Computationally intelligent recognition of characters and symbols addresses a wide range of applications including foreign language translation and chemical formula identification. The combination of intelligent learning and optimization algorithms with layered neural structures offers powerful techniques for character recognition. These techniques were originally developed by Sandia National Laboratories for pattern and spectral analysis; however, their ability to optimize vast amounts of data make them ideal for character recognition. An adaptation of the Neural Network Designer soflsvare allows the user to create a neural network (NN_) trained by a genetic algorithm (GA) that correctly identifies multiple distinct characters. The initial successfid recognition of standard capital letters can be expanded to include chemical and mathematical symbols and alphabets of foreign languages, especially Arabic and Chinese. The FIN model constructed for this project uses a three layer feed-forward architecture. To facilitate the input of characters and symbols, a graphic user interface (GUI) has been developed to convert the traditional representation of each character or symbol to a bitmap. The 8 x 8 bitmap representations used for these tests are mapped onto the input nodes of the feed-forward neural network (FFNN) in a one-to-one correspondence. The input nodes feed forward into a hidden layer, and the hidden layer feeds into five output nodes correlated to possible character outcomes. During the training period the GA optimizes the weights of the NN until it can successfully recognize distinct characters. Systematic deviations from the base design test the network's range of applicability. Increasing capacity, the number of letters to be recognized, requires a nonlinear increase in the number of hidden layer neurodes. Optimal character recognition performance necessitates a minimum threshold for the number of cases when genetically training the net. And, the amount of noise significantly degrades character recognition efficiency, some of which can be overcome by adding noise during training and optimizing the form of the network's activation fimction.
The goal of the 7th International Meshing Roundtable is to bring together researchers and developers from industry, academia, and government labs in a stimulating, open environment for the exchange of technical information related to the meshing process. In the past, the Roundtable has enjoyed significant participation from each of these groups from a wide variety of countries.
Use of nature`s laboratory for scientific analysis of complex systems is a largely untapped resource for understanding long-term disposal of hazardous materials. The Waste Isolation Pilot Plant (WIPP) in the US is a facility designed and approved for storage of transuranic waste in a salt medium. Isolation from the biosphere must be ensured for 10,000 years. Natural analogs provide a means to interpret the evolution of the underground disposal setting. Investigations of ancient sites where manmade materials have experienced mechanical and chemical processes over millennia provide scientific information unattainable by conventional laboratory methods. This paper presents examples of these pertinent natural analogs, provides examples of features relating to the WIPP application, and identifies potential avenues of future investigations. This paper cites examples of analogical information pertaining to the Hallstatt salt mine in Austria and Wieliczka salt mine in Poland. This paper intends to develop an appreciation for the applicability of natural analogs to the science and engineering of a long-term disposal facility in geomedia.
The United States Department of Energy Office of Environmental Management's (DOE/EM's) National Spent Nuclear Fuel Program (NSNFP), through a collaboration between Sandia National Laboratories (SNL) and Idaho National Engineering and Environmental Laboratory (INEEL), is conducting a systematic Nuclear Dynamics Consequence Analysis (NDCA) of the disposal of SNFs in an underground geologic repository sited in unsaturated tuff. This analysis is intended to provide interim guidance to the DOE for the management of the SNF while they prepare for final compliance evaluation. This report presents results from a Nuclear Dynamics Consequence Analysis (NDCA) that examined the potential consequences and risks of criticality during the long-term disposal of spent nuclear fuel owned by DOE-EM. This analysis investigated the potential of post-closure criticality, the consequences of a criticality excursion, and the probability frequency for post-closure criticality. The results of the NDCA are intended to provide the DOE-EM with a technical basis for measuring risk which can be used for screening arguments to eliminate post-closure criticality FEPs (features, events and processes) from consideration in the compliance assessment because of either low probability or low consequences. This report is composed of an executive summary (Volume 1), the methodology and results of the NDCA (Volume 2), and the applicable appendices (Volume 3).
