Search NASA⌕ Search

SEARCH · Search NASA

Results for “hybrid algorithm”

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 145 records · Page 8

The Case for Strong Scaling in Deep Learning: Training Large 3D CNNs with Hybrid Parallelism

Here, we present scalable hybrid-parallel algorithms for training large-scale 3D convolutional neural networks. Deep learning-based emerging scientific workflows often require model training with large, high-dimensional samples, which can make training much more costly and even infeasible due to excessive memory usage. We solve these challenges by extensively applying hybrid parallelism throughout the end-to-end training pipeline, including both computations and I/O. Our hybrid-parallel algorithm extends the standard data parallelism with spatial parallelism, which partitions a single sample in the spatial domain, realizing strong scaling beyond the mini-batch dimension with a larger aggregated memory capacity. We evaluate our proposed training algorithms with two challenging 3D CNNs, CosmoFlow and 3D U-Net. Our comprehensive performance studies show that good weak and strong scaling can be achieved for both networks using up to 2K GPUs. More importantly, we enable training of CosmoFlow with much larger samples than previously possible, realizing an order-of-magnitude improvement in prediction accuracy.

97 MATHEMATICS AND COMPUTING↗

Automatic Transformation of MPI Programs to Asynchronous, Graph-Driven Form

The goals of this project are to develop new, scalable, high-fidelity algorithms for atomic-level simulations and program transformations that automatically restructure existing applications, enabling them to scale forward to Petascale systems and beyond. The techniques enable legacy MPI application code to exploit greater parallelism though increased latency hiding and improved workload assignment. The techniques were successfully demonstrated on high-end scalable systems located at DOE laboratories. Besides the automatic MPI program transformations efforts, the project also developed several new scalable algorithms for ab-initio molecular dynamics, including new massively parallel algorithms for hybrid DFT and new parallel in time algorithms for molecular dynamics and ab-initio molecular dynamics. These algorithms were shown to scale to very large number of cores, and they were designed to work in the latency hiding framework developed in this project. The effectiveness of the developments was enhanced by the direct application to real grand challenge simulation problems covering a wide range of technologically important applications, time scales and accuracies. These included the simulation of the electronic structure of mineral/fluid interfaces, the very accurate simulation of chemical reactions in microsolvated environments, and the simulation of chemical behavior in very large enzyme reactions.

97 MATHEMATICS AND COMPUTING↗

Manufacturing Demonstration Facility: Development and Evaluation of Hybrid Manufacturing Toolpaths

The integration of additive manufacturing (AM) capabilities on Computer Numerical Control (CNC) systems allows for the expansion of additive manufacturing to a wide range of part and tool repair operations. This multi-tasking integration, termed hybrid manufacturing, has been researched by others in the past, and Autodesk has been critical in developing process planning and toolpath algorithms for hybrid systems. Objectives and Tasks: Hybrid manufacturing systems enable both additive and subtractive capabilities in a single manufacturing workcell. These systems have the potential to impact a variety of industries, including the tool and die industry due to their repair, refurbishment, and complex geometry manufacturing capabilities. While there has been significant development of toolpath planning for both subtractive and additive processes independently, there has been little, if any, development of hybrid toolpath planning to integrate both processes during the manufacturing design and toolpath generation stage of a product’s lifecycle. Furthermore, additive toolpath planning has been limited to planar manufacturing, but this limitation could be overcome as hybrid CNC machines have multi-axis control. The objectives of this research include: - Development and demonstration of independent three-, four-, and five-axis toolpath generation algorithms for both additive and subtractive processes, and - Development, demonstration, and integration of three-, four-, and five-axis hybrid process planning and toolpath generation algorithms for hybrid additive and subtractive processes. The team will leverage the widely used Autodesk Fusion 360 product design and manufacturing (CAD/CAM) platform to achieve these objectives. Autodesk will provide the expertise in CAD tools, as well as access to their new CAD/CAM manufacturing tools (3-, 4-, and 5-axis milling, additive toolpath generation). ORNL will provide expertise in additive manufacturing toolpath generation, process planning, and manufacturing validation. By the end of the program, the team will have developed and validated multi-axis milling, additive manufacturing, and hybrid manufacturing on an industrial hybrid CNC system (Mazak 500-VC).

