Browsing by Author "Fuh-Gwo Yuan, Committee Member"

Now showing 1 - 7 of 7

Bayesian Based Structural Health Management and An Uncertainty Analysis Technique Utilizing Support Vector Machine
(2007-09-13) Cao, Yingfang; Fuh-Gwo Yuan, Committee Member; Mohammad N. Noori, Committee Chair; Hamid Krim, Committee Member; Gregory D. Buckner, Committee Member
A Differential-Based Multiple Bit Rate PSK Receiver: Theory, Architecture, and SOI CMOS Implementation
(2004-10-07) Yuce, Mehmet Rasit; Fuh-Gwo Yuan, Committee Member; Ralph Cavin, Committee Member; Numan Dogan, Committee Member; Wentai Liu, Committee Chair; Gianluca Lazzi,, Committee Member
The development of telecommunications electronics with low-power and low-mass will be significant for future deep-space communications. The design of a receiver for deep-space communication requires the receiver to be robust against frequency variations due to Doppler effect in addition to radiation tolerance and low-power consumption. This dissertation reports a very low-power differential-based phase-shift keying (PSK) receiver that is targeted at deep-space and satellite communications, on both architectural and implementation levels. The power consumption of the PSK baseband circuit alone is less than 100 μW, which is significantly better than previously reported designs. Another major feature that has not been previously offered for PSK modulation is the use of 1-bit analog-to-digital converter (ADC) with sub-sampling front-end. The receiver uses double differential detection with traditional PSK modulation in the baseband to eliminate the impact of Doppler shift. Furthermore, the baseband can be employed in IF-sampling and direct sub-sampling front-end. Both front-ends offer minimal power consumption and differ from many traditional ones by eliminating some existing problems such as DC offset, dc voltage drifts and 1/f noise. The receiver also incorporates digital decimation stages to accommodate variable bit rates, and therefore it is highly programmable. The ability to support a wide range of data rates is an important feature of the receiver. This is achieved via digital channel selection by means of digital signal processing (DSP). The baseband and an analog part of the receiver are realized in 0.35 μm Silicon-on-Insulator (SOI) CMOS. SOI Complementary Metal Oxide Semiconductor (CMOS) technology is used mainly because it is a radiation hardness process. SOI technology is currently the most attractive choice in transceiver designs due to its advantages in both speed and power over standard CMOS because of lower parasitic capacitances. The designed baseband circuit consumes a power as low as 90.6 μW from a 1.1 V power supply. The analog part of the designed test chip consists of a two-stage differential IF amplifier that consumes a power of 0.8 mW from a 2.5 V supply. The primary goal of the proposed receiver is to achieve a higher integration at chip level, therefore resulting in significant size, power, and mass reductions for orbiter-lander communications while still meeting the system-level constraints.
Fault Detection/Isolation and Fault Tolerant Control for Hypersonic Vehicle.
(2010-05-14) Cai, Xuejing; Fen Wu, Committee Chair; Branda Nowell, Committee Member; Paul Ro, Committee Member; Gregory Buckner, Committee Member; Fuh-Gwo Yuan, Committee Member
An Inelastic Analysis Methodology for Bonded Joints with Shear Deformable, Anisotropic Adherends
(2004-02-19) Smeltzer, Stanley S. III; Eric C. Klang, Committee Chair; Timothy G. Clapp, Committee Member; Fuh-Gwo Yuan, Committee Member; Jeffrey W. Eischen, Committee Member
The development of a one-dimensional analysis method for evaluating adhesively bonded joints composed of anisotropic adherends and adhesives that exhibit nonlinear material behavior is presented. The strain and resulting stress fields in a general, bonded joint overlap are determined by using a variable-step, finite-difference solution algorithm to iteratively solve a system of first-order differential equations. Applied loading is given as a system of combined extensional, bending, and shear loads that are applied to the edge of the joint overlap. Adherends are assumed to behave as linear, cylindrically-bent plates using classical laminated plate theory that includes the effects of first-order transverse shear deformation. This provides a capability for modeling differences in the transverse shear modulus between each adherend. Using a total plasticity theory and a modified von-Mises yield criterion, inelastic material behavior is modeled in the adhesive layer. Results for the proposed method are verified using the single-lap joint geometry against previous results from the literature and shown to be in excellent agreement. Convergence of the strain and stress fields determined using the finite-difference solver are described as a function of the number of evaluation points along the length of the joint. Additionally, design studies using the single-lap joint are presented that investigate the effects of changes to the joint overlap, adherend thickness, laminate stacking sequence of the adherend, adherend material properties, and adhesive material properties. Results from the design studies established a nonlinear relationship between changes in the bending and axial stiffness of the adherends due to laminate ply manipulations and a reduction in the inelastic adhesive strain and shear stress responses. Additionally, analyses performed on the bonded joint models that had a difference in the transverse shear stiffness between the upper and lower adherends displayed a minimal effect on the adhesive strain and stress responses.