A mechanical characterization of frozen silty soils has been conducted to support computer modeling of penetrators. The soils were obtained from the Eilson AFB (Alaska) vicinity. Quasi-static testing with a multiaxial system in a cold room and intermediate strain rate testing with a split Hopkinson pressure bar were conducted. Maximum stresses achieved were slightly above 1 GPa, apparently limiting the observed behavior primarily to elastic compression and pore crushing phenomena. Lower temperatures seem to increase the strength of the material markedly, although not by a simple factor. Lower temperatures and higher strain rates increase the apparent Young's and bulk moduli as well (an increase of {approximately} a factor of two is observed for strain rate increasing from 0.001 s{sup {minus}1} to 800 s{sup {minus}1}). The strength also depends strongly on strain rate. Increasing the strain rate from 0.001 {sup {minus}1} to 0.07 {sup {minus}1} increases the strength by a factor of five to ten (to values of order 1 GPa). However,only a small increase in strength is seen as strain rate is increased to {approximately} 10{sup 2}--10{sup 3} s{sup {minus}1}. The reliability of the strength measurements at strain rates< 1 s{sup {minus}1} is decreased due to details of the experimental geometry, although general trends are observable. A recipe is provided for a simulant soil based on bentonite, sand, clay-rich soil and water to fit the {approximately} 6% air-filled porosity, density and water content of the Alaska soils, based on benchtop mixing and jacketed compression testing of candidate mixes.
The purpose of this document is to present a strategy for effectively using SMSS (Sea.leable Mass Storage System) and to distribute a simple implementation of this strategy. This work was done as a stopgap memure to ~lOW ~ ~~yst to USe the storage Power of SMSS in the absence of a more user friendly interface. The features and functionality discussed in this document represent a minimum set of capabilities to allow a useful archiving interface functionality. The implementation presented is the most basic possible and would benefit significantly from an organized support and documentation effort.
This report summarizes a three-year project to characterize and improve the ride quality of the Department of Energy (DOE) tractor/trailer. A high-fidelity computer model was used to simulate the vibrational response in the passenger compartment of the truck due to a common roadway environment. It is the intensity of this response that is indicative of the ride quality of the vehicle. The computational model was then validated with experimental tests using a novel technique employing both lab-based modal tests and modal data derived using the Natural Excitation Technique (NExT). The validated model proved invaluable as a design tool. Utilizing the model in a predictive manner, modifications to improve ride quality were made to both the existing vehicle and the next-generation design concept. As a result, the next-generation fleet of tractors (procurement process begins in FY98) will incorporate elements of a successful model-based design for improved truck ride.
As feature sizes of Integrated Circuits (ICs) continue to shrinlL the sensitivity of these devices, particularly SRAMS and DR4Ms, to natural radiation is increasing. The radiation can lead to the uncontrolled deposition of charge within an IC, which ean alter, for example, the memoty state of a bit and thereby produce what is edled a `SOW error, or Single Event Upset (SEU). The response of ICS to natural background radiation is therefore of great coneem regarding the reliability of Mure devices. In this paper, we present results where Ion Beam Induced Charge Collection (TBICC) technique was used to simulate neutron-induced Si recoil dlkcts in IC test structures. The present wo~ wnducted at the San& National Laboratories, uses a 10 MeV Carbon mierobeam with 1 pm spot to scan test structures on specifically designed ICS. The test structure contains junctions typical of S RAMS and DRAMs. Charge is eolleeted from different areas of the IC under various conditions of junction back bias. The data are digitized and displayed as 3D images combined with KY) coordination. With the aid of IC layout informatio~ the 3D images are sepamted into difTerent layers to allow the identification of charge collection etlciency in the test structures. An analysis of the charge collection efficiency from dillerent test areas is given.
