A two-stage self-organizing neural network architecture has been applied to object recognition in Synthetic Aperture Radar imagery. The first stage performs feature extraction and implements a two-layer Neocognitron. The resulting feature vectors are presented to the second stage, an ART 2-A classifier network, which clusters the features into multiple target categories. Training is performed off-line in two steps. First, the Neocognitron self-organizes in response to repeated presentations of an object to recognize. During this training process, discovered features and the mechanisms for their extraction are captured in the excitatory weight patterns. In the second step, Neocognitron learning is inhibited and the ART 2-A classifier forms categories in response to the feature vectors generated by additional presentations of the object to recognize. Finally, all training is inhibited and the system tested against a variety of objects and background clutter. In this paper we report the results of our initial experiments. The architecture recognizes a simulated tank vehicle at arbitrary azimuthal orientations at a single depression angle while rejecting clutter and other object returns. The neural architecture has achieved excellent classification performance using 20 clusters.
The effects of orientation and stress on chemically prepared Pb(Zr,Ti)O3 (PZT) film properties have been determined. Systematic modification of the underlying substrate technology has made it possible to fabricate suites of films that have various degrees of orientation at a constant stress level, and to fabricate films that are in different states of stress but have similar orientation. Highly oriented films of the following compositions have been fabricated: PZT 60/40, PZT 40/60, and PZT 20/80. Remanent polarizations ( approximately=60 mu C/cm2) greater than those of the best bulk polycrystalline ferroelectrics were obtained for PZT 40/60 films that were under compression and highly.
This paper presents results of a set of numerical experiments performed bo benchmark the Cell-Centered Implicit Continuous-fluid Eulerian (CCICE), and to determine their limitations as flow solvers for water entry and water exit simulations.
Translations of two pioneering Russian papers on antenna theory are presented. The first paper provides a treatise on finite-length dipole antennas; the second paper addresses infinite-length, impedance-loaded transmitting antennas.
This paper will include a brief overview of the components of the QUICKSILVER suite and its current modeling capabilities. As time permits, results from sample applications will be shown, including time animations of simulation results.
The UNIX LANs in 1500 are experiencing explosive growth. The individual departments are creating LANs to address their particular needs; however, at the same time, shared software tools between the departments are becoming more common. It is anticipated that users will occasionally need access to various department software and/or LAN services, and that support personnel may carry responsibilities which require familiarization with multiple environments. It would be beneficial to users and support personnel if the various department environments share some basic similarities, allowing somewhat transparent access. This will become more important when departments share specific systems, as 1510 and 1550 have proposed with an unclassified UNIX system. Therefore, standards/conventions on the department LANs and the central site systems have to be established to allow for these features. it should be noted that the goal of the UEC is to set standards/conventions which affect the users and provide some basic structure for software installation and maintenance; it is not the intent that all 1500 LANs be made identical at an operating system and/or hardware level. The specific areas of concern include: (1) definition of a non-OS file structure; (2) definition of an interface for remote mounted file systems; (3) definition of a user interface for public files; (4) definition of a basic user level environment; and (5) definition of documentation requirements for public files (shared software). Each of these areas is addressed in this paper.
This document contains implementation details for the Sandia Management Restructure Study Team (MRST) Prototype Information System, which resides on a Sun SPARC II workstation employing the INGRES RDBMS. The INGRES/Windows 4GL application editor was used to define the components of the two user applications which comprise the system. These specifications together with the MRST information model and corresponding database definition constitute the MRST Prototype Information System technical specification and implementation description presented herein. The MRST Prototype Information System represents a completed software product which has been presented to the Management Restructure Study Team to support the management restructing processes at Sandia National Laboratories.
Finite element analyses of oil-filled caverns were performed to investigate the effects of cavern depth on surface subsidence and storage loss, a primary performance criteria of SPR caverns. The finite element model used for this study was axisymmetric, approximating an infinite array of caverns spaced at 750 ft. The stratigraphy and cavern size were held constant while the cavern depth was varied between 1500 ft and 3000 ft in 500 ft increments. Thirty year simulations, the design life of the typical SPR cavern, were performed with boundary conditions modeling the oil pressure head applied to the cavern lining. A depth dependent temperature gradient of 0.012{degrees}F/ft was also applied to the model. The calculations were performed using ABAQUS, a general purpose of finite element analysis code. The user-defined subroutine option in ABAQUS was used to enter an elastic secondary creep model which includes temperature dependence. The calculations demonstrated that surface subsidence and storage loss rates increase with increasing depth. At lower depths the difference between the lithostatic stress and the oil pressure is greater. Thus, the effective stresses are greater, resulting in higher creep rates. Furthermore, at greater depths the cavern temperatures are higher which also produce higher creep rates. Together, these factors result in faster closure of the cavern. At the end of the 30 year simulations, a 1500 ft-deep cavern exhibited 4 percent storage loss and 4 ft of subsidence while a 3000 ft-deep cavern exhibited 33 percent storage loss and 44 ft of subsidence. The calculations also demonstrated that surface subsidence is directly related to the amount of storage loss. Deeper caverns exhibit more subsidence because the caverns exhibit more storage loss. However, for a given amount of storage loss, nearly the same magnitude of surface subsidence was exhibited, independent of cavern depth.