Modeling and Design of a Novel Cooling Device for Microelectronics using Piezoelectric Resonating Beams
(2003-12-29) Wu, Tao; Paul I. Ro, Committee Chair; Andrey V. Kuznetsov, Committee Member; Fuh-Gwo Yuan, Committee Member; Paul D. Franzon, Committee Member
As thermal management in microelectronics becomes more and more important in insuring the reliable operation, a novel and effective cooling device by smart materials such as piezoelectric bimorph needs to be developed. Investigation of modeling and design of piezoelectric resonating structures was conducted. A dynamic performance prediction method was proposed to calculate tip deflections at resonances and investigate the effect of finite stiffness bonding layer in piezoelectric bimorph. Considering the product of resonance frequency and dynamic tip deflection as a performance merit, the effects of length and location of the actuators on passive piezoelectric structures as well as the boundary conditions were analyzed for generating acoustic streaming which may be used for cooling microelectronic components. The cooling effects generated by vibrating non-slot and slotted piezoelectric bimorphs were experimentally investigated. A prototype, which is comprised of a piezoelectric bimorph actuator, an aluminum block with commercial cartridge heater served as heat source, four micrometer heads to adjust the gap size between bimorph and heat source, was constructed. Validated finite element analyses were employed to simulate the vibration characteristics including the natural frequencies and mode shapes of the proposed prototype. Setting the operation frequency at the fundamental resonance frequency, the cooling effects were measured by the temperature drops of the heat source above the vibrating bimorph. Electric field applied on the bimorph and the gap between heat source and actuator were adjusted to find out the best cooling result. Heat transfer coefficients between the heat source and vibrating bimorphs were calculated by ANSYS steady state thermal analysis and the lumped energy balance method. Air flow patterns around the bimorph actuator were visualized using particle tracking velocimetry (PTV) as well. The experiments showed that there exists an optimal gap between the heat source and the vibrating bimorph which brings the maximum temperature drop and the cooling effect increases with the electric field strength. The enhancement of heat transfer between the heat source and the non-slot bimorph can be up to 210% with the acoustic streaming generated by the bimorph vibration. The presence of slots in the bimorphs may enhance the mixing of streams outside and inside the channel resulting in an amplified heat transfer performance. However, the number, location and size of slots may influence the vibration characteristics and the formation of swirling streaming in the channel between the heat source and the bimorph. Finally, the heat transfer coefficient of the prototyped cooling device in terms of mean Nusselt number was correlated as a function of streaming Reynolds number. This study may provide useful information on modeling the vibration characteristics of piezoelectric actuators and designing the miniature cooling device utilizing bimorph vibrations.
Novel Finite Element Methods for Wave Propagation Modeling
(2004-10-12) Yue, Bin; Murthy N. Guddati, Committee Chair; Fuh-Gwo Yuan, Committee Member; G. Kumar Mahinthakumar, Committee Member; Vernon C. Matzen, Committee Member; M. Shamimur Rahman, Committee Member
The phenomenon of wave propagation is encountered in various engineering problems related to earthquake engineering, nondestructive evaluation and acoustics. Due to the complex material and geometrical features, many of these wave propagation problems are modeled using numerical methods such as the finite element method. Most numerical methods, due to their approximate nature, incur errors in the solution. In the context of wave propagation, these errors can be classified as amplitude and dispersion errors. Of these, dispersion error tends to have more severe effect on the accuracy due to its accumulative nature. Although it is possible to reduce the dispersion error by mesh refinement, such refinement imposes unrealistic computational cost even for medium-sized problems. In light of this, researchers have long sought efficient methods that reduce the dispersion error without any mesh refinement, but such efforts have only been partially successful. This dissertation develops efficient finite element methods for simulation of time-harmonic as well as transient wave propagation. For time harmonic waves, most existing dispersion reducing methods are limited to square meshes and homogeneous acoustic media. This dissertation develops two novel finite element methods that are applicable to unstructured meshes, as well as to heterogeneous media. They are the Local mesh-dependent augmented Galerkin finite element methods and the modified integration rules. Compared with existing methods, the proposed methods have higher convergence rate while maintaining low computational cost. When applied to elastic waves, the modified integration rules can reduce the dispersion error for either longitudinal or transverse wave, but not both. In the context of transient wave propagation, the spatial error of dispersion is coupled with temporal error resulting from time discretization. This dissertation focuses on reducing these errors by utilizing the modified integration rules and a modified time integration scheme. All the existing methods have second order convergence rates, while that of the proposed method has fourth order convergence. Numerical examples are utilized to illustrate the accuracy of the proposed method.