42 ENGINEERING↗

Full wave modeling of radio-frequency beams in tokamaks in the electron cyclotron frequency range

Simulation of full wave, without paraxial approximation, high-resolution solution of wave equations in frequency domain in the electron cyclotron resonance (ECR) frequency range for realistic Tokamak plasma parameters became possible by using recently formulated hybrid iterative algorithm [Svidzinski et al., Phys. Plasmas 25, 082509 (2018)] for numerically solving discretized wave equations. This approach combines time evolution and iterative relaxation techniques into iteration cycles. This algorithm is implemented in 2D code FullWave, solving wave equations in Tokamaks in cold and hot plasma models, and it has been tested in 3D full wave iterative RF beams simulation tool, which is presently being developed to model 3D ECRH RF beams in fusion devices using dynamic grid adaptation. The results of 2D full wave modeling, assuming specified toroidal mode number, of ECRH RF beams in DIII-D plasma, performed in the cold and hot plasma models for outboard and top launch scenarios using FullWave are presented. Nonlocal hot plasma response model, based on accurate numerical solution of linearized Vlasov equation, is used to model beam propagation and absorption in the 2nd electron cyclotron harmonic region. Demonstration of capability of the hybrid iterative algorithm to model ECRH RF beams in 3D is made by simulating a substantial part of realistic beam in DIII-D, launched from outboard side of the machine. All relevant physics of RF beam propagation, most of which is not captured in paraxial approximation, such as beam's divergence, interference between the X and O modes in the beam, X-O mode conversion, beam splitting into the X and O mode beams, transformation of beam's cross section, and absorption at the 2nd electron cyclotron harmonic, is captured in the simulations. A numerical technique to find an optimal beam polarization at the launcher to launch a nearly pure X or O mode beam in plasma is developed and tested.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Scaling tests of a new algorithm for DFT hybrid-functional calculations on Trinity Haswell

We show scaling results for materials of interest in Sandia Radiation-Effects and High-Energy-Density-Physics Mission Areas. Each timing is from a self-consistent calculation for bulk material. Two timings are given: (1) walltime for the construction of the CR exchange operator (Exchange-Operator) and (2) walltime for everything else (non-Exchange-Operator).

97 MATHEMATICS AND COMPUTING↗

Ground state energy and magnetization curve of a frustrated magnetic system from real-time evolution on a digital quantum processor

Models of interacting many-body quantum systems that may realize new exotic phases of matter, notably quantum spin liquids, are challenging to study using even state-of-the-art classical methods such as tensor network simulations. Quantum computing provides a promising route for overcoming these difficulties to find ground states, dynamics, and more. In this paper, we argue that recently developed hybrid quantum-classical algorithms based on real-time evolution are promising methods for solving a particularly important model in the search for spin liquids, the antiferromagnetic Heisenberg model on the two-dimensional kagome lattice. We show how to construct efficient quantum circuits to implement time evolution for the model and to evaluate key observables on the quantum computer, and we argue that the method has favorable scaling with increasing system size. We then restrict to a 12-spin star plaquette from the kagome lattice and a related 8-spin system, and we give an empirical demonstration on these small systems that the hybrid algorithms can efficiently find the ground state energy and the magnetization curve. For these demonstrations, we use four levels of approximation: exact state vectors, exact state vectors with statistical noise from sampling, noisy classical emulators, and (for the 8-spin system only) real quantum hardware, specifically the Quantinuum H1-1 processor; for the noisy simulations and hardware demonstration, we also employ error mitigation strategies based on the symmetries of the Hamiltonian. Our results strongly suggest that these hybrid algorithms present a promising direction for studying quantum spin liquids and more generally for resolving important unsolved problems in condensed matter theory and beyond.

