integration. Different parameters were analyzed to establish the sensitivity of a rotor containing a crack whereby whirl orbits, displacement, velocity, acceleration, and the unbalance phase angle were all studied in order to determine the best graphical output for detecting the early growth of a breathing crack. Dynamic simulations were undertaken with a finite element representation of a 500 MW generator at an operating speed of 3000 rpm with cracks of 25 and 10 percent of the diameter, located near a retaining ring. Comparisons of the crack model with another available from literature show good agreement in the vertical direction, which is a reassurance for both models, and indicate a better performance of the developed model in the horizontal direction. The simulations indicate that the harmonics of the unbalance phase angle can be a useful aid, since its higher components are more sensitive to the presence of a crack, than conventional displacement components. The technique is capable of sensing the early stages of a crack with the considerable advantage of not being affected by the presence of any other fault condition such as coupling misalignment, rubbing, etc. Dissert. Abstr. N91-12087 Florida Univ., Gainesville. EFFECT OF DESIGN PARAMETERS ON THE STABILITY OF MULTIMASS GEARED SYSTEMS SUPPORTED ON FLUID FILM BEARINGS AND SUBJECTED TO UNBALANCE AND TORSIONAL EXCITATION Ph.D. Thesis Seunghoon Ro 1989 194 P Avail: Univ. Microfilms Order No. DA9028564 Many high speed mechanical systems incorporate gearing for speed reduction. The stability of multi-rotor geared systems supported on oil film bearings, is investigated, taking into consideration the coupling between torsional and lateral dynamics. Emphasis is on the analysis of the interaction between the combined torsional and whirl instabilities. The result of simulation in this study checked the previously published empirical approach for predicting nonharmonic whirl of geared systems which results from the inertial coupling at the gear mesh. The feasibility of inducing a lateral and torsional instability to neutralize an anticipated unstable condition is investigated. The possibility of suppressing the instabilities by controlling the parameters of the oil film bearings is also considered. It is interesting to note that the coupling between the torsional and lateral oscillations can induce torsional instability due to unbalance excitation. Similarly, a torsional disturbance can induce whirl. A procedure is given to guide the designer in selecting the system parameters which insure its safe operation in critical situations where instabilities or resonances are expected. It is shown that torsional instability can be avoided by inducing a tuned whirl using an unbalance excitation in a system supported on either ball bearings or oil film bearings. Dissert. Abstr. N91-12088 Aston Univ., Birmingham (England). Kamaljeet Singh Sagoo 1988 335 p Avail: Univ. Microfilms Order No. BRDX89652 The feasibility of applying surface coating/treatments to seal materials, in order to provide improved seal performance is investigated. Various surface coating/treatment methods were considered; these included electroless nickel plating, ion plating, plasma nitriding, thermal spraying, and high temperature diffusion processes. The best wear resistance was conferred by the sprayed tungsten carbide/nickel/tungsten-chromium carbide deposit, produced by the high energy plasma spraying (Jet-Kote) process. In general, no correlation was found between hardness and wear resistance or surface finish and friction. Wear mechanisms observed in the investigation included adhesive transfer, oxidation, plastic deformation, ploughing, fracture, and delamination. Corrosion resistance was evaluated by Tafel extrapolation, linear polarization, and anodic potentiodynamic polarization techniques. The best corrosion performance was exhibited by an electroless nickel/titanium nitride duplex coating. The surface coating/treatments were ranked using a systematic method, which also considered other properties such as, adhesion, internal stress, and resistance to thermal cracking. The sealing behavior of surface coated/treated seals was investigated on an industrial seal testing rig. The best sealing performances were exhibited by the Jet-Kote and electroless nickel silicon carbide composite coated seals. The failure of the electroless nickel and electroless nickel/titanium nitride duplex coated seals was due to inadequate adhesion of the deposits to the substrate. Abrasion of the seal faces was the principal wear mechanism. For operation in an environment similar to the experimental system employed (acidified salt solution) the Jet-Kote deposit appears to be the best compromise. Dissert. Abstr. N91-12089 Council for National Academic Awards (England). TURBINE BLADE HEAT TRANSFER COEFFICIENT DETERMINATION USING OPTICAL PYROMETRY Ph.D. Thesis