General Electric

About: General Electric is a company organization based out in Boston, Massachusetts, United States. It is known for research contribution in the topics: Turbine & Signal. The organization has 76365 authors who have published 110557 publications receiving 1885108 citations. The organization is also known as: General Electric Company & GE.

Performance characterization of a high-power dual active bridge DC-to-DC converter

Abstract: The performance of a high-power, high-power-density DC-to-DC converter based on the single-phase dual active bridge (DAB) topology is described. The dual active bridge converter has been shown to have very attractive features in terms of low device and component stresses, small filter components, low switching losses, high power density and high efficiency, bidirectional power flow, buck-boost operation, and low sensitivity to system parasitics. For high output voltages, on the order of kilovolts, a cascaded output structure is considered. The effects of snubber capacitance and magnetizing inductance on the soft switching region of control are discussed. Various control schemes are outlined. Coaxial transformer design techniques have been utilized to carefully control leakage inductance. The layout and experimental performance of a prototype 50 kW 50 kHz unit operating with an input voltage of 200 V DC and an output voltage of 1600 V DC are presented. >

Single-Shot Refinement Neural Network for Object Detection

Abstract: For object detection, the two-stage approach (e.g., Faster R-CNN) has been achieving the highest accuracy, whereas the one-stage approach (e.g., SSD) has the advantage of high efficiency. To inherit the merits of both while overcoming their disadvantages, in this paper, we propose a novel single-shot based detector, called RefineDet, that achieves better accuracy than two-stage methods and maintains comparable efficiency of one-stage methods. RefineDet consists of two inter-connected modules, namely, the anchor refinement module and the object detection module. Specifically, the former aims to (1) filter out negative anchors to reduce search space for the classifier, and (2) coarsely adjust the locations and sizes of anchors to provide better initialization for the subsequent regressor. The latter module takes the refined anchors as the input from the former to further improve the regression accuracy and predict multi-class label. Meanwhile, we design a transfer connection block to transfer the features in the anchor refinement module to predict locations, sizes and class labels of objects in the object detection module. The multitask loss function enables us to train the whole network in an end-to-end way. Extensive experiments on PASCAL VOC 2007, PASCAL VOC 2012, and MS COCO demonstrate that RefineDet achieves state-of-the-art detection accuracy with high efficiency. Code is available at https://github.com/sfzhang15/RefineDet.

A review of normal tissue hydrogen NMR relaxation times and relaxation mechanisms from 1-100 MHz: dependence on tissue type, NMR frequency, temperature, species, excision, and age.

Abstract: The longitudinal (T1) and transverse (T2) hydrogen (1H) nuclear magnetic resonance (NMR) relaxation times of normal human and animal tissue in the frequency range 1-100 MHz are compiled and reviewed as a function of tissue type, NMR frequency, temperature, species, in vivo versus in vitro status, time after excision, and age. The dominant observed factors affecting T1 are tissue type and NMR frequency (V). All tissue frequency dispersions can be fitted to the simple expression T1 = AVB in the range 1-100 MHz, with A and B tissue-dependent constants. This equation provides as good or better fit to the data as previous more complex formulas. T2 is found to be multicomponent, essentially independent of NMR frequency, and dependent mainly on tissue type. Mean and raw values of T1 and T2 for each tissue are tabulated and/or plotted versus frequency and the fitting parameters A, B and the standard deviations determined to establish the normal range of relaxation times applicable to NMR imaging. The mechanisms for tissue NMR relaxation are reviewed with reference to the fast exchange two state (FETS) model of water in biological systems, and an overview of the dynamic state of water and macromolecular hydrogen compatible with the frequency, temperature, and multicomponent data is postulated. This suggests that 1H tissue T1 is determined predominantly by intermolecular (possibly rotational) interactions between macromolecules and a single bound hydration layer, and the T2 is governed mainly by exchange diffusion of water between the bound layer and a free water phase. Deficiencies in measurement techniques are identified as major sources of data irreproducibility.