This economic analysis compares human and robotic TRUPACT unloading at the Waste Isolation Pilot Plant. Robots speed up the unloading process, reduce human labor requirements, and reduce human exposure to radiation. The analysis shows that benefit/cost ratios are greater than one for most cases using government economic parameters. This suggests that robots are an attractive option for the TRUPACT application, from a government perspective. Rates of return on capital investment are below 15% for most cases using private economic parameters. Thus, robots are not an attractive option for this application, from a private enterprise perspective.
Proceedings of the 35th International Power Sources Symposium
Clark, N.H.
Technologies that use carbon and mixed metal oxides as the electrode material have been pursued for the purpose of producing high-reliability double-layer capacitors (DLCs). The author demonstrates their environmental stability in temperature, shock, vibration, and linear acceleration. She reviews the available test data for both types of DLCs under these stress conditions. This study suggests that mixed metal oxides and carbon-based double-layer capacitors can survive robust environments if packaged properly, and that temperature decreases performance of double-layer capacitors.
We describe a simple engineering model applicable to stand-off “Whipple bumper” shields, which are used to protect space-based assets from impacts by orbital debris particles. The model provides a framework for analyzing: 1) the parameter limits governing the penetration and breakup or decomposition of the hypervelocity debris particle; 2) the behavior of the induced debris cloud, including its velocity and divergence; and 3) the design and optimization of the stand-off shield for a specific threat and level of protection required. The model is normalized to actual stand-off debris shield experiments and multi-dimensional numerical simulations at impact velocities of ~10 km/s. The subsequent analysis of a current space station shield design suggests that: 1) for acceptable levels of protection, stand-off shields can be significantly thinner than previously thought; and 2) with the proper balance between shield thickness and stand-off distance, the total shield mass can be reduced substantially.
A series of experiments has been performed on the Sandia Hypervelocity Launcher to determine the performance limits of conventional Whipple shields against representative 0.8 g aluminum orbital debris plate-like fragments with velocities of 7 and 10 km/s. Supporting diagnostics include flash X-rays, high speed photography and transient digitizers for timing correlation. Two Whipple shield designs were tested with either a 0.030 cm or a 0.127 cm thick front sheet and a 0.407 cm thick backsheet separated by 30.5 cm. These two designs bracket the ballistic penetration limit curve for protection against these debris simulants for 7 km/s impacts.
This document describes the Temperature Monitoring System for the RHEPP project at Sandia National Laboratories. The system is designed to operate in the presence of severe repetitive high voltage and electromagnetic fields while providing real time thermal data on component behavior. The thermal data is used in the design and evaluation of the major RHEPP components such as the magnetically switched pulse compressor and the linear induction voltage adder. Particular attention is given to the integration of commercially available hardware and software components with a custom written control program. While this document is intended to be a reference guide, it may also serve as a template for similar applications. 3 refs.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Greenberg, David S.
The Choice Coordination Problem with k alternatives (k-CCP) was introduced by Rabin in 1982 [Rab82]. The goal is to design a wait-free protocol for n asynchronous processes which causes all correct processes to agree on one out of k possible alternatives. The agreement on a single choice is complicated by the fact that there is no a priori agreement on names for the alternatives. Furthermore processes must state their choice and do all communication via registers associated with the alternatives. We exactly characterize when the k-CCP can be solved deterministiclly, prove upper and lower space bounds for deterministic solutions, and provide a randomized protocol which is significantly better than the deterministic lower bound.
This bibliography contains 34 references concerning utilizing benchmarking in the management of businesses. Books and articles are both cited. Methods for gathering and utilizing information are emphasized. (GHH)
Final Program and Paper Summaries for the 1992 Digital Signal Processing Workshop, DSPWS 1992
Jakowatz Jr., C.V.; Thompson, P.A.