M. H. Tothill 1989 270 p Avail: Univ. Microfilms Order No. BRDX89761 A technique was developed for measuring local heat transfer rates from the surfaces of gas turbine blades. The technique is based on the optical measurement of transient temperatures and requires no mechanical contact. It was designed to take direct measurements from turbine blades while the engine is running and so be instrumental in turbine development. A high pressure, high temperature, turbine cascade rig was designed, constructed and developed to provide bench mark data, using well tried methods employing buried thermocouples, from which the optical technique was validated. A computational method for reducing the transient data into local heat transfer coefficients based on one-dimensional heat flux theory was devised. Dissert. Abstr. N91-12090# Argonne National Lab., IL. R. R. Sekar, W. W. Murr, T. J. Marcihiak, J. E. Schaus, J. N. and Power Plant Congress and Exposition, Milwaukee, WI, 10-13 Sep. 1990 (Contract W-31-109-eng-38) (DE90-017660; CONF-9009113-1) Avail: NTIS HC/MF A02 Analytical studies of oxygen-enriched diesel engine combustion have indicated the various benefits as well as the need for using cheaper fuels with water addition. To verify analytical results, a series of single cylinder diesel engine tests were conducted to investigate the concepts of oxygen enriched air (OEA) for combustion with water emulsified fuels. Cylinder pressure traces were obtained for inlet oxygen levels of 21 to 35 percent and fuel emulsions with water contents of 0 to 20 percent. Data for emulsified fuels included No. 2 and No. 4 diesel fuels. The excess oxygen for the tests was supplied from compressed bottled oxygen connected to the intake manifold. The cylinder pressure data was collected with an AVL pressure transducer and a personal computer based data logging system. The crank angle was measured with an optical encoder. In each data run, 30 consecutive cycles were recorded and later averaged for analysis. The data analysis was done with a heat release analysis code written for a personal computer. The results indicate that water emulsified fuels consistently gave a slight improvement in thermal efficiency and a greater portion of the fuel energy was released in the early part of the combustion process compared with base fuels. OEA reduced the ignition delay and measurably changed the shape of the calculated heat release rate and cumulative heat release diagrams. Comparative cylinder gas temperatures, computed from measured cylinder pressures, are presented in an effort to explain changes observed in the emissions of nitrogen oxides (NO(x)), smoke, and particulate matters. The data indicates that smoke and particulate emissions decrease and NO(x) increases when intake O2 level is increased. DOE N91-12091# Ohio State Univ., Columbus. Dept. of Electrical Engineering. AUTOMATED HANDLING AND ASSEMBLING OF NON-RIGID OBJECTS Semiannual Report No. 1, 15 Jan. - 15 Jul. 1990 Yuan F. Zheng 15 Jul. 1990 66 p (Contract N00014-90-J-1516) (AD-A224585) Avail: NTIS HC/MF A04 CSCL 20/11 Strategies for automatic assembly of deformable objects are presented. A flexible beam mating with a rigid hole is taken as the target of investigation. First, the deflection behavior of a deformable beam is studied. Then the geometric and force conditions for successfull assembly of rigid parts are reviewed. Based on these conditions of assembly, strategies for assembling deformable beams are proposed. The strategies are defined by the motion trajectory of the tool which handles the beam. It is found that when the clearance between the beam and the hole is loose, the desired trajectory should be the same as the curve formed by the deformed beam. When the clearance is tight, position and orientation of the tool should be adjusted, following the involute of the deformed beam curve. Experimental results are finally given to prove that the proposed strategies are effective. GRA N91-12092# Pennsylvania State Univ., University Park. Applied FUNDAMENTAL HYDRODYNAMICS RESEARCH Quarterly S. A. Abdallah, M. L. Billet, H. L. Petrie, P. J. Morris, and S. (AD-A225030) Avail: NTIS HC/MF A03 CSCL 20/4 The hydrodynamics research conducted under this program falls into two basic areas: (1) Turbomachinery: To develop an improved understanding of the fluid mechanics and acoustics associated with low-speed turbomachines and marine propulsors. To employ this knowledge to the development of improved propulsor and turbomachine design methods. (2) Drag reduction: To develop fundamental understanding of the mechanisms that cause drag on bodies and surfaces and to explore novel methods to reduce drag. Under each thrust area, one or more projects are conducted under the direction of the principal investigator who initiated the given task. All tasks are designed to provide results that will improve the scientific understanding of