SEARCH · Engineering Papers
Results for “Modularity”
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.
Modular Spiking Neural Circuits for Mapping Long Short-Term Memory on a Neurosynaptic Processor
Due to the distributed and asynchronous nature of neural computation through low-energy spikes, brain-inspired hardware systems offer high energy efficiency and massive parallelism. One such platform is the IBM TrueNorth neurosynaptic system. Recently, TrueNorth compatible representation learning algorithms have emerged, achieving close to the state-of-the-art performance in various data sets. However, its application in temporal sequence processing models, such as recurrent neural networks (RNNs), is still only at the proof of concept level. There is an inherent difficulty in capturing temporal dynamics of an RNN using spiking neurons, which is only exasperated by the hardware constraints in connectivity and synaptic weight resolution. This paper presents a design flow that overcomes these difficulties and maps a special case of recurrent networks called long short-term memory (LSTM) onto a spike-based platform. The framework utilizes various approximation techniques, such as activation discretization, weight quantization, and scaling and rounding, spiking neural circuits that implement the complex gating mechanisms, and a store-and-release technique to enable neuron synchronization and faithful storage. While the presented techniques can be applied to map LSTM to any spiking neural network (SNN) simulator/emulator, here we choose the TrueNorth chip as the target platform by adhering to its hardware constraints. Three LSTM applications, parity check, extended Reber grammar, and question classification, are evaluated. The tradeoffs among accuracy, performance, and energy tradeoffs achieved on TrueNorth are demonstrated. Finally, this is compared with the performance on an SNN platform without hardware constraints, which represents the upper bound of the achievable accuracy.
PhytoPET: Design and initial results of modular PET for plant biology
We have developed a positron emission tomography (PET) system designed specifically for plant imaging in Phytotron at Duke University. Initial evaluation of a PhytoPET system to image the biodistribution of the positron emitting tracer {sup 11}C in live plants is presented. The mechanical arrangement of the detectors is designed to accommodate the unpredictable and random distribution in space of the plant parts such as stems, leaves, and roots. Prototyping such a system requires a completely new PET system design strategy different from preclinical and clinical applications. This PhytoPET design allows flexible arrangements of PET detectors based on individual standalone detector modules built from single 5 cm × 5 cm Hamamatsu H8500 position sensitive photomultiplier tubes. Each H8500 is coupled to a L YSO:Ce scintillator array composed of 48 × 48 elements that are 10 mm thick with a 1 mm pitch. Initial results provide planar PET images from two different arrangements (1 × 4 or 2 × 2) for flexible imaging capability.
Simulation of Modular Multilevel Converter System Via an Analytical Approach
Explore the source record for details and available documents.
Modular High Field Quadrupole Design for Electron-Ion Collider
Explore the source record for details and available documents.
Economic Implementation of 99.98% Efficiency in Natural-Cooling Modular MVDC SSCBs
Not provided.
Cell Capacitor Voltage Switching-Cycle Balancing Control for Modular Multilevel Converters
Not provided.
Development of Packaging and Modular Control of Thermoelectric Clothes Dryer, With Performance Evaluation
This paper provides an overview of a thermoelectric heat pump clothes dryer which was developed with the aim of reducing the significant primary energy consumption attributed to residential electric clothes drying in the United States (623 TBtu/yr). The use of thermoelectric modules in place of the conventional electric resistance heater resulted in a 40% reduction in the energy consumption of the system, compared to the minimum energy efficiency standard. This was achieved for the first time for a standard test load of 8.45 lb, using a clothes dryer prototype with a thermoelectric heat pump module as the sole heating mechanism. The current experimental prototype was developed after extensive modeling, system design and control optimization, and experimental system-level evaluation of control parameters. The demonstration of improved energy consumption has laid the foundation for future development of this technology.
BioBits™ Explorer: A modular synthetic biology education kit
Hands-on demonstrations greatly enhance the teaching of science, technology, engineering, and mathematics (STEM) concepts and foster engagement and exploration in the sciences. While numerous chemistry and physics classroom demonstrations exist, few biology demonstrations are practical and accessible due to the challenges and concerns of growing living cells in classrooms. We introduce BioBits™ Explorer, a synthetic biology educational kit based on shelf-stable, freeze-dried, cell-free (FD-CF) reactions, which are activated by simply adding water. The FD-CF reactions engage the senses of sight, smell, and touch with outputs that produce fluorescence, fragrances, and hydrogels, respectively. We introduce components that can teach tunable protein expression, enzymatic reactions, biomaterial formation, and biosensors using RNA switches, some of which represent original FD-CF outputs that expand the toolbox of cell-free synthetic biology. The BioBits™ Explorer kit enables hands-on demonstrations of cutting-edge science that are inexpensive and easy to use, circumventing many current barriers for implementing exploratory biology experiments in classrooms.
De novo design of protein homo-oligomers with modular hydrogen-bond network-mediated specificity
Not Available
Modular Engineering of L-Tyrosine Production in Escherichia coli
Explore the source record for details and available documents.
Submicrosecond linear pulse transformer for 800 kV voltage with modular low-inductance primary power supply
A pulsed power source with voltage amplitude up to 800 kV for fast charging (350–400 ns) of the forming line of a high-current nanosecond accelerator is developed. The source includes capacitive energy storage and a linear pulse transformer. The linear transformer consists of a set of 20 inductors with circular ferromagnetic cores surrounded by primary windings inside of which a common stock adder of voltage with film-glycerol insulation is placed. The primary energy storage consists of ten modules, each of which is a low-inductance assembly of two capacitors with a capacitance of 0.35 μF and one gas switch mounted in the same frame. The total energy stored in capacitors is 5.5 kJ at the operating voltage of 40 kV. According to test results, the parameters of the equivalent circuit of the source are the following: shock capacitance = 17.5 nF, inductance = 2 μH, resistance = 3.2 Ω.
A layered approach for modular container construction and orchestration in HPC environments.
Abstract not provided.
Modular Aquatic Simulation System 1D
MASS1 simulates open channel hydrodynamics and transport in branched channel networks, using cross-section averaged forms of the continuity, momentum, and convection diffusion equations. Thermal energy transport (temperature), including meteorological influences is supported. The thermodynamics of total dissolved gas (TDG) can be directly simulated. MASS1 has been developed over the last 20 years. It is currently being used on DOE projects that require MASS1 to beopen source. Hence, the authors would like to distribute MASS1 in source form.
Modular Finite Element Methods
MFEM is a free, lightweight, scalable C++ library for finite element methods. The goal of MFEM is to enable research and development of scalable finite element discretization and solver algorithms through general finite element abstractions, accurate and flexible visualization, and tight integration with the hypre library. Conceptually, MFEM can be viewed as a finite element toolbox that provides the building blocks for developing finite element algorithms in a manner similar to that of MATLAB for linear algebra methods.