The critical energy inkstructures include gas, OL and electric power. These Mrastructures are complex and interdependent nmvorks that are vital to the national secwiy and social well being of our nation. Many electric power systems depend upon gas and oil, while fossil energy delive~ systems depend upon elecnic power. The control mechanisms for these Mrastructures are often referred to as SCADA (Supmivry CkmdandDaU Ac@itz&z) systems. SCADA systems provide remote monitoring and centralized control for a distributed tmnsportation infmsmucture in order to facilitate delivery of a commodi~. AIthough many of the SCADA concepts developed in this paper can be applied to automotive mmsponation systems, we will use transportation to refer to the movement of electrici~, gas, and oil. \ Recently, there have been seveml reports suggesting that the widespread and increasing use of SCADA for control of energy systems provides an increasing opportuni~ for an advers~ to cause serious darnage to the energy inbstmcturei~. This damage could arise through cyber infiltration of the SCADA networks, by physically tampering with the control networks, or through a combination of both means. SCADA system threats decompose into cyber and physical threats. One solution to the SCADA security problem is to design a standard for a highly secure KA.DA system that is both cyber, and physdly secure. Not all-physical threats are possible to guard again% but of those threats that are, high security SCADA provides confidence that the system will continue to operate in their presence. One of the most important problems in SCADA securi~ is the relationship between the cyber and physical vulnerabilities. Cyber intrusion increases physical Vulnerabilities, while in the dual problem physical tampering increases cyber vulnerabilit.ies. There is potential for feedback and the precise dynamics need to be understood. As a first step towards a stan~ the goal of this paper is to facilitate a discussion of the requirements analysis for a highly secure SCADA system. The fi-arnework for the discussion consists of the identification of SCADA security investment areas coupled with the tradeoffs that will force compromises in the solution. For example, computational and bandwidth requirements of a security standard could force the replacement of entire SCADA systems. The requirements for a real-time response in a cascading electric power failure could pose limitations on authentication and encryption mechanisms. The shortest path to the development of a high securi~ SC.ADA standard will be achieved by leveraging existing standards efforts and ensuring that security is being properly addressed in those standards. The Utility Communications Architecture 2.o (UC@, for real-time utili~ decision control, represents one such standard. The development of a SCADA secwiy specification is a complex task that will benefit from a systems engineering approach.
The key to insight is coupling the power of the computer with unique skills of the human. At Sandia National Laboratories' Interaction Laboratory, we call this teraflop visualization. We are concentrating research in three main area: 1) using the computer as a facility for authoring content, 2) adding the physics to model real behaviors, and 3) allowing the human to utilize the improved precision and resolution provided by this new class of compute power.
Demonstrating compliance with the applicable regulations for the Waste Isolation Pilot Plant (WIPP) requires an assessment of the long-term performance of the disposal system. Scenario development is one starting point of this assessment, and generates inquiry about the present state and future evolution of the disposal system. Scenario development consists of four tasks: (1) identifying and classifying features, events and processes (FEPs), (2) screening FEPs according to well-defined criteria, (3) forming scenarios (combinations of FEPs) in the context of regulatory performance criteria and (4) specifying of scenarios for consequence analysis. The development and screening of a comprehensive FEP list provides assurance that the identification of significant processes and events is complete, that potential interactions between FEPs are not overlooked, and that responses to possible questions are available and well documented. Two basic scenarios have been identified for the WIPP: undisturbed performance (UP) and disturbed performance (DP). The UP scenario is used to evaluate compliance with the Environmental Protection Agency's (EPA's) Individual Dose (40 CFR Section 191-15) and Groundwater Protection (40 CFR Section 191-24) standards and accounts for all natural-, waste- and repository-induced FEPs that survive the screening process. The DP scenario is required for assessment calculations for the EPA's cumulative release standard (Containment Requirements, 40 CFR Section 191-13) and accounts for disruptive future human events, which have an uncertain probability of occurrence, in addition to the UP FEPs.
Refractory ternary nitride films for diffusion barriers in microelectronics have been grown using chemical vapor deposition. Thin films of titanium-silicon-nitride, tungsten-boron-nitride, and tungsten-silicon-nitride of various compositions have been deposited on 150 mm Si wafers. The microstructure of the films are either fully amorphous for the tungsten based films, or nauocrystalline TiN in an amorphous matrix for titanium-silicon-nitride. All films exhibit step coverages suitable for use in future microelectronics generations. Selected films have been tested as diffusion barriers between copper and silicon, and generally perform extremely weH. These fiIms are promising candidates for advanced diffusion barriers for microelectronics applications. The manufacturing of silicon wafers into integrated circuits uses many different process and materials. The manufacturing process is usually divided into two parts: the front end of line (FEOL) and the back end of line (BEOL). In the FEOL the individual transistors that are the heart of an integrated circuit are made on the silicon wafer. The responsibility of the BEOL is to wire all the transistors together to make a complete circuit. The transistors are fabricated in the silicon itself. The wiring is made out of metal, currently aluminum and tungsten, insulated by silicon dioxide, see Figure 1. Unfortunately, silicon will diffuse into aluminum, causing aluminum spiking of junctions, killing transistors. Similarly, during chemical vapor deposition (CVD) of tungsten from ~fj, the reactivity of the fluorine can cause "worn-holes" in the silicon, also destroying transistors. The solution to these problems is a so-called diffusion barrier, which will allow current to pass from the transistors to the wiring, but will prevent reactions between silicon and the metal.