various hydrodynamic phenomena associated with the operation of submerged bodies and surfaces. This report documents the technical progress realized during the first quarter of FY-90 for the projects currently approved under this program. N91-12093# GRA Technische Univ., Berlin (Germany, F.R.). Inst. fuer Luft- und Raumfahrt. MOTION EQUATIONS AND STABILITY ANALYSIS OR SHAFT OF SINGLE-BLADED AND TWO-BLADED WIND TURBINES WITH FE-MODELED TOWER [BEWEGUNGSGLEICHUNGEN UND STABILITAETSANALYSE EIN- UND ZWEIFLUEGLIGER WINDTURBINEN MIT FE-MODELLIERTEM TURM ODER -MAST] Johannes Ackva Aug. 1989 82 p In GERMAN (ILR-Mitt-236(1990); ETN-90-97909) Avail: NTIS HC/MF A05 Two types of wind turbines are studied: one is the rigid two bladed, whose rotor is composed of two blades, rigidly fixed at the hub, the other is the pendulum bladed, whose rotor is composed of one rigid blade with counterweight or two rigid blades. For their mechanical models, the linear periodic time variant motion equations are installed and examined as far as stability is concerned, with the Floquet Method. The Finite Element Method is used in the modeling. The results show that the rigid two bladed is less susceptible to instability than the pendulum-bladed. ESA N91-12094# Naval Postgraduate School, Monterey, CA. STATISTICAL APPROACH TO FAULT DETECTION OF GEARS M.S. Thesis John D. Robinson Dec. 1989 204 P (AD-A225386) Avail: NTIS HC/MF A10 CSCL 13/9 The cost associated with machinery maintenance is a major portion of operating expenses. Vibration analysis, used as a method of monitoring the condition of machinery, provides a means to identify machinery faults before significant levels of damage occur. The results of research using statistical parameters of the vibration signal produced by machinery is presented. The statistical parameters investigated included the mean, mean square, variance, and coefficients of skewness and kurtosis. These values are compared with results of spectrum analysis techniques similar to methods used by the United States Navy in machinery programs. The investigation focuses on the use of band limited statistical parameters used as a fault detection technique that permits machinery operation to be categorized as satisfactory or unsatisfactory. GRA N91-12095# Massachusetts Inst. of Tech., Cambridge. Artificial Intelligence Lab. THE GRASPING PROBLEM: TOWARD TASK-LEVEL PROGRAMMING FOR AN ARTICULATED HAND M.S. Thesis Nancy S. Pollard May 1990 96 p (Contracts N00014-85-K-0124; N00014-86-K-0685) (AD-A225712; AI-TR-1214) Avail: NTIS HC/MF A05 CSCL 12/9 This report presents a system for generating a stable, feasible, and reachable grasp of a polyhedral object. A set of contact points on the object is found that can result in a stable grasp; a feasible grasp is found in which the robot contacts the object at those contact points; and a path is constructed from the initial configuration of the robot to the stable, feasible final grasp configuration. The algorithm described in the report is designed for the Salisbury hand mounted on a Puma 560 arm, but a similar approach could be used to develop grasping systems for other robots. Simulations show that the system can generate a wide range of grasps in difficult situations. GRA N91-11829# (England). British Aerospace Dynamics Group, Bristol SLIP COEFFICIENTS FOR SHEAR JOINTS: THE EFFECTS OF DYNAMIC LOADING AND SURFACE TREATMENT R. J. Baylis In ESA, Space Applications of Advanced Structural Materials Jun. 1990 p 117-122 (For primary document see N91-11812 03-24) Copyright Avail: NTIS HC/MF A19 38 QUALITY ASSURANCE AND RELIABILITY Includes product sampling procedures and techniques; and quality control. N91-12096# China Nuclear Information Centre, Beijing. (DE90-638372; CNIC-00292; SINRE-0020) Avail: NTIS (US By using the acoustic emission technique to monitor the fractures on analogous pressure pipes of the primary circuit which has had cracks and loading with pressure was investigated. The dynamic process, from cracking to fracturing, was recorded by the acoustic emission technique. Comparing with the conventional method, this method gives more informations, such as pre-cracking, cracking growing, fast fracturing, and the pressure values at different phases. During testing time a microcomputer was used for real-time data processing and locating the fracturing position. These data are useful for the mechanical analysis of the reactor components. DOE N91-12097# Polytechnic Inst. of New York, Brooklyn. The objective of this research effort has been to develop algorithms for in situ location and identification, by ultrasound, of flaws in plates or laminated layered elastic materials. Achieving this objective requires detailed knowledge of the excitation, propagation, scattering and detection of high frequency sound waves in the unflawed and flawed environments. Based on an understanding of these fundamental wave phenomena, one