Trident Scholar Abstracts 2000

An Earth impact with an asteroid has potentially devastating consequences.  In order to avoid global destruction, astronomers have been attempting to map out the orbits of all possibly hazardous Near-Earth Asteroids (NEAs) for the past two decades.  Although there are a number of astronomers devoted to discovering new NEAs, there are very few who are tracking these discoveries.  Therefore many recently identified NEAs are routinely lost.
This Trident Scholar project had two main objectives.  First, astronomical research at the U. S. Naval Academy hinges on the successful operation of a 20" Cassegrain telescope with a Photometrics 1024 x 1024 Charge Coupled Device (camera), as well as several computers to control these devices and process the data.  As a result of the work accomplished in this project, the U.S. Naval Academy now possesses a fully-functioning, state-of-the-art observatory, including newly acquired and developed software for tracking NEAs and analyzing astronomical observations in general.
The scientific goal of this project focused upon tracking several NEAs over a period of approximately two to three months in order to determine their orbits.  In addition, lightcurves were observed for a subset of these NEAs in order to find their rotational periods and their general shapes.  This was done by taking observations of each asteroid for at least one full night and then plotting intensity versus time.
The accomplishments realized in this Trident Scholar project will benefit midshipmen and Academy faculty who wish to track NEAs or begin a new field of astronomical research using the fully functional observatory.

In this project, identification and control of the Integrated Power System (IPS) for the next generation surface combatant ship (DD-21) was developed using a feed-forward, back-propagation neural network.  This neural network provided fault tolerant stabilization and control of an Integrated Power System (IPS).  Neural networks can be robust in the sense that they are not disabled by incomplete or inconsistent information.  The ability of a neural network to adapt to unforeseen eventualities such as flooding, fire, and combat casualties was investigated for a complex, interactive power system.  The IPS investigated in this project might provide integrated propulsion and ship service power generation and distribution for the next generation of the U.S. Navy surface combatant ships known as DD-21.  These solid state power systems involve nonlinear dynamics which can lead to negative impedance instability and voltage collapse.
This research represents an initial step toward unifying nonlinear, negative impedance stabilization with robust neural network fault detection and isolation.  The Naval Sea Systems Integrated Power System and the Office of Naval Research Electrically Reconfigurable Ship Programs motivated this research.

This project studied the flow field characteristics responsible for lateral instabilities observed on the F-18E/F aircraft in the Power Approach (PA) configuration.  Discovered during the aircraft's initial testing, these instabilities were observed when the aircraft exceeded twelve degrees of attack.  This problematic behavior came to be known as "PA wing drop", and it was corrected by the closure of a vent on the aircraft body.  However, it was not understood why the problem occurred, or the mechanics of how it was solved.
In response to a Naval Air Systems Command (NAVAIR) tasking to better understand the flow mechanics responsible for the wing drop problem and its solution, this project was initiated.  The methodology of the study used computational fluid dynamics to solve for the airflow properties over the aircraft with both the vent open and closed.  Elements of the study included:  1) the construction of a detailed computer model of the aircraft in both configurations; 2) the generation of a computational grid encompassing the aircraft model and its surroundings; and 3) the computation of the airflow properties over the aircraft at different angles of attack and at different stabilator positions using a full Navier-Stokes flow solver.  The computational results were then used to analyze the airflow over the aircraft in order to understand why the wing drop occurred and why the solution worked.  This project required the use of industry standard flow solvers run on local and remote computers, and was performed in collaboration with members of the Naval Air Systems Team at Patuxent River, Maryland, and the NASA Langley Research Center in Virginia.

The objective of this project was to characterize engine combustion-pressure-measurement errors caused by flame-induced transducer thermal stress.  Currently, engine-combustion-pressure data are obtained using piezoelectric pressure transducers.  Overall, these transducers perform well, however, the combustion flame causes a significant error.  The transducer's temperature rapidly increases when the flame arrives, causing its diaphragm to momentarily distort thereby changing its output.  The error associated with this phenomenon is called thermal shock.  To date, most of the efforts to reduce thermal shock have focused on designing and mounting the transducers to minimize its magnitude.  While these efforts have yielded improvements, the problem remains.
In this work, thermal shock was quantified in an engine using two pressure transducers.  One was exposed to the combustion gasses throughout the engine cycle and was therefore subjected to thermal shock.  The second was not exposed to the combustion gasses until combustion was complete; consequently it was protected from thermal shock.  The two measurements were compared over a variety of operating conditions in order to quantify in-cylinder thermal shock.  Because the reference transducer could not be exposed until combustion was complete, these measurements could not characterize thermal shock behavior at the instant of flame arrival.  To study this, a thermal shock simulator was built which exposed the transducer to an intermittent atmospheric flame.  Since the atmospheric pressure was constant, any change in the transducer's output during this exposure could be attributed to thermal shock.  The heat flux from the simulator flame was measured and adjusted to best re-create transient-flame-arrival conditions expected within an engine at the transducer location.