Shock equation of state and strength data have been obtained on the explosive PBXW-128 over the pressure range O-3 GPa using gun impact techniques and velocity interferometry diagnostics. Nonlinear shock-velocity-versus-particle velocity behavior is observed. Possible mechanisms are discussed and a Hug.oniot equation of slate model for the data is provided.
High sensitivity acoustic wave chemical microsensors are being developed on GaAs substrates. These devices take advantage of the piezoelectric properties of GaAs as well as its mature microelectronics fabrication technology and nascent micromachining technology. The design, fabrication, and response of GaAs SAW chemical microsensors are reported. Functional integrated GaAs SAW oscillators, suitable for chemical sensing, have been produced. The integrated oscillator requires 20 mA at 3 VK, operates at frequencies up to 500 MHz, and occupies approximately 2 mmz. Discrete GaAs sensor components, including IC amplifiers, SAW delay lines, and IC phase comparators have been fabricated and tested. A temperature compensation scheme has been developed that overcomes the large temperature dependence of GaAs acoustic wave devices. Packaging issues related to bonding miniature flow channels directly to the GaAs substrates have been resolved. Micromachining techniques for fabricating FPW and TSM microsensors on thin GaAs membranes are presented and GaAs FPW delay line performance is described. These devices have potentially higher sensitivity than existing GaAs and quartz SAW sensors.
Safety analysts frequently must provide results that are based on sparse (or even no) data. When data (or more data) become available, it is important to utilize the new information optimally in improving the analysis results. Two methods for accomplishing this purpose are Bayesian analysis, where "prior" probability distributions are modified to become "posterior" distributions based on the new data, and hybrid (possibilistic/probabilistic analysis) where possibilistic "membership" portrays the subjectivity involved and the probabilistic analysis is "frequentist." Each of these approaches has interesting features, and it is advantageous to compare and contrast the two. In addition to describing and contrasting these two approaches, we will discuss how features of each can be combined to give new advantages neither offers by itself.
Our investigations of both new and field-aged photovoltaic modules have indicated that, in general, today's commercially available modules area highly reliable product. However, by using new test procedures, subtle failure mechanisms have also been identified that must be addressed in order to achieve 30-year module lifetimes. This paper summarizes diagnostic test procedures, results, and implications of in-depth investigations of the performance and durability characteristics of commercial modules after long-term field exposure. A collaborative effort with U.S. module manufacturers aimed at achieving 30-year module lifetimes is also described.
The influence of changes in the contracted Gaussian basis set used for ab initio calculations of nuclear magnetic resonance (NMR) phosphorous chemical shift anisotropy (CSA) tensors was investigated. The isotropic chemical shitl and chemical shift anisotropy were found to converge with increasing complexity of the basis set at the Hartree-Fock @IF) level. The addition of d polarization function on the phosphorous nucIei was found to have a major impact of the calculated chemical shi~ but diminished with increasing number of polarization fimctions. At least 2 d polarization fimctions are required for accurate calculations of the isotropic phosphorous chemical shift. The introduction of density fictional theory (DFT) techniques through tie use of hybrid B3LYP methods for the calculation of the phosphorous chemical shift tensor resulted in a poorer estimation of the NMR values, even though DFT techniques result in improved energy and force constant calculations. The convergence of the W parametem with increasing basis set complexity was also observed for the DFT calculations, but produced results with consistent large deviations from experiment. The use of a HF 6-31 l++G(242p) basis set represents a good compromise between accuracy of the simulation and the complexity of the calculation for future ab initio calculations of 31P NMR parameters in larger complexes.
This report provides summary descriptions of Energetic Materials (EM) Research and Development activities performed at Sandia National Laboratories and funded through the Department of Energy DP-10 Program Office in FY97 and FY98. The work falls under three major focus areas: EM Chemistry, EM Characterization, and EM Phenomenological Model Development. The research supports the Sandia component mission and also Sandia's overall role as safety steward for the DOE Nuclear Weapons Complex.
The Technical Safety Requirements (TSR) document is prepared and issued in compliance with DOE Order 5480.22, Technical Safety Requirements. The bases for the TSR are established in the ACRRF Safety Analysis Report issued in compliance with DOE Order 5480.23, Nuclear Safety Analysis Reports. The TSR identifies the operational conditions, boundaries, and administrative controls for the safe operation of the facility.