may then attempt to construct analytical models with accompanying algorithms, so as to parametrize the nondestructive evaluation (nde) problem in terms of good observables. During the contract period, carefully selected prototype problems have been investigated to determine good observables for particular flawed environments. Two major phases have received attention: (1) phenomena within a flat layered plate, especially beam-to-mode conversion, and the consequent interaction with a weak debonding flaw; (2) characterization of transducer outputs in terms of good wave objects that facilitate coupling into and out of the plate environment. The analytical tools rely on spatial and spectral domain formulations, and they comprise self-consistent hybrid beam-mode methods; complex source modeling of Gaussian beams, with complex ray tracing to track such beams; and decomposition of general wavefields into Gaussian beams. Thus, Gaussian beams, which are good observables are central to the problem strategy. N91-12098# Rolls-Royce Ltd., Derby (England). GRA I. J. Shellard, C. Baxter, and A. T. Jones 15 Jun. 1990 13 p (PNR-90746; ETN-90-97953) Copyright Avail: NTIS HC/MF A03 SCORPIO is an inspection system used in precision casting facilities to measure the orientation of the crystal axes with respect to the design axes within single crystal turbine components. The system is used in the production environment and as such operated either by production quality control personnel or totally robotically. The factors which created the need for SCORPIO and the history of its technical development, with focus on the effect that the decision to market the product commercially has had on the specification and the manner in which the project has progressed, are described. ESA N91-12099 Michigan Univ., Ann Arbor. PARAMETRIC RESONANCE OF ARBITRARILY LAYERED COMPOSITE CIRCULAR CYLINDRICAL SHELLS Ph.D. Thesis Alan Argento 1989 170 p Avail: Univ. Microfilms Order No. DA9013850 A composite circular cylindrical shell subjected to harmonic axial loading is studied. The shell is composed of arbitrarily layered fiber reinforced laminae. Lamination theory is used to develop the stiffness properties of the overall shell structure. Coupling between bending, stretching, and twisting in the shell's response can result because of the generality of the layering. The shell is modeled using thin shell theory. An element of the shell in a deformed configuration gives the equations of motion, one of which is nonlinear. Linearization is achieved via a perturbation technique which involves splitting the solution into a pre-instability response and a perturbed response. The equations of motion for the perturbed response of the shell are first partially uncoupled. Fourier series' are then used to suppress spatial dependence and generate a set of Mathieu-type differential equations. The static and dynamic stability, as well as the free vibration natural frequency of a response described by a set of Mathieu equations each involve the numerical calculation of a determinant. This is done by solving the associated eigenvalue problem. The unperturbed response is assumed to be axisymmetric and inertialness; the unperturbed response equations of motion are then solved. Spatial variation in these solutions is allowed, in contrast to the commonly used assumption of a constant membrane state. The effects on parametric resonance of including unperturbed response spatial dependence is studied. The variation of the parametric resonance frequency with lamina orientation is studied. Dissert. Abstr. N91-12100 Cornell Univ., Ithaca, NY. New analytical and experimental techniques from dynamical systems theory are combined with a geometrically exact rod theory to yield insight into the source of complex dynamical phenomena observed in a thin prismatic steel rod. Because of the geometry of the system, it is expected that the motions remain planar. However, experiments show that planar motions become unstable in wedged-shaped regions of the forcing frequency, forcing amplitude parameter plane, with each wedge centered on a resonant frequency of the system. The motions inside of these wedges are observed to be nonplanar and chaotic. A family of asymmetric bending-torsion nonlinear modes are found experimentally, and the frequency/amplitude characteristic of the family is obtained. Other phenomena discovered include dynamic two-well behavior and energy cascading from high to low frequencies. The fractal dimension of the attracting sets in different resonant wedges are estimated from experimental scalar time series using a numerical code. Starting with a geometrically exact, linearly elastic rod theory, a model system of two nonlinearly coupled partial differential equations is derived. The dynamics of a two-mode model incorporating a single bending and a single torsional mode are explored, and a simple mechanical