In this paper we take a new look at the tomographic formulation of spotlight mode synthetic aperture radar (SAR), so as to include the case of targets having three-dimensional structure. This bridges the work of David C. Munson and his colleagues, who first described SAR in terms of two-dimensional tomography, with Jack Walker`s original derivation of spotlight mode SAR imaging via Doppler analysis. The main result is to demonstrate that the demodulated radar return data from a spotlight mode collection represent a certain set of samples of the three-dimensional Fourier transform of the target reflectivity function, and to do so using tomographic principles instead of traditional Doppler arguments. We then show that the tomographic approach is useful in interpreting the two-dimensional SAR image of a three-dimensional scene. In particular, the well-known SAR imaging phenomenon commonly referred to as layover is easily explained in terms of tomographic projection. 4 refs.
The unit cell shape of thick frequency selective surfaces, or dichroic plate, is dependent on its frequency requirements. One aperture shape may be chosen to give wider bandwidths, and another chosen for sharper frequency roll-off. This is analogous to circuits where the need for differing frequency response determines the circuit topology. Acting as spatial frequency filters, dichroics are a critical component in supporting the Deep Space Network (DSN) for spacecraft command a control up links as well as spacecraft down links. Currently these dichroic plates separate S-band at 2.0--232 GHz from X-band at 8.4--8.45 GHz. But new spacecraft communication requirements are also calling for an up link frequency at 7.165 GHz. In addition future spacecraft such as Craft/Casssini will require dichroics effectively separating K{sub a}-band frequencies in the 31--35 GHz range. The requirements for these surfaces are low transmission loss of < 0.1 dB at high power levels. Also is important to maintain a minimal relative phase shift between polarizations for circular polarization transmission. More current work has shown the successful demonstration of design techniques for straight, rectangular apertures at an incident angle of 30{degrees}. The plates are air-filled due to power dissipation and noise temperature considerations. Up-link frequency powers approach 100 kW making dielectrics undesirable. Here we address some of the cases in which the straight rectangular shape may have limited usefulness. For example, grating lobes become a consideration when the bandwidth required to include the new frequency of 7.165 GHz conflicts with the desired incident angle of 30{degrees}. For this case, the cross shape`s increased packing density and bandwidth could make it desirable. When a sharp frequency response is required to separate two closely space K{sub a}-band frequencies, the stepped rectangular aperture might be advantageous. 5 refs.
Phase II of the Long Valley Exploratory Well was completed to a depth of 7588 feet in November 1991. The drilling comprised two sub-phases: (1) drilling 17-1/2 inch hole from the Phase I casing shoe at 2558 feet to a depth of 7130 feet, plugging back to 6826 feet, and setting 13-3/8 inch casing at 6825 feet, all during August--September 1991; and (2) returning in November to drill a 3.85-inch core hole deviated out of the previous wellbore at 6868 feet and extending to 7588 feet. Ultimate depth of the well is planned to be 20,000 feet, or at a bottomhole temperature of 500{degrees}C, whichever comes first. Total cost of this drilling phase was approximately $2.3 million, and funding was shared about equally between the California Energy Commission and the Department of Energy. Phase II scientific work will commence in July 1992 and will be supported by DOE Office of Basic Energy Sciences, DOE Geothermal Division, and other funding sources.
Several closed form trajectory solutions have been developed for low-thrust interplanetary flight and used with patched conies for analysis of combined propulsion systems. The solutions provide insight into alternative types of Mars missions, and show considerable mass savings for fast crewed missions with outbound trip times on the order of 90-100 days.
Nuclear Thermal Propulsion (NTP) has been identified as a critical technology in support of the NASA Space Exploration Initiative (SEI). In order to safely develop a reliable, reusable, long-lived flight engine, facilities are required that will support ground tests to qualify the nuclear rocket engine design. Initial nuclear fuel element testing will need to be performed in a facility that supports a realistic thermal and neutronic environment in which the fuel elements will operate at a fraction of the power of a flight weight reactor/engine. Ground testing of nuclear rocket engines is not new. New restrictions mandated by the National Environmental Protection Act of 1970, however, now require major changes to be made in the manner in which reactor engines are now tested. These new restrictions now preclude the types of nuclear rocket engine tests that were performed in the past from being done today, A major attribute of a safely operating ground test facility is its ability to prevent fission products from being released in appreciable amounts to the environment. Details of the intricacies and complications involved with the design of a fuel element ground test facility are presented in this report with a strong emphasis on safety and economy.