97 MATHEMATICS AND COMPUTING↗

Hybridization of IMC and PID control structures based on filtered GPC using genetic algorithm

Classical and advanced hybridization methods are employed in the control literature to calibrate the tuning parameters of internal model control (IMC) and proportional-integral-derivative (PID) controllers. This paper presents an alternative tuning design for the generalized predictive controller (GPC) that is based on both the positional process model and cost function. The proposed method involves the selection of an integral polynomial weighting filter for reference and output signals to deal with reference tracking, disturbance rejection, model-plant mismatch (MPM), minimum control energy and closed-loop robustness. The filter positional GPC (FP-GPC) obtained is transformed into a two-degree-of-freedom polynomial filtered RST structure and then into filtered internal model control (F-IMC) and PID controllers. Also, a multi-objective optimization based on genetic algorithm is applied to tune the filter parameters. Numerical and experimental essays show the effectiveness of the proposed control methodology.

97 MATHEMATICS AND COMPUTING↗

Quantum Local Search with the Quantum Alternating Operator Ansatz

We present a new hybrid, local search algorithm for quantum approximate optimization of constrained combinatorial optimization problems. We focus on the Maximum Independent Set problem and demonstrate the ability of quantum local search to solve large problem instances on quantum devices with few qubits. This hybrid algorithm iteratively finds independent sets over carefully constructed neighborhoods and combines these solutions to obtain a global solution. We study the performance of this algorithm on 3-regular, Community, and Erdős-Rényi graphs with up to 100 nodes.

Tomesh, Teague↗

Integrating a ponderomotive guiding center algorithm into a quasi-static particle-in-cell code based on azimuthal mode decomposition

High fidelity modeling of plasma based acceleration (PBA) requires the use of three dimensional, fully nonlinear, and kinetic descriptions based on the particle-in-cell (PIC) method. In PBA an intense particle beam or laser (driver) propagates through a tenuous plasma whereby it excites a plasma wave wake. Three-dimensional PIC algorithms based on the quasi-static approximation (QSA) have been successfully applied to efficiently model the interaction between relativistic charged particle beams and plasma. In a QSA PIC algorithm, the plasma response to a charged particle beam or laser driver is calculated based on forces from the driver and self-consistent forces from the QSA form of Maxwell's equations. These fields are then used to advance the charged particle beam or laser forward by a large time step. Since the time step is not limited by the regular Courant-Friedrichs-Lewy (CFL) condition that constrains a standard 3D fully electromagnetic PIC code, a 3D QSA PIC code can achieve orders of magnitude speedup in performance. Recently, a new hybrid QSA PIC algorithm that combines another speedup technique known as an azimuthal Fourier decomposition has been proposed and implemented. This hybrid algorithm decomposes the electromagnetic fields, charge and current density into azimuthal harmonics and only the Fourier coefficients need to be updated, which can reduce the algorithmic complexity of a 3D code to that of a 2D code. Modeling the laser-plasma interaction in a full 3D electromagnetic PIC algorithm is very computationally expensive due the enormous disparity of physical scales to be resolved. In the QSA the laser is modeled using the ponderomotive guiding center (PGC) approach. We describe how to implement a PGC algorithm compatible for the QSA PIC algorithms based on the azimuthal mode expansion. Here this algorithm permits time steps orders of magnitude larger than the cell size and it can be asynchronously parallelized. Details on how this is implemented into the QSA PIC code that utilizes an azimuthal mode expansion, QPAD, are also described. Benchmarks and comparisons between a fully 3D explicit PIC code (OSIRIS), as well as a few examples related to laser wakefield acceleration, are presented.

97 MATHEMATICS AND COMPUTING↗

Hardware-accelerated Components for Hybrid Computing Systems

We present a study on the use of component technology for encapsulating platform-specific hardwareaccelerated algorithms on hybrid HPC systems. Our research shows that component technology can have significant benefits from a software engineering pointof- view to increase encapsulation, portability and reduce or eliminate platform dependence for hardwareaccelerated algorithms. As a demonstration of this concept, we discuss our experience in designing, implementing and integrating an FPGA-accelerated kernel for Polygraph, an application in computational proteomics.