analogue is described. Nonlinear modes analogous to those found experimentally in the rod are discovered numerically in the two-mode model. Numerically obtained Poincare sections for a related Hamiltonian system reveal that the nonlinear modes are born by a pitchfork bifurcation in the energy. Dissert. Abstr. N91-12101 Illinois Univ., Chicago. ON THE FOUNDATIONS OF BEAM THEORY Ph.D. Thesis Avail: Univ. Microfilms Order No. DA9015746 The asymptotic parameter expansion technique is employed to carry out a rigorous examination of existing formulations for the elastic response of beam type bodies. Advantages are that no prior assumptions regarding stress and displacement variations are needed and that the interior and boundary layer problems are defined in a consistent way. Both the outer expansion which is valid in the interior of the beam, and the inner expansion valid in the end zones, are developed. The boundary conditions for the outer expansion are constructed with the aid of auxiliary solutions which are canonical and without analytical form in general. Although the asymptotic approach has proved to be of great value in the theoretical analysis of elastic bodies and in developing some benchmark solutions, the resulting formulations have previously not been used as a basis for engineering beam theories. An engineering beam theory is derived based on the asymptotic outer expansion. Comparisons between the present theory and other engineering beam theories, such as the Timoshenko theory, are carried out through numerical analysis. Dissert. Abstr. N91-12102 Cornell Univ., Ithaca, NY. TRANSIENT STRESS WAVES IN TRUSSES AND FRAMES Samuel Moss Howard 1990 170 p The dynamics of lattice-type structures in the form of planar trusses and frames are investigated in terms of axial and flexural waves which propagate along members and scatter at structural joints. Theoretical and experimental results are presented for pin-jointed trusses and frames with rigid joints. The axial waves are considered along first, whose reflection and transmission at joints are related by analytical scattering coefficients. The complex reverberations of waves within a structure are calculated in the frequency domain with a technique called the reverberation method, which is implemented on a digital computer. Transient waves are then computed by Fourier synthesis. Results of extensive laboratory experiments are also presented. Techniques were developed for the generation and measurement of broadband stress waves in truss structures, using foil strain gages, and an elaborate digital data processing system. The results of these experiments are compared to theoretical simulations based on the axial wave theory. A general theory including both axial and flexural waves is developed, where the latter are predicted with either the Euler-Bernoulli or Timoshenko theories of bending, which introduce two modes of flexural waves in each member. First, general scattering coefficients for all three modes of waves at a joint are derived, and are shown to agree closely with experiments done on a single joint. The reverberation method is then generalized to include these general scattering coefficients, with the simplification that the mode conversion from flexural waves to axial waves at each joint is negligible. Simulated axial waves based on this assumption are found to be in much better agreement with the aforementioned experimental data over a longer duration of observation. Dissert. Abstr. N91-12103 Bristol Univ. (England). NONLINEAR ANALYSIS AND TESTING OF ADHESIVE JOINTS UNDER IMPACT AND QUASI-STATIC LOADING Ph.D. Thesis J. A. Harris 1983 276 p Avail: Univ. Microfilms Order No. BRDX88646 The current standard test method for determining the impact performance of adhesives was assessed, both theoretically and experimentally. The shortcomings of the test are shown to limit the usefulness of the results. An impact test was devised based on the single lap joint, to enable a clearer understanding of the impact performance of adhesive joints to be gained. Although joint strength was found to be insensitive to loading rate, the mechanical properties of the adhesives were shown to be very dependent on rate. Tests were carried out on bulk specimens over a range of quasi-static and impact rates for both pure shear and uniaxial tensile loading. An analysis of the single lap joint was carried out using a large displacement finite element method. By including the elasto-plastic properties of the adhesives and adherends into the finite element analysis, stress and strain distributions were determined in the various joints over a range of applied load. From these, together with appropriate failure criteria based on the uniaxial tensile properties of the adhesives, reasonable predictions of the locus of failure as well as joint strength were made. = A research program was initiated to study and isolate