Although valve-regulated lead-acid (VRLA) batteries have served in stationary applications for more than a decade, proprietary concerns of battery manufacturers and users and varying approaches to record-keeping have made the data available on performance and life relatively sparse and inconsistent. Such incomplete data are particularly detrimental to understanding the cause or causes of premature capacity loss (PCL) reported in VRLA batteries after as little as two years of service. The International Lead Zinc Research Organization (ILZRO), in cooperation with Sandia National Laboratories, has initiated a multi-phase project to characterize relationships between batteries, service conditions, and failure modes; establish the degree of correlation between specific operating procedures and PCL; identify operating procedures that mitigate PCL; identify best-fits between the operating requirements of specific applications and the capabilities of specific VRLA technologies; and recommend combinations of battery design, manufacturing processes, and operating conditions that enhance VRLA performance and reliability. This paper, prepared before preliminary conclusions were possible, presents the surveys distributed to manufacturers and end-users; discusses the analytic approach; presents an overview of the responses to the surveys and trends that emerge in the early analysis of the data; and previews the functionality of the database being constructed. The presentation of this paper will include preliminary results and information regarding the follow-on workshop for the study.
Tasks such as the localization of chemical sources, demining, perimeter control, surveillance and search and rescue missions are usually performed by teams of people. At least conceptually, large groups of relatively cheap mobile vehicles outfitted with sensors should be able to automatically accomplish some of these tasks. Sandia National Labs is currently developing a swarm of semi-autonomous all terrain vehicles for remote cooperative sensing applications. This paper will describe the capabilities of this system and outline some of its possible applications. Cooperative control and sensing strategies will also be described. Eight Roving All Terrain Lunar Explorer Rovers (RATLERs) have been built at Sandia as a test platform for cooperative control and sensing applications. This paper will first describe the hardware capabilities of the RATLER system. Then it will describe the basic control algorithm for GPS based navigation and obstacle avoidance. A higher level cooperative control task will then be described.
Ross (1990) developed an analytical relationship to calculate the diversion length of a tilted fine-over-coarse capillary barrier. Oldenburg and Pruess compared TOUGH2 simulation results to the diversion length predicted by Ross` formula using upstream and harmonic weighting. The results were mixed. The qualitative agreement is reasonable but the quantitative comparison is poor, especially for upstream weighting. The proximity of the water table to the fine-coarse interface at breakthrough has been proposed as a possible reason for the poor agreement. In the present study, the Oldenburg and Pruess problem is extended to address the water table issue. When the water table is sufficiently far away from the interface at breakthrough, good qualitative and quantitative agreement is obtained using upstream weighting.
In this work, the mesoscale processes of consolidation, deformation and reaction of shocked porous energetic materials are studied using shock physics analysis of impact on a collection of discrete HMX crystals. High resolution three-dimensional CTH simulations indicate that rapid deformation occurs at material contact points causing large amplitude fluctuations of stress states having wavelengths of the order of several particle diameters. Localization of energy produces hot-spots due to shock focusing and plastic work near grain boundaries as material flows to interstitial regions. These numerical experiments demonstrate that hot-spots are strongly influenced by multiple crystal interactions. Chemical reaction processes also produce multiple wave structures associated with particle distribution effects. This study provides new insights into the micromechanical behavior of heterogeneous energetic materials strongly suggesting that initiation and reaction of shocked heterogeneous materials involves states distinctly different than single jump state descriptions.
An overview of cookoff modeling at Sandia National Laboratories is presented aimed at assessing the violence of reaction following cookoff of confined energetic materials. During cookoff, the response of energetic materials is known to involve coupled thermal/chemical/mechanical processes which induce thermal damage to the energetic material prior to the onset of ignition. These damaged states enhance shock sensitivity and lead to conditions favoring self-supported accelerated combustion. Thus, the level of violence depends on the competition between pressure buildup and stress release due to the loss of confinement. To model these complex processes, finite element-based analysis capabilities are being developed which can resolve coupled heat transfer with chemistry, quasi-static structural mechanics and dynamic response. Numerical simulations that assess the level of violence demonstrate the importance of determining material damage in pre- and post-ignition cookoff events.