stage Launch Vehicle

Dry weights for single-stage launch vehicles that incorporate nontangent, developed contour bulkheads are estimated and compared to a baseline vehicle with 1.414 aspect ratio ellipsoidal bulkheads. Weights, volumes, and heights of optimized bulkhead designs are computed using a preliminary design bulkhead analysis code. The dry weights of vehicles that incorporate the optimized bulkheads are predicted using a vehicle weights and sizing code. Two optimization approaches are employed. A structural-level method, where the vehicle's three major bulkhead regions are optimized separately and then incorporated into a model for computation of the vehicle dry weight, predicts a reduction of4365 lb (2.2 %) from the 200,679-lb baseline vehicle dry weight. In the second, vehicle-level, approach, the vehicle dry weight is the objective function for the optimization. For the vehicle-level analysis, modified bulkhead designs are analyzed and incorporated into the weights model for computation of a dry weight. The optimizer simultaneously manipulates design variables for all three bulkheads to reduce the dry weight. The vehicle-level analysis predicts a dry weight reduction of 5129 lb, a 2.6% reduction from the baseline weight. Based on these results, nontangent, developed contour bulkheads may provide substantial weight savings for single stage vehicles.

Wu, K. Chauncey↗

TPSAS-NF1676L-13638-DND

Dry weights for single-stage launch vehicles that incorporate nontangent, developed contour bulkheads are estimated and compared to a baseline vehicle with 1.414 aspect ratio ellipsoidal bulkheads. Weights, volumes, and heights of optimized bulkhead designs are computed using a preliminary design bulkhead analysis code. The dry weights of vehicles that incorporate the optimized bulkheads are predicted using a vehicle weights and sizing code. Two optimization approaches are employed. A structural-level method, where the vehicle's three major bulkhead regions are optimized separately and then incorporated into a model for computation of the vehicle dry weight, predicts a reduction of 4365 Ib (2.2%) from the 200,679-lb baseline vehicle dry weight. In the second, vehicle-level, approach, the vehicle dry weight is the objective function for the optimization. For the vehicle-level analysis, modified bulkhead designs are analyzed and incorporated into the weights model for computation of a dry weight. The optimizer simultaneously manipulates design variables for all three bulkheads to reduce the dry weight. The vehicle- level analysis predicts a dry weight reduction of 5129 Ib, a 2.6% reduction from the baseline weight. Based on these results, nontangent, developed contour bulkheads may provide substantial weight savings for single stage vehicles.

K Chauncey Wu↗

Cloud Optimized Data Formats

Cloud computing offers the promise of being able to analyze Big Data earth Observations at scale, by allowing scientists to deploy many nodes at once to analyze the data. However, in order to take full advantage of cloud scalability, it is often necessary to reorganize and reformat the data to enable fine-grained, parallel access to the data in Web Object Storage. NASA recently conducted a study of several formats that are optimized for analysis in the cloud: Parquet, zarr, HDF (Hierarchical Data Format) in the Cloud, and Cloud-Optimized GeoTIFF (Tagged Image File Format). They were compared against non-cloud-optimized formats, netCDF (network Common Data Form) and GeoTIFF, with criteria based both on stewardship and analysis performance.

Christopher Lynnes↗

Program For Optimization Of Antenna Structures

JPL-ANTOPT computer program designed to help engineers optimize microwave antenna structures by use of path-length- and pointing-error-analysis calculations. Program is package of codes and subroutines added to MSC/NASTRAN program. Written in MSC/NASTRAN's(TM) Direct Matrix Abstraction Programming (DMAP) language.

Strain, Douglas↗

TORCH Computational Reference Kernels - A Testbed for Computer Science Research

For decades, computer scientists have sought guidance on how to evolve architectures, languages, and programming models in order to improve application performance, efficiency, and productivity. Unfortunately, without overarching advice about future directions in these areas, individual guidance is inferred from the existing software/hardware ecosystem, and each discipline often conducts their research independently assuming all other technologies remain fixed. In today's rapidly evolving world of on-chip parallelism, isolated and iterative improvements to performance may miss superior solutions in the same way gradient descent optimization techniques may get stuck in local minima. To combat this, we present TORCH: A Testbed for Optimization ResearCH. These computational reference kernels define the core problems of interest in scientific computing without mandating a specific language, algorithm, programming model, or implementation. To compliment the kernel (problem) definitions, we provide a set of algorithmically-expressed verification tests that can be used to verify a hardware/software co-designed solution produces an acceptable answer. Finally, to provide some illumination as to how researchers have implemented solutions to these problems in the past, we provide a set of reference implementations in C and MATLAB.