the factors responsible for scale effects in the tensile strength of graphite/epoxy composite laminates. Four layups were chosen with appropriate stacking sequences so as to highlight individual and interacting failure modes. Four scale sizes were selected for investigation including full scale size, 3/4, 2/4, and 1/4, with n to 4, 3, 2, and 1, respectively. The full scale specimen sizes was 32 piles thick as compared to 24, 16, and 8 piles for the 3/4, 2/4, and 1/4 specimen sizes respectively. Results were obtained in the form of tensile strength, stress-strain curves and damage development. Problems associated with strength degradation with increasing specimen sizes are isolated and discussed. Inconsistencies associated with strain measurements were also identified. Enhanced x ray radiography was employed for damage evaluation, following step loading. It was shown that fiber dominated layups were less sensitive to scaling effects compared to the matrix dominated layups. Author N91-12105 Kentucky Univ., Lexington. NONLINEAR THERMOELASTIC ANALYSIS OF LAMINATED Avail: Univ. Microfilms Order No. DA9023339 The use of anisotropic composites in applications involving severe thermal environments has increased dramatically in recent years. Geometrically nonlinear static and dynamic thermoelastic problems of laminated plates, cylindrical panels and shallow doubly-curved shells are the subject of the present investigation. The geometric nonlinearity considered here is that implied by the von Karman theory. Plate and shell behaviors are investigated using the first-order shear deformation theory (FSDT). The Love first geometric assumption is employed to simplify the shell formulations. First, the limitations of using the FSDT and Love's first approximation in predicting the responses of laminated shells to both mechanical and thermal loadings are examined. Results based on the shell theory are compared with the corresponding three-dimensional elasticity solutions. The quasi-static, large-deflection behaviors of laminated panels subject to a uniform temperature rise are analyzed using the finite element method. The structural behaviors of shallow panels are seen to be characterized by the existence of instabilities (bifurcation and snapping). Tracing the post-bifurcation and post-limit equilibrium paths represents an essential part of the nonlinear analysis. Some of the laminates considered undergo two instabilities prior to stress failure. Effects of the stacking sequence and shell geometries upon the critical and failure temperatures of the laminates are investigated. Nonlinear thermally-induced vibrations of laminated panels subject to rapid heating are also examined. The finite element technique is employed to discretize the space domain, and the resulting nonlinear differential equations are solved numerically by the Newmark method. The role of inertia is seen to be important in cases of shallow and thin laminates. N91-12106 Cornell Univ., Ithaca, NY. Dissert. Abstr. TIME DEPENDENCE OF THE DISPLACEMENT FIELDS Avail: Univ. Microfilms Order No. DA9027107 A shear-lag formulation is presented which investigates the character of the stress redistributions in the vicinity of single fiber failures in simple, tape-like composites. These composites are composed of small numbers of parallel, linearly elastic fibers embedded in a matrix lamina. The matrix material, in shear, creeps according to a nonlinear, power law constitutive model, which has exponents for both time and shear stress, and a simple memory of the shear stress history. Under remotely applied tensile loads, the center fiber in the symmetric model composite fails, and the load it carried is taken up in some manner by the remaining intact fibers, at least in the neighborhood of the break. The matrix transfers loads from one fiber to the next via shear tractions, so that at some distance from the fiber failure, the broken fiber is fully reloaded and the overloads in the intact fibers are eliminated. The character, physical length, and time dependence of this stress redistribution are governed by the constitutive behavior of the matrix material. An equivalent compression problem is devised, whose governing equations are transformed into self-similar form and solved. The tension problem solutions are recovered from the compression problem results. The self-similar character of these solutions is a consequence matrix constitutive behavior, however, the exponents for the self-similar growth of the deformation zone, in both time and composite stress, are different from those of the matrix constitutive law. The time dependent overload region lengths, matrix shear stresses, fiber displacements, fiber loads, and fiber strain distributions for several values of the matrix creep exponents are presented. Also included are approximations for these quantities under limiting cases of the creep exponents, which