Synthesis and Characterization of Sol-Gel Nanocomposites Demonstrating Enhanced Mechancial Properties
(2007-10-13) Mosher, Brian Patrick; Fuh-Gwo Yuan, Committee Member; Angus Kingon, Committee Member; Taofang Zeng, Committee Chair; William Roberts, Committee Co-Chair
The mild reaction conditions of the sol-gel process allow incorporation of an inorganic component into organic materials, making it very favorable for the synthesis of organicinorganic nanocomposite materials. However, researchers still strive to produce materials with the unique properties of inorganic compounds which also possess the mechanical properties of organic polymers. We have developed several sol-gel derived nanocomposite silicate materials using both inorganic precursors and organic-inorganic precursors, which show superior mechanical properties due to nanoscale structural reinforcement. The materials reported were developed in three separate investigations, although the synthesis techniques are relatively similar. STRUCTURAL EVOLUTION OF PARTICLE REINFORCED ORGANIC-INORGANIC WATER-BASED NANOCOMPOSITE MATERIALS We present for the first time, the use of Al(ClO4)3, an effective catalyst for synthesis of 3-Glycidoxypropyltrimethoxysilane (GPTMS) - based nanocomposite materials via sol-gel chemistry. Aluminum perchlorate (Al(ClO4)3) simultaneously serves as catalyst for epoxy polymerization and methoxysilane hydrolysis at room temperature. Catalytic effects of Al(ClO4)3 were demonstrated through nuclear magnetic resonance (NMR) of precursor sols and structural investigation of resulting films by Fourier transform infrared spectroscopy (FTIR). Our method incorporates nanophase particulate reinforcement; Ludox® TMA colloidal silica, which possesses surface silanols (Si-OH), further activate the epoxy groups and expedite organic polymerization. The silica nanoparticles chemically link with the organic-inorganic network through Si-O-Si and Si-O-C bonds. Nanoparticles reinforcement serves to strengthen the network and enhance mechanical properties such as microhardness and abrasion resistance while maintaining the unique optical properties of these materials. Furthermore, the synthesis is water-based, providing for an environmentally friendly synthetic route to hybrid materials. We present a general strategy for synthesis of particle reinforced nanocomposite organic-inorganic sol-gels that can serve as a building block for synthesis of more advanced hybrid materials. PARTICLE-REINFORCED WATER-BASED ORGANIC-INORGANIC NANCOMPOSITE COATINGS FOR TAILORED APPLCATIONS Based on the concepts in the work described above, we further developed synthetic routes to three organic-inorganic coatings with nanoparticle reinforcement, which serves to enhance mechanical properties. The films are sol-gel derived using non-ionic surfactant, with aluminum perchlorate (Al(ClO4)3) as a catalyst and 3-Glycidoxypropyltrimethoxysilane (GPTMS) as precursor. Through the aid of nanoparticle colloids and a minute amount of catalyst, dense, hard and monolithic materials are obtained. Incorporating metal oxide nanoparticles brings forth unique properties, such as absorbing harmful UV radiation. Silica colloid composites provide greatly enhanced mechanical properties without modifying the unique optical properties of inorganic materials. Water-based synthesis of these coatings is straightforward and produces very few harmful byproducts, making them ideal materials in industry. The materials presented are relatively hard and abrasion resistant with very good adhesion; two of the coatings are UV absorbent. Various colloids can be employed in our methods to tailor properties and resulting materials may serve applications such as optical, protective, catalytic, guest-host, and multifunctional coatings. ENHANCING MECHANICAL PROPRETIES OF SILICA AEROGELS THROUGH NANOENGINEERING Furthermore, we have developed a novel method to prepare modified silica aerogels, in which a small amount of water-soluble inorganic synthetic nanocomposite is added (Laponite® RDS). The molecular-level synergism between silica nanoparticles and molecular cross-linkers inverts the relative host-guest roles in glass-polymer composites, leading to new stronger and more robust low-density materials. After drying with supercritical CO2, the materials were characterized by 3-point bending, transient hotwire techniques, bulk density measurements, transmission electron microscopy (TEM), and Brunauer, Emmett and Teller (BET) method. Transient hotwire methods confirm that the nanocomposite materials retain the ultra low thermal conductivity of pure silica aerogels.