97 MATHEMATICS AND COMPUTING↗

Uncertainty Aware Structural Topology Optimization Via a Stochastic Reduced Order Model Approach

This work presents a stochastic reduced order modeling strategy for the quantification and propagation of uncertainties in topology optimization. Uncertainty aware optimization problems can be computationally complex due to the substantial number of model evaluations that are necessary to accurately quantify and propagate uncertainties. This computational complexity is greatly magnified if a high-fidelity, physics-based numerical model is used for the topology optimization calculations. Stochastic reduced order model (SROM) methods are applied here to effectively 1) alleviate the prohibitive computational cost associated with an uncertainty aware topology optimization problem; and 2) quantify and propagate the inherent uncertainties due to design imperfections. A generic SROM framework that transforms the uncertainty aware, stochastic topology optimization problem into a deterministic optimization problem that relies only on independent calls to a deterministic numerical model is presented. This approach facilitates the use of existing optimization and modeling tools to accurately solve the uncertainty aware topology optimization problems in a fraction of the computational demand required by Monte Carlo methods. Finally, an example in structural topology optimization is presented to demonstrate the effectiveness of the proposed uncertainty aware structural topology optimization approach.

Aguilo, Miguel A.↗

Neural network accelerator for quantum control

Efficient quantum control is necessary for practical quantum computing implementations with current technologies. Conventional algorithms for determining optimal control parameters are computationally expensive, largely excluding them from use outside of the simulation. Existing hardware solutions structured as lookup tables are imprecise and costly. By designing a machine learning model to approximate the results of traditional tools, a more efficient method can be produced. Such a model can then be synthesized into a hardware accelerator for use in quantum systems. In this study, we demonstrate a machine learning algorithm for predicting optimal pulse parameters. This algorithm is lightweight enough to fit on a low-resource FPGA and perform inference with a latency of 175 ns and pipeline interval of 5 ns with > 0.99 gate fidelity. In the long term, such an accelerator could be used near quantum computing hardware where traditional computers cannot operate, enabling quantum control at a reasonable cost at low latencies without incurring large data bandwidths outside of the cryogenic environment.

43 PARTICLE ACCELERATORS↗

Deliver Signal Phase and Timing (SPAT) for Energy Optimization of Vehicle Cohort Via Cloud-Computing and LTE Communications

Predictive Signal Phase and Timing (SPAT) message set is one fundamental building block for vehicle-to-infrastructure (V2I) applications such as Eco-Approach and Departure (EAD) at traffic signal controlled urban intersections. Among the two complementary communication methods namely short-range sidelink (PC5) and long-range cellular radio link (Uu), this paper documents the work with long-range link: the complete data chain includes connecting to the traffic signals via existing backhaul communication network, collecting the raw signal phase state data, predicting the signal state changes and delivering the SPAT data via a geofenced service to requests over HTTP protocols. An Application Programming Interface (API) library is developed to support various cellular data transmission reduction and latency improvement techniques. An emulation-based algorithm is applied to predict the traffic signal state changes to provide adequate prediction horizon (e.g., at minimum 2 minutes) for the cohort energy optimization. In fact, the same connectivity and SPAT delivery methodology has been applied to traffic signalized intersections nationwide in the United States upon public agency approvals for access to their firewalled traffic control network and signal control systems or directly to individual controllers. This methodology proves its effectiveness and potential for rapid growth of such SPAT deliveries at mass production scale without needing infrastructure hardware retrofit or excessive communication means. To support the energy optimization of light and heavy-duty vehicle cohorts of mixed automation and propulsion systems (EV, ICE and hybrid), the connection and SPAT deliveries at two sites were completed, including public roads in Washtenaw County, Michigan and closed track test sites at American Center for Mobility (ACM) in Ypsilanti, Michigan. However, only closed test track results at ACM will be presented in this paper. A neuroevolution based optimizer is developed and implemented to control the speed of a vehicle cohort with different propulsion systems and automation levels. Closed track tests showed significant energy savings of the cohort operation.

99 GENERAL AND MISCELLANEOUS↗