approach rigid-perfectly plastic matrix behavior. Dissert. Abstr. N91-12107 North Carolina State Univ., Raleigh. A BOUNDARY ELEMENT METHOD (BEM) SOLUTION FOR A Avail: Univ. Microfilms Order No. DA9023419 A numerical solution procedure is developed using the boundary element method (BEM) to predict the steady-state and the transient responses of a coupled arbitrarily shaped three-dimensional viscoelastic structure-acoustic cavity system. As a special application, the procedure is applied to investigate a plugged ear-canal system under external excitations. A newly explored method, the fractional operator constitutive equation, is chosen to model the mechanical behavior of the viscoelastic structure. The fractional operator model is shown to provide an accurate analytical representation of the actual viscoelastic material used in this research. A Laplace transformed viscoelastic BEM formulation, incorporating the fractional operator constitutive equation, is developed and coupled with a Laplace transformed BEM formulation for the acoustic cavity to formulate the complete numerical solution procedure. The use of the Laplace transformation facilitates the development by removing the time dependence of the system. A boundary discretization scheme using a constant triangular element is employed in the development. The transient response of the system is obtained by numerical inversion of the solution in the Laplace transformed domain. The steady-state response is obtained from the transformed solution by letting delta = ix omega in the formulation where delta is the transform parameter, i sq root of -1, and omega is a frequency of interest. The BEM numerical procedure is implemented into a computer program. The results from the use of this computer program on several case studies are presented. Dissert. Abstr. = N91-12108 Cornell Univ., Ithaca, NY. VIBRATION SUPPRESSION OF FLEXIBLE STRUCTURES USING COLOCATED VELOCITY FEEDBACK AND NONLOCAL ACTUATOR CONTROL Ph.D. Thesis Pei-Yen Chen 1990 222 p Avail: Univ. Microfilms Order No. DA9027039 A generalized colocated velocity feedback control system is proposed as an active damper which never pumps energy into the structure, and is applied to suppress the vibration in large flexible structures. Some fundamental characteristics of this system are exploited, as well as locations of poles and zeros for the open loop, and root loci for the closed loop. By minimizing a quadratic objective function defined by structural states and control forces, a suitable optimization procedure is presented to determine the optimal feedback gains in the form of general, symmetric, diagonal, and proportional matrices where the robustness property is preserved. A 6.5 m long experimental space truss with rigid joints was manufactured to implement the concept of generalized colocated velocity feedback control. The self-equilibrated internal control forces are generated through a new magnetic actuators with a high force-to-mass ratio. The corresponding velocity signals are picked up by magnetic sensors which were designed as an integral part of the actuators. In order to transmit these nonlocal torque-free control forces, an actuator mechanism was also invented. All of the results from theoretical analysis, numerical simulation, and experimental testing demonstrate that the transient vibration of the lowest modes in the experimental truss can be suppressed efficiently by using this control strategy with an optimal feedback gain. A comparison of the performance between colocated and noncolocated controls is examined. In addition, the effects on the robustness of this system due to non-ideal conditions are investigated, such as dynamic characteristics of control elements, unmodelled masses, nonlinear buckling behavior of structural members, and saturation limits on control forces. Dissert. Abstr. N91-12109 Vanderbilt Univ., Nashville, TN. Avail: Univ. Microfilms Order No. DA9026461 A versatile finite element model is introduced which makes it possible to study subsurface, frictional Mode II cracks subjected to rolling contact loading, under general conditions of plane strain. The model may address the linear-elastic or linear-elastickinematic-plastic material behaviors. The influence of system geometry and type of loading on the crack driving force is also evaluated. The results of the finite element model are compared with analytical solutions for the non-zero crack face friction coefficient, and for an elastic material confirm the soundness of the model. In agreement with experimental results, the model predicts a faster propagation rate from the trailing tip. The increase in crack tip driving force due to the onset of cyclic plasticity is found to be easily related to material constants. The crack tip driving force, delta K(sub II), obtained by using the finite element model, agrees with well known analytical solutions for the |