MicroElectroMechanical Systems (MEMS) are silicon microchips that have both electrical and mechanical components. MEMS process electrical information gathered from mechanical components which are about the diameter of a human hair (less than 100 microns). With the use of MEMS, machinery and computers can be linked, allowing for machine monitoring on a detailed level and providing for accurate and timely responses to changes in operating conditions.  The application of MEMS technology to scroll compressors is being explored by the National Institute of Standards and Technology for the Copeland Corporation, a manufacturer of scroll compressors for heating, ventilation, and air conditioning equipment.
This project investigated the use of MEMS technology to detect vibrations associated with abnormal scroll compressor operation.  The mechanical component on the MEMS chip was a vibration sensitive micro-cantilever beam.  The MEMS chip was tested for circuit operability and to determine the piezoresistive vibration sensitivity of various length cantilevers.  A model for the vibration sensitivity, which is a function of piezoresistive vibration sensitivity, was developed using the dimensions and properties of the piezoresistive cantilevers.  Through the use of the vibration sensitivity model and experimentation, the piezoresistive vibration sensitivity of the cantilevers was determined and used to identify the most sensitive cantilever design.  From the results of this project, design of the MEMS chip will be modified, allowing for optimum detection of vibrations associated with abnormal scroll compressor operation.

Studies have found that the passage of a charged particle through a dynamic random access memory (DRAM) can cause a bit flip (1 to 0 or 0 to1), also referred to as a single event upset (SEU). This is more noticeable in newer, denser DRAMs which are much more sensitive to radiation.  SEUs are also more common at higher altitudes, where the neutron and pion fluxes are found to be as much as several hundred times greater than at sea level.  For this reason, IBM, Boeing, the Department of Defense, and other government and commercial organizations have performed numerous studies on the phenomenon aimed at reducing the SEU effect in aircraft, missiles, and satellites which use DRAMs.
Many of the previous models developed to characterize the SEU are not applicable to modern high density chips.  This project has developed a new and improved model which applies to the higher density chips based on SEU cross-section, particle flux, and particle energy data taken from a wide range of experiments.
This study also identifies the nuclear reactions, chip characteristics, and particle environments which affect a DRAMs SEU rate.  From this model, the SEU rates of various commercial off-the-shelf (COTS) DRAMs were calculated at various altitudes, latitudes and longitudes.  These rates were used to identify which DRAMs are the most and least sensitive radiation.  Those DRAMs with the lowest expected SEU rates will be more reliable in aircraft systems while those with the highest expected SEU rates can potentially be used in the development of a smaller lightweight neutron detection system.
 
 

A multiple aperture camera system has been constructed that integrates six individual images from wide-angle video cameras into one unified, computer-generated image.  The form and function of this prototype system was inspired by research and development activities in the same area at the Naval Air Warfare Center (NAWC), Patuxent River, Maryland.  The NAWC effort is focused on improving air combat situational awareness in future aircraft through the use of improved infrared cameras in an aircraft-mounted multiple aperture system.
This research has investigated the complex process of transforming distorted wide-angle images taken from multiple standard video cameras into a cohesive non-distorted image that may be moved throughout a scene by computer program alone.  A prototype camera system has been constructed.  It consists of six wide-angle standard video cameras mounted in opposing directions along three orthogonal axes.  A camera-computer system combines a portion of the detected images to form a new "virtual image" that appears to have been taken from a single camera pointed in an arbitrary direction.  A fully field-operational system of this type could replace a camera on a moving mount such as those found in a submarine periscope or in the pan and tilt system of an attack helicopter's targeting camera.  Some of the many advantages of this system are:  a spherical field of view that may be viewed in an arbitrary direction without distortion, the potential for multiple operators to see images in different directions at the same time, and the ability to mount this system on many different platforms.