97 MATHEMATICS AND COMPUTING↗

A matrix-algebraic formulation of distributed-memory maximal cardinality matching algorithms in bipartite graphs

We describe parallel algorithms for computing maximal cardinality matching in a bipartite graph on distributed-memory systems. Unlike traditional algorithms that match one vertex at a time, our algorithms process many unmatched vertices simultaneously using a matrix-algebraic formulation of maximal matching. This generic matrix-algebraic framework is used to develop three efficient maximal matching algorithms with minimal changes. The newly developed algorithms have two benefits over existing graph-based algorithms. First, unlike existing parallel algorithms, cardinality of matching obtained by the new algorithms stays constant with increasing processor counts, which is important for predictable and reproducible performance. Second, relying on bulk-synchronous matrix operations, these algorithms expose a higher degree of parallelism on distributed-memory platforms than existing graph-based algorithms. We report high-performance implementations of three maximal matching algorithms using hybrid OpenMP-MPI and evaluate the performance of these algorithm using more than 35 real and randomly generated graphs. On real instances, our algorithms achieve up to 200 × speedup on 2048 cores of a Cray XC30 supercomputer. Even higher speedups are obtained on larger synthetically generated graphs where our algorithms show good scaling on up to 16,384 cores.

97 MATHEMATICS AND COMPUTING↗

21st century locomotive technology: quarterly technical status report 26

Parasitic losses due to hybrid sodium battery thermal management do not significantly reduce the fuel saving benefits of the hybrid locomotive. Optimal thermal management trajectories were converted into realizable algorithms which were robust and gave excellent performance to limit thermal excusions and maintain fuel savings.

25 ENERGY STORAGE↗

Advanced Models and Fast Simulation Algorithms for MARS Plants

With the successful commercialization of modular multilevel converters (MMC) and rapid increase in photovoltaic (PV) penetration and energy storage systems (ESS), one of the research areas being explored dynamically is integration of PV systems and ESS systems to MMCs. On this front, an integrated development of PV plants, ESS, and alternating- and direct-current (ac/dc) systems through Multi-Port Autonomous Reconfigurable Solar (MARS) plant is proposed and being researched upon. However, simulation of MARS systems on traditional existing software takes very long time (order of days) due to the presence of a large number of states and non-linear devices. Therefore, development of fast high fidelity models is of paramount importance to test the hardware design, control design and planning of power system expansions. In this context, two research objectives are addressed in this paper: firstly, an ultra fast single- or multi- CPU simulation algorithm to simulate the MARS system based on state-space models, hybrid discretization algorithm with a relaxation technique that reduces the imposed computational burden and a multi-rate method is presented. Secondly, a reduced order model of MARS is also presented. Both the developed techniques are validated with respect to reference PSCAD/EMTDC model.

Marthi, Phani Ratna Vanamali↗

FORCE Development Status Update: Vertical Integration and Benchmarking of System Dynamics

Recent efforts to establish effective models for grid energy analysis, especially given the increase in variable renewable energy (VRE) sources and the economic challenges faced by traditional nuclear energy, have generated new technological considerations. One effort to improve the economic viability of nuclear power involves investigating integrated energy systems (IES) which include secondary energy systems that introduce flexibility and secondary market possibilities to existing and perceived future nuclear energy generation technologies. To analyze the technical and economic potential of IES, the Framework for Optimization of Resources and Economics (FORCE) tool suite was developed through a collaboration among national laboratories. Within the FORCE tool suite, the Holistic Energy Resource Optimization Network (HERON) provides algorithms for analyzing the long-term viability of potential IES technologies, while HYBRID provides algorithms and models to achieve high-resolution analysis of coupling physics over a short time period. Continued maturing of the FORCE tool suite requires further interconnections between the various tools in the suite in order to ensure consistent analysis. Analyses performed by applying HYBRID to transient process modeling should be easily harvestable as inputs to HERON analyses. The first item in this status update is a demonstration of an automated data pipeline for loading data from HYBRID into HERON analyses. Application of HYBRID results to HERON, as part of using the FORCE tool suite, relies on robust modeling assumptions for the various models included in HYBRID. The second result of this status update is the benchmarking and validation of cost and operational data, with a particular focus on natural gas energy generators. These generators are benchmarked with a focus on contrasting them with proposed thermal energy storage